The Oxford Handbook of Evolutionary Psychology and Behavioral Endocrinology 0190649739, 9780190649739

Table of contents :
Short Contents
About the Editors
Contributors
Table of Contents
1. Integrating Mechanisms and Functions to Understand Behavior • Lisa L. M. Welling and Todd K. Shackelford
Part 1: Development and Survival
2. Hormones and Behavior: A Life History Perspective • Maren N. Vitousek and Laura A. Schoenle
3. Sex Differences in Primate Social Relationships During Development • Joyce F. Benenson
4. Sex Differences in Cognition: Evidence for the Organizational–Activational Hypothesis • Elizabeth Hampson
5. Involvement of the Sex Hormones in Learning and Memory • Kelsy S. J. Ervin and Elena Choleris
6. Endocrine Disruptors and Other Environmental Influences on Hormone Action • Laura N. Vandenberg
Part 2: Reproductive Behavior
7. Investigating the Ovulatory Cycle: An Overview of Research and Methods • Lisa L. M. Welling and Robert P. Burriss
8. Reproductive Behavior in the Human Male • Stefan M. M. Goetz, Glenn Weisfeld, and Samuele Zilioli
9. Mate Preferences Across the Lifespan • Lynda G. Boothroyd and Jovana Vukovic
10. The Influence of Maternal Stress and Child Maltreatment on Offspring • LillyBelle K. Deer, Kristin Bernard, and Camelia E. Hostinar
11. Evolution and Human Fatherhood • Adam H. Boyette and Lee T. Gettler
12. Hormones, Sexual Orientation, and Gender Identity • Nicholas C. Neibergall, Alex J. Swanson, and Francisco J. Sánchez
13. Intersexual and Intrasexual Competition and Their Relation to Jealousy • Abraham P. Buunk, Karlijn Massar, Pieternel Dijkstra, and Ana Maria Fernandez
14. Synthetic Hormones: The Influence of Hormonal Contraceptives and Hormone Replacement Therapy on Aspects of Women’s Mating Psychology • Amanda C. Hahn and Kelly D. Cobey
Part 3: Social and Affective Behavior
15. The Endocrinology of Social Relationships and Affiliation • Christine Anderl, Shimon Saphire-Bernstein, and Frances S. Chen
16. Hierarchy and Testosterone: How Can Testosterone Promote Upward Mobility in Status Hierarchies? • Shawn N. Geniole and Justin M. Carré
17. Competition, Dominance, and Social Hierarchy • Kathleen V. Casto and Pranjal H. Mehta
18. Oxytocin: An Evolutionary Framework • Nicholas M. Grebe and Steven W. Gangestad
19. Social Bond Paradoxes • Lynea R. Witczak, Trenton C. Simmons, and Karen L. Bales
20. Stress Hormones, Physiology, and Behavior • Justin K. Mogilski, Anna Wysocki, Simon D. Reeve, Virginia E. Mitchell, Jenna Lunge, and Lisa L. M. Welling
21. Hormones, Circadian Rhythms, and Mental Health • Yasmine-Marie Cissé, Jeremy C. Borniger, and Randy J. Nelson
22. Hormones and Major Depressive Disorder • Mark A. Ellenbogen, Virginia Tsekova, and Lisa Serravalle
23. Sex Differences in Anxiety Disorders • Teresa A. Piggott, Alexandra N. Duran, Isha Jalnapurkar, Tyler Kimm, Stephanie Linscheid, and Melissa K. Allen
24. Future Directions in Human Behavioral Endocrinology • Lisa L. M. Welling and Todd K. Shackelford
Name Index
Subject Index

Citation preview

The Oxford Handbook of Evolutionary Psychology and Behavioral Endocrinology

Oxford Library of Psychology Area Editors: Clinical Psychology David H. Barlow Cognitive Neuroscience Kevin N. Ochsner and Stephen M. Kosslyn Cognitive Psychology Daniel Reisberg Counseling Psychology Elizabeth M. Altmaier and Jo-Ida C. Hansen Developmental Psychology Philip David Zelazo Health Psychology Howard S. Friedman History of Psychology David B. Baker Methods and Measurement Todd D. Little Neuropsychology Kenneth M. Adams Organizational Psychology Steve W. J. Kozlowski Personality and Social Psychology Kay Deaux and Mark Snyder

OX F O R D L I B R A RY O F P S YC H O LO G Y

The Oxford Handbook of Evolutionary Psychology and Behavioral Endocrinology Edited by

Lisa L. M. Welling Todd K. Shackelford

1 2019

1 Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and certain other countries. Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America. © Oxford University Press 2019 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms agreed with the appropriate reproduction rights organization. Inquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above. You must not circulate this work in any other form and you must impose this same condition on any acquirer. Library of Congress Cataloging-in-Publication Data Names: Welling, Lisa L. M., editor. | Shackelford, Todd K. (Todd Kennedy), 1971–editor. Title: The Oxford handbook of evolutionary psychology and behavioral endocrinology/edited by Lisa L. M. Welling, Todd K. Shackelford. Other titles: Handbook of evolutionary psychology and behavioral endocrinology Description: New York NY: Oxford University Press, [2018] | Series: Oxford library of psychology | Includes bibliographical references and index. Identifiers: LCCN 2018015650 | ISBN 9780190649739 (hardcover: alk. paper) Subjects: LCSH: Evolutionary psychology. | Human behavior—Endocrine aspects. Classification: LCC BF698.95 .O946 2018 | DDC 155.7—dc23 LC record available at https://lccn.loc.gov/2018015650

135798642

Printed by Sheridan Books, Inc., United States of America

S H O RT C O N T E N T S

About the Editors  vii Contributors ix Table of Contents  xiii Chapters 1–442 Name Index  443 Subject Index  446

v

A B O U T T H E E D I TO R S

Lisa L. M. Welling is an Associate Professor of Psychology at Oakland University in Rochester, Michigan. Dr. Welling received her PhD in 2008 from the University of Aberdeen in the United Kingdom. She heads the Welling Research Laboratory, where her and her students’ research focuses on behaviors related to sexual selection, with a special focus on hormonal mediation. Todd K. Shackelford is Distinguished Professor and Chair of the Department of Psychology at Oakland University. He received his PhD in evolutionary psychology in 1997 from the University of Texas at Austin. Much of Dr. Shackelford’s research addresses sexual conflict between men and women, with a special focus on testing hypotheses derived from sperm competition theory.

vii

C O N T R I B U TO R S

Melissa K. Allen McGovern Medical School Houston, TX, USA Christine Anderl University of British Columbia Vancouver, British Columbia, Canada Karen L. Bales University of California—Davis Davis, CA, USA Joyce F. Benenson Emmanuel College Boston, MA, USA Kristin Bernard Stony Brook University Stony Brook, NY, USA Lynda G. Boothroyd Durham University Durham, England, UK Jeremy C. Borniger The Ohio State University Columbus, OH, USA Adam H. Boyette Duke University Durham, NC, USA Robert P. Burriss University of Basel Basel, Switzerland Abraham P. Buunk University of Groningen Groningen, the Netherlands; Netherlands Interdisciplinary Demographic Institute The Hague, The Netherlands Justin M. Carré Nipissing University North Bay, Ontario, Canada Kathleen V. Casto University of Oregon Eugene, OR, USA

Frances S. Chen University of British Columbia Vancouver, British Columbia, Canada Elena Choleris University of Guelph Guelph, Ontario, Canada Yasmine-Marie Cissé The Ohio State University Columbus, OH, USA Kelly D. Cobey Ottawa Hospital Research Institute Ottawa, Ontario, Canada; University of Ottawa Ottawa, Ontario, Canada; University of Stirling Stirling, Scotland LillyBelle K. Deer University of California—Davis Davis, CA, USA Pieternel Dijkstra The Netherlands Alexandra N. Duran McGovern Medical School Houston, TX, USA Mark A. Ellenbogen Concordia University Montreal, Quebec, Canada Kelsy S. J. Ervin University of Guelph Guelph, Ontario, Canada Ana Maria Fernandez University of Santiago, Chile Santiago, Chile Steven W. Gangestad University of New Mexico Albuquerque, NM, USA Shawn N. Geniole University of Vienna Vienna, Austria

ix

Lee T. Gettler University of Notre Dame Notre Dame, IN, USA Stefan M. M. Goetz Wayne State University Detroit, MI, USA Nicholas M. Grebe Duke University Durham, NC, USA Amanda C. Hahn Humboldt State University Arcata, CA, USA Elizabeth Hampson University of Western Ontario London, Ontario, Canada Camelia E. Hostinar University of California—Davis Davis, CA, USA Isha Jalnapurkar University of Massachusetts Medical School Boston, MA, USA Tyler Kimm McGovern Medical School Houston, TX, USA Stephanie Linscheid University of Michigan Medical School Ann Arbor, MI, USA Jenna Lunge Oakland University Rochester, MI, USA Karlijn Massar Maastricht University Maastricht, the Netherlands Pranjal H. Mehta University College London London, England, UK Virginia E. Mitchell Oakland University Rochester, MI, USA Justin K. Mogilski Oakland University Rochester, MI, USA Nicholas C. Neibergall University of Missouri—Columbia Columbia, MO, USA

x Contributors

Randy J. Nelson The Ohio State University Columbus, OH, USA Teresa A. Piggott McGovern Medical School Houston, TX, USA Simon D. Reeve Oakland University Rochester, MI, USA Francisco J. Sánchez University of Missouri—Columbia Columbia, MO, USA Shimon Saphire-Bernstein University of California—Los Angeles Los Angeles, CA, USA Laura A. Schoenle Virginia Tech Blacksburg, VA, USA Lisa Serravalle Concordia University Montreal, Quebec, Canada Todd K. Shackelford Oakland University Rochester, MI, USA Trenton C. Simmons University of California—Davis Davis, CA, USA Alex J. Swanson University of Missouri—Columbia Columbia, MO, USA Virginia Tsekova Concordia University Montreal, Quebec, Canada Laura N. Vandenberg University of Massachusetts—Amherst Amherst, MA, USA Maren N. Vitousek Cornell University Ithaca, NY, USA; Virginia Tech Blacksburg, VA, USA Jovana Vukovic Broward College Fort Lauderdale, FL, USA Glenn Weisfeld Wayne State University Detroit, MI, USA

Lisa L. M. Welling Oakland University Rochester, MI, USA Lynea R. Witczak University of California—Davis Davis, CA, USA

Anna Wysocki Oakland University Rochester, MI, USA Samuele Zilioli Wayne State University Detroit, MI, USA

Contributors

xi

TA B L E O F C O N T E N T S

1. Integrating Mechanisms and Functions to Understand Behavior  1 Lisa L. M. Welling and Todd K. Shackelford

Part 1 

• 

Development and Survival

2. Hormones and Behavior: A Life History Perspective  13 Maren N. Vitousek and Laura A. Schoenle 3. Sex Differences in Primate Social Relationships During Development  27 Joyce F. Benenson 4. Sex Differences in Cognition: Evidence for the Organizational–Activational  Hypothesis  43 Elizabeth Hampson 5. Involvement of the Sex Hormones in Learning and Memory  67 Kelsy S. J. Ervin and Elena Choleris 6. Endocrine Disruptors and Other Environmental Influences on   Hormone Action  87 Laura N. Vandenberg

Part 2 

• 

Reproductive Behavior

7. Investigating the Ovulatory Cycle: An Overview of Research   and Methods  109 Lisa L. M. Welling and Robert P. Burriss 8. Reproductive Behavior in the Human Male  125 Stefan M. M. Goetz, Glenn Weisfeld, and Samuele Zilioli 9. Mate Preferences Across the Lifespan  143 Lynda G. Boothroyd and Jovana Vukovic 10. The Influence of Maternal Stress and Child Maltreatment   on Offspring  161 LillyBelle K. Deer, Kristin Bernard, and Camelia E. Hostinar 11. Evolution and Human Fatherhood  179 Adam H. Boyette and Lee T. Gettler 12. Hormones, Sexual Orientation, and Gender Identity  201 Nicholas C. Neibergall, Alex J. Swanson, and Francisco J. Sánchez 13. Intersexual and Intrasexual Competition and Their Relation   to Jealousy  215 Abraham P. Buunk, Karlijn Massar, Pieternel Dijkstra, and Ana Maria Fernandez xiii

14. Synthetic Hormones: The Influence of Hormonal Contraceptives   and Hormone Replacement Therapy on Aspects of Women’s   Mating Psychology  237 Amanda C. Hahn and Kelly D. Cobey

Part 3 

• 

Social and Affective Behavior

15. The Endocrinology of Social Relationships and Affiliation  259 Christine Anderl, Shimon Saphire-Bernstein, and Frances S. Chen 16. Hierarchy and Testosterone: How Can Testosterone Promote Upward   Mobility in Status Hierarchies?  281 Shawn N. Geniole and Justin M. Carré 17. Competition, Dominance, and Social Hierarchy  295 Kathleen V. Casto and Pranjal H. Mehta 18. Oxytocin: An Evolutionary Framework  317 Nicholas M. Grebe and Steven W. Gangestad 19. Social Bond Paradoxes  335 Lynea R. Witczak, Trenton C. Simmons, and Karen L. Bales 20. Stress Hormones, Physiology, and Behavior  351 Justin K. Mogilski, Anna Wysocki, Simon D. Reeve, Virginia E. Mitchell, Jenna Lunge, and Lisa L. M. Welling 21. Hormones, Circadian Rhythms, and Mental Health  367 Yasmine-Marie Cissé, Jeremy C. Borniger, and Randy J. Nelson 22. Hormones and Major Depressive Disorder  381 Mark A. Ellenbogen, Virginia Tsekova, and Lisa Serravalle 23. Sex Differences in Anxiety Disorders  405 Teresa A. Piggott, Alexandra N. Duran, Isha Jalnapurkar, Tyler Kimm, Stephanie Linscheid, and Melissa K. Allen 24. Future Directions in Human Behavioral Endocrinology  433 Lisa L. M. Welling and Todd K. Shackelford Name Index  443 Subject Index  446

xiv

Table of Contents

CH A PT E R

1

Integrating Mechanisms and Functions to Understand Behavior

Lisa L. M. Welling and Todd K. Shackelford

Abstract Evolutionary psychology and behavioral endocrinology provide complementary perspectives on interpreting human behavior and psychology. Hormones can function as underlying mechanisms that influence behavior in functional ways. Understanding these proximate mechanisms can inform ultimate explanations of human psychology. This chapter introduces this edited volume by first discussing evolutionary perspectives in behavioral endocrinology. It then briefly addresses three broad topic areas of behavioral endocrinology: (1) development and survival, (2) reproductive behavior, and (3) social and affective behavior. It provides examples of research within each of these areas and describes potential adaptations. The chapter concludes with a discussion on the importance of integrating mechanisms with function when investigating human behavior and psychology. Key words:  evolutionary psychology, behavioral endocrinology, mechanisms, function, behavior

Hormones are chemical messengers secreted into the bloodstream or tissue fluid system by specialized cells. They travel from their source until they reach target cells with specialized receptors, to which they bind to evoke a response. The three main classes of hormones are steroids, peptide hormones, and monoamines. Steroid hormones in humans include estrogens (e.g., estradiol), progestins (e.g., progesterone), androgens (e.g., testosterone), and corticosteroids (e.g., cortisol). They are lipophilic and must bind to carrier proteins (e.g., sex hormone–binding globulin) to circulate in the bloodstream. Peptides (e.g., oxytocin) and monoamines (e.g., melatonin) are made of one (monoamines) or more (peptides) amino acids. They are hydrophilic and travel freely in the bloodstream. Hormones can change gene expression (e.g., Béchet, 1986), influence signaling pathways (e.g., Goglia, Moreno, & Lanni,  1999), and/or change cell sensitivity by increasing (i.e., upregulating) or decreasing (i.e., down-regulating) the number of receptors for that or another hormone. For example, pregnancy hormones cause an up-regulation

of oxytocin receptors in the uterus during the third trimester (Gimpl & Fahrenholz, 2001), whereas elevated levels of insulin cause a down-regulation of insulin receptors (Fröjdö, Vidal, & Pirola, 2009). In addition to these biological functions, hormones may serve other evolved purposes under certain circumstances by influencing human behavior. Evolutionary psychology and behavioral endocrinology have progressed as relatively i­ndependent fields over the last few decades. Evolutionary psychologists seek to explain human behavior and psychology in terms of functional outcomes of natural and sexual selection across our evolutionary past. Specifically, evolutionary psychology stands on the supposition that selection pressures designed the adaptations of the mind (as well as the body) and  these adaptations produced behavior through psychological, neurological, and physiological mechanisms (Tooby & Cosmides, 1992, 2005). Similar to physical adaptations, such as how the human foot is designed to afford long-distance running (e.g., Bramble & Lieberman, 2004; Raichlen, Armstrong, 1

& Lieberman,  2011; Rolian, Lieberman, Hamill, Scott, & Werbel, 2009), evolutionary psychologists maintain that behavior and psychological traits (e.g., memory, learning, perception, personality, preferences) are the result of psychological mechanisms that evolved because they served an adaptive function during our ancestral past (e.g., Buss,  2005; Pinker, 2002). Hypotheses are generated accordingly and tested against alternative explanations using both Western and cross-cultural samples (discussed in Tooby & Cosmides, 2005). Behavioral endocrinologists investigate the underlying hormonal and neuroendocrine ­mechanisms that influence or regulate behavior in humans and other animals. Behavioral endocrinology is the scientific study of the bidirectional relationship between hormones and behavior, and of the interactions among the brain, the endocrine system, hormones, and ­behavior (Nelson, 2010). Although the field arguably dates back to Berthold’s (1849) landmark experiment on castrated chicks, which provided the first formal evidence that the testes produce substances that influence both physiology and behavior, Frank Beach, who wrote the book Hormones and Behavior (1948) and cofounded the journal of the same name, first branded and defined the discipline more than a century later (Beach, 1975). Behavioral endocrinologists examine relationships with behaviors of interest using measured (e.g., Denson, Mehta, & Ho Tan, 2013; Hahn, DeBruine, Fisher, & Jones, 2015; Welling et al., 2007, 2008), estimated (e.g., Garver-Apgar, Gangestad, & Thornhill,  2008; Jones et al., 2005; Lukaszewski & Roney, 2009), or manipulated (e.g., Carré et al.,  2015; Donaldson, Welling, & Reeve,  2017; Welling, Moreau, Bird, Hansen, & Carré,  2016; Welling, Puts, Roberts, Little, & Burriss, 2012) circulating hormone levels, or through investigating the associated biological circuitry (e.g., receptor-binding sites, gene expression) of these interactions (e.g., Arango et al., 1990; Farfel, Kleven, Woolverton, Seiden, & Perry, 1992; Israel, 2016). Hormones influence gene expression or cellular function, and increase the likelihood of specific behaviors occurring under certain circumstances by affecting sensory (i.e., input) systems, central nervous system processing systems, and/or peripheral effectors (i.e., output systems such as striated muscles; Nelson, 2010). Scholars of both evolutionary psychology and behavioral endocrinology often study interactions using interdisciplinary methods. Both investigate questions and/or borrow techniques from neuro­ science, anthropology, social psychology, genetics, 2

biology, zoology, and other disciplines. The complementary nature of these areas is increasingly recognized; annual conferences hosted by the Human Behavior and Evolution Society and the Society for Behavioral Neuroendocrinology include a variety of speakers who are investigating the hormonal basis of behavior from an evolutionary perspective. In other words, researchers are acknowledging that integrating mechanisms in our understanding of function provides additional insight that produces more complete theories and hypotheses. Focusing on underlying mechanisms moves evolution-minded researchers from asking ultimate (“Why?”) to proximate (“How?”) questions. Ultimate explanations outline the function of a trait or ­behavior, whereas proximate explanations outline how underlying mechanisms interact with the environment to produce that trait or behavior (Tinbergen, 1963). For example, an ultimate explanation for birds singing is that birds that sang over that species’ ancestral past were more likely to attract a mate and pass on their genes (i.e., singing conferred a reproductive advantage), whereas a proximate explanation for birds singing is that seasonal fluctuations in testosterone level cause singing behavior during the breeding season. Proximate explanations deal with mechanisms, whereas ultimate explanations deal with evolved functions. Certainly, understanding how a m ­ echanism works informs the understanding of why a mechanism came to exist in the first place, and hypotheses about why mechanisms exist inform research on how they function. Therefore, both proximate and ultimate explanations should be considered when investigating behavior.

Broad Areas of Behavioral Endocrinology

The current chapter discusses evolutionary perspectives in behavioral endocrinology by introducing and briefly reviewing the three broad topic areas that are the focus of this volume: (1) development and survival, (2) reproductive behavior, and (3) social and affective behavior. Within each of these topic areas, chapter authors discuss research and theory in important subareas. We conclude this chapter with a note on the importance of integrating mechanisms (proximate questions) with function (ultimate questions) to disentangle the how and why of behavior.

Development and Survival

An organism’s investment in growth, reproduction, and survival defines its life history strategy (e.g., Roff, 2002; reviewed in Vitousek & Schoenle, this volume), and hormones govern multiple biological processes

Integrating Mechanisms and Functions to Understand Behavior

related to development from conception onward. Differential hormone exposure in utero affects sexual behavior later in life in multiple species (reviewed in Clark & Galef, 1995). Behavioral sex differences are noticeable early on (reviewed in Benenson, this volume) and likely stem, at least in part, from this differential hormone exposure (see Collaer & Hines, 1995). Across primate species, young males are more likely than females to engage in rough-and-tumble play and confrontations over status (e.g., Meaney, Stewart, & Beatty, 1985). Work with rhesus and vervet monkeys documents sex-differentiated toy preferences similar to those documented in human children, with females preferring dolls and cooking pots more than males and males preferring cars and balls more than females (Alexander & Hines, 2002; Hassett, Siebert, & Wallen,  2008). These preferences are unlikely to be about the objects themselves, given that monkeys have no use for cooking pots or cars, but rather are likely about movement, with males preferring toys that move actively through space more than females do. Correspondingly, human girls are more likely than boys to help their parents with child-rearing tasks for their siblings and take more pleasure in caring for infants (e.g., Edwards, 2002). These sorts of sex-conforming behaviors are weaker among girls exposed to testosterone prenatally, such as those girls born with congenital adrenal hyperplasia (reviewed in Bailey & Zucker, 1995). The Organizational-Activational Hypothesis posits that sex steroid hormones permanently organize neural circuitry during early development (i.e., via exposure to specific steroids at critical ­developmental periods) to give rise to sexually ­ differentiated behaviors that are activated by steroid hormone ­ ­exposure in adulthood (reviewed in Hampson, this volume). Most evidence for this hypothesis comes from animal work (e.g., Phoenix et al., 1959), but evidence in humans is accumulating (e.g., Heil, Kavšek, Rolke, Beste, & Jansen, 2011; Vuoksimaa et al., 2010). Indeed, several sex differences in cognition have been identified; for example, women tend to excel at object location memory tasks (Voyer, Postma, Brake, & Imperato-McGinley, 2007), whereas men tend to excel at mental rotation tasks (Voyer, Voyer, & Bryden, 1995), among other sex differences (see, e.g., Kimura, 2004). These sex differences often appear or increase around puberty when sex hormone levels increase (e.g., Hyde, Mezulis, & Abramson, 2008), lending support to the proposed organizing (and later activating) role of sex steroids in behavior. Similarly, the influence of estrogens in activating aspects of learning and memory has r­eceived a great

deal of attention in recent years (reviewed in Ervin & Choleris, this volume; see also Duarte-Guterman, Yagi, Chow, & Galea, 2015), although most work has been done with rodents (e.g., Kim, Szinte, Boulware, & Frick, 2016; Luine, 2015). Together, this work highlights how underlying hormonal mechanisms can evolve to influence multiple domains. Sex differences in play behavior and adult cognition appear to develop under the influence of sex steroids (e.g., Hampson & Rovet, 2015; Hines, 2003; Puts et al., 2008; Williams et al., 1990) and likely reflect different selection pressures on males and females (see Cashdan & Gaulin, 2016). For instance, travel over large ranges may have been more important for males due to different reproductive interests (Miner, Gurven, Kaplan, & Gaulin,  2014), which would lead to greater selection for visuospatial abilities in males compared to females (e.g., Halpern, 2013). Others have hypothesized that the female advantage in object location memory reflects a sex-linked division of labor in our ancestral past, with females engaging in comparatively more gathering than males (e.g., Neave, Hamilton, Hutton, Tildesley, & Pickering,  2005; Silverman, Choi, & Peters, 2007). Thus, differences in hormonal profile are related to evolved differences in behavior, although it is worth noting that evolved hormonal mechanisms can be affected by modern pollutants called endocrine-disrupting chemicals (reviewed in Vandenberg, this volume), which may alter important hormonal functions, including those associated with reproductive behavior (e.g., Colborn, vom Saal, & Soto, 1993). These mechanisms work between individuals (e.g., sex differences) but also within individuals as levels of endogenous hormones fluctuate. Next, we review how changes in these hormones within an individual functionally influence changes in behavior that promote reproduction and parenting.

Reproductive Behavior

Sexual selection involves members of one sex choosing members of the other sex to mate with and competing with members of the same sex for access to potential mates. Over the last few decades, increasing evidence suggests that women’s fertile status is not entirely concealed, as had been  previously supposed (see, e.g., Gangestad & Thornhill, 2008; Gilldersleeve, Haselton, & Fales, 2014a,  2014b). Rather, hormonal fluctuations ­associated with ovulation are related to shifts in ­attractiveness (e.g., Miller & Maner, 2011; Pipitone & Gallup, 2008; Puts et al., 2013; Roberts et al., Welling and Shackelford

3

2004), sexual motivation (e.g., Gangestad, Thornhill, & Garver,  2002; Haselton & Gangestad, 2006; Roney & Simmons, 2013), and preferences for male traits putatively associated with underlying genetic quality (e.g., Gangestad, Simpson, Cousins, Garver-Apgar, & Christensen, 2004; Havlíček, Roberts, & Flegr, 2005; Little & Jones, 2012; Welling et al., 2007). Within-subject changes in measured or estimated sex hormones associated with the fertile phase of the menstrual cycle may underlie physical, psychological, and behavioral changes that increase conception probability and the likelihood of reproducing with high-quality mates (reviewed in Welling & Burriss, this volume). Relatedly, women report greater jealousy (Buunk & Van Brummen-Girigori,  2016; Cobey et al., 2012) and engage in more intrasexual competition (e.g., Fisher, 2004; Lucas, Koff, & Skeath, 2007; Piccoli, Foroni, & Carnaghi, 2013) when maximally fertile (for an in-depth review of the hormonal correlates of jealousy and intrasexual competition, see Buunk, Massar, Dijkstra, & Fernandez, this volume). Because jealousy and i­ntrasexual competition facilitate securing and ­retaining a mate (e.g., Buss,  2005), these findings lend support to the premise that within-woman variation in sex steroid hormones influences mating psychology and behavior. Moreover, synthetic hormones such as those found in the hormonal contraceptive pill may alter aspects of mate choice (e.g., Roberts et al., 2014) and behavior (e.g., Welling et al., 2012), indicating that exogenous hormones also impact behavior and may disrupt evolved functions of ­endogenous hormones (reviewed in Hahn & Cobey, this volume). Sex steroids can serve similar functions in men (reviewed in Goetz, Weisfeld, & Zilioli, this volume); testosterone in men amplifies risk-taking behaviors in the service of mating effort (Ronay & von Hippel, 2010), increases mate-seeking behavior (van der Meij, Almela, Buunk, Fawcett, & Salvador, 2012), and decreases interest in infants (Zilioli et al., 2016). Likewise, men’s testosterone is negatively correlated with relationship length (Gray et al., 2004), relationship commitment (Edelstein, van Anders, Chopik, Goldey, & Wardecker, 2014), marriage and fatherhood (Gettler, McDade, Agustin, Feranil, & Kuzawa, 2013; Gettler, McDade, Feranil, & Kuzawa, 2011; Gray, Ellison, & Campbell, 2007; Gray, Kahlenberg, Barrett, Lipson, & Ellison, 2002; see also Boyette & Gettler, this volume, for a review on fatherhood), and involvement in parenting (e.g., Gettler, McDade, Agustin, Feranil, & Kuzawa, 2015; Gettler, McKenna, 4

Agustin, McDade, & Kuzawa, 2012). Given the association between testosterone and aggression ­ (e.g., Archer, 2006), this research suggests that testosterone functionally decreases across a long-term relationship and in response to fatherhood in order to decrease aggressive and mate-seeking behaviors and, in turn, to increase resource allocation toward parenting effort (i.e., investment in one’s mate and children; see also, e.g., Bribiescas, 2001). Mate preferences in both sexes likely evolved to facilitate reproduction with individuals possessing “good” genes and other traits (e.g., kindness) indicative of fecundity and good parenting potential (reviewed in, e.g., Little, 2015). To be sure, stress during gestation and poor parenting can have a ­severely negative impact on offspring well-being and survival (e.g., Gilbert et al., 2015; Pratchett & Yehuda, 2011; Taylor, Guterman, Lee, & Rathouz, 2009; for a review, see Deer, Bernard, & Hostinar, this volume), making mate choice an important component of reproductive outcomes. Preferences for these traits emerge very early; newborn infants spend longer looking at faces judged as attractive by adults compared to faces judged as relatively unattractive (Slater et al., 1998; Slater, Quinn, Hayes, & Brown, 2000), although childhood preferences do not become fully mature until around age 14 years (e.g., Boothroyd, Meins, Vukovic, & Burt, 2014; for a review of mate preferences across the lifespan, see Boothroyd & Vukovic, this volume). The onset of adult mate preferences peripubertally implicates a hor­ monal component, such as dehydroepiandrosterone (e.g., Boothroyd et  al.,  2014; Saxton, DeBruine, Jones, Little, & Roberts, 2009), and further evinces the connection between hormones and reproductive behavior (e.g., Jones et al., 2005; Welling et al., 2008). Yet, sex steroids and other hormones serve a multitude of other social and affective functions. Next, we briefly review some findings on the bidirectional relationships between hormones, affect, and social behavior.

Social and Affective Behavior

The relationships between hormones and affect are among the most well established of the many research areas within behavioral endocrinology. For example, when someone colloquially refers to a person as “hormonal,” they are often referencing sudden changes in mood or general negative affect. This pejorative use aside, hormones do influence multiple aspects of affect and social behavior, both positive (e.g., affiliation; Brown et al.,  2009) and negative (e.g., aggression; Carré, Putnam, &

Integrating Mechanisms and Functions to Understand Behavior

McCormick, 2009). Affiliation, which consists of close interpersonal relationships (e.g., parent–child, romantic partners) and the behaviors necessary to establish and maintain those relationships (e.g., caregiving, trustworthiness; Feldman,  2012), is a commonly studied element of social behavior (­reviewed by Anderl, Saphire-Bernstein, & Chen, this volume). Social/affiliative behaviors are related to steroid ­concentrations, such as cortisol level (e.g., higher salivary cortisol levels are associated with less social networking; Koernienko, Clemans, Out, & Granger, 2014), but the majority of research has ­focused on the mediating influence of the peptide hormone oxytocin. Oxytocin increases in response to affectionate contact with infants (e.g., Apter-Levi, Zagoory-Sharon, & Feldman, 2014) and romantic partner empathy (Schneiderman, Kanat-Maymon, Ebstein, & Feldman, 2014), and is positively correlated with reports of relationship quality (e.g., Holt-Lunstad, Birmingham, & Light, 2015). Some findings are somewhat contradictory, however. For example, oxytocin has been found to both increase (Kosfeld, Heinrichs, Zak, Fischbacher, & Fehr, 2005) and decrease (Bartz et al.,  2011) trust during an economic game. This contradictory evidence suggests that hormones can be co-opted across social contexts to accomplish specific behavioral goals under specific conditions (e.g., nurturing behavior characterized by high oxytocin and low testosterone vs. sexual behavior characterized by high oxytocin and high testosterone; van Anders, Goldey, & Kuo, 2011; see also Witczak, Simmons, & Bales, this volume, for a discussion of social bond paradoxes). Oxytocin may function more broadly as  a component of a threat management system, whereby it enhances the salience of social cues in response to perceived threats to important social relationships and motivates the individual to repair possible damage (see Grebe & Gangestad, this volume). In this way, oxytocin would serve a prosocial function, but may be expressed differently depending on the circumstances. Other hormones, particularly testosterone and cortisol, influence antisocial behaviors. Circulating testosterone influences status-seeking (e.g., Eisenegger, Haushofer, & Fehr, 2011), competitive (e.g., Carré et al.,  2009), and aggressive (e.g., Klinesmith, Kasser, & McAndrew, 2006) behaviors (reviewed in Geniole & Carré, this volume). These behaviors help define social hierarchies within ­primates, and placement within these hierarchies is  reflected in an individual member’s circulating testosterone and cortisol levels (e.g., Kornienko,

Schaefer, Weren, Hill, & Granger, 2016; reviewed in Casto & Mehta, this volume). Specifically, in primate species in which hierarchical rank is maintained with aggression, dominant individuals are under more stress and have higher cortisol levels (e.g., Muehlenbein & Watts, 2010), whereas in primate species in which the subordinates are suppressed with intimidation, the subordinates are under more stress and have higher cortisol levels (e.g., Sapolsky, 2005). Although acute stress can be functional in that it fuels survival-related behaviors and motivates relationship reparation, chronic stress and the associated hormones can have detrimental long-term impacts on health and well-being, including decreased physical and mental health ­(reviewed in Mogilski et al., this volume). Indeed, hormonal changes may precede or accompany ­several mental health disorders, such as depressive disorders (reviewed in Ellenbogen, Tsekova, & ­ Serravalle, this volume) and anxiety disorders (­reviewed in Pigott et al., this volume). For example, chronic dysregulation of the hypothalamic-pituitaryadrenal (HPA) axis resulting in abnormally high cortisol concentrations is commonly found in ­individuals with depressive disorders (e.g., Carroll et  al.,  2007). Dysregulation of diurnal hormone levels can have a similar impact on mental health (reviewed in Cissé, Borniger, & Nelson, this volume). This research demonstrates that mental health issues can manifest physiologically as a downstream consequence of hormonal ­mechanisms being out of balance.

Conclusion

Behavioral endocrinology and evolutionary psychology encompass the topics covered in this chapter and many more. Evolutionary psychology, in particular, has at times been mischaracterized as supporting or justifying negative behaviors (e.g., rape, aggression; discussed in Nicolas & Welling, 2015). These critiques are often based in the naturalistic fallacy—the idea that what is ought to be (i.e., what is natural is good; see Hagen,  2005). Such criticisms are unfounded, however, because studying or postulating why something exists does not mean something ought to exist. Put another way, because a trait or behavior is natural does not make it morally defensible. Similarly, evolutionary psychology and behavioral endocrinology have at times been mistakenly assumed to support genetic determinism (i.e., the view that human behavior is entirely controlled by genes/biology). For example, the cross-cultural finding that men are, on average, Welling and Shackelford

5

more aggressive than women and that testosterone is associated with aggression can be misinterpreted as claiming that men are biologically incapable of controlling their aggression (reviewed in Campbell, 2012). But genetically determined mechanisms do not necessarily translate into specific behaviors (see Hagen,  2005); for instance, the extent to which men engage in intimate partner violence varies greatly depending on the gender equality of their culture (Archer, 2006), showing that environment plays a role in shaping behavioral responses. By the same token, although feelings of pathogen disgust likely evolved to facilitate contagion avoidance (Tybur, Lieberman, Kurzban, & DeScioli,  2013), people (e.g., healthcare professionals, sanitation workers) can and do purposefully come into contact with disease-causing agents. Thus, mechanisms such as hormones may predispose individuals to behave in certain ways given a particular environment, but that does not mean biology is destiny. The human brain and body consist of adaptations that evolved in ancestral environments, but to understand why they work, it helps to understand how they work. In addition to their more basic biological functions, hormones influence our behavior in many domains, including our developmental, survival-related, reproductive, social, and affective behaviors. Viewing human behavior through an evolutionary lens informs our understanding of ­ the selection pressures faced by our species in our ­ancestral past. Incorporating hormonal mechanisms into that evolutionary perspective allows researchers to  integrate proximate explanations into ultimate causal reasoning. More fundamentally, understanding both the proximate and ultimate causes of behavior helps us to understand why traits change over time.

References

Alexander, G., & Hines, M. (2002). Sex differences in response to children’s toys in nonhuman primates (Cercopithecus aethiops sabaeus). Evolution and Human Behavior, 23, 467–479. Apter-Levi, Y., Zagoory-Sharon, O., & Feldman, R. (2014). Oxytocin and vasopressin support distinct configurations of social synchrony. Brain Research, 1580, 124–132. Arango, V., Ernsberger, P., Marzuk, P. M., Chen, J., Tierney, H., Stanley, M., . . . Mann, J.  J. (1990). Autoradiographic demonstration of increased serotonin 5-HT2 and β-adrenergic receptor binding sites in the brain of suicide victims. Archives of General Psychiatry, 47(11), 1038–1047. Archer, J. (2006). Cross-cultural differences in physical aggression between partners: A social-role analysis. Personality and Social Psychology Review, 10(2), 133–153. Bailey, J.  M., & Zucker, K.  J. (1995). Childhood sex-typed behavior and sexual orientation: A conceptual analysis and quantitative review. Developmental Psychology, 31(1), 43–55.

6

Bartz, J., Simeon, D., Hamilton, H., Kim, S., Crystal, S., Braun, A., . . . Hollander, E. (2011). Oxytocin can hinder trust and cooperation in borderline personality disorder. Social Cognitive and Affective Neuroscience, 6(5), 556–563. Beach, F. A. (1948). Hormones and behavior. New York, NY: Paul Hoeber. Beach, F.  A. (1975). Behavioral endocrinology: An emerging discipline. American Scientist, 63, 178–187. Béchet, D. (1986). Control of gene expression by steroid hormones. Reproduction, Nutrition, & Development, 26(5A), 1025–1055. Berthold, A.  A. (1849). Transplantation der Hoden [The transplantation of the testes]. Archiv für Anatomie, Physiologie und Wissenschaftliche Medicin, 16(1), 42–46. Boothroyd, L. G., Meins, E., Vukovic, J., & Burt, D. M. (2014). Developmental changes in children’s facial preferences. Evolution and Human Behavior, 35(5), 376–383. Bramble, D.  M., & Lieberman, D.  E. (2004). Endurance running and the evolution of Homo. Nature, 432, 345–352. Bribiescas, R. G. (2001). Reproductive ecology and life history of the human male. American Journal of Physical Anthropology, 116, 148–176. Brown, S. L., Fredrickson, B. L., Wirth, M. M., Poulin, M. J., Meier, E.  A., Heaphy, E.  D., . . . Schultheiss, O.  C. (2009). Social closeness increases salivary progesterone in humans. Hormones and Behavior, 56(1), 108–111. Buss, D.  M. (2005). Handbook of evolutionary psychology. Hoboken, NJ: Wiley. Buunk, A. P., & Van Brummen-Girigori, O. (2016). Menstrual cycle effects on jealousy: A study in Curaçao. Evolution, Mind and Behavior, 14(1), 43–54. Campbell, A. (2012). The study of sex differences: Feminism and biology. Zeitschrift für Psychologie, 220, 137–143. Carré, J. M., Ortiz, T. L., Labine, B., Moreau, B. J. P., Viding, E., Newmann, C.  S., & Goldfarb, B. (2015). Digit ratio (2D:4D) and psychopathic traits moderate the effect of exogenous testosterone on socio-cognitive processes in men. Psychoneuroendocrinology, 62, 319–326. Carré, J.  M., Putnam, S.  K., & McCormick, C.  M. (2009). Testosterone responses to competition predict future aggressive behaviour at a cost to reward in men. Psychoneuroendocrinology, 34(4), 561–570. Carroll, B.  J., Cassidy, F., Naftolowitz, D., Tatham, N.  E., Wilson, W.  H., Iranmanesh, A., . . . Veldhuis, J.  D. (2007). Pathophysiology of hypercortisolism in depression. Acta Psychiatrica Scandinavica, 115(s433), 90–103. Cashdan, E., & Gaulin, S. J. C. (2016). Why go there? Evolution of mobility and spatial cognition in women and men. Human Nature, 27(1), 1–15. Clark, M., & Galef, B.  G. (1995). Prenatal influences on reproductive life history strategies. Trends in Ecology and Evolution, 10, 151–153. Cobey, K. D., Buunk, A. P., Roberts, C. S., Klipping, C. Appels, N., Zimmerman, Y., . . . Pollet, T. (2012). Reported jealousy differs as a function of menstrual cycle stage and contraceptive pill use: A within-subjects investigation. Evolution and Human Behavior, 33, 395–401. Colborn, T., vom Saal, F. S., & Soto, A. M. (1993). Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environmental Health Perspectives, 101(5), 378–384. Collaer, M.  L., & Hines, M. (1995). Human behavioral sex differences: A role for gonadal hormones during early development? Psychological Bulletin, 118(1), 55–107.

Integrating Mechanisms and Functions to Understand Behavior

Denson, T. F., Mehta, P. H., & Ho Tan, D. (2013). Endogenous testosterone and cortisol jointly influence reactive aggression in women. Psychoneuroendocrinology, 38, 416–424. Donaldson, S. H., Welling, L. L. M., & Reeve, S. D. (2017). The influence of hormone replacement therapy on mating psychology among post-menopausal women. Personality and Individual Differences, 115, 13–18. Duarte-Guterman, P., Yagi, S., Chow, C., & Galea, L.  A.  M. (2015). Hippocampal learning, memory, and neurogenesis: Effects of sex and estrogens across the lifespan in adults. Hormones and Behavior, 74, 37–52. Edelstein, R. S., van Anders, S. M., Chopik, W. J., Goldey, K. L., & Wardecker, B.  M. (2014). Dyadic associations between testosterone and relationship quality in couples. Hormones and Behavior, 65, 401–407. Edwards, C. P. (2002). Behavioral sex differences in children of diverse cultures: The case of nurturance to infants. In M.  Pereira & L.  Fairbanks (Eds.), Juvenile primates: Life history, development, and behavior (pp. 327–338). Chicago, IL: University of Chicago Press. Eisenegger, C., Haushofer, J., & Fehr, E. (2011). The role of testosterone in social interaction. Trends in Cognitive Sciences, 15(6), 263–271. Farfel, G. M., Kleven, M. S., Woolverton, W. L., Seiden, L. S., &  Perry, B.  D. (1992). Effects of repeated injections of cocaine on catecholamine receptor binding sites, dopamine transporter binding sites and behavior in rhesus monkey. Brain Research, 578(1–2), 235–243. Feldman, R. (2012). Oxytocin and social affiliation in humans. Hormones and Behavior, 61(3), 380–391. Fisher, M. L. (2004). Female intrasexual competition decreases female facial attractiveness. Proceedings of the Royal Society B: Biological Sciences, 271, 283–285. Fröjdö, S., Vidal, H., & Pirola, L. (2009). Alterations of insulin signaling in type 2 diabetes: A review of the current evidence from humans. Biochimica et Biophysica Acta, 1792(2), 83–92. Gangestad, S. W., Simpson, J. A., Cousins, A. J., Garver-Apgar, C. E., & Christensen, P. N. (2004). Women’s preferences for male behavioral displays change across the menstrual cycle. Psychological Science, 15(3), 203–207. Gangestad, S.  W., & Thornhill, R. (2008). Human oestrus. Proceedings of the Royal Society B: Biological Sciences, 275, 991–1000. Gangestad, S.  W., Thornhill, R., & Garver, C.  E. (2002). Changes in women’s sexual interests and their partners’ materetention tactics across the menstrual cycle: Evidence for shifting conflicts of interest. Proceedings of the Royal Society B: Biological Sciences, 269(1494), 975–982. Garver-Apgar, C. E., Gangestad, S. W., & Thornhill, R. (2008). Hormonal correlates of women’s mid-cycle preference for the scent of symmetry. Evolution and Human Behavior, 29(4), 223–232. Gettler, L. T., McDade, T. W., Agustin, S. S., Feranil, A. B., & Kuzawa, C. W. (2013). Do testosterone declines during the transition to marriage and fatherhood relate to men’s sexual behavior? Evidence from the Philippines. Hormones and Behavior, 64(5), 755–763. Gettler, L. T., McDade, T. W., Agustin, S. S., Feranil, A. B., & Kuzawa, C. W. (2015). Longitudinal perspectives on fathers’ residence status, time allocation, and testosterone in the Philippines. Adaptive Human Behavior and Physiology, 1(2), 124–149.

Gettler, L. T., McDade, T. W., Feranil, A. B., & Kuzawa, C. W. (2011). Longitudinal evidence that fatherhood decreases testosterone in human males. Proceedings of the National Academy of Sciences of the United States of America, 108(29), 16194–16199. Gettler, L. T., McKenna, J. J., Agustin, S. S., McDade, T. W., & Kuzawa, C. W. (2012). Does cosleeping contribute to lower testosterone levels in fathers? Evidence from the Philippines. PLoS One, 7(9), e41559. Gilbert, L.  K., Breiding, M.  J., Merrick, M.  T., Thompson, W. W., Ford, D. C., Dhingra, S. S., & Parks, S. E. (2015). Childhood adversity and adult chronic disease: An update from ten states and the District of Columbia, 2010. American Journal of Preventive Medicine, 48(3), 345–349. Gildersleeve, K., Haselton, M.  G., & Fales, M.  R. (2014a). Do women’s mate preferences change across the ovulatory cycle? A meta-analytic review. Psychological Bulletin, 140(5), 1205–1259. Gildersleeve, K., Haselton, M. G., & Fales, M. (2014b). ­Metaanalyses and p-curves support robust cycle shifts in mate preferences: Response to Wood & Carden and Harris, Pashler & Mickes (2014). Psychological Bulletin, 40, 1272–1280. Gimpl, G., & Fahrenholz, F. (2001). The oxytocin receptor system: Structure, function, and regulation. Physiological Reviews, 81(2), 629–683. Goglia, F., Moreno, M., & Lanni, A. (1999). Action of thyroid hormones at the cellular level: The mitochondrial target. FEBS Letters, 452(3), 115–120. Gray, P. B., Chapman, J. F., Burnham, T. C., McIntyre, M. H., Lipson, S.  F., & Ellison, P.  T. (2004). Human male pair  bonding and testosterone. Human Nature—An Inter­ disciplinary Biosocial Perspective, 15, 119–131. Gray, P. B., Ellison, P. T., & Campbell, B. C. (2007). Testosterone and marriage among Ariaal men of northern Kenya. Current Anthropology, 48(5), 750–755. Gray, P. B., Kahlenberg, S. M., Barrett, E. S., Lipson, S. F., & Ellison, P. T. (2002). Marriage and fatherhood are associated with lower testosterone in males. Evolution and Human Behavior, 23(3), 193–201. Hagen, E.  H. (2005). Controversial issues in evolutionary psychology. In D.  Buss (Ed.), Handbook of evolutionary psychology (pp. 145–173). Hoboken, NJ: Wiley. Hahn, A. C., DeBruine, L. M., Fisher, C., & Jones, B. C. (2015). The reward value of infant facial cuteness tracks withinsubject changes in women’s salivary testosterone. Hormones and Behavior, 67, 54–59. Halpern, D. (2013). Sex differences in cognitive abilities. East Sussex, UK: Psychology Press. Hampson, E., & Rovet, J.  F. (2015). Spatial function in adolescents and young adults with congenital adrenal hyperplasia: Clinical phenotype and implications for the androgen hypothesis. Psychoneuroendocrinology, 54, 60–70. Haselton, M.  G., & Gangestad, S.  W. (2006). Conditional expression of women’s desires and men’s mate guarding across the ovulatory cycle. Hormones and Behavior, 49(4), 509–518. Hassett, J.  M., Siebert, E.  R., & Wallen, K. (2008). Sex differences in rhesus monkey toy preferences parallel those of children. Hormones and Behavior, 54(3), 359–364. Havlíček, J., Roberts, S.  C., & Flegr, J. (2005). Women’s preference for dominant male odour: Effects of menstrual cycle and relationship status. Biology Letters, 1, 256–259.

Welling and Shackelford

7

Heil, M., Kavšek, M., Rolke, B., Beste, C., & Jansen, P. (2011). Mental rotation in female fraternal twins: Evidence for intrauterine hormone transfer? Biological Psychology, 86, 90–93. Hines, M. (2003). Brain gender. Oxford, UK: Oxford University Press. Holt-Lunstad, J., Birmingham, W.  C., & Light, K.  C. (2015). Relationship quality and oxytocin: Influence of stable and modifiable aspects of relationships. Journal of Social and Personal Relationships, 32, 472–490. Hyde, J. S., Mezulis, A. H., & Abramson, L. Y. (2008). The ABCs of depression: Integrating affective, biological, and cognitive models to explain the emergence of the gender difference in depression. Psychological Review, 115(2), 291–313. Israel, S. (2016). Toward a mechanistic understanding of how variation in the oxytocin receptor gene shapes individual differences in brain and social behavior. Biological Psychiatry, 80(2), e7–e9. Jones, B.  C., Little, A.  C., Boothroyd, L., DeBruine, L.  M., Feinberg, D. R., Law Smith, M. J., . . . Perrett, D. I. (2005). Commitment to relationships and preferences for femininity and apparent health in faces are strongest on days of the menstrual cycle when progesterone level is high. Hormones and Behavior, 48(3), 283–290. Kim, J., Szinte, J. S., Boulware, M. I., & Frick, K. M. (2016). 17βEstradiol and agonism of G-protein-coupled estrogen receptor enhance hippocampal memory via different cell-signaling mechanisms. Journal of Neuroscience, 36, 3309–3321. Kimura, D. (2004). Human sex differences in cognition, fact, not predicament. Sexualities, Evolution & Gender, 6(1), 45–53. Klinesmith, J., Kasser, T., & McAndrew, F.  T. (2006). Guns, testosterone, and aggression: An experimental test of a mediational hypothesis. Psychological Science, 17(7), 568–571. Kornienko, O., Clemans, K. H., Out, D., & Granger, D. A. (2014). Hormones, behavior, and social network analysis: Exploring associations between cortisol, testosterone, and  network structure. Hormones and Behavior, 66(3), 534–544. Kornienko, O., Schaefer, D.  R., Weren, S., Hill, G.  W., & Granger, D. A. (2016). Cortisol and testosterone associations with social network dynamics. Hormones and Behavior, 80, 92–102. Kosfeld, M., Heinrichs, M., Zak, P. J., Fischbacher, U., & Fehr, E. (2005). Oxytocin increases trust in humans. Nature, 435(7042), 673–676. Little, A. C. (2015). Attraction and human mating. In V. ZeiglerHill, L.  L.  M.  Welling, & T.  K.  Shackelford (Eds.), Evolutionary perspectives on social psychology (pp. 319–332). New York, NY: Springer Publishing. Little, A. C., & Jones, B. C. (2012). Variation in facial masculinity and symmetry preferences across the menstrual cycle is moderated by relationship context. Psychoneuroendocrinology, 37(7), 999–1008. Lucas, M.  M., Koff, E., & Skeath, S. (2007). Pilot study of relationship between fertility risk and bargaining. Psychological Reports, 101(1), 302–310. Luine, V. (2015). Recognition memory tasks in neuroendocrine research. Behavioral Brain Research, 285, 158–164. Lukaszewski, A. W., & Roney, J. R. (2009). Estimated hormones predict women’s mate preferences for dominant personality traits. Personality and Individual Differences, 47(3), 191–196. Meaney, M.  J., Stewart, J., & Beatty, W.  W. (1985). Sex differences in social play: The socialization of sex roles. Advances in the Study of Behavior, 15, 1–58.

8

Miller, S. L., & Maner, J. K. (2011). Ovulation as a male mating prime: Subtle signs of women’s fertility influence men’s mating cognition and behavior. Journal of Personality and Social Psychology, 100(2), 295–308. Miner, E. J., Gurven, M., Kaplan, H., & Gaulin, S. J. C. (2014). Sex difference in travel is concentrated in adolescence and tracks reproductive interests. Proceedings of the Royal Society of London B, 281, 20141476. Muehlenbein, M.  P., & Watts, D.  P. (2010). The costs of dominance: testosterone, cortisol and intestinal parasites in wild male chimpanzees. Biopsychosocial Medicine, 4, 21–33. Neave, N., Hamilton, C., Hutton, L., Tildesley, N., & Pickering, A. T. (2005). Some evidence of a female advantage in object location memory using ecologically valid stimuli. Human Nature, 16(2), 146–163. Nelson, R. J. (2010). Hormones and behavior: Basic concepts. In M.  D.  Breed & J.  Moore (Eds.), Encyclopedia of animal behavior (pp. 97–105). Amsterdam, Netherlands: Elsevier. Nicolas, S. C. A., & Welling, L. L. M. (2015). The Darwinian mystique? Synthesizing evolutionary psychology and feminism. In V.  Zeigler-Hill, L.  L.  M.  Welling, & T.  K. Shackelford (Eds.), Evolutionary perspectives on social psychology (pp. 203–212). New York, NY: Springer Publishing. Phoenix, C. H., Goy, R. W., Gerall, A. A., & Young, W. C. (1959). Organizing action of prenatally administered testosterone propionate on the tissues mediating mating behavior in the female guinea pig. Endocrinology, 65, 369–382. Piccoli, V., Foroni, F., & Carnaghi, A. (2013). Comparing group dehumanization and intra-sexual competition among normally ovulating women and hormonal contraceptive users. Personality and Social Psychology Bulletin, 39, 1600–1609. Pinker, S. (2002). The blank slate. New York, NY: Viking. Pipitone, R.  N., & Gallup, G.  G. (2008). Women’s voice attractiveness varies across the menstrual cycle. Evolution and Human Behavior, 29(4), 268–274. Pratchett, L.  C., & Yehuda, R. (2011). Foundations of posttraumatic stress disorder: Does early life trauma lead to  adult posttraumatic stress disorder? Development and Psychopathology, 23(2), 477–491. Puts, D. A., Bailey, D. H., Cárdenas, R. A., Burriss, R. P., Welling, L.  L.  M., Wheatley, J.  R., & Dawood, K. (2013). Women’s attractiveness changes with estradiol and progesterone across the ovulatory cycle. Hormones and Behavior, 63, 13–19. Puts, D. A., McDaniel, M. A., Jordan, C. L., & Breedlove, S. M. (2008). Spatial ability and prenatal androgens: Meta-analyses of congenital adrenal hyperplasia and digit ratio (2d:4d) studies. Archives of Sexual Behavior, 37(1), 100–111. Raichlen, D.  A., Armstrong, H., & Lieberman, D.  E. (2011). Calcaneus length determines running economy: Implications for endurance running performance in modern humans and Neandertals. Journal of Human Evolution, 60(3), 299–308. Roberts, S. C., Havlíček, J., Flegr, J., Hruskova, M., Little, A. C., Jones, B. C., . . . Petrie, M. (2004). Female facial attractiveness increases during the fertile phase of the cycle. Proceedings of the Royal Society of London B, 271, S270–S272. Roberts, S.  C., Little, A.  C., Burriss, R.  P., Cobey, K.  D., Klapilová, K., Havlicek, J., . . . Petrie, M. (2014). Partner choice, relationship satisfaction, and oral contraception. Psychological Science, 25(7), 1497–1503. Roff, D.  A. (2002). Life history evolution. Sunderland, MA: Sinauer Associates. Rolian, C., Lieberman, D.  E., Hamill, J., Scott, J.  W., & Werbel, W. (2009). Walking, running and the evolution of

Integrating Mechanisms and Functions to Understand Behavior

short toes in humans. Journal of Experimental Biology, 212, 713–721. Ronay, R., & von Hippel, W. (2010). The presence of an attractive woman elevates testosterone and physical risk taking in young men. Social Psychological and Personality Science, 1, 57–64. Roney, J. R., & Simmons, Z. L. (2013). Hormonal predictors of women’s sexual desire in normal menstrual cycles. Hormones and Behavior, 63, 636–645. Sapolsky, R.  M. (2005). The influence of social hierarchy on primate health. Science, 308, 648–652. Saxton, T. K., DeBruine, L. M., Jones, B. C., Little, A. C., & Roberts, S. C. (2009). Face and voice attractiveness judgments change during adolescence. Evolution and Human Behavior, 30(6), 398–408. Schneiderman, I., Kanat-Maymon, Y., Ebstein, R.  P., & Feldman, R. (2014). Cumulative risk on the oxytocin receptor gene (OXTR) underpins empathic communication difficulties at the first stages of romantic love. Social Cognitive and Affective Neuroscience, 9, 1524–1529. Silverman, I., Choi, J., & Peters, M. (2007). The hunter-gatherer theory of sex differences in spatial abilities: Data from 40 countries. Archives of Sexual Behavior, 36(2), 261–268. Slater, A., Quinn, P. C., Hayes, R., & Brown, E. (2000). The role of facial orientation in newborn infants’ preference for attractive faces. Developmental Science, 3(2), 181–185. Slater, A., Von der Schulenburg, C., Brown, E., Badenoch, M., Butterworth, G., Parsons, S., & Samuels, C. (1998). Newborn infants prefer attractive faces. Infant Behavior & Development, 21(2), 345–354. Taylor, C.  A., Guterman, N.  B., Lee, S.  J., & Rathouz, P.  J. (2009). Intimate partner violence, maternal stress, nativity, and risk for maternal maltreatment of young children. American Journal of Public Health, 99(1), 175–183. Tinbergen, N. (1963). On the aims of methods of ethology. Zeitschfrift für Tierpsychologie, 20, 410–433. Tooby, J., & Cosmides, L. (1992). The psychological foundations of culture. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 19–136). New York, NY: Oxford University Press. Tooby, J., & Cosmides, L. (2005). Conceptual foundations of evolutionary psychology. In D. Buss (Ed.), Handbook of evolutionary psychology (pp. 5–67). Hoboken, NJ: Wiley. Tybur, J. M., Lieberman, D., Kurzban, R., & DeScioli, P. (2013). Disgust: Evolved function and structure. Psychological Review, 120(1), 65–84.

van Anders, S. M., Goldey, K. L., & Kuo, P. X. (2011). The Steroid/ Peptide Theory of Social Bonds: Integrating testosterone and peptide responses for classifying social behavioral contexts. Psychoneuroendocrinology, 36(9), 1265–1275. van der Meij, L., Almela, M., Buunk, A. P., Fawcett, T. W., & Salvador, A. (2012). Men with elevated testosterone levels show more affiliative behaviours during interactions with women. Proceedings of the Royal Society B, 279, 202–208. Voyer, D., Postma, A., Brake, B., & Imperato-McGinley, J. (2007). Gender differences in object location memory: A meta-analysis. Psychonomic Bulletin & Review, 14(1), 23–38. Voyer, D., Voyer, S., & Bryden, M. P. (1995). Magnitude of sex differences in spatial abilities: A meta-analysis and consideration of critical variables. Psychological Bulletin, 117(2), 250–270. Vuoksimaa, E., Kaprio, J., Kremen, W. S., Hokkanen, L., Viken, R. J., Tuulio-Henriksson, A., & Rose, R. J. (2010). Having a male co-twin masculinizes mental rotation performance in females. Psychological Science, 21, 1069–1071. Welling, L.  L.  M., Jones, B.  C., DeBruine, L.  M., Conway, C. A., Law Smith, M. J., Little, A. C., . . . Al-Dujaili, E. A. (2007). Raised salivary testosterone in women is associated with increased attraction to masculine faces. Hormones and Behavior, 52, 156–161. Welling, L. L. M., Jones, B. C., DeBruine, L. M., Smith, F. G., Feinberg, D. R., Little, A. C., & Al-Dujaili, E. A. S. (2008). Men report stronger attraction to femininity in women’s faces when their testosterone levels are high. Hormones and Behavior, 54, 703–708. Welling, L.  L.  M., Moreau, B.  J.  P., Bird, B.  M., Hansen, S., &  Carré, J.  M. (2016). Exogenous testosterone increases men’s perceptions of their own physical dominance. Psychoneuroendocrinology, 64, 136–142. Welling, L. L. M., Puts, D. A., Roberts, S. C., Little, A. C., & Burriss, R. P. (2012). Hormonal contraceptive use and mate retention behavior in women and their male partners. Hormones and Behavior, 61, 114–120. Williams, C.  L., Barnett, A.  L., & Meck, W.  B. (1990). Organizational effects of early gonadal secretions on sexual differentiation in spatial memory. Behavioral Neuroscience, 104, 84–97. Zilioli, S., Ponzi, D., Henry, A., Kubicki, K., Nickels, N., Wilson, M. C., & Maestripieri, D. (2016). Interest in babies negatively predicts testosterone responses to sexual visual stimuli among heterosexual young men. Psychological Science, 27, 114–118.

Welling and Shackelford

9

P A R T 

Development and Survival

1

CH A PT E R

2

Hormones and Behavior A Life History Perspective

Maren N. Vitousek and Laura A. Schoenle

Abstract Hormones mediate the expression of life history traits—phenotypic traits that contribute to lifetime fitness (i.e., reproductive timing, growth rate, number and size of offspring). The endocrine system shapes phenotype by organizing tissues during developmental periods and by activating changes in behavior, physiology, and morphology in response to varying physical and social environments. Because hormones can simultaneously regulate many traits (hormonal pleiotropy), they are important mediators of life history trade-offs among growth, reproduction, and survival. This chapter reviews the role of hormones in shaping life histories with an emphasis on developmental plasticity and reversible flexibility in endocrine and life history traits. It also discusses the advantages of studying hormone–behavior interactions from an evolutionary perspective. Recent research in evolutionary endocrinology has provided insight into the heritability of endocrine traits, how selection on hormone systems may influence the evolution of life histories, and the role of hormonal pleiotropy in driving or constraining evolution. Keywords:  life history, trade-offs, hormones, reproductive timing, growth, endocrine traits, fitness

What Is Life History?

The life history of an organism is defined as the ­pattern of investment it makes in growth, reproduction, and survival throughout its lifetime (Ricklefs, 1977; Roff, 2002; Stearns, 1992). Life histories are composed of different stages, which are characterized by differing patterns of investment. For example, during ontogeny, juvenile great apes devote most of their energy to growth and survival. Once they reach sexual maturity, investment in growth is replaced by investment in reproduction. All organisms have a life history, but the sequence of stages, and the particular challenges that they entail, can differ dramatically. Insects like the Monarch butterfly, which undergo complete metamorphosis, proceed through multiple stages of ontogeny (egg, larval, and pupal) before reaching adulthood. The life ­history of many species involves entering the same life history stage repeatedly, for example, multiparous individuals e­ngaging in repeated reproductive

a­ ttempts, or species that migrate annually between wintering and breeding grounds. Life history traits are specific behavioral, morphological, or physiological traits that influence the rate or timing of life history stages, or the location in which they occur. In birds, important life history traits include the timing of clutch initiation and the number of eggs laid in a clutch. Because these traits influence organisms’ lifetime fitness in different ways depending on the environment, life history traits show distinct patterns across environments. For example, birds that inhabit more seasonal environments, where breeding seasons are shorter and each pair is likely to have only a single reproductive ­attempt per season, invest more in each attempt by laying more eggs per clutch (Cardillo,  2002; Jetz, Sekercioglu, & Bohning-Gaese, 2008). Because resources are finite, life histories are shaped by fundamental trade-offs among allocation to different key components (Roff,  2002; Stearns, 13

1992). Thus, organisms often differ in their “pace of life” (Ricklefs & Wikelski, 2002). Short-lived species, like many insects and rodents, often invest an enormous amount in one or a few reproductive a­ ttempts. In contrast, long-lived species, like African elephants and many primates, delay reproduction in favor of a long period of growth, before investing heavily in the care of a small number of young, often over an extended period of time. Such trade-offs among investment in different life history traits or stages also operate within populations. For example, investment in immune function (which can be crucial for survival) often impairs reproduction (Bonneaud et al., 2003); conversely, when reproductive effort increases, immune investment declines (Ardia, Schat, & Winkler, 2003; Knowles, Nakagawa, & Sheldon, 2009). As a result of these trade-offs, individuals within a population can differ—sometimes dramatically—in their life history.

How Do Hormones Mediate Phenotypic Expression?

Hormones are key mediators of life history. They influence the expression of life history traits, initiate transitions among life history stages, and mediate trade-offs among investment in different components of life history. Mechanistically, hormones can change the probability that a behavior will be performed, or an individual’s physiological state, through two main modes of action: organizational and activational ­effects (see also Hampson, this volume). During development, hormones organize phenotypes by ­ shaping tissue structure and function or by inducing epigenetic changes (i.e., long-term alterations in the transcriptional potential of a cell). Hormones’ organizational effects, which are often irreversible, are a form of developmental plasticity; a single g­ enotype could lead to different phenotypes depending on hormone concentrations during development (Arnold, 2009). For example, exposure to gonadal hormones during puberty contributes to the development of adult sexual behavior in rodents. Males castrated during adolescence attempt fewer matings than intact males and can fail to successfully mate as adults, even when supplemented with testosterone (K.  M.  Schulz & Sisk, 2016; Sisk & Zehr, 2005). Hormones play an activational role when they reversibly influence the probability of trait expression. Hormone-regulated phenotypic flexibility allows organisms to respond rapidly to changes in the physical or social environment (Arnold, 2009). For ­instance, glucocorticoid hormones increase in ­ response to 14

Hormones and Behavior

challenges (e.g., psychological, physiological, or social stressors) and activate a suite of behaviors that can help organisms respond effectively to challenges. In many species, moderate stressors increase foraging behavior (Landys, Ramenofsky, & Wingfield, 2006); in humans and rodents, elevated glucocorticoids promote the intake of specific food types, often those with high fat or sugar content (Dallman, 2009; Dallman et al., 2004; la Fleur, 2006). For most organisms this “stress eating” phenomenon is likely to be ­adaptive, as it supports the acquisition of energy required to deal with a challenge. The stressors facing modern humans, however, are largely psychological in nature, and are rarely accompanied by food limitation. Instead, this pattern of eating has been implicated in generating a variety of health problems, from obesity to heart disease. Hormones mediate the expression of life history traits, and other phenotypic traits, by binding to receptors on or inside cells. Thus, trait expression is affected not only by circulating hormone concentrations but also by other factors that influence how effectively hormones reach receptors, and how they change cellular function and gene expression following binding (e.g., carrier/transport protein ­expression, receptor affinity and location, and the presence of cofactors). Hormone concentrations and other regulatory factors produce multiple variants of hormone–trait relationships. For example, trait expression can be linear and dose dependent, remain constant across a range of hormone concentrations (step function), and/or peak at intermediate concentrations (inverted-U function) (AdkinsRegan, 2005). Hormones frequently influence the expression of  multiple traits simultaneously. Such pleiotropic actions are an important mechanism, enabling ­hormones to coordinate the multitude of traits that make up an individual’s life history strategy. Pleiotropic actions can all contribute to inducing the same phenotypic state—for instance, when females of ­ many mammal species give birth, oxytocin induces uterine contractions, initiates the release of milk, and primes females for postpartum maternal behavior (Gimpl & Fahrenholz, 2001). These actions of oxytocin all support reproduction. Pleiotropy can also result in the same hormone producing positive and negative effects on a life history trait. For example, estrogens support egg production in birds, but at the same time can suppress red blood cell production (Wagner, Prevolsek, Wynne-Edwards, & Williams, 2008; Wagner, Stables, & Williams, 2008).

The ­resulting reduction in red blood cell counts, ­hematocrit, and hemoglobin (Wagner, Stables, & Williams, 2008) can impair reproductive success (Fronstin, Christians, & Williams, 2015). From embryonic development through senescence, endocrine phenotype shapes the physiology and behaviors that underlie life history traits. Further­ more, hormones facilitate the developmental plasticity and trait flexibility that enable individuals to alter life history traits to match the environment. In the following section we present specific examples of the endocrine mediation of life history traits to illustrate the importance of endocrine physiology to survival and reproduction.

The Endocrine Mediation of Life History: Examples Reproductive Rate: Sexual Signals and Mate Choice

Life histories are intimately linked with reproductive rate because of the inherent trade-off between ­investment in reproduction and survival. One of the ways that hormones influence reproductive rate is by changing how much individuals invest in acquiring mates. Hormones play a central role in mediating the development and expression of many diverse behaviors and morphological traits used to advertise to potential mates. They also influence how these signals are perceived and evaluated by competitors and by members of the opposite sex—thus yielding their ultimate effect on reproductive success. Organizational effects of exposure to sex steroids during development have long been known to shape the expression of secondary sexual traits. The pioneering work of Arnold Berthold in the 1840s revealed that the male-typical plumage, colorful ­ comb, and sexual behavior of roosters are absent in individuals castrated as chicks, but can be restored by reimplanting testes in the abdomen. More recent work has revealed that natural variation in the amount or timing of exposure to androgens during development can also impact sexual signal development and expression. For example, male rodents that develop in utero in between two other male embryos are ­exposed to higher testosterone levels than those that develop between two females. As adults, they differ in their sexual, aggressive, and parental behavior (­reviewed in Clark & Galef, 1995) and develop ­olfactory signals that are perceived as more attractive to females (Clark & Galef, 1994). Exposure to elevated glucocorticoid levels during development also can profoundly affect the development of sexual

s­ignals, both through their direct effects and by ­altering the concentration or temporal dynamics of other hormones. A particularly well-studied example of this phenomenon is the complex learned songs that songbirds use to advertise to conspecifics (e.g., Nowicki, Peters, & Podos, 1998). Males exposed to developmental stress, or to experimentally elevated glucocorticoids during development, show a reduced ability to learn the complex songs used in sexual ­interactions (Bell et al. 2018; Spencer, Buchanan, Goldsmith, & Catchpole, 2003). Females show a distinct preference for the song of nondevelopmentally stressed males (Spencer et al., 2005). Hormones also activate the production and ­expression of dynamic signal traits—those whose expression levels change over time—in adulthood. Androgens are well-known mediators of a diversity of signal types, including bright or distinctively colored feathers and skin, behavioral displays, courtship vocalizations, and olfactory signals (Evans, Goldsmith, & Norris,  2000; Gosling & Roberts, 2001; Mougeot, Dawson, Redpath, & Leckie, 2005; Wikelski, Steiger, Gall, & Nelson, 2005). Androgens can also be involved in maintaining the honesty of sexual advertisements through pleiotropic effects on immune function (Foo, Nakagawa, Rhodes, & Simmons, 2016) or other aspects of phenotype (Wingfield, Lynn, & Soma, 2001). Yet androgens are by no means ubiquitous mediators of sexual signals; the expression of many signal types is ­mediated by other hormones, or by the combined actions of multiple hormones. For example, the vocal signals of midshipman fish are mediated by the opposing effects of steroids (11-ketotestosterone and cortisol) and nonapeptides (arginine vasotocin and isotocin) (Bass, 2008; Goodson & Bass, 2000). Hormones also influence mating success by altering how sexual signals are perceived and evaluated. The hypothalamic-pituitary-gonadal (HPG) axis, which regulates the production of gonadal hormones (e.g., estrogens and androgens), plays a particularly important role in mediating mating preferences. Female tungara frogs and zebra finches increase their response to the acoustic displays of males when they are dosed with gonadal hormones (Lynch, Crews, Ryan, & Wilczynski, 2006; Vyas, Harding, Borg, & Bogdan,  2009). Gonadotropin-releasing hormone (GnRH), which constitutes the first step of the HPG axis, can influence females’ willingness to mate (Sakuma & Pfaff, 1980), as well as changing how prior experience influences mating decisions (Okuyama et al., 2014). Vitousek and Schoenle

15

Because organisms in stressful environments often have a lower residual reproductive value (i.e., a lower chance of reproducing successfully in the future), they are more likely to benefit from investing in a current reproductive attempt rather than waiting for the uncertain possibility that they will reproduce again. It is thus unsurprising that glucocorticoid hormones also appear to influence how accepting females are of potential mates. This can occur through both organizational and activational effects. For example, female European starlings that are developmentally stressed show less of a preference for the song of conspecifics over heterospecifics and differ in their immediate early gene response to song in the auditory forebrain (Farrell, Neuert, Cui, & Macdougall-Shackleton, 2015). In rats and frogs, females with experimentally elevated circulating glucocorticoids are more accepting of males that display nonpreferred signals (Davis & Leary, 2015; Kavaliers & Ossenkopp,  2001). In some cases, ­investment in mate choice may be more closely ­related to the response to acute stressors than baseline hormone levels. Female Galapagos marine iguanas down-regulate glucocorticoid production during mate choice—an extraordinarily costly process in which females can lose over 20 percent of their body mass (Vitousek, Mitchell, Woakes, Niemack, & Wikelski,  2007; Vitousek, Mitchell, Romero, Awerman, & Wikelski, 2010). Natural variation in baseline glucocorticoid levels is unrelated to mate selectivity, but females that mount a stronger glucocorticoid response to acute stress invest less effort in mate choice (Vitousek & Romero, 2013).

Parental Investment and Offspring Fitness

Following mating, individuals of one or both sexes face a series of decisions about how much to invest in their developing offspring—decisions that can have major consequences for offspring survival, and for parents’ ability to engage and invest in future reproductive attempts. It is often clear from casual observation that individuals differ enormously in the extent to which they invest in their offspring. These differences can include the extent to which developing embryos are provisioned with nutrients and immune compounds, postnatal feeding rates, nest defense, and affiliation behaviors like grooming and licking offspring. The hormones that activate and mediate parental investment differ across species and between the sexes, but steroid hormones (estrogens, progesterone, glucocorticoids, and testosterone), prolactin, and nonapeptides are commonly involved (Adkins-Regan, 2005; Lynn, 2016). Less is 16

Hormones and Behavior

known about whether variation in endocrine system function drives individual variation in parental investment, but clues are emerging in several systems. Stressor exposure during development has been linked to differences in parental investment in a number of species. A remarkable demonstration of this effect can be seen in the children born in the western Netherlands in 1945. During this period, a Nazi embargo cut off food transport into the region, and its inhabitants suffered a several-month period of acute starvation. Once the Netherlands were ­liberated by the Allies, food was again plentiful. Exposure to this prenatal stressor had lasting effects not only on the health and senescence rates of the individuals that were born from exposed mothers but also on their own patterns of parental investment (L. C. Schulz, 2010). As adults, prenatally exposed women reproduced sooner and gave birth to babies with more body fat; their offspring also suffered from poor health in adulthood (Painter et al., 2008). Experimental work in other species has begun to pinpoint the mechanisms linking stressor exposure to transgenerational changes in parental investment and other life history traits. In rats, the amount of licking and grooming that young receive from their mothers shapes their own parental behavior as adults (Francis, Diorio, Liu, & Meaney, 1999). Rats groomed at higher rates in their youth are also more attentive mothers to their own young. Work by Michael Meaney and colleagues has determined that these changes occur through glucocorticoidmediated ­epigenetic programming (Weaver et al., 2004). Grooming activates the serotonin systems of offspring and increases glucocorticoid receptor gene transcription. As adults, offspring reared by highgrooming mothers have reduced methylation in the glucocorticoid receptor gene promoter region, which leads to greater glucocorticoid receptor expression in the brain and a more muted hormonal stress response (van Hasselt et al., 2012; Hellstrom, Dhir, Diorio, & Meaney,  2012; Weaver et al.,  2004). Such effects are not limited to mammals; maternal stressor exposure and maternal care have been causally linked with endocrine function, gene expression, behavior, and cognitive function in a wide variety of taxa (e.g., Andrewartha & Burggren, 2012; Boogert, Zimmer, & Spencer, 2013; Nyman, Fischer, AubinHorth, & Taborsky 2018; Marasco, Herzyk, Robinson, & Spencer, 2016; Roche, McGhee, & Bell, 2012). It has also recently been discovered that fathers’ experiences can program aspects of offspring phenotype through stress-induced changes in sperm microRNA (Rodgers, Morgan, Leu, & Bale, 2015).

Understanding the extent to which hormones function as transgenerational mediators of phenotype and life history trade-offs, and the situations under which such effects are evolutionarily advantageous, are exciting and active areas of investigation.

Hormones Mediate Life History Transitions

Transitioning between life history stages involves coordinated changes in the expression of morphological, physiological, and behavioral traits. This phenotypic flexibility enables an animal to cope with fluctuating environmental conditions (e.g., seasons, drought) or perform different functions (e.g., reproduction, growth). Hormones can regulate the onset of life history transitions in response to environmental and internal cues, allowing individuals to time the transition to match current conditions (Wingfield, 2008). The morphological transformations that occur throughout metamorphosis might be the most ­extreme example of phenotypic plasticity during a life history transition. Metamorphosis has been widely studied in insects and amphibians, but also occurs in other taxa including fish, invertebrates, fungi, and plants (Bishop et al., 2006). In insects, metamorphosis is regulated primarily by antagonistic interactions between juvenile hormones and ecdysteroids (Mitra, 2013; Truman & Riddiford, 2002). Ecdysteroids support growth and development (Di Cara & King-Jones,  2013), whereas juvenile hormones typically restrict metamorphosis, resulting in the maintenance of juvenile traits (Mitra,  2013). The hormones’ release and, thus, the timing of metamorphosis depend on multiple factors, including temperature, energy stores, and the often photoregulated circadian clock (Di Cara & King-Jones, 2013). The endocrinology of amphibian metamorphosis parallels that of insects in that it involves ­antagonistic interactions between endocrine signals that promote development (thyroid hormones) and those that inhibit metamorphosis (prolactin) (Flatt, Moroz, Tatar, & Heyland,  2006; Galton,  1992; Kikuyama, Kawamura, Tanaka, & Yamamoto,1993; but see Huang & Brown, 2000). The optimal timing of transitioning between life history stages is likely to be influenced by the quality of the environment. In amphibians, environmental stressors affect the timing of metamorphosis through their effect on glucocorticoid hormones. For example, pond ­desiccation—which is typically more costly to larval stage individuals than to adults—accelerates metamorphosis by increasing glucocorticoid levels, which act in synergy with thyroid hormones to advance

metamorphosis (Denver,  2009,  2013; Denver & Middlemis-Maher,  2010). However, accelerating meta­morphosis involves life history trade-offs: Indi­ viduals that metamorphose at a smaller size also reach a smaller maximal body size (Denver, 2013, 2009). Smaller individuals are less likely to survive to sexual maturity, and reproduce later (Berven, 1990; Semlitsch, Scott, & Pechmann, 1988). In species that reproduce seasonally, appropriately timing reproduction can be critical to success (Hahn & Macdougall-Shackleton,  2008; van Noordwijk, Mccleery, & Perrins, 1995; Perrins, 1970; Verhulst & Nilsson, 2008). As a result, many species rely on environmental cues to determine when to transition into breeding condition (Wingfield, 2008). Hormonal responses to changing day length are fundamental to reproductive readiness in many vertebrates, particularly those living at high latitudes (Dawson, King, Bentley, & Ball,  2001; Gerlach & Aurich,  2000; Kriegsfeld, Ubuka, Bentley, & Tsutsui,  2015). In birds and mammals, patterns of melatonin release change with day length. In spring and summer breeders, long duration and high levels of melatonin secretion inhibit reproduction via suppression of the HPG axis that regulates production of the reproductive steroids (Falcon et al., 2009). Environmental conditions are not perfectly linked to cyclic cues, like photoperiod. If organisms depend on a single predictable cue to time transitions, they risk a mismatch between life history stage and environmental conditions (Wingfield, 2008). Hormonal flexibility allows individuals to fine-tune life history transitions in alignment with a fluctuating environment (Wingfield, 2015). The hypothalamic-­ pituitary-adrenal (HPA) axis, which leads to the production of glucocorticoids, regulates behavioral and physiological responses to challenges (e.g., low food availability, high predation risk; Sapolsky, Romero, & Munck,  2000). The HPA axis plays a critical role in regulating the onset of reproduction in arctic-breeding birds, which encounter unpredictable weather and food availability at their breeding grounds (Ramenofsky & Wingfield,  2017). If conditions are poor (e.g., inclement weather, low food availability), elevated glucocorticoids suppress reproductive behaviors. However, glucocorticoids’ effects may be buffered by other components or ­mediators of HPA axis activity, including the carrier protein corticosteroid-binding globulin (Breuner & Orchinik, 2002). By reducing the amount of free, circulating glucocorticoids, corticosteroid-binding globulin might limit the extent of reproductive ­suppression, allowing birds to initiate or continue Vitousek and Schoenle

17

breeding once conditions improve (Lynn, Breuner, & Wingfield, 2003; Ramenofsky & Wingfield, 2006; Wingfield et al., 2004). Differences among species in the extent to which  they base life history transitions on flexible (e.g., temperature, food availability) versus fixed (e.g., photoperiod) cues could have significant consequences for patterns of resource availability, as well as the ability of a given species to persist in changing environments. For example, larger mammals living in middle to high latitudes (e.g., caribou, marmots) often initiate breeding based on photoperiod alone, but depend on resources that fluctuate according to environmental conditions. Because of this mismatch, large mammals are predicted to suffer greater consequences from the changing climate than smaller ­rodents (e.g., voles, mice), which typically utilize other environmental cues, like food availability, to time reproduction (Bronson, 2009).

Hormones and Senescence

Organisms cannot escape death, but life history strategy influences their rates of aging, or senescence. Lifespan is inextricably linked to investment patterns early in life: Organisms that invest heavily in rapid growth and reproduction tend to senesce quickly and have short lifespans, whereas those that extend growth over a longer time and/or invest more in self-maintenance typically live longer. Across species, the timing of senescence, and thus lifespan, correlates with a suite of life history traits along the pace-of-life continuum, including growth rate, age at sexual maturity, and fecundity (Metcalfe & Monaghan, 2003; Stearns, 1992). Pleiotropic effects of hormones may integrate life history phenotypes along the pace-of-life continuum (Ketterson & Nolan, 1992; Ricklefs & Wikelski, 2002). Insulin-like growth factor-1 (IGF-1) mediates life history trade-offs among growth, fecundity, and survival by promoting tissue growth and development at a cost to longevity (Dantzer & Swanson, 2012; Stewart & Rotwein,  1996). The IGF-1-mediation of  trade-offs has been observed in mouse models (­reviewed in Bartke et al., 1998; Berryman, Sandahl Christiansen, Johannsson, Thorner, & Kopchick, 2008; Chistyakova,  2008); for example, Snell and Ames dwarf mice, which have lower levels of circulating IGF-1 than control mice, are characterized by small size, reduced fertility, and an over 45 percent increase in longevity (Bartke et al.,  1998). Similar patterns have been observed in domestic animals and humans. The lifespan of domestic dogs covaries with size and IGF-1 concentrations, with giant 18

Hormones and Behavior

breeds having shorter lifespans and higher concentrations of IGF-1 than small breeds. An observational study of small and giant breeds demonstrated that a single mutation in the IGF-1 gene of small breeds is likely the source of the phenotypic variation in size and longevity (Sutter et al.,  2007). Studies of humans have found that among older women, those with lower IGF-1 concentrations (Milman et al.,  2014) or reduced activity in the IGF-1 signaling pathway (van Heemst et al., 2005) live longer. Responding effectively to substantial challenges can be crucial for survival. However, the increase in glucocorticoids that supports adaptive responses to challenges can also accelerate senescence (Garrido, 2011; Haussmann & Marchetto,  2010; Sapolsky, Krey, & McEwen, 1986). The relationship between glucocorticoids and senescence may be mediated by changes in telomere length (mechanisms discussed in Haussmann & Marchetto, 2010). Telomeres are noncoding DNA sequences located at the ends of chromosomes, and telomere shortening eventually leads to cellular aging and death (Aubert & Lansdorp, 2008; Haussmann & Marchetto,  2010). Among birds and mammals, species whose telomeres shorten more quickly have shorter lifespans (Haussmann et  al., 2003). In some cases, telomere length can predict an individual’s likelihood of survival (e.g., in humans [e.g., Bakaysa et al.,  2007; Kimura et al., 2008] and birds [e.g., Haussmann, Winkler, & Vleck, 2005; Heidinger et al.,  2012; Pauliny, Wagner, Augustin, Szép, & Blomqvist, 2006]). For example, women that self-identify as highly stressed have significantly shorter telomeres than less stressed ­ women—in one study, the difference between groups was equivalent to 9 to 17 years of telomere attrition due to aging (Epel et al., 2004). Stressor exposure can also have transgenerational effects on telomere length. Parents can pass shortened telomeres to their offspring directly (via germ cells), or offspring telomere length and attrition rate can be influenced by exposure to parental glucocorticoids during development or stress-induced reductions in parental care (Haussmann & Heidinger, 2015). The multigenerational links among stress, telomere length, and lifespan could have important implications for humans. In the United States, stressor exposure (social, financial, and environmental) is correlated with low socioeconomic status, as well as with being a racial minority. Likely as a result, individuals in these groups face poorer health (Senn, Walsh, & Carey, 2014; D. R. Williams, 1999) and a lower life expectancy

(Chetty et al.,  2016; Meara, Richards, & Cutler, 2008). It is not yet clear whether these results are driven by within-individual and/or transgenerational effects of stressor exposure; however, experimental work in animal systems suggests that the physiological costs of socioeconomic status and race could be passed from parent to child (Baum et al., 1999; Cavigelli & Chaudhry, 2012).

an evolutionary perspective on the endocrine mediation of life histories

Because hormone systems mediate many traits ­important for survival and reproduction, selection operating on these traits could help to match the life history of a population to its environment. But to what extent are endocrine traits shaped by selection? For selection to operate on any trait, it must be variable, heritable, and linked with fitness. Individual variation in hormone concentrations and other ­endocrine traits is widespread (Kempenaers, Peters, & Foerster, 2008; T. D. Williams, 2008). As discussed earlier, hormones can have dose-dependent effects on life history traits in ways that could influence lifetime fitness. Less is known about the heritability of endocrine traits, but a growing number of studies are investigating the heritability of hormone concentrations. Artificial selection experiments (Evans, Roberts, Buchanan, & Goldsmith, 2006; Pottinger & Carrick, 1999), twin studies (Ring et al., 2005; Steptoe, Jaarsveld, Semmler, Plomin, & Wardle, 2009), and estimates in free-living populations (Jenkins, Vitousek, Hubbard, & Safran, 2014; King, Cline, & Hubbard, 2004) all indicate that steroid hormone levels have low to moderate heritability (Cox, McGlothlin & Bonier 2016). These findings are also consistent with large-scale phylogenetic comparisons that have found links between circulating hormone concentrations and life history strategies across species. For example, higher circulating testosterone concentrations are correlated with shorter breeding season length in amphibians (Eikenaar et al., 2012), reptiles (Eikenaar et al., 2012), and birds (Goymann et al., 2004; Hau, Ricklefs, Wikelski, Lee, & Brawn, 2010). Among birds, species with higher brood values (i.e., high value of the current reproductive effort relative to future opportunities) release lower concentrations of glucocorticoids in response to stressors (Bókony et al., 2009). In addition, both endocrine and life history traits can covary with environmental characteristics including net primary production and latitude (Jessop, Woodford, & Symonds, 2013; Scharf et al., 2015),

suggesting that environmental variation can generate selection on endocrine phenotypes. While the emerging picture suggests that endocrine variation can be shaped by selection, determining when and how selection operates on specific traits remains a substantial challenge. One of the reasons for this difficulty relates to the high level of phenotypic plasticity of many endocrine traits. In addition to developmental plasticity, most ­endocrine traits show reversible flexibility, changing across seasons and environments (Clinchy, Sheriff, & Zanette, 2013; Romero, 2004), in response to social interactions (Fuxjager & Marler, 2009), and intrinsic ­condition (e.g., body condition, immune activation; Cook, Connor, Mcconnachie, Gilmour, & Cooke, 2012; see Figure 2.1). Because of this, it can be difficult, particularly in natural populations, to determine whether individuals show consistent or heritable variation in endocrine trait expression in a given set of contexts. Nevertheless, a growing number of studies on free-living animals have found that circulating hormone levels are individually repeatable within years—and even across years ­ (­reviewed in Taff, Schoenle & Vitousek 2018). Genes

Developmental Environment

Current Environment

Endocrine Phenotype

Regulatory Cross-Talk

Hormone-Trait Relationships

Life History Trait Expression & Flexibility

Figure 2.1  The endocrine regulation of life history traits. Genes and the developmental and current environments influence life history trait expression via effects on endocrine phenotype (neuroendocrine architecture and activation), hormone–trait relationships, and regulatory cross-talk within and among physiological networks. The developmental environment stimulates semipermanent organi­zational changes in tissue structure and cellular transcription potential that contribute to phenotypic expression throughout an organism’s life. Current environmental conditions, which can shift rapidly, induce corresponding changes in physiology that are often but not always reversible. Within an organism, expression of an endocrine phenotype both influences and is affected by other regulatory networks and the relationships among hormones and traits. Together, this constellation of traits and interactions mediates the expression and flexibility of life histories.

Vitousek and Schoenle

19

As several recent conceptual papers have ­ ighlighted, individual differences in the capacity or h propensity to shift endocrine expression across contexts could be an important predictor of fitness (Bonier & Martin, 2016; Hau, Casagrande, Ouyang, & Baugh, 2016; Taff & Vitousek, 2016). However, most endocrine research examines circulating hormone levels in one or more discrete contexts but does not determine the presence or functional significance of variation in endocrine flexibility across contexts. Individual differences in both the level of endocrine trait expression and its flexibility can be assessed using a modified reaction norm approach (Figure 2.2), which has been widely used for other flexible traits (e.g., behaviors; Brommer, Kontiainen, & Pietiäinen, 2012; Dingemanse, Kazem, Réale, & Wright, 2010). In this approach an endocrine trait, like hormone concentration, is measured in the same individual multiple times across an environmental or social gradient, or across discrete contexts (Taff & Vitousek, 2016). This approach enables the separate quantification of the level of trait expression (or elevation of the reaction norm) and its flexibility (the slope of change over time; Figure 2.2). Research on behavioral flexibility shows that this distinction is important, as the elevation and slope of behavioral reaction norms are in some cases genetically distinct traits that can be differently ­ shaped by selection (Han & Brooks, 2014; Nussey, Wilson, & Brommer, 2007). Another benefit of the reaction norm approach is that it can be used to decouple differences in distinct components of ­ ­endocrine flexibility, including the scope (or magnitude) of a response and its speed (i.e., the rapidity (B)

(C)

(D)

Endocrine Expression

(A)

with which hormone systems respond to a change in context; Taff & Vitousek,  2016). Furthermore, this approach helps to control for some of the ­confounds and misinterpretations associated with ­correlating labile, condition-dependent traits with fitness (Bonier & Martin,  2016). Once a reaction norm approach has identified the elevation and flexibility of an endocrine response, these metrics can be evaluated for variation among individuals, heritability, and correlations with fitness. Several recent studies have used a reaction norm approach to quantify individual variation in endocrine flexibility across environments or challenges (e.g., Furtbauer, Pond, Heistermann, & King, 2015; Lendvai et al., 2014). Variation in the scope of flexibility of endocrine responses is also suggested by studies that have used pharmacological compounds to measure individual differences in the potential to mount an endocrine response (e.g., GnRH [Jawor et al., 2006, Needham, Dochtermann, & Grieves, 2017], adrenocorticotropic hormone [ACTH; Sheriff, Dantzer, Delehanty, Palme, & Boonstra, 2011]). Even less is known about variation in the speed of flexibility, but some evidence suggests that individuals differ in the speed of the glucocorticoid and catecholamine stress responses (Baugh, Oers, Naguib, & Hau, 2013; Koolhaas, Boer, Buwalda, & Reenen, 2007). To date, no studies have addressed the heritability of endocrine traits across environmental or social gradients, but ongoing research in tree swallows indicates that the scope of the glucocorticoid response to restraint stress is heritable and predicts both survival and reproductive success (Stedman et al., 2017; Vitousek et al., 2018). Future research

Environmental Gradient or Change in Context Figure 2.2  Endocrine reaction norms. Each panel illustrates two hypothetical individuals (or genotypes) whose endocrine trait expression (e.g., circulating hormone levels, receptor expression) is measured across a gradient of environmental conditions, or changes in context. The mean level of trait expression across contexts is considered to be the elevation of the reaction norm. Endocrine flexibility is defined as the slope of change over the gradient. (A) Individuals differ in endocrine expression but show no flexibility across contexts. (B) Individuals differ in endocrine expression but show the same amount of flexibility across contexts. (C) Individuals show similar endocrine expression in one context but differ in both mean trait expression and endocrine flexibility across contexts. (D) Individuals have similar mean trait expression but differ in flexibility. (Adapted from Taff, C. C., & Vitousek, M. N. [2016]. Endocrine flexibility: Optimizing phenotypes in a dynamic world? Trends in Ecology & Evolution, 6, 476–488.)

20

Hormones and Behavior

that addresses the reaction norms of endocrine ­expression will be important for understanding how hormones mediate life h ­ istories, and how distinct components of these reaction norms are shaped by selection. Although circulating hormones are the easiest traits to measure repeatedly, they are only one of many traits that influence the ultimate effect of endocrine activation. Thus, it is also vital to begin to address the potential for selection to shape the expression and flexibility of other endocrine traits, including receptor and carrier protein expression. Understanding how selection shapes endocrine systems will also require a more explicit focus on context-dependence in hormone-fitness relationships (Schoenle, Zimmer & Vitousek, 2018). While many conceptual models of endocrine regulation incorporate life history trade-offs (e.g., Ketterson & Nolan, 1992; Wingfield & Sapolsky, 2003; Zera & Harshman, 2001), few quantitative analyses of hormone-fitness relationships have explicitly incorporated life history. Incorporating such context-dependence may help to clarify why hormone-­fitness relationships often differ within and across populations.

Conclusion and Future Directions

From development through senescence, endocrine traits organize and activate many of the behavioral and physiological traits central to life history. Hormonal regulation of life history is taxonomically widespread; although we have focused here on vertebrates and insects, hormones also mediate life history traits in a diversity of taxa, from echinoderms to plants (e.g., Finch & Rose, 1995; Heyland & Moroz, 2006). Both laboratory and field-based research has provided fascinating insights into the mechanisms by which environmental and social stimuli are transduced into endocrine signals, and the specific processes by which these signals regulate transitions among life history stages and mediate investment in different stages. Yet despite this progress, much remains to be determined about how hormone systems evolve, how they are shaped by the environments and experiences of an individual, and how they interact with other regulatory systems to influence life histories. One of the most fundamental gaps in our ­understanding of life histories concerns how and why individuals within a population differ in their life history strategy. Within-population variations in life history strategies are widespread, yet we still lack a clear understanding of the mechanistic basis of this variation for many (T. D. Williams, 2008, 2012). Likewise, individuals can differ in life history

flexibility—the extent to which their life history strategy changes over time. This flexibility could have important fitness consequences: When resources or social environments change, individuals that can alter the timing of life history stages, or their investment in a given stage, could have a significant advantage (Wingfield, 2015). Consistent betweenindividual differences in plasticity have been found in some life history traits. For example, like many birds, the great tit adjusts the timing of breeding each year based on spring temperatures (Nussey, Postma, Gienapp, & Visser, 2005). Birds that breed earlier in the year have access to more insect prey but also face the risk that spring storms or cold fronts will destroy their nests. Long-term studies in great tits have found that some individuals show consistently greater plasticity in the onset of breeding than others, advancing breeding more in warmer springs and delaying it in cooler years. At least in some populations, this flexibility is a heritable trait that can be shaped by selection (Husby, Visser, & Kruuk, 2011; Nussey et al., 2005). Thus, as springs become warmer and more variable, selection could operate not only on the mean timing of breeding but also on the flexibility of this important life history trait. Despite the potential importance of life history plasticity, relatively few studies have explored whether life history traits show consistent, heritable plasticity (Brommer, 2013). Even fewer have addressed the mechanistic basis of differences in plasticity (Taff & Vitousek, 2016). Determining the extent to which variation in life history flexibility reflects ­underlying differences in the endocrine ­mediators of life history or in other mechanisms—including sensory perception or integration—is an important future direction. Another major frontier in endocrine research lies in dissecting the intricate coregulatory networks that connect neuroendocrine systems with one another and with other physiological elements (e.g., immune system, metabolism; Martin & Cohen, 2015). Because of the extraordinary complexity of these systems, they have been difficult to tackle with ­traditional methods; the vast majority of endocrine research to date has focused on the role of one or a small number of hormone systems in mediating a given trait. Yet the e­ xistence of these vast regulatory networks suggests that the activation of one hormonal cascade can create ripple effects that influence a diversity of physiological systems. Physiological ­regulatory networks may be particularly important mediators of the suites of integrated traits whose expression levels must be coordinated—like the ­ suites of traits that make up a life history strategy. Vitousek and Schoenle

21

Understanding the dynamics of these physiological regulatory networks may therefore provide important insights into the regulation of life histories. Because of the sometimes staggering number of moving parts involved in physiological regulatory networks, their measurement and analysis continue to present major challenges; however, advances in statistical and analytical approaches have made a network-based understanding of physiological regulation a realistic goal (Martin & Cohen, 2015). Finally, a major goal for future research is to ­integrate advances from both laboratory and fieldbased animal research to gain insight into how the endocrine mediation of life history impacts human health. Research on mammalian model systems is widely recognized to be relevant for human health, as it ­enables the careful dissection of mechanisms, and the experimental determination of their links with ­ specific behaviors and physiological health outcomes, in ways that are not possible to test in humans. However, studies in more genetically diverse populations, particularly those that involve free-living organisms in their natural environments, provide much greater insight into the causes and consequences of variation in trait expression. It is precisely this variation that is likely to be an important causal factor in many human health problems, from stressor-induced reproductive suppression to the effect of socioeconomic status on longevity. Future research that integrates model and natural systems approaches, drawing on the unique strengths of each, could shed light on major challenges in human health.

References

Adkins-Regan, E. (2005). Hormones and social behavior. Princeton, NJ: Princeton University Press. Andrewartha, S. J., & Burggren, W. W. (2012). Transgenerational variation in metabolism and life-history traits induced by maternal hypoxia in Daphnia magna. Physiological and Biochemical Zoology, 85, 625–634. Ardia, D.  R., Schat, K.  A., & Winkler, D.  W. (2003). Repro­ ductive effort reduces long-term immune function in breeding tree swallows (Tachycineta bicolor). Proceedings of the Royal Society B, 270, 1679–1683. Arnold, A. P. (2009). The organizational—activational hypothesis as the foundation for a unified theory of sexual differentiation of all mammalian tissues. Hormones and Behavior, 55, 570–578. Aubert, G., & Lansdorp, P.  M. (2008). Telomeres and aging. Physiological Reviews, 88, 557–579. Bakaysa, S. L., Mucci, L. A., Slagboom, P. E., Boomsma, D. I., McClearn, G. E., Johansson, B., & Pedersen, N. L. (2007). Telomere length predicts survival independent of genetic influences. Aging Cell, 6, 769–774. Bartke, A., Brown-Borg, H.  M., Bode, A.  M., Carlson, J., Hunter, W. S., & Bronson, R. T. (1998). Does growth hormone

22

Hormones and Behavior

prevent or accelerate aging? Experimental Gerontology, 33, 675–687. Bass, A. H. (2008). Steroid-dependent plasticity of vocal motor systems: Novel insights from teleost fish. Brain Research Reviews, 57, 299–308. Baugh, A.  T., van Oers, K., Naguib, M., & Hau, M. (2013). Initial reactivity and magnitude of the acute stress response associated with personality in wild great tits (Parus major). General and Comparative Endocrinology, 189, 96–104. Baum, A., Garofalo, J. P., & Yali, A. M. (1999). Socioeconomic status and chronic stress: Does stress account for SES effects on health? Annals of the New York Academy of Sciences, 896, 131–144. Bell, B. A., Phan, M. L, Meillere, A., Evans, J. K., Leitner, S., Vicario, D. S., & Buchanan, K. L. (2018). Influence of earlylife nutritional stress on songbird memory formation. Proceedings of the Royal Society of London B, 285, 20181270. Berryman, D.  E., Sandahl Christiansen, J., Johannsson, G., Thorner, M. O., & Kopchick, J. J. (2008). Role of the GH/ IGF-1 axis in lifespan and healthspan: Lessons from animal models. Growth Hormone & IGF Research, 18, 455–471. Berven, K. A. (1990). Factors affecting population fluctuations in larval and adult stages of the wood frog (Rana sylvatica). Ecology, 71, 1599–1608. Bishop, C.  D., Erezyilmaz, D.  F., Flatt, T., Georgiou, C.  D., Hadfield, M.  G., Heyland, A., . . . Youson, J.  H. (2006). What is metamorphosis? Integrative and Comparative Biology, 46, 655–661. Bókony, V., Lendvai, A.  Z., Liker, A., Angelier, F., Wingfield, J. C., & Chastel, O. (2009). Stress response and the value of reproduction: Are birds prudent parents? American Naturalist, 173, 589–598. Bonier, F., & Martin, P. R. (2016). How can we estimate natural selection on endocrine traits? Lessons from evolutionary biology. Proceedings of the Royal Society B, 283, 20161887. Bonneaud, C., Mazuc, J., Gonzalez, G., Haussy, C., Chastel, O., Faivre, B., & Sorci, G. (2003). Assessing the cost of mounting an immune response. American Naturalist, 161, 367–379. Boogert, N. J., Zimmer, C., & Spencer, K. A. (2013). Pre- and post-natal stress have opposing effects on social information use. Biology Letters, 9, 20121088. Breuner, C. W., & Orchinik, M. (2002). Plasma binding proteins as mediators of corticosteroid action in vertebrates. Journal of Endocrinology, 175, 99–112. Brommer, J. E. (2013). Phenotypic plasticity of labile traits in the wild. Current Zoology, 59, 485–505. Brommer, J. E., Kontiainen, P., & Pietiäinen, H. (2012). Selection on plasticity of seasonal life-history traits using random regression mixed model analysis. Ecology and Evolution, 2, 695–704. Bronson, F. H. (2009). Climate change and seasonal reproduction in mammals. Proceedings of the Royal Society B, 364, 3331– 3340. Cardillo, M. (2002). The life-history basis of latitudinal diversity gradients: How do species traits vary from the poles to the equator? Journal of Animal Ecology, 71, 79–87. Cavigelli, S.  A., & Chaudhry, H.  S. (2012). Social status, glucocorticoids, immune function, and health: Can animal studies help us understand human socioeconomic-statusrelated health disparities? Hormones and Behavior, 62, 295–313. Chetty, R., Stepner, M., Abraham, S., Lin, S., Scuderi, B., Turner, N., Bergeron, A., & Cutler, D. (2016). The association

between income and life expectancy in the United States. Journal of the American Medical Association, 315, 1750–1766. Chistyakova, O.  V. (2008). Signaling pathway of insulin and insulin-like growth factor 1 (IGF-1) as a potential regulator of lifespan. Journal of Evolutionary Biochemistry and Physiology, 44, 1–11. Clark, M., & Galef, B.  G. (1995). Prenatal influences on reproductive life history strategies. Trends in Ecology and Evolution, 10, 151–153. Clark, M. M., & Galef, B. G. (1994). A male gerbil’s intrauterine position affects female response to his scent marks. Physiology and Behavior, 55, 1139–1141. Clinchy, M., Sheriff, M. J., & Zanette, L. Y. (2013). Predatorinduced stress and the ecology of fear. Functional Ecology, 27, 56–65. Cook, K. V, Connor, C. M. O., Mcconnachie, S. H., Gilmour, K. M., & Cooke, S. J. (2012). Condition dependent intraindividual repeatability of stress-induced cortisol in a freshwater fish. Comparative Biochemistry and Physiology, Part A, 161, 337–343. Cox, R. M., McGlothlin, J. W., and Bonier, F. (2016) Hormones as mediators of phenotypic and genetic integration: an evolutionary genetics perspective. Integrative and Comparative Biology, 56, 126–137. Dallman, M. F. (2009). Stress-induced obesity and the emotional nervous system. Trends in Endocrinology & Metabolism, 21, 159–165. Dallman, M. F., Fleur, S. E. L. A., Pecoraro, N. C., Gomez, F., Houshyar, H., & Akana, S.  F. (2004). Minireview: Glucocorticoids-food intake, abdominal obesity, and wealthy nations in 2004. Endocrinology, 145, 2633–2638. Dantzer, B., & Swanson, E. M. (2012). Mediation of vertebrate life histories via insulin-like growth factor-1. Biological Reviews, 87, 414–429. Davis, A.  G., & Leary, C.  J. (2015). Elevated stress hormone diminishes the strength of female preferences for acoustic signals in the green treefrog. Hormones and Behavior, 69, 119–122. Dawson, A., King, V. M., Bentley, G. E., & Ball, G. F. (2001). Photoperiodic control of seasonality in birds. Journal of Biological Rhythms, 16, 365–380. Denver, R.  J. (2009). Stress hormones mediate environmentgenotype interactions during amphibian development. General and Comparative Endocrinology, 164, 20–31. Denver, R.  J. (2013). Neuroendocrinology of amphibian meta­ morphosis. Animal metamorphosis (1st ed., Vol. 103). Elsevier. Denver, R.  J., & Middlemis-Maher, J. (2010). Lessons from evolution: Developmental plasticity in vertebrates with complex life cycles. Journal of Developmental Origins of Health and Disease, 1, 282–291. Di Cara, F., & King-Jones, K. (2013). How clocks and hormones act in concert to control the timing of insect development. Current Topics in Developmental Biology, 105, 1–36. Dingemanse, N.  J., Kazem, A., Réale, D.  J.  F., & Wright, J. (2010). Behavioural reaction norms: Animal personality meets individual plasticity. Trends in Ecology and Evolution, 25, 81–89. Eikenaar, C., Husak, J., Escallón, C., Moore, I. T., Eikenaar, C., Husak, J., Escallo, C. & . . . Moore, I. T. (2012). Variation in testosterone and corticosterone in amphibians and reptiles: Relationships with latitude, elevation, and breeding season length. American Naturalist, 180, 642–654. Epel, E.  S., Blackburn, E.  H., Lin, J., Dhabhar, F.  S., Adler, N. E., Morrow, J. D., & Cawthon, R. M. (2004). Accelerated

telomere shortening in response to life stress. Proceedings of the National Academy of Sciences of the United States of America, 101, 17312–17315. Evans, M. R., Goldsmith, A. R., & Norris, S. R. A. (2000). The effects of testosterone on antibody production and plumage coloration in male house sparrows (Passer domesticus). Behavioral Ecology and Sociobiology, 47, 156–163. Evans, M. R., Roberts, M. L., Buchanan, K. L., & Goldsmith, A.  R. (2006). Heritability of corticosterone response and changes in life history traits during selection in the zebra finch. Journal of Evolutionary Biology, 19, 343–352. Falcon, J., Besseau, L., Fuentes, M., Sauzet, S., Magnanou, E., & Boeuf, G. (2009). Structural and functional evolution of the pineal melatonin system in vertebrates. Annals of the New York Academy of Sciences, 1163, 101–111. Farrell, T.  M., Neuert, M.  A.  C., Cui, A., & MacdougallShackleton, S. A. (2015). Developmental stress impairs a female songbird’s behavioural and neural response to a sexually selected signal. Animal Behaviour, 102, 157–167. Finch, C.  E., & Rose, M.  R. (1995). Hormones and the physiological architecture of life history evolution. Quarterly Review of Biology, 70, 1–52. Flatt, T., Moroz, L.  L., Tatar, M., & Heyland, A. (2006). Comparing thyroid and insect hormone signaling. Integrative and Comparative Biology, 46, 777–794. Foo, Y. Z., Nakagawa, S., Rhodes, G., & Simmons, L. W. (2016). The effects of sex hormones on immune function: a metaanalysis. Biological Reviews, 92, 551–571. Francis, D., Diorio, J., Liu, D., & Meaney, M.  J. (1999). Nongenomic transmission across generations of maternal behavior and stress responses in the rat. Science, 286, 1155 Fronstin, R.  B., Christians, J.  K., & Williams, T.  D. (2015). Experimental reduction of hematocrit affects reproductive performance in European starlings. Functional Ecology, 30, 398–409. Furtbauer, I., Pond, A., Heistermann, M., & King, A. J. (2015). Personality, plasticity and predation: Linking endocrine and behavioural reaction norms in stickleback fish. Functional Ecology, 29, 931–940. Fuxjager, M.  J., & Marler, C.  A. (2009). How and why the winner effect forms: Influences of contest environment and species differences. Behavioral Ecology, 21, 37–45. Galton, V. A. (1992). The role of thyroid hormone in amphibian metamorphosis. Trends in Endocrinology & Metabolism, 3, 96–100. Garrido, P. (2011). Aging and stress: Past hypotheses, present approaches and perspectives. Aging and Disease, 2, 80–99. Gerlach, T., & Aurich, J.  E. (2000). Regulation of seasonal reproductive activity in the stallion, ram and hamster. Animal Reproduction Science, 58, 197–213. Gimpl, G., & Fahrenholz, F. (2001). The oxytocin receptor system: Structure, function, and regulation. Physiological Reviews, 81, 629–683. Goodson, J.  L., & Bass, A.  H. (2000). Forebrain peptides modulate sexually polymorphic vocal circuitry. Nature, 403, 769–772. Gosling, L. M., & Roberts, S. C. (2001). Scent-marking by male mammals: Cheat-proof signals to competitors and mates. Advances in the Study of Behavior, 30, 169–217. Goymann, W., Moore, I. T., Scheuerlein, A., Hirschenhasuer, K., Grafen, A., & Wingfield, J.  C. (2004). Testosterone in tropical birds: Effects of environmental and social factors. American Naturalist, 164, 327–334.

Vitousek and Schoenle

23

Hahn, T. P., & Macdougall-Shackleton, S. A. (2008). Adaptive specialization, conditional plasticity and phylogenetic history in the reproductive cue response systems of birds. Proceedings of the Royal Society B, 363, 267–286. Han, C. S., & Brooks, R. C. (2014). Long-term effect of social interactions on behavioral plasticity and lifetime mating success. American Naturalist, 183, 431–444. Hau, M., Casagrande, S., Ouyang, J. Q., & Baugh, A. T. (2016). Glucocorticoid-mediated phenotypes in vertebrates: Multi­ level variation and evolution. Advances in the Study of Behavior, 48, 41–115. Hau, M., Ricklefs, R. E., Wikelski, M., Lee, K. A., & Brawn, J.  D. (2010). Corticosterone, testosterone and life-history strategies of birds. Proceedings of the Royal Society B, 277, 3203–3212. Haussmann, M. F., & Heidinger, B. J. (2015). Telomere dynamics may link stress exposure and ageing across generations. Biology Letters, 11, 20150396. Haussmann, M.  F., & Marchetto, N.  M. (2010). Telomeres: Linking stress and survival, ecology and evolution. Current Zoology, 56, 714–728. Haussmann, M.  F., Winkler, D.  W., Reilly, K.  M.  O., Huntington, C. E., Nisbet, I. C. T., & Vleck, C. M. (2003). Telomeres shorten more slowly in long-lived birds and mammals than in short-lived ones. Proceedings of the Royal Society B, 270, 1387–1392. Haussmann, M.  F., Winkler, D.  W., & Vleck, C.  M. (2005). Longer telomeres associated with higher survival in birds. Biology Letters, 1, 212–214. Heidinger, B. J., Blount, J. D., Boner, W., Griffiths, K., Metcalfe, N. B., & Monaghan, P. (2012). Telomere length in early life predicts lifespan. Proceedings of the National Academy of Sciences of the United States of America, 109, 1743–1748. Hellstrom, I.  C., Dhir, S.  K., Diorio, J.  C., & Meaney, M.  J. (2012). Maternal licking regulates hippocampal glucocor­ ticoid receptor transcription through a thyroid hormone— serotonin—NGFI-A signalling cascade. Philosophical Tran­ sactions of the Royal Society B, 367, 2495–2510. Heyland, A., & Moroz, L.  L. (2006). Signaling mechanisms underlying metamorphic transitions in animals. Integrative and Comparative Biology, 46, 743–759. Huang, H., & Brown, D. D. (2000). Prolactin is not a juvenile hormone in Xenopus laevis metamorphosis. Proceedings of the National Academy of Sciences of the United States of America, 97, 195–199. Husby, A., Visser, M.  E., & Kruuk, L.  E.  B. (2011). Speeding up  microevolution: The effects of increasing temperature selection and genetic variance in a wild bird population. PLoS Biology, 9, e1000585. Jawor, J.  M., Mcglothlin, J.  W., Casto, J.  M., Greives, T.  J., Snajdr, E.  A., Bentley, G.  E., & Ketterson, E.  D. (2006). Seasonal and individual variation in response to GnRH challenge in male dark-eyed juncos (Junco hyemali). General and Comparative Endocrinology, 149, 182–189. Jenkins, B. R., Vitousek, M. N., Hubbard, J. K., & Safran, R. J. (2014). An experimental analysis of the heritability of variation in glucocorticoid concentrations in a wild avian population. Proceedings of the Royal Society B: Biological Sciences, 281, 20141302. Jessop, T.  S., Woodford, R., & Symonds, M.  R.  E. (2013). Macrostress: Do large-scale ecological patterns exist in the glucocorticoid stress response of vertebrates? Functional Ecology, 27, 120–130.

24

Hormones and Behavior

Jetz, W., Sekercioglu, C. H., & Bohning-Gaese, K. (2008). The worldwide variation in avian clutch size across species and space. PLoS Biology, 6, e303. Kavaliers, M., & Ossenkopp, K.-P. (2001). Corticosterone rapidly reduces male odor preferences in female mice. NeuroReport, 12, 2999–3002. Kempenaers, B., Peters, A., & Foerster, K. (2008). Sources of individual variation in plasma testosterone levels. Proceedings of the Royal Society B, 363, 1711–1723. Ketterson, E.  D., & Nolan Jr., V. (1992). Hormones and life histories: An integrative approach. American Naturalist, 140, S33–S62. Kikuyama, S., Kawamura, K., Tanaka, S., & Yamamoto, K. (1993). Aspects of amphibian metamorphosis: Hormonal control. International Review of Cytology, 143, 105–148. Kimura, M., Hjelmborg, J.  V.  B., Gardner, J.  P., Bathum, L., Brimacombe, M., Lu, X., Christiansen, L., Vaupel, J.  W., Aviv, A., & . . . Christensen, K. (2008). Telomere length and mortality: A study of leukocytes in elderly Danish twins. American Journal of Epidemiology, 167, 799–806. King, R. B., Cline, J. H., & Hubbard, C. J. (2004). Heritable variation in testosterone levels in male garter snakes (Thamnophis sirtalis). Journal of Zoology, 264, 143–147. Knowles, S.  C.  L., Nakagawa, S., & Sheldon, B.  C. (2009). Elevated reproductive effort increases blood parasitaemia and decreases immune function in birds: A meta-regression approach. Functional Ecology, 23, 405–415. Koolhaas, J. M., De Boer, S. F., Buwalda, B., & Van Reenen, K. (2007). Individual variation in coping with stress: A multidimensional approach of ultimate and proximate mechanisms. Brain, Behavior and Evolution, 70, 218–226. Kriegsfeld, L. J., Ubuka, T., Bentley, G. E., & Tsutsui, K. (2015). Seasonal control of gonadotropin-inhibitory hormone (GnIH) in birds and mammals. Frontiers in Neuroendocrinology, 37, 65–75. la Fleur, S. E. (2006). The effects of glucocorticoids on feeding behavior in rats. Physiology & Behavior, 89, 110–114. Landys, M. M., Ramenofsky, M., & Wingfield, J. C. (2006). Actions of glucocorticoids at a seasonal baseline as compared to stress-related levels in the regulation of periodic life processes. General and Comparative Endocrinology, 148, 132–149. Lendvai, Á. Z., Ouyang, J.  Q., Schoenle, L.  A., Fasanello, V., Haussmann, M.  F., Bonier, F., & Moore, I.  T. (2014). Experimental food restriction reveals individual differences in corticosterone reaction norms with no oxidative costs. PLoS One, 9, e110564. Lynch, K. S., Crews, D., Ryan, M. J., & Wilczynski, W. (2006). Hormonal state influences aspects of female mate choice in the Túngara frog (Physalaemus pustulosus). Hormones and Behavior, 49, 450–457. Lynn, S.  E. (2016). Endocrine and neuroendocrine regulation of fathering behavior in birds. Hormones and Behavior, 77, 237–248. Lynn, S. E., Breuner, C. W., & Wingfield, J. C. (2003). Shortterm fasting affects locomotor activity, corticosterone, and corticosterone binding globulin in a migratory songbird. Hormones and Behavior, 43, 150–157. Marasco, V., Herzyk, P., Robinson, J., & Spencer, K. A. (2016). Pre- and post-natal stress programming: Developmental exposure to glucocorticoids causes long-term brain-region specific changes to transcriptome in the precocial Japanese quail. Journal of Neuroendocrinology, 28(215), 3955–3964.

Martin, L. B., & Cohen, A. A. (2015). Physiological regulatory networks: The orchestra of life? In L.  B.  Martin, C.  K. Ghalambor, & H. A. Woods (Eds.), Integrative organismal biology (pp. 137–152). Hoboken, NJ: John Wiley & Sons. Meara, E. R., Richards, S., & Cutler, D. M. (2008). The gap gets bigger: Changes in mortality and life expectancy, by education, 1981–2000. Health Affairs, 27, 350–360. Metcalfe, N.  B., & Monaghan, P. (2003). Growth versus life­ span: Perspectives from evolutionary ecology. Experimental Gerontology, 38, 935–940. Milman, S., Atzmon, G., Huffman, D. M., Wan, J., Crandall, J.  P., Cohen, P., & Barzilai, N. (2014). Low insulin-like growth factor-1 predicts survival in humans with exceptional longevity. Aging Cell, 13, 769–771. Mitra, A. (2013). Cinderella’s new shoes—How and why insects remodel their bodies between life stages. Current Science, 104, 1028–1036. Mougeot, F., Dawson, A., Redpath, S. M., & Leckie, F. (2005). Testosterone and autumn territorial behavior in male red grouse Lagopus lagopus scoticus. Hormones and Behavior, 47, 576–584. Needham, K. B., Dochtermann, N. A., and Grieves, T. J. (2017). Consistent individual variation in day, night, and GnRHinduced testosterone concentrations in house sparrows (Passer domesticus). General and Comparative Endocrinology, 246, 211–217. Nowicki, S., Peters, S., & Podos, J. (1998). Song learning, early nutrition and sexual selection in songbirds. American Zoologist, 38, 179–190. Nussey, D. H., Postma, E., Gienapp, P., & Visser, M. E. (2005). Selection on heritable phenotypic plasticity in a wild bird population. Science, 310, 304–306. Nussey, D. H., Wilson, A. J., & Brommer, J. E. (2007). The evolutionary ecology of individual phenotypic plasticity in  wild populations. Journal of Evolutionary Biology, 20, 831–844. Nyman, C., Fischer, S., Aubin-Horth, N., & Taborsky, B. Evolutionarily conserved neural signature of early life stress affects animal social competence. Proceedings of the Royal Society B, 285, 20172344. Okuyama, T., Yokoi, S., Abe, H., Isoe, Y., Suehiro, Y., Imada, H., . . . Takeuchi, H. (2014). A neural mechanism underlying mating preferences for familiar individuals in medaka fish. Science, 343, 91–94. Painter, R. C., Osmond, C., Gluckman, P., Hanson, M., Phillips, D.  I.  W., & Roseboom, T.  J. (2008). Transgenerational effects of prenatal exposure to the Dutch famine on neonatal adiposity and health in later life. International Journal of Obstetrics and Gynaecology, 115, 1243–1249. Pauliny, A., Wagner, R. H., Augustin, J., Szép, T., & Blomqvist, D. (2006). Age-independent telomere length predicts fitness in two bird species. Molecular Ecology, 15(6), 1681–1687. http://doi.org/10.1111/j.1365-294X.2006.02862.x Perrins, C. M. (1970). The timing of birds’ breeding seasons. Ibis, 112(2), 242–255. Pottinger, T. G., & Carrick, T. R. (1999). Modification of the plasma cortisol response to stress in rainbow trout by selective breeding. General and Comparative Endocrinology, 132, 122–132. Ramenofsky, M., & Wingfield, J.  C. (2006). Behavioral and physiological conflicts in migrants: The transition between migration and breeding. Journal of Ornithology, 147, 135–145.

Ramenofsky, M., & Wingfield, J.  C. (2017). Regulation of complex behavioural transitions: Migration to breeding. Animal Behaviour, 124, 299–306. Ricklefs, R.  E. (1977). On the evolution of reproductive strategies in birds: Reproductive effort. American Naturalist, 111, 453–478. Ricklefs, R.  E., & Wikelski, M. (2002). The physiology/lifehistory nexus. Trends in Ecology & Evolution, 17, 462–468. Ring, H. Z., Lessov, C. N., Reed, T., Marcus, R., Holloway, L., & Swan, G. E. (2005). Heritability of plasma sex hormones and hormone binding globulin in adult male twins. Journal of Clinical Endocrinology and Metabolism, 90, 3653–3658. Roche, D. P., McGhee, K. E., & Bell, A. M. (2012). Maternal predator-exposure has lifelong consequences for offspring learning in threespined sticklebacks. Biology Letters, 8, 932–935. Rodgers, A. B., Morgan, C. P., Leu, N. A., & Bale, T. L. (2015). Transgenerational epigenetic programming via sperm microRNA recapitulates effects of paternal stress. Proceedings of the National Academy of Sciences of the United States of America, 112, 13699–13704. Roff, D.  A. (2002). Life history evolution. Sunderland, MA: Sinauer Associates. Romero, L. M. (2004). Physiological stress in ecology: Lessons from biomedical research. Trends in Ecology & Evolution, 19, 249–255. Sakuma, Y., & Pfaff, D. W. (1980). LH-RH in the mesencephalic central grey can potentiate lordosis reflex of female rats. Nature, 283, 566–567. Sapolsky, R.  M., Krey, L.  C., & McEwen, B.  S. (1986). The neuroendocrinology of stress and aging: The glucocorticoid cascade hypothesis. Endocrine Reviews, 7, 284–301. Sapolsky, R. M., Romero, L. M., & Munck, A. U. (2000). How do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocrine Reviews, 21, 55–89. Scharf, I., Feldman, A., Novosolov, M., Pincheira-donoso, D., Das, I., Böhm, M., . . . Meiri, S. (2015). Late bloomers and baby boomers: Ecological drivers of longevity in squamates and the tuatara. Global Ecology and Biogeography, 24, 396–405. Schoenle, L.  A., Zimmer, C., & Vitousek, M.  N. (2018). Understanding context dependence in glucocorticoid-fitness relationships: the role of the nature of the challenge, the intensity and frequency of stressors, and life history. Integrative and Comparative Biology, 58, 777–789. Schulz, K. M., & Sisk, C. L. (2016). The organizing actions of adolescent gonadal steroid hormones on brain and behavioral development. Neuroscience and Biobehavioral Reviews, 70, 148–158. Schulz, L.  C. (2010). The Dutch Hunger Winter and the developmental origins of health and disease. Proceedings of the National Academy of Sciences of the United States of America, 107(39), 16757–16758. Semlitsch, R.  D., Scott, D.  E., & Pechmann, J.  H.  K. (1988). Time and size at metamorphosis related to adult fitness in Ambystoma talpoideum. Ecology, 69(1), 184–192. Senn, T. E., Walsh, J. L., & Carey, M. P. (2014). The mediating roles of perceived stress and health behaviors in the relation between objective, subjective, and neighborhood socio­ economic status and perceived health. Annals of Behavioral Medicine, 48, 215–224. Sheriff, M. J., Dantzer, B., Delehanty, B., Palme, R., & Boonstra, R. (2011). Measuring stress in wildlife: Techniques for quantifying glucocorticoids. Oecologia, 166, 869–887.

Vitousek and Schoenle

25

Sisk, C. L., & Zehr, J. L. (2005). Pubertal hormones organize the adolescent brain and behavior. Frontiers in Neuroendocrinology, 26, 163–174. Spencer, K. A., Buchanan, K. L., Goldsmith, A. R., & Catchpole, C.  K. (2003). Song as an honest signal of developmental stress in the zebra finch (Taeniopygia guttata). Hormones and Behavior, 44, 132–139. Spencer, K. A., Wimpenny, J. H., Buchanan, K. L., Lovell, P. G., Goldsmith, A. R., & Catchpole, C. K. (2005). Developmental stress affects the attractiveness of male song and female choice in the zebra finch (Taeniopygia guttata). Behavioral Ecology and Sociobiology, 58, 423–428. Stearns, S. C. (1992). The evolution of life histories. Oxford, UK: Oxford University Press. Stedman, J.  M., Hallinger, K.  K., Winkler, D.  W., & Vitousek  M.  N. 2017. Heritable variation in circulating glucocorticoids and endocrine flexibility in a free-living songbird. Journal of Evolutionary Biology, 30(9): 1724–1735. Steptoe, A., van Jaarsveld, C. H. M., Semmler, C., Plomin, R., & Wardle, J. (2009). Heritability of daytime cortisol levels and cortisol reactivity in children. Psychoneuroendocrinology, 34, 273–280. Stewart, C. E. H., & Rotwein, P. (1996). Growth, differentiation, and survival: Multiple physiological functions for insulinlike growth factors. Physiological Reviews, 76, 1005–1026. Sutter, N.  B., Bustamante, C.  D., Chase, K., Gray, M.  M., Zhao, K., Zhu, L., . . . Ostrander, E.  A. (2007). A single IGF1 allele is a major determinant of small size in dogs. Science, 316, 112–115. Taff, C.  C., & Vitousek, M.  N. (2016). Endocrine flexibility: Optimizing phenotypes in a dynamic world? Trends in Ecology & Evolution, 6, 476–488. Taff, C.  C., Schoenle, L.  A., & Vitousek, M.  N. (2018). The repeatability of glucocorticoids: a review and meta-analysis. General and Comparative Endocrinology, 260, 136–145. Truman, J. W., & Riddiford, L. M. (2002). Endocrine insights into the evolution of metamorphosis in insects. Annual Review of Entomology, 47, 467–500. van Hasselt, F. N., Cornelisse, S., Zhang, T. Y., Meaney, M. J., Velzing, E.  H., Krugers, H.  J., & Joels, M. (2012). Adult hippocampal glucocorticoid receptor expression and dentate synaptic plasticity correlate with maternal care received by individuals early in life. Hippocampus, 22, 255–266. van Heemst, D., Beekman, M., Mooijaart, S. P., Heijmans, B. T., Brandt, B.  W., Zwaan, B.  J., & Slagboom, P.  E. (2005). Reduced insulin/IGF-1 signalling and human longevity. Aging Cell, 4, 79–85. van Noordwijk, A.  A.  J., Mccleery, R.  H., & Perrins, C.  M. (1995). Selection for the timing of great tit breeding in relation to caterpillar growth and temperature. Journal of Animal Ecology, 64, 451–458. Verhulst, S., & Nilsson, J. Å. (2008). The timing of birds’ breeding seasons: A review of experiments that manipulated timing of breeding. Proceedings of the Royal Society B, 363, 399–410. Vitousek, M. N., Mitchell, M. A., Romero, L. M., Awerman, J., & Wikelski, M. (2010). To breed or not to breed: Physiological correlates of reproductive status in a facultatively biennial iguanid. Hormones and Behavior, 57, 140–146.

26

Hormones and Behavior

Vitousek, M.  N., Mitchell, M.  A., Woakes, A.  J., Niemack, M. D., & Wikelski, M. (2007). High costs of female choice in a lekking lizard. PLoS One, 2, e567. Vitousek, M. N., & Romero, L. M. (2013). Stress responsiveness predicts individual variation in mate selectivity. General and Comparative Endocrinology, 187, 32–38. Vitousek M. N., Taff, C. C., Hallinger, K. K., Zimmer, C., & Winkler, D.  W. 2018. Hormones and fitness: evidence for trade-offs in glucocorticoid regulation across contexts. Frontiers in Ecology and Evolution, 6(42), 1–14. Vyas, A., Harding, C., Borg, L., & Bogdan, D. (2009). Acoustic characteristics, early experience, and endocrine status interact to modulate female zebra finches’ behavioral responses to songs. Hormones and Behavior, 55, 50–59. Wagner, E.  C., Prevolsek, J.  S., Wynne-Edwards, K.  E., & Williams, T.  D. (2008). Hematological changes associated with egg production: Estrogen dependence and repeatability. Journal of Experimental Biology, 211, 400–408. Wagner, E.  C., Stables, C.  A., & Williams, T.  D. (2008). Hematological changes associated with egg production: Direct evidence for changes in erythropoiesis but a lack of resource dependence? Journal of Experimental Biology, 211, 2960–2968. Weaver, I.  C.  G., Cervoni, N., Champagne, F.  A., Alessio, A. C. D., Sharma, S., Seckl, J. R., . . . Meaney, M. J. (2004). Epigenetic programming by maternal behavior. Nature Neuroscience, 7, 847–854. Wikelski, M., Steiger, S. S., Gall, B., & Nelson, K. N. (2005). Sex, drugs, and mating role: Testosterone-induced phenotypeswitching in Galapagos marine iguanas. Behavioral Ecology, 16, 260–268. Williams, D. R. (1999). Race, socioeconomic status, and health. The added effects of racism and discrimination. Annals of the New York Academy of Sciences, 896, 173–188. Williams, T.  D. (2008). Individual variation in endocrine systems: Moving beyond the “tyranny of the golden mean.” Proceedings of the Royal Society B, 363, 1687–1698. Williams, T. D. (2012). Hormones, life-history, and phenotypic variation: Opportunities in evolutionary avian endocrinology. General and Comparative Endocrinology, 176, 286–295. Wingfield, J.  C. (2008). Organization of vertebrate annual cycles: Implications for control mechanisms. Proceedings of the Royal Society B, 363, 425–441. Wingfield, J. C. (2015). Coping with change: A framework for environmental signals and how neuroendocrine pathways might respond. Frontiers in Neuroendocrinology, 37, 89–96. Wingfield, J. C., Lynn, S., & Soma, K. K. (2001). Avoiding the “costs” of testosterone: Ecological bases of hormone-behavior interactions. Brain, Behavior and Evolution, 57, 239–251. Wingfield, J. C., Owen-Ashley, N. T., Benowitz-Fredericks, M., Lynn, S.  E., Hahn, T.  P., Wada, H., . . . Romero, L.  M. (2004). Arctic spring: The arrival biology of migrant birds. Acta Zoologica Sinica, 50, 948–960. Wingfield, J. C., & Sapolsky, R. M. (2003). Reproduction and resistance to stress: when and how. Journal of Neuro­ endocrinology, 15, 711–724. Zera, A. J., & Harshman, L. G. (2001). The physiology of life history trade-offs in animals. Annual Review of Ecological Systems, 32, 95–126.

CH A PT E R

3

Sex Differences in Primate Social Relationships During Development

Joyce F. Benenson

Abstract The chapter focuses on the social development of immature primates across several species in the Cercopithecidae and Hominidae families, in particular rhesus macaques, the great apes, and humans. The chapter provides an overview of critical factors that characterize the rearing environments of immature females and males, including social structure, residence patterns, and dominance relations. Regardless of rearing environment, consistent sex differences in immatures occur in relationships with mothers, adult males, same-sex peers, and infants. Additionally, sex differences regularly are found in rates of development, quests for dominance, frequency of social play, and rate and intensity of direct aggression across species. Key words:  Sex differences, primates, social relationships, development, social behavior, dominance, aggression, social play

In Non-Human Primates the development of ­immature females’ and males’ social behavior has ­received relatively little attention. Instead, researchers have focused on sex differences in adults’ social structures or patterns of relationships (Kappeler & van Schaik, 2002). In humans, however, the reverse has occurred. Sex differences in myriad immature social behaviors have been documented, but adult social structure remains little studied. The adult social structure of a primate species provides an explanatory framework for understanding the behavior of the ­individuals within the community. Moreover, it provides the context for socialization of immatures. At the same time, sex differences in immature behaviors help ­illuminate both the phylogenetic adaptations and proximate constraints and opportunities provided by the community that contribute to adults’ patterns of relationships. When consistent sex differences are found from early in life across closely ­related species, this suggests strong adaptive pressures, both phylogenetic and environmental, produce sexdifferentiated patterns of relationships.

A major difficulty in investigating sex differences in immature animals’ social relationships is that the behaviors that have been investigated in one species often are not examined in another. Nonetheless, the behavior of young members of our own species provides a valuable starting point because the majority of research on sex differences in social structures has been conducted on human children (Homo sapiens). This permits comparisons with data that are available for immatures of living members of the most closely related species. Old World monkeys (Cercopithecidae), particularly macaques and specifically rhesus monkeys, who share over 90 percent of their genes with humans, have been studied extensively, so they provide a rich source of data for comparisons with humans (Suomi, 1997). Further, because the four great apes—orangutans (Pongiae), gorillas (Gorilla), bonobos (Pan paniscus), and chimpanzees (Pan troglodytes)—human’s closest living genetic relatives, share over 98 percent of humans’ genes (Prüfer et al., 2012), they also provide critical comparisons. The focus here will be 27

confined to reviewing regularities in sex differences in the social behavior of immatures across some ­species of Cercopithecidae and Hominidae families, particularly rhesus macaques and the great apes. Sexual behavior, atypical behaviors, and development in atypical environments will be excluded, and, to the greatest extent possible, social behavior will be included from wild as opposed to captive groups. Sex differences in social behavior in adulthood are to be expected in mammals because females typically take more responsibility than males for raising offspring, whereas males invest more in mating (e.g., Emlen & Oring,  1977). Thus, adult female primates must emphasize satisfying energy and safety needs to be able to produce offspring, lactate, provide protection, and socialize offspring, possibly for a number of years. In contrast, adult males compete to attain access to fertile females. Whether sex differences arise in immatures that reflect these basic sex-differentiated goals provides insight into both potential biological and socialization influences that cannot easily be disentangled. At the same time, these fundamental sex differences in social behavior must be understood within the context of stage of development, social structure of the species, mating systems, and dominance relations (Kappeler & van Schaik,  2002; Ostner & Schülke, 2014; Sterck, Watts, & van Schaik, 1997). Stages of development. Immature primates undergo three stages of development: infancy (before weaning), the juvenile period (before puberty), and ­adolescence (after puberty commences; Bogin, 1999). There are species differences, however, with humans’ juvenile period seeming to be divided into early (before feeding independence) and middle (before puberty) childhood. Likewise, orangutan males enter into a period of delayed adolescence, which can last from 10 to 20 years, although they are fertile during this period (Rijksen, 1978). Humans and apes have more prolonged periods of development compared to other primate species. One principle that appears to apply across primate species is that females develop more rapidly than males, with females attaining puberty a full two years earlier in the great apes and humans (Bogin, 1999). Therefore, studies of sex differences in immatures that compare females and males of the same age must be interpreted with caution as age cannot be  disentangled from sex. However, adolescence often lasts longer for males than females in humans (Hochberg, 2011), chimpanzees (Goodall, 1986), 28

bonobos (Kanō,  1992), and orangutans, where subadults form a class in between adolescents and fully flanged adult males (Rijksen, 1978). Somewhat paradoxically, although development of secondary sexual characteristics often is separated from fertility by a relatively longer interval for ­females than males (Hochberg,  2011), once females produce their first offspring, their social behaviors change dramatically, whereas the same is not true for males. Thus, a corollary to the more rapid development of  females is that males gain a prolonged time as immatures. Community social structure, residence patterns, and mating systems. Critical to understanding the development of an individual’s social behavior is the social environment into which it is born and socialized. For infants, social interactions are limited to one other individual at a time, most often the mother holding, touching, caressing, carrying, sniffing, and inspecting it (Fairbanks, 2002). As an infant matures, however, the social organization or numbers of individuals who are present, sex ratio, and spatiotemporal organization of a primate society, as well as the relationships between adults in the community or social structure (Kappeler & van Schaik, 2002), influence the environment in which it is raised. The social structure of a species is defined by proximity between individuals, association patterns, grooming, support in agonistic coalitions, and food sharing (Dunbar, 1988/2013). It is determined by relationships that depend on a partner’s kinship, sex, sta­tus, friendship, and age (Bissonnette et al.,  2015). Although most primates live in large multimale, multifemale groups, orangutans are semisolitary, and gorillas live in harems. In many species, social structure differs for ­females and males (Kappeler & van Schaik, 2002). Relation­ ships between females depend on the degree of ­ompetition over food or protection from predators. When food is clumped, females usually form coalitions, primarily with female kin, to defend food patches. When food is scattered or the threat of predation is small, females may be solitary. Females’ relationships within a species are determined by their residence patterns (philopatry), preference for kin (nepotism), and tolerance or despotism in dominance relationships (Sterck et al., 1997). When females continue to reside in a community throughout their lives, they form long-term, ­differentiated bonds, primarily with kin, but also with female friends, and these can last a lifetime and enhance reproductive success (Silk, Alberts, & Altmann, 2003).

Sex Differences in Primate Social Rel ationships

In contrast, relationships between males often depend on access to fertile females. The extent to which one male can gain access to all reproductive females appears to play a large role in determining whether males are solitary, form differentiated relationships, or interact in groups (Ostner & ­ Schülke, 2014). Because matings cannot be shared, males’ relationships are almost always more directly and immediately competitive than females’ relationships. When several females reside together, males may benefit from forming coalitions to aid one another in rank changing or leveling coalitions vis-à-vis other males within the community or in repelling takeovers by other male coalitions from outside the community. In larger communities, when hostile intergroup interactions occur, males benefit from widespread cooperation against external communities. Whether males form opportunistic and temporary associations or long-term differentiated bonds depends on a number of factors including reproductive skew within the community. Residence patterns play an important role in social structure, and they too are highly sex differentiated. When only one sex disperses, which is often the case, each sex will differ markedly in how much its development will be influenced by the patterns of relationships it has established during childhood. Where females remain in their natal communities throughout their lives, they are surrounded by kin and hence more likely than males to form long-term bonds with same-sex kin and friendly nonkin. At adolescence, males must disperse, and the developmental processes that prepare a male and/or produce this permanent separation from a male’s kin and familiar conspecifics remain little understood. Likewise, where males are philopatric, females must disperse at adolescence. The preparation for this major transition to an unfamiliar community apart from kin and familiar conspecifics has not been explored. It is highly likely that residence patterns exert a powerful influence on the development of immatures, as one sex derives the safety and familiarity of spending its entire life in the community in which it was born, and the other sex must join a new community as an individual with all the stress that such a transition entails. Understanding the proximate factors that lead one sex to disperse at adolescence is an area that is ripe for further investigation of developmental processes. In most Cercopithecine species, a number of matrilines consisting of several generations of ­females will structure the infant and juvenile lives of both sexes, and the adolescent and adult lives of

females (Chapais & Berman,  2004). Infants are born into a community composed of female kin who have strongly differentiated bonds. This community occupies a territory, although the community may travel to other territories over time. In contrast, males disperse at adolescence and face an increased threat of mortality, as high as 50 percent, when they attempt to enter a new troop. Orangutans appear to be female philopatric, with males dispersing farther from mothers at adolescence (Arora et al.,  2012). Nonetheless, orangutans are semisolitary and female adults meet with their mature daughters only in large fruiting trees. An adult female’s territory ­unusually overlaps with those of other adult females, so that as a juvenile develops, he or she comes into contact with the offspring of other nearby mothers, some of whom are older sisters. Although immature males too will meet the adult females and their offspring of nearby territories, by adolescence they will move farther away from this familiar ­environment. In marked contrast, gorillas form harems with one dominant silverback male and several (usually unrelated) females who do not get along easily, unless they are sisters (Robbins et al.,  2004). In mountain gorillas, sometimes more than one male is present (Harcourt & Stewart, 2007). Both female and male infants grow up physically close to their father, the silverback, and their own mothers and will travel together as a harem or small group. Neither sex forms relationships with unrelated ­ females before puberty. At adolescence, both males and ­females emigrate, though in some mountain gorillas the silverback will allow one or two males to remain. Bonobos and chimpanzees reside in multimale, multifemale communities with prescribed territories, similar to humans. Infants and juveniles therefore grow up surrounded by many adults of both sexes. In both bonobo and chimpanzee communities, males remain in their natal communities throughout their lives, so that some of the adolescent and adult males are related. In contrast, adolescent ­females disperse. Female bonobos transfer to ­another community between 6 and 10 years of age, even before adolescence (Furuichi et al., 2012), whereas female chimpanzees transfer after estrus commences. However, in a few communities or cases, females will spend their entire lives in their natal communities (Pusey, 1990). In bonobo communities, females and their ­ adolescent and adult sons form the core of the community. Females cooperate to share food and  occasionally dominate males, but a female’s Benenson

29

c­ losest ally is her son, though adult females form differentiated bonds (Moscovice et al.,  2017). A young adolescent female who emigrates to a new community, however, often finds an older resident female with whom she develops a bond, which helps her become integrated into the community (Tokuyama & Furuichi, 2016). Offspring of female bonobos spend time in mixed-sex parties with their mothers and older brothers and with unrelated females and their sons. Male bonobos do not form bonds with anyone but their mothers until their mothers die, after which they may bond. All members of the community spend time together in large mixed-sex groups, though members of the groups frequently change (Hohmann & Fruth, 2002). In contrast, unrelated and related males form the core of chimpanzee communities. Adult females are subordinate to both adolescent and adult males. In many, but not all, chimpanzee communities, adult females live in core territories, whereas males travel across the entire community (Boesch & BoeschAchermann, 2000). An adolescent female who emigrates to a new community often is protected by resident adult males because adult resident f­emales may attack her when she attempts to emigrate. Infants grow up in close proximity to their mothers and older siblings but do not spend much time with other adults. It is only when their mothers join parties, which happens increasingly as weaning approaches and the mothers come into  estrus, that they join other members of the community. It is then that they meet the adolescent and adult males and unrelated adult females, though  in smaller parties than in bonobo communities. Human societies are often considered male philopatric as well, though it is clear that there is much more flexibility in dispersal patterns than in most other primate species. Unrelated and related men often form the core of the wider community (Rodseth, Wrangham, Harrigan, & Smuts,  1991). Unlike other species, however, human mothers do not take sole responsibility for raising their children. Older infants, children and adolescents almost always are cared for by other kin and affines, but particularly by fathers, grandmothers, and older siblings (Konner,  2010; Sear & Mace,  2008). Frequently, a mother will move close to her own mother after she gives birth, so the infant is cared for by a grandmother and her kin. When a mother divorces, she also endeavors to move closer to her own mother, so that she can obtain assistance in raising her children (Alvarez, 2004). 30

Sex Differences in Quests for Dominance and Status

Despite the vast species differences in social environments in which immatures are raised, dominance determines access to physical and social ­resources across species and sexes. In most species, adult dominance rank exerts a strong effect on survival of offspring and subsequent reproductive success, because rank increases either access to higher quality foods and safer territories for females or more mating opportunities for males (Clutton-Brock & Huchard, 2013). One principle that applies across species is that, even as immatures, males are more likely than ­females to engage in direct one-on-one confrontations over status beginning in late infancy or the early juvenile period and continuing through adolescence (Meaney, Stewart, & Beatty,  1985). The reason that females do not engage in as many direct contests for status is likely related to the greater costs that potential physical injuries could inflict on females—who produce and raise offspring over many years—relative to males, who can attain equivalent reproductive success in a briefer period (Campbell, 1999). It is also possible that females are more likely than males to attain status in differing ways, such as by relying more on alliances, especially with kin, or by competing indirectly over time for resources, territory, and allies, thereby obviating the need for direct one-on-one contests (Clutton-Brock & Huchard, 2013). In female philopatric Old World monkeys, such as macaques, vervets, and baboon species, both female and male infants rapidly learn about their mothers’ dominance status within the community through observing the outcomes over her contests with other mothers, her aggressive reactions toward other mothers’ infants and older offspring, and other mothers’ aggressive reactions toward them. Female dominance ranks are stable within and ­between matrilines in a community over many generations. Thus, a daughter enters a rigid, lifelong dominance hierarchy in which her position is predetermined under most circumstances. An infant female can expect to rank where her mother does and often slightly above her older sisters. The mechanism that allows the intergenerational transmission of status is that daughters are aided in coalitions by their mothers and other female kin (Chapais & Berman, 2004; Silk, 2009). As infant females ­develop into juveniles, coalitions form between mothers and daughters, and those who are higher ranked

Sex Differences in Primate Social Rel ationships

have greater access to food, space, and other limited resources (Chapais & Gauthier, 2002). Nevertheless, an individual immature male can dominate a female, unless she can form a coalition. Another principle that appears to apply across these species is that many intrinsic assets and skills, including age, size, strength, intelligence, motivation, and ability to form new alliances with a range of individuals, determine males’ more than females’ status (Bercovitch,  1997; Goodall,  1986). These assets and skills will develop for males as they grow and learn, and form differing coalitions. A corollary therefore is that a male’s status fluctuates more than a female’s status does. For many females in female philopatric species, relative status is stable throughout their lives, beginning even before they are born in terms of the food and safety their pregnant mothers can access. In the great apes, the same principles apply: As immatures, males engage in more direct confrontations over status. Further, a number of assets and skills that fluctuate with both an individual’s developmental stage and which individuals are present appear to influence males’ more than females’ status. Little is known about status quests in immature orangutans, most likely because of their relatively solitary nature, although adolescent males and unflanged males are subordinate to flanged adult males (Rijksen,  1978). In gorillas, by adolescence, all male adolescent gorillas will dominate all females (Watts & Pusey, 2002), but they all are subordinate to the silverback. Adolescent female gorillas will submit to adolescent males and the silverback, but they may not submit to unrelated female adults. In male bonobos, competition for dominance begins in adolescence, when adolescent males a­ ttempt to enter the linear adult male dominance hierarchy. Unusually, what partially determines rank in adolescent and adult male bonobos is their mothers’ rank, similar to the situation for immature females in female philopatric Old World monkeys. Mothers maintain close proximity to their adolescent sons and support them in conflicts. The rank of an adolescent male therefore is highly correlated with his mother’s rank until she dies, after which he drops in rank (Kanō, 1992). Little is known about whether daughters benefit from their mothers’ dominance status before they emigrate, though this seems likely as adult females form coalitions against males during food competition. What is clear, however, is that after an adolescent female emigrates to a new community, she is at the bottom of the female

­ ominance hierarchy and depends on an older d resident female’s support to integrate her into the community (Tokuyama & Furuichi,  2016). With increasing age and length of residence, adolescent females attain higher status, so that relative rank remains constant as older residents die and younger ones move up the hierarchy. In chimpanzees, as adolescence approaches, every male learns to dominate every female in the community, including female adults. Older juvenile and adolescent male chimpanzees begin to compete for status, which is based on the same intrinsic factors, including age, size, strength, intelligence, motivation, and social alliances, as in the female philopatric old world monkeys (Goodall, 1986). Again, as with Old World monkeys, chimpanzee males’ rank depends on intrinsic factors, acquired skills, developmental changes, and alliances. Dominance status peaks at age 21 when chimpanzee males are at the height of their physical prowess, but there are continual rank shifts as a male joins differing coalitions in his quest to defeat higher ranked individuals. Female chimpanzees usually emigrate at adolescence, though in some communities they remain in residence throughout their lives. Because adult females form hierarchies (Pusey, Williams, & Goodall, 1997), a female who remains with her mother in her natal community likely obtains the same advantages as daughters in female philopatric Old World monkeys. Mothers and adolescent daughters can form coalitions to ensure that daughters gain access to territories richer in food (Foerster et al., 2016). In most communities, however, ­females emigrate to a new community at adolescence, and status occurs through queueing. Increasing age or length of residence determines status, so that newcomer female adolescents begin with the poorest quality territories and must queue for richer territories by awaiting the deaths of older resident females. Thus, in both male philopatric bonobos and chimpanzees, just as in female philopatric Old World monkeys, relative rank remains stable for females, so that, unlike males, continual direct contests over rank do not occur. In humans, in hunter-gather societies that consist of small bands that frequently change in composition, relations between adults are relatively egalitarian (Boehm, 1999). When communities become larger and more stable, however, clear hierarchies emerge. Differences in socioeconomic status strongly influence infant mortality, health, and longevity (Marmot & Wilkinson, 2006; Sapolsky, 2004). In Benenson

31

modern humans, before the end of infancy, human males dominate females, who withdraw during mixed-sex dyadic contests (Jacklin & Maccoby, 1978). Thus, although it has not been directly ­reported in most species, in both female philopatric macaque species (Cords & Aureli, 2002) and humans (Charlesworth & La Freniere, 1983), an individual male dominates an individual female, even as imma­ tures of roughly the same size and strength. Limited evidence suggests that across species, high-ranked mothers invest more in immatures of the sex that is philopatric. In a few studies in female philopatric species, high-ranked mothers invest more than low-ranked mothers in daughters rather than in sons, presumably because the daughters serve as current and future allies (Silk, Altmann, & Alberts, 2006). High-ranked adult females may also form a coalition to attack a low-ranked female’s infant, but only a female infant (Silk & Boyd, 1983). Nonetheless, in female philopatric species in which males emigrate at adolescence, high-status mothers can also invest in sons by forming coalitions with them against other community members. Consequently, it appears that sons of higher status mothers often remain longer than sons of lower status mothers in their natal communities before undergoing the stress of emigration (Lee & Johnson, 1992). A few studies likewise show that in male philopatric species, high-ranked females invest more in sons than in daughters as predicted by the TriversWillard hypothesis (Trivers & Willard, 1973). For example, in one community, dominant chimpanzee mothers invested up to two years longer in sons than daughters compared to nondominant mothers (Boesch & Boesch-Achermann,  2000). Because almost all females in this community emigrated at adolescence, mothers can aid only their adult sons in conflicts. However, in chimpanzee communities where daughters often remain for life, mothers do not invest differentially in either sex. In these communities, a mother’s status can positively ­influence both her son’s and her daughter’s status (Goodall, 1986). Whether this influences a daughter’s decision to emigrate remains unknown. In humans, in many cultures, mothers invest more in sons than daughters, perhaps because humans are male philopatric (Rodseth et al., 1991). Regardless of the reason, evidence that boys are the preferred sex comes from cross-cultural findings that daughters are more likely than sons to be n ­ eglected or killed in childhood (Sen,  1990). Although there is some evidence that high-ranked families invest more in 32

sons than daughters compared to low-ranked families (Dickemann, 1979), few empirical studies exist.

Physical and Verbal Contact With Mothers

Mothers take primary responsibility for keeping their offspring alive, and humans are one of the few primate species in which they receive assistance from others. Across species, mothers invest to a similar extent in daughters and sons. At the same time, another principle that guides development is that immature females spend more time than males in close proximity to mothers. This is because immature males are more likely than females to reduce physical proximity to mothers. In female philopatric rhesus macaques, infants of both sexes maintain physical contact with their mothers about 70 percent of the time for the first two months (Kulik, Amici, Langos, & Widdig, 2015). By the second year of life, however, daughters spend more time than sons in close proximity to their mothers. Beginning in the early juvenile period and continuing into adolescence, females increase their grooming of mothers and other female kin, and in some species, even males in the troop. In contrast, from infancy through adolescence, males spend more time grooming male kin and nonkin (Roney & Maestripieri, 2003). After they emigrate, adolescent males who survive the transition then begin grooming females in their new troops. In orangutans, who spend the longest time alone with their mothers, mothers and daughters appear to have more positive interactions compared with mothers and sons. Further, in this female philopatric species, daughters remain closer than sons to their mothers throughout life (Galdikas, 1995). It is infant and juveniles, however, who take responsibility for maintaining close proximity to mothers. Juvenile female orangutans are less submissive to mothers as they approach adulthood than before, and their mothers seem to be more tolerant of their daughters than their sons (van Adrichem, Utami, Wich, van Hooff, & Sterck,  2006). In gorillas, j­uvenile females spend more time closer to their mothers than juvenile males do (Watts & Pusey, 2002). As adolescents, males spend much of their time alone as they leave their harems (Harcourt & Stewart, 2007). An important question is whether immature female bonobos maintain closer proximity than males to mothers, given the unusual nature of the community in which unrelated adult females bond more than unrelated males, but mothers form close lifetime bonds with sons. No evidence exists yet to answer this question.

Sex Differences in Primate Social Rel ationships

Immature chimpanzee females maintain closer proximity than sons to their mothers beginning early in life. Chimpanzee male infants begin to travel independently from their mothers earlier than ­females do, and by age 3 years distance themselves more than females from their mothers (Lonsdorf et  al.,  2014). However, during the juvenile years, ­females spend 100 percent of their time and males spend almost 90 percent of their time in close proximity to their mothers (Pusey, 1990). Juvenile females groom their mothers and siblings almost exclusively and the grooming is more reciprocal, whereas males groom their mothers about 67 percent of the time while also grooming other members of the community, including any adult males who may be present. Adolescent female chimpanzees who do not emigrate remain in close proximity to their mothers. In contrast, males increasingly spend time away from their mothers and with other female adults and males (Watts & Pusey,  2002). Young chimpanzee females also learn to fish for termites using sticks by observing their mothers and learning their mothers’ method, whereas males do not learn this technique from their mothers (Lonsdorf, Eberly, & Pusey, 2004). Beginning in early adolescence, however, a chimpanzee mother is less likely to reciprocate her daughter’s than her son’s grooming (Goodall, 1986). This likely pushes her daughter to emigrate. Nevertheless, time spent in close proximity to mothers remains high for females until they emigrate, whereas ­adolescent males spent about one-third of their time alone, and the remainder with adolescent or adult males or unrelated females in estrus (Goodall, 1986; Pusey, 1990). Just before emigrating, adolescent ­females often spend most of their time alone or with males. Those adolescent females who remain in their natal communities vary between spending time with their mothers, alone, and when they are in estrus, with males. Adolescent male chimpanzees who remain in their natal communities suddenly leave their mothers’ sides and begin to integrate themselves into the adolescent male peer group, which maintains ­ proximity to the adult males of the community. Grooming rates toward mothers diminish and grooming of other adolescent males and adult males suddenly increases (Pusey, 1990). Individual adolescent males often obtain support from an adult male who wishes to forge a future alliance, who may be (but is not always) a brother (Goodall,  1986). Adolescent males confront an approach–avoidance dilemma regarding adult males. Adult males do

not wish an adolescent male to usurp their positions in the dominance hierarchy, so adolescent males become targets of adult males’ brutal attacks. At the same time, adolescent males are highly motivated to assume their place in the adult community and must therefore compete for a position. Adult males benefit however from increasing the size of the male community because of their participation in intercommunity aggression in which victory depends on strength in numbers (Ostner & Schülke, 2014). In humans, there is some evidence that even by the end of the first year of life, girls spend more time than boys close to their mothers (Goldberg & Lewis, 1969). By early childhood, this sex difference becomes marked and continues into adolescence (Schlegel & Barry, 1991; Whiting et al., 1988; Lancy, 2015). Across cultures, daughters more than sons cooperate with mothers by caring for their younger siblings and completing household tasks. In contrast, boys are found farther away from mothers and homes. Even when fathers are present, boys are found on the periphery of men’s groups. Nonetheless, adolescents of both sexes increase their distance from mothers, although sons distance themselves more than daughters. Sons also are more likely than daughters to dominate mothers (Conger, Patterson, & Ge,  1995). Moreover, across diverse societies, even when daughters move to another community, they are more likely than sons to maintain their emotional closeness to mothers (Troll, 1987). Greater communication between mothers and daughters versus sons is another potential principle. Beginning in infancy, macaque females and males differ in their frequencies and intensities of calls (Wallen,  2005). Females exhibit vocalizations earlier and, after separation from or rejection by their mothers, call more frequently and for longer intervals than males do. Further, their vocalizations are more pleasant and softer and hence probably more attractive to mothers. Whether this sex difference appears in ape species is unknown. Like rhesus ­macaques, however, beginning in infancy human females are more advanced in their verbal skills than males (Bouchard, Trudeau, Sutton, Boudreault, & Deneault,  2009), and this continues throughout childhood (Kimura, 2000). Accordingly, beginning in early childhood, human mothers converse with their daughters more than with their sons (Leaper, Anderson, & Sanders, 1998).

Interaction With Same-Sex Peers

Sex segregation and asymmetry in preference for samesex peers. Across most primate species, immature Benenson

33

individuals of each sex prefer to interact with samesex peers. Sex segregation begins early as immatures begin to interact with peers, usually by the end of infancy (Meredith, 2013). This same-sex preference likely occurs due to differing behavioral preferences of each sex that emerge early in life. An important corollary, however, is that immature males exhibit a greater preference than females for interaction with same-sex peers (Meaney et al., 1985). By definition, this partially results from the fact that immature ­females spend more time than males with mothers and other female kin and friends. However, it is not simply a reflection of less time available for interaction with same-sex peers because there are many other classes of individuals with whom an immature could affiliate. In rhesus monkeys, who leave their mothers’ bodies after a few months, both sexes interact most frequently with those of the same sex and age, but female infants spend more time than male infants with female kin of all ages (Brown & Dixson, 2000; Mitchell, 1979). In contrast, males spend more time interacting with same-sex peers (Bernstein, Judge, & Ruehlmann,  1993). Ape species exhibit similar patterns. Orangutans meet one another as they become juveniles and explore the periphery of their mothers’ territory. Adolescent and subadult males have been reported to spend time together, whereas groupings of adolescent females have not been ­reported unless male adolescents or subadults are also present (Rijksen, 1978). Even in infancy, infant gorilla males prefer to play with same-sex peers more than females do (Maestripieri & Ross, 2004). Bachelor groups of unrelated juvenile and adolescent gorilla males who leave their natal groups also occur, but there are no female bachelorette counterparts (Watts, 2000). No evidence exists at present for groups of same-sex males or females in immature bonobos. Only males whose mothers have died interact in ­all-male groups, and in most cases, these are adult males (Kanō, 1992). Nonetheless, an adolescent male whose mother had died would join the all-male group. Although it would be expected that immature female bonobos would interact with same-sex peers given the dominance and coalitionary power of females in bonobo communities, this has not yet been examined. Adolescent chimpanzee males, however, form groups after they permanently leave their mothers’ sides. These all-male adolescent groups compete for dominance and form a way station as  each male begins to integrate himself into the adult male dominance hierarchy (Goodall,  1986). 34

Adolescent females do not form all-female groups either before emigrating or after having emigrated to a new community. In humans, beginning at the end of infancy, children universally segregate themselves by sex. By the end of early childhood, males do this more strongly than females do (Maccoby, 1998), which produces an asymmetry whereby boys prefer to interact with same-sex peers more than girls do (Fouts, Hallam, & Purandare, 2013; Lancy, 2015; Munroe & Romney, 2006). The asymmetry is so strong that beginning in early childhood boys are less accepting than girls of boys who engage in feminine activities, such as doll play or domestic activities, or play with girls (Ruble, Martin, & Berenbaum,  2006). This continues cross-culturally in adolescents when males who do not fit gender stereotyped norms ­experience more difficulties in peer relations than their female counterparts (de Vries, Steensma, Cohen-Kettenis, VanderLaan, & Zucker,  2016). Whether nonhuman primates make this type of gender distinction based on behavior or other types of stimuli has not been investigated. Size of same-sex peer networks. Another principle is that when males interact with same-sex peers, both juvenile and adolescent males form larger same-sex social networks than their female counterparts do (Lancy, 2015; Mitchell, 1979). In particular, males interact with more peers simultaneously, whereas females are more likely to interact with only one ­individual or a small clique. In female philopatric species, adolescent males form temporary bachelor groups and may enter new communities together. Their alliances, along with their physical strength, can increase their chances of successful assimilation into their new communities (Schülke, Bhagavatula, Vigilant, & Ostner, 2010). In Old World monkeys, by 6 months of age, j­uveniles spend most of their time with same-sex peers. Males however, spend time in larger peer groups than females, with both central and peripheral males (Roney & Maestripieri, 2003). Further, after they emigrate, adolescent males often temporarily join adolescent male peer groups until they singly or jointly join a new community (Suomi, 1997). In contrast, females spend more time with female kin and friends of differing generations and interact with fewer same-sex peers at a time. Although little evidence is available, what does exist suggests that even in semisolitary orangutans, juvenile and adolescent males interact with slightly larger same-sex peer groups than females, whereas

Sex Differences in Primate Social Rel ationships

females interact in mixed-sex groups or with one other female at a time (Rijksen, 1978). In gorillas, adolescent males occasionally form same-sex groups (Watts, 2000). Chimpanzee males, but not females, regularly form single-sex adolescent peer groups (Goodall, 1986). With increasing age, juvenile chimpanzees of both sexes spend less time with their mothers, but only adolescent males join an adolescent peer group until they enter the adult male community. In humans, by the end of infancy, males cluster in larger groups than females (Fabes, Martin, & Hanish, 2003). Beginning in middle childhood and continuing through adolescence across diverse cultures, the size of same-sex peer groups is larger for males than females (Fine, 1980; Lancy, 2015; SavinWilliams, 1980; Schlegel & Barry, 1991). In contrast, immature females are more likely to interact with one close same-sex friend at a time or in a small clique of females. Social play. Most primates play, particularly during the late infancy and juvenile stages, after which its frequency declines (Fagen,  1993). Play likely provides rehearsal for competing, reconciling conflicts, and cooperating with same-sex peers, skills that will be valuable throughout life. There are many forms of social play, including play fighting, chasing, and initiating and receiving bids to play. Solitary play and play with infants are not included here in the definition of social play. Infant and juvenile males play more than females across primate species, and the sex difference continues through adolescence (Mitchell,  1979; ­ Roney & Maestripieri,  2003). This may result simply from the fact that immature males who interact with peers play, so that peer interaction and social play may be synonymous (Meaney et al., 1985). Nonetheless, it is possible that peer interaction and social play are more likely to overlap completely for males than females. Thus, female peers might interact in other ways, such as by examining objects or spending time with one another’s mothers, without engaging in play. In rhesus monkeys, males compared to females engage in more stationary and chasing play, give and receive more invitations to play, and engage in more rough-and-tumble play or play fighting (Brown & Dixson, 2000). Rough-and-tumble play requires high energy, involves grasping and wrestling, and universally occurs more frequently in juvenile males than females across macaque species, regardless of rearing environment (Wallen,  1996, 2005). Play

fighting in particular appears to assist in rehearsing motoric and cognitive skills required for establishing status based on physical assets and social skills, which in adulthood will influence reproductive success (Meaney et al., 1985). In macaque species, infants and juveniles begin by engaging in reciprocal interactions involving active play with same-age, same-sex unrelated individuals. Eventually, this leads to dominance contests where only one individual wins, but only for males (Meaney et al., 1985). Sex differences in social play in the great apes are less consistent, but, when they are found, infant and juvenile males play more than females. Greater involvement in social play has been reported in immature orangutans (Rijksen, 1978), gorillas (Maestripieri & Ross, 2004), and chimpanzees (Lonsdorf et al., 2014). In chimpanzees, male i­nfants and juveniles initiate play using tactile gestures more than females do (Fröhlich, Wittig, & Pika, 2016). In humans too, beginning in early childhood across diverse cultures, males play more than females who are more engaged than males in providing assistance with household responsibilities (Lancy, 2015; Whiting et al., 1988). Further, beginning at the end of infancy, males consistently engage in more rough-and-tumble play than females do (Fry, 2005), whereas females engage in more parallel play than males do (Barbu et al., 2011). Direct aggression. In most primate species, males engage in more frequent and more intense direct verbal and physical aggression than females across the lifespan. Further, as the intensity of the aggression and concomitant potential for harm increases, the magnitude of the sex difference also increases. In macaque species, for example, aggression increases from infancy through adolescence for males, but levels off during the juvenile period for females. Beginning in infancy, males are more aggressive than females, and this sex difference becomes particularly pronounced in adolescence (Lonsdorf & Ross, 2012). Adolescent males are the most aggressive class of individual in the troop, and they form coalitions to defeat individual males. They also are the biggest recipients of aggression. Although males are dominant over all females by adolescence, support from maternal kin allows females to counter an individual male’s aggression. Little is known about aggression in immature orangutans, though the limited evidence suggests that male juveniles are more aggressive than females in captivity (Nadler & Braggio,  1974). In mountain gorillas, the severity of aggression increases with age in both sexes, but immature males are more aggressive Benenson

35

than females (Watts & Pusey, 2002). Bonobo males increase their aggression markedly in adolescence, as they compete for status in the male hierarchy, with the support of their mothers (Kanō, 1992). In marked contrast, adolescent female bonobos who have just emigrated almost never engage in aggressive behavior. As n ­ ewcomers, they are quiet and rarely even counterattack. Whether individual adolescent females are as aggressive as individual males in their natal communities before they emigrate has not been examined. However, it is likely that aggression is a matter of physical strength, and as happens with adult bonobos, only through coalitions can females dominate a male. In chimpanzees, frequency of aggressive play peaks between 2 and 4 years and males are more aggressive than females (Goodall,  1986). Intensity of aggression peaks in late adolescence and early adulthood when males are most likely to become the highest ranked member of their communities. Adolescent males become aggressive toward adult females as they begin to work their way up the male dominance hierarchy. Adolescent females joining a new community often confront severe aggression from high-ranked female residents. In response, they appear to associate with adult males who provide protection (Boesch & Boesch-Achermann, 2000). Similar to chimpanzees, across diverse cultures in humans, frequency of aggression peaks between 2 and 4 years of age (Coie & Dodge, 1998; Konner, 2010), and intensity of aggression peaks in adolescence and early adulthood (Daly & Wilson, 2001). At all ages, males are more directly aggressive than females (Archer,  2004). The magnitude of the sex difference increases in direct proportion to the ­intensity of the aggression (Moffitt, Caspi, Rutter, & Silva, 2001). Other forms of aggression may be as high or higher in females, particularly in adolescence (Archer, 2006), but direct aggression is invariably higher in males beginning in infancy.

Interaction With Adult Males

Paternal care is relatively rare in primates (Cabrera & Tamis-LeMonda,  2013). Therefore, immatures in most primate species have less contact with adult males than adult females. This does not mean, however, that adult males never assist mothers with carrying, grooming, supporting, and protecting ­infants and juveniles. Nevertheless, they engage in these activities much less frequently than mothers do, and there is large variance across species with respect to the degree of investment by adult males in immatures. 36

Paternity certainty likely increases male investment in immatures. For example, in gorillas, the silverback (who is virtually always the father of the offspring in his harem) can provide extended care to his offspring (Harcourt & Stewart, 2007). In many species, however, adult males provide important sources of protection from infanticide to any infant in the community, although they also are most likely to commit infanticide (Boesch & BoeschAchermann, 2000). Further, in males of species that compete against other communities, adult males protect all immatures in the community (Ostner & Schülke, 2014). Adult males often exhibit a preference for male immatures over female immatures. This likely has to do with their mutual enjoyment of play. In most macaque species, adult males exhibit greater interest in male than female infants (Meaney et al., 1985). Adult males spend more time and play more with sons than daughters. Concomitantly, beginning in the juvenile period, young males groom adult males more than females do (Roney & Maestripieri, 2003). However, as immatures reach adolescence, adult males become much less tolerant as adolescent males become competitors for status and mating opportunities. Lack of tolerance for adolescent males occurs both in male and female philopatric societies. In orangutans, adult males are solitary, but they will tolerate adolescent and unflanged males during feeding (Rijksen,  1978). In gorilla harems, in contrast, with age infants and juveniles of both sexes spend increasing amounts of time near the silverback male, who most likely is their father, and away from their mothers. By adolescence, however, most females spend more time than males near the silverback. When immature males reach adolescence, the silverback begins to attack them more than females (Watts & Pusey, 2002). In bonobos, some evidence suggests that only adult males and not adult females are allowed by  mothers to carry their infants on their backs (Kanō,  1992). Further, adult male bonobos will play with adolescent males in bouts of wrestling (Kanō,  1992). As the adolescent male approaches adulthood, however, he is attacked by the adult males. Consequently, older adolescent males remain on the periphery of the community except when their mothers can provide coalitional support. Adult male chimpanzees will play with infants and sometimes even attempt to adopt orphaned ­juveniles (Goodall, 1986). Furthermore, chimpanzee infant males spend more time than infant f­emales with adult males, as their mothers are more likely

Sex Differences in Primate Social Rel ationships

than mothers of females to maintain close proximity to males in the community (Murray et al., 2014). Whether this occurs because of male infants’ greater attraction to adult males, mothers’ greater desire to  integrate their sons than daughters into the ­community, or adult males’ greater interest in infant males is unknown. Juvenile chimpanzee males, however, spend increasing amounts of time near the males of the community by actively persuading their mothers to spend more time with adult males. By ­adolescence, males’ time in the presence of adult males and away from their mothers increases further; adolescent chimpanzee males groom adult males more than adolescent females do (Nishida, 1988). As stated, the adult male community becomes less welcoming to adolescent males than it was to juvenile males, who could not challenge them for dominance (Pusey, 1990). However, it is often the case that an adult male will form a bond with an adolescent male and help integrate him into the community. Often, the adult male is the adolescent male’s brother, but not always (Goodall,  1986). Older ­adolescents eventually join the adult male community and engage in shared activities. In contrast, ­juvenile and adolescent chimpanzee females avoid adult males until they emigrate to a new community (Watts & Pusey, 2002). Even in humans, who are one of the rare pairbonded primate species (which increases paternity certainty), fathers are estimated to invest in offspring at medium or high levels only 40 percent of the time (Barry & Paxson,  1971). Again, fathers spend more time with sons than daughters and are less likely to divorce their wives when they have sons (Lundberg, 2005). Moreover, fathers engage in more gross motor play with infant sons than daughters and initiate more physical contact with sons than daughters (Lamb, 2004; Parke, 1996). Beginning in early childhood, fathers will also spend more time and play more with sons than daughters. Adolescent sons are more likely than daughters to join their fathers, other male relatives, and unrelated adult males in the community in common activities (Schlegel & Barry, 1991). In many cultures, however, adolescent males must first undergo initiation rites that often involve injury before they are allowed to join the adult male community (Sosis, Kress, & Boster, 2007).

Interaction With Infants

Across primate species, immature females demonstrate greater interest in infants than males do. This almost certainly occurs because practice with

infant care increases a mother’s chances of keeping her infant alive. A mother learns the appropriate ­motoric repertoire for handling a vulnerable infant and rehearses her future role as a mother (Meaney et al., 1985). For female philopatric species, young ­females attempt to babysit higher ranked females’ infants. This provides parenting rehearsal and also enhances the probability that the higher ranked adult female will provide reciprocal support to her babysitter (Smith, 2005). The exceptions to greater interest in infants by females than males occur in the few species in which males provide as much care as or more care than mothers, such as in some New World monkeys. Studies of rhesus macaques in the first three years of life show large sex differences in interest in ­infants, with females more likely to touch, groom, initiate proximity, embrace, hold, play with, chase, harass, and kidnap infants, and the sex differences increase in magnitude after the first year (Herman, Measday, & Wallen, 2003). In some macaque species, mothers are careful to permit only older siblings to handle infants, and almost all of these are daughters. Female immature gorillas also exhibit greater interest than males in infants (Meder,  1990). Likewise, in chimpanzees, female juveniles and young adolescents are more likely than their male counterparts to care for infants, especially their siblings (Watts & Pusey, 2002). In chimpanzees, females even engage in maternal pretend play- cradling and making nests for sticks as if they were infants (Kahlenberg & Wrangham, 2010). Across cultures, in humans too, beginning ­between 2 and 6 years of age, girls spend more time than boys caring for their infant siblings. Girls also work harder than boys helping their mothers with child care and other household responsibilities especially when their mothers require more help, such as in poorer households (Lancy, 2015). Further, the evidence suggests that between 3 and 11 years, girls enjoy caring for infants more than boys do (Edwards, 2002; Lancy, 2015; Whiting et al., 1988). Thus, it appears that girls are not simply being more compliant with commands to look after infants, but actually take pleasure in the activity.

Conclusion and Future Directions

This review outlines several principles that apply across multiple primate species from the Cerco­ pithecidae and Hominidae families. The first principles concern rate of development. Immature ­females develop more rapidly than males and maintain closer proximity to mothers than males. Males also Benenson

37

experience a prolonged adolescence compared with females. Another set of principles apply to the establishment of dominance ranks. Immature males more than ­females compete in one-on-one contests over status. Outcomes appear to be based on a number of developing physical characteristics, including age, size, strength, and motivation, as well as social and cognitive skills that are learned. Ranks fluctuate as the characteristics that determine outcomes develop. In contrast, immature females’ status is relatively constant, based on their mothers’ rank when mothers are present as coalition partners, or on age and time spent in the community. Finally, a third group of principles describe relationships with same-sex peers. Immature males spend more time than females with same-sex peers, form larger same-sex peer groups, engage in more social play, and exhibit more direct aggression. Immature individuals of both sexes ­interact with adult males less than adult females, but immature males likely interact with adult males more than immature females do. In contrast, juvenile and adolescent females prefer to care for and play with infants more than their male counterparts do. This review includes only a relatively few species, and in particular those that are most closely related to humans. Moreover, even within the large family of Cercopithecidae, differences exist between species that do not always follow the principles that are outlined here. For example, male and female juvenile blue monkeys (Cercopithecus mitis) both remain in close proximity to mothers, and juvenile males avoid adult males (Cords, Sheehan, & Ekernas, 2010). Nonetheless, the principles regarding sex differences in social relationships apply across social structures, residence and dispersal patterns, mating systems, and dominance relations for most species. Some of the most exciting future research questions concern understanding differences in the ­socialization of females and males. Although sex differences in humans are regularly attributed to sex-typed socialization, little empirical evidence suggests the mechanisms by which this occurs. Examining the behaviors of other species in their natural habitats affords a way of testing hypotheses in a simpler environment that may apply to humans. Knowledge of closely related species can also ­illuminate phylogenetic adaptations or convergent ­solutions that can further understanding of human sex-typed behaviors. Finally, future research questions that seem particularly valuable to explore ­include elucidating the factors that differentially attract immature females versus males to adults of each sex, specifying the variables that promote 38

c­ ooperation between unrelated immature same-sex peers, examining the role that competition may play in relations between immature female same-sex peers, and uncovering the proximate processes that lead adolescents to emigrate to a new community.

References

Alvarez, H. (2004). Residence groups among hunter-gatherers: A  view of the claims and evidence for patrilocal bands. In B.  Chapais & C.  Berman (Eds.), Kinship and behavior in primates (pp. 400–442). Oxford, UK: Oxford University Press. Archer, J. (2004). Sex differences in aggression in real-world settings: A meta-analytic review. Review of General Psychology, 8(4), 291–322. Archer, J. (2006). Cross-cultural differences in physical aggression between partners: A social-role analysis. Personality and Social Psychology Review, 10(2), 133–153. doi:10.1207/ s15327957pspr1002_3 Arora, N., Van Noordwijk, M.  A., Ackermann, C., Willems, E.  P., Nater, A., Greminger, M., . . . Pamungkas, J. (2012). Parentage-based pedigree reconstruction reveals female matrilineal clusters and male-biased dispersal in nongregarious Asian great apes, the Bornean orang-utans (Pongo pygmaeus). Molecular Ecology, 21(13), 3352–3362. Barbu, S., Cabanes, G., & Le Maner-Idrissi, G. (2011). Boys and  girls on the playground: Sex differences in social development are not stable across early childhood. PloS One, 6(1), e16407. Barry, H., & Paxson, L. M. (1971). Infancy and early childhood: Cross-cultural codes 2. Ethnology, 10(4), 466–508. Bercovitch, F.  B. (1997). Reproductive strategies of rhesus macaques. Primates, 38(3), 247–263. Bernstein, I. S., Judge, P. G., & Ruehlmann, T. E. (1993). Sex differences in adolescent rhesus monkey (Macaca mulatta) behavior. American Journal of Primatology, 31(3), 197–210. Bissonnette, A., Perry, S., Barrett, L., Mitani, J. C., Flinn, M., Gavrilets, S., & de Waal, F.  B.  M. (2015). Coalitions in theory and reality: A review of pertinent variables and processes. Behaviour, 152(1), 1–56. Boehm, C. (1999). Hierarchy in the forest: Egalitarianism and the evolution of human altruism. Cambridge, MA; Harvard University Press. Boesch, C., & Boesch-Achermann, H. (2000). The chimpanzees of the Taï forest: Behavioural ecology and evolution. New York: Oxford University Press. Bogin, B. (1999). Evolutionary perspective on human growth. Annual Review of Anthropology, 28, 109–153. Bouchard, C., Trudeau, N., Sutton, A., Boudreault, M-C., & Deneault, J. (2009). Gender differences in language development in French Canadian children between 8 and 30 months of age. Applied Psycholinguistics, 30(4), 685–707. Brown, G.  R., & Dixson, A.  F. (2000). The development of behavioural sex differences in infant rhesus macaques (Macaca mulatta). Primates, 41(1), 63–77. Cabrera, N.  J., & Tamis-LeMonda, C.  S. (2013). Handbook of father involvement: Multidisciplinary perspectives. New York: Routledge. Campbell, A. (1999). Staying alive: Evolution, culture, and women’s intrasexual aggression. Behavioral and Brain Sciences, 22(2), 203–252. doi:10.1017/s0140525x99001818 Chapais, B., & Berman, C. M. (2004). Kinship and behavior in primates. New York: Oxford University Press.

Sex Differences in Primate Social Rel ationships

Chapais, B., & Gauthier, C. (2002). Early agonistic experience and the onset of matrilineal rank acquisition in Japanese macaques. In M.  Pereira & L.  Fairbanks (Eds.), Juvenile primates: Life history, development and behavior (pp. 245–258). Chicago: University of Chicago Press. Charlesworth, W.  R., & La Freniere, P. (1983). Dominance, friendship, and resource utilization in preschool children’s groups. Ethology and Sociobiology, 4(3), 175–186. Clutton-Brock, T.  H., & Huchard, E. (2013). Social com­ petition and selection in males and females. Philosophical Transactions of the Royal Society B: Biological Sciences, 368(1631), 20130074. Coie, J. D., & Dodge, K. A. (1998). Aggression and antisocial behavior. In W. Damon & N. Eisenberg (Eds.), Handbook of child psychology (Vol. 5, pp. 779–862). Hoboken, NJ: John Wiley & Sons. Conger, R. D., Patterson, G. R., & Ge, X. (1995). It takes two to replicate: A mediational model for the impact of parents’ stress on adolescent adjustment. Child Development, 66(1), 80–97. Cords, M., & Aureli, F. (2002). Patterns of reconciliation among juvenile long-tailed macaques. In M. Pereira & L. Fairbanks (Eds.), Juvenile primates: Life history, development and behavior (pp. 271–284). Chicago: University of Chicago Press. Cords, M., Sheehan, M. J., & Ekernas, L. S. (2010). Sex and age differences in juvenile social priorities in female philopatric, nondespotic blue monkeys. American Journal of Primatology, 72(3), 193–205. Daly, M., & Wilson, M. (2001). Risk-taking, intrasexual compe­ tition, and homicide. Nebraska Symposium on Motivation, 47, 1–36. de Vries, A.  L.  C., Steensma, T.  D., Cohen-Kettenis, P.  T., VanderLaan, D.  P., & Zucker, K.  J. (2016). Poor peer relations predict parent- and self-reported behavioral and emotional prob­lems of adolescents with gender dysphoria: A cross-national, cross-clinic comparative analysis. European Child & Adolescent Psychiatry, 25(6), 579–588. doi:10.1007/ s00787-015-0764-7 Dickemann, M. (1979). The ecology of mating systems in hypergynous dowry societies. Social Science Information, 18(2), 163–195. Dunbar, R. I. M. (1988/2013). Primate social systems. Beckenham, Kent: Croom Helm. Edwards, C.  P. (2002). Behavioral sex differences in children of  diverse cultures: The case of nurturance to infants. In M.  Pereira & L.  Fairbanks (Eds.) Juvenile primates: Life history, development, and behavior (pp. 327–338). Chicago, IL: University of Chicago Press. Emlen, S. T., & Oring, L. W. (1977). Ecology, sexual selection, and the evolution of mating systems. Science (New York, N.Y.), 197(4300), 215–223. Fabes, R.  A., Martin, C.  L., & Hanish, L.  D. (2003). Young children’s play qualities in same, other, and mixed sex peer groups. Child Development, 74(3), 921–932. Fagen, R. (1993). Primate juveniles and primate play. In M.  Pereira & L.  Fairbanks (Eds.), Juvenile primates: Life history, development, and behavior (pp. 182–196). Chicago: University of Chicago Press. Fairbanks, L.  A. (2002). Juvenile vervet monkeys: Establishing relationships and practicing skills for the future. In M. Pereira & L.  Fairbanks (Eds.) Juvenile primates: Life history, development and behavior (pp. 211–227). Chicago, IL: University of Chicago Press.

Fine, G.  A. (1980). The natural history of preadolescent male friendship groups. In H.  Foot, C.  Chapman, & J.  Smith (Eds.), Friendship and social relations in children (pp. 293–320). Piscataway, NJ: Transaction Publishers. Foerster, S., Franz, M., Murray, C. M., Gilby, I. C., Feldblum, J.  T., Walker, K.  K., & Pusey, A.  E. (2016). Chimpanzee females queue but males compete for social status. Scientific Reports, 6, 35404. Fouts, H.  N., Hallam, R.  A., & Purandare, S. (2013). Gender segregation in early-childhood social play among the Bofi foragers and Bofi farmers in Central Africa. American Journal of Play, 5(3), 333. Fröhlich, M., Wittig, R. M., & Pika, S. (2016). Play-solicitation gestures in chimpanzees in the wild: Flexible adjustment to social circumstances and individual matrices. Royal Society Open Science, 3(8), 160278. Fry, D. (2005). Rough-and-tumble social play in humans. In A. P. Smith & K. Peter (Eds.), The nature of play (pp. 54–85). New York, NY: Guilford. Furuichi, T., Idani, G., Ihobe, H., Hashimoto, C., Tashiro, Y., Sakamaki, T., . . . Kuroda, S. (2012). Long-term studies on wild bonobos at Wamba, Luo Scientific Reserve, DR Congo: Towards the understanding of female life history in  a malephilopatric species. In  P.  Kappeler & D.  Watts (Eds.) Longterm field studies of primates (pp. 413–433). Berlin: Springer. Galdikas, B. M. F. (1995). Reflections of Eden: My life with the orangutans of Borneo. New York: Little Brown. Goldberg, S., & Lewis, M. (1969). Play behavior in the year-old infant: Early sex differences. Child Development, 40, 21–31. Goodall, J. (1986). The chimpanzees of Gombe: Patterns of behavior. Cambridge, MA: Harvard University Press. Harcourt, A. H., & Stewart, K. J. (2007). Gorilla society: Conflict, compromise, and cooperation between the sexes. Hawthorne, NY: Aldine de Gruyter. Herman, R.  A., Measday, M.  A., & Wallen, K. (2003). Sex differences in interest in infants in juvenile rhesus monkeys: Relationship to prenatal androgen. Hormones and Behavior, 43(5), 573–583. Hochberg, Z. (2011). Evo-devo of child growth: Treatise on child  growth and human evolution. New York, NY: WileyBlackwell. Hohmann, G., & Fruth, B. (2002). Dynamics in social organization of bonobos (Pan paniscus). In C. Boesch, G. Hohmann, & L. Marchant (Eds.) Behavioural diversity in chimpanzees and bonobos (pp. 138–150). Cambridge University Press. Jacklin, C.  N., & Maccoby, E.  E. (1978). Social behavior at thirty-three months in same-sex and mixed-sex dyads. Child Development, 49(3), 557–569. Kahlenberg, S. M., & Wrangham, R. W. (2010). Sex differences in chimpanzees’ use of sticks as play objects resemble those of children. Current Biology, 20(24), R1067–R1068. Kanō, T. (1992). The last ape: Pygmy chimpanzee behavior and ecology. Palo Alto: Stanford University Press. Kappeler, P.  M., & van Schaik, C.  P. (2002). Evolution of primate social systems. International Journal of Primatology, 23(4), 707–740. Kimura, D. (2000). Sex and cognition. Cambridge, MA: MIT Press. Konner, M. (2010). The evolution of childhood: Relationships, emotion, mind. Cambridge, MA: Belknap Press. Kulik, L., Amici, F., Langos, D., & Widdig, A. (2015). Sex differences in the development of social relationships in rhesus macaques (Macaca mulatta). International Journal of Primatology, 36(2), 353–376.

Benenson

39

Lamb, M. E. (2004). The role of the father in child development. Hillsdale, NJ: John Wiley & Sons. Lancy, D. (2015). The anthropology of childhood. New York: Cambridge University Press. Leaper, C., Anderson, K. J., & Sanders, P. (1998). Moderators of gender effects on parents’ talk to their children: A ­metaanalysis. Developmental Psychology, 34(1), 3–27. doi:10.1037/ 0012-1649.34.1.3 Lee, P. C., & Johnson, J. A. (1992). Sex differences in alliances, and the acquisition and maintenance of dominance status among immature primates. In A.  H.  Harcourt & F.  B.  M. de Waal (Eds.), Coalitions and alliances in humans and other animals (pp. 391–414). Oxford, UK: Oxford University Press. Lonsdorf, E., Markham, A. C., Heintz, M. R., Anderson, K. E., Ciuk, D.  J., Goodall, J., & Murray, C.  M. (2014). Sex differences in wild chimpanzee behavior emerge during infancy. PLoS One, 9(6), 9. doi:10.1371/journal.pone.0099099 Lonsdorf, E., Eberly, L. E., & Pusey, A. E. (2004). Sex differences in learning in chimpanzees. Nature, 428(6984), 715–716. Lonsdorf, E.  V., & Ross, S.  R. (2012). Socialization and development of behavior. In J. Mitani, J. Call, P. Kappeler, R. Palombit, & J. Silk (Eds.), The evolution of primate societies (pp. 245–268). Chicago: University of Chicago Press. Lundberg, S. (2005). Sons, daughters, and parental behaviour. Oxford Review of Economic Policy, 21(3), 340–356. Maccoby, E. E. (1998). The two sexes: Growing up apart, coming to­gether. Cambridge, MA: Belknap Press/Harvard University Press. Maestripieri, D., & Ross, S.  R. (2004). Sex differences in play among western lowland gorilla (Gorilla gorilla gorilla) infants: Implications for adult behavior and social structure. American Journal of Physical Anthropology, 123(1), 52–61. Marmot, M. G., & Wilkinson, R. G. (2006). Social determinants of health. Oxford, UK: Oxford University Press. Meaney, M. J., Stewart, J., & Beatty, W. W. (1985). Sex differences in social play: The socialization of sex roles. Advances in the Study of Behavior, 15, 1–58. Meder, A. (1990). Sex differences in the behaviour of immature captive lowland gorillas. Primates, 31(1), 51–63. Meredith, S.  L. (2013). Identifying proximate and ultimate causation in the development of primate sex-typed social behavior. In Building babies (pp. 411–433). Berlin: Springer. Mitchell, G. (1979). Behavioral sex differences in non-human primates. New York: Van Nostrand Reinhold. Moffitt, T.  E., Caspi, A., Rutter, M., & Silva, P.  A. (2001). Sex  differences in antisocial behaviour: Conduct disorder, delinquency, and violence in the Dunedin Longitudinal Study. New York, NY: Cambridge University Press. Moscovice, L. R., Douglas, P. H., Martinez-Iñigo, L., Surbeck, M., Vigilant, L., & Hohmann, G. (2017). Stable and fluc­ tuating social preferences and implications for cooperation among female bonobos at LuiKotale, Salonga National Park, DRC. American Journal of Physical Anthropology, 163, 158–172. Munroe, R. L., & Romney, A. K. (2006). Gender and age differences in same-sex aggregation and social behavior: A four-culture study. Journal of Cross-Cultural Psychology, 37(1), 3–19. Murray, C. M., Lonsdorf, E. V., Stanton, M. A., Wellens, K. R., Miller, J. A., Goodall, J., & Pusey, A. E. (2014). Early social exposure in wild chimpanzees: Mothers with sons are more gregarious than mothers with daughters. Proceedings of the National Academy of Sciences of the United States of America, 111(51), 18189–18194.

40

Nadler, R. D., & Braggio, J. T. (1974). Sex and species differences in captive reared juvenile chimpanzees and orang-utans. Journal of Human Evolution, 3(6), 541–550. Nishida, T. (1988). Development of social grooming between mother and offspring in wild chimpanzees. Folia Primatologica, 50(1–2), 109–123. Ostner, J., & Schülke, O. (2014). The evolution of social bonds in primate males. Behaviour, 151(7), 871–906. Parke, R.  D. (1996). Fatherhood. Cambridge, MA: Harvard University Press. Prüfer, K., Munch, K., Hellmann, I., Akagi, K., Miller, J.  R., Walenz, B., . . . Winer, R. (2012). The bonobo genome compared with the chimpanzee and human genomes. Nature, 486(7404), 527–531. Pusey, A. (1990). Behavioural changes at adolescence in chimpanzees. Behaviour, 115(3), 203–246. Pusey, A., Williams, J., & Goodall, J. (1997). The influence of  dominance rank on the reproductive success of female chimpanzees. Science (New York, N.Y.), 277(5327), 828–831. Rijksen, H.  D. (1978). A field study on Sumatran orang utans (Pongo pygmaeus abelii Lesson 1827). Ecology, behaviour and conservation. Unpublished doctoral dissertation. Wageningen, The Netherlands. Robbins, M.  M., Bermejo, M., Cipolletta, C., Magliocca, F., Parnell, R. J., & Stokes, E. (2004). Social structure and lifehistory patterns in western gorillas (Gorilla gorilla gorilla). American Journal of Primatology, 64(2), 145–159. doi:10.1002/ ajp.20069 Rodseth, L., Wrangham, R. W., Harrigan, A. M., & Smuts, B. B. (1991). The human community as a primate society. Current Anthropology, 32(3), 221–254. doi:10.1086/203952 Roney, J.  R., & Maestripieri, D. (2003). Social development and  affiliation. In D.  Maestripieri (Ed.) Primate psychology (pp. 171–204). Cambridge, MA: Harvard University Press. Ruble, D.  N., Martin, C.  L., & Berenbaum, S.  A. (2006). Gender development. In N.  Eisenberg, W.  Damon, R.  M.  Lerner, N.  Eisenberg, W.  Damon, & R.  M.  Lerner (Eds.), Handbook of child psychology: Social, emotional, and personality development (Vol. 3, 6th ed., pp. 858–932). Hoboken, NJ: John Wiley & Sons. Sapolsky, R.  M. (2004). Social status and health in humans and  other animals. Annual Review of Anthropology, 33, 393–418. Savin-Williams, R.  C. (1980). Social interactions of adolescent females in natural groups. In H. C. Foot, A. J. Chapman, & J. R. Smith (Eds.), Friendship and social relations in children (pp. 343–364). New Brunswick, NJ: Transaction. Schlegel, A., & Barry III, H. (1991). Adolescence: An anthropological inquiry. New York, NY: Free Press. Schülke, O., Bhagavatula, J., Vigilant, L., & Ostner, J. (2010). Social bonds enhance reproductive success in male macaques. Current Biology, 20(24), 2207–2210. Sear, R., & Mace, R. (2008). Who keeps children alive? A review of the effects of kin on child survival. Evolution and Human Behavior, 29(1), 1–18. Sen, A. (1990). More than 100 million women are missing. New York Review of Books, 37, 123–149. Silk, J. B. (2009). Nepotistic cooperation in non-human primate groups. Philosophical Transactions of the Royal Society B: Biological Sciences, 364(1533), 3243–3254. Silk, J. B., Alberts, S. C., & Altmann, J. (2003). Social bonds of female baboons enhance infant survival. Science, 302(5648), 1231–1234.

Sex Differences in Primate Social Rel ationships

Silk, J. B., Altmann, J., & Alberts, S. C. (2006). Social relationships among adult female baboons (Papio cynocephalus) I. Variation in the strength of social bonds. Behavioral Ecology and Sociobiology, 61(2), 183–195. Silk, J.  B., & Boyd, R. (1983). Cooperation, competition, and  mate  choice in matrilineal macaque groups. In Social behavior of female vertebrates (pp. 316–348). New York: Academic Press. Smith, H.  J. (2005). Parenting for primates. Cambridge, MA: Harvard University Press. Sosis, R., Kress, H.  C., & Boster, J.  S. (2007). Scars for war: Evaluating alternative signaling explanations for cross-cultural variance in ritual costs. Evolution and Human Behavior, 28(4), 234–247. Sterck, E. H. M., Watts, D. P., & van Schaik, C. P. (1997). The evolution of female social relationships in nonhuman primates. Behavioral Ecology and Sociobiology, 41(5), 291–309. Suomi, S. J. (1997). Early determinants of behaviour: Evi­dence from primate studies. British Medical Bulletin, 53(1), 170–184. Tokuyama, N., & Furuichi, T. (2016). Do friends help each other? Patterns of female coalition formation in wild bonobos at Wamba. Animal Behaviour, 119, 27–35. Trivers, R.  L., & Willard, D.  E. (1973). Natural selection of parental ability to vary the sex ratio of offspring. Science, 179(4068), 90–91.

Troll, L. E. (1987). Gender differences in cross-generation networks. Sex Roles, 17(11), 751–766. van Adrichem, G. G. J., Utami, S. S., Wich, S. A., van Hooff, J. A., & Sterck, E. H. M. (2006). The development of wild immature Sumatran orangutans (Pongo abelii) at Ketambe. Primates, 47(4), 300–309. Wallen, K. (1996). Nature needs nurture: The interaction of hormonal and social influences on the development of behavioral sex differences in rhesus monkeys. Hormones and Behavior, 30(4), 364–378. Wallen, K. (2005). Hormonal influences on sexually differen­tiated behavior in nonhuman primates. Frontiers in Neuroendocrino­ logy, 26(1), 7–26. Watts, D.  P. (2000). Causes and consequences of variation in male mountain gorilla life histories and group membership. In P. Kappeler (Ed.) Primate males (pp. 169–179). Cambridge, UK: Cambridge University Press. Watts, D.  P., & Pusey, A.  E. (2002). Behavior of juvenile and adolescent great apes. In Juvenile primates: Life history, development, and behavior (2nd ed., pp. 148–167). Chicago, IL: University of Chicago Press. Whiting, B.  B., Edwards, C., Ember, C.  R., Erchak, G.  M., Harkness, S., Munroe, R. L., . . . Seymour, S. (1988). Children of different worlds: The formation of social behavior. Cambridge, MA: Harvard University Press.

Benenson

41

CH A PT E R

4

Sex Differences in Cognition Evidence for the Organizational–Activational Hypothesis

Elizabeth Hampson

Abstract Organizational and activational effects of sex steroids were first discovered in laboratory animals, but these concepts extend to hormonal actions in the human central nervous system. This chapter begins with a brief overview of how sex steroids act in the brain and how the organizational-activational hypothesis originated in the field of endocrinology. It then reviews common methods used to study these effects in humans. Interestingly, certain cognitive functions appear to be subject to modification by sex steroids, and these endocrine influences may help explain the sex differences often seen in these functions. The chapter considers spatial cognition as a representative example because the spatial family of functions has received the most study by researchers interested in the biological roots of sex differences in cognition. The chapter reviews evidence that supports an influence of both androgens and estrogens on spatial functions, and concludes with a glimpse of where the field is headed. Keywords:  cognitive function, spatial, estrogen, androgen, estradiol, mental rotation

Over the past few decades, evidence from laboratory animals has amply demonstrated that sex steroids, through their actions at various points in the lifespan, affect the morphology and function of the central nervous system (CNS). These effects are exerted in a region-specific manner in the CNS and underlie many sexual dimorphisms that have been identified in other species, including differences that are expressed at a behavioral level. Although these so-called organizational and activational effects of steroids are widespread in the CNS of laboratory animals, their application to the human brain is less well mapped. This chapter will survey the principles of sex steroid action that have been derived from animal research and then discuss their translation to understanding certain sex differences found in humans. Nowhere has this translation been more controversial than in understanding whether these fundamental principles can be applied to higher order cortical functions, including processes related to cognition.

How Do Hormones Act in the Brain?—A Brief Overview

Steroid hormones are small lipid-based molecules that have a characteristic four-ring (“sterane”) structure and are derived from cholesterol. They can easily diffuse across cell membranes, including neurons, but are able to exert physiological effects only where cells express receptors for a particular steroid. Most of the major hormones produced by the gonads are steroids, including testosterone (T) and its more potent offspring, dihydrotestosterone (DHT). Both T and DHT act largely by binding to the androgen receptor. The estrogen 17β-estradiol (which we will simply call “estradiol”) is a steroid too, and is the dominant form of estrogen produced by the ovaries in women of reproductive age. It acts by binding to estrogen receptors (ERs), of which there are at least three types (Galea, Frick, Hampson, Sohrabji, & Choleris,  2017). Progesterone is another gonadal steroid, but we will have less to say about it in the

43

present chapter because it has not been prominently implicated in cognition (for a review of its known effects in the CNS, see Schumacher et al., 2014). The classic “genomic” mechanism of steroid hormone action is one whereby a particular steroid (such as DHT, for example) crosses from the bloodstream where it is carried, into cells, then binds with intracellular receptors if present (in our example the type of receptor that would need to be present is the androgen receptor). The steroid–receptor complex is then translocated to the cell’s nucleus, where it binds to acceptor sites on the DNA and initiates an increase or decrease in local gene transcription. The ability to regulate gene transcription is an enormously powerful mechanism, allowing steroids to control a vast array of gene products made by cells. If the cell is a neuron, a steroid might influence structural proteins made by the cell or the production of particular enzymes (such as those involved in neurotransmitter synthesis or metabolism). In recent years, it has been discovered that certain steroids (e.g., estradiol) act not only through classic genomic mechanisms but also by binding to membrane receptors associated with the plasma membrane (see Galea et al., 2017; Schumacher et al.,  2014). Membrane receptors are believed to mediate the rapid effects of steroids that are sometimes seen, which occur too rapidly to be mediated by a genomic mechanism. Thus, steroid hormones can act via both genomic and sometimes nongenomic mechanisms to influence cellular activity. Importantly, a number of different steroids can exert effects within the CNS. This is possible because particular regions of the brain seem to be programmed to express steroid receptors. Animal studies have demonstrated that receptors are not expressed equivalently across all parts of the brain, nor are they necessarily expressed at all points in an animal’s lifespan, but when present they allow for effects to occur under the guidance of particular steroids that can either permanently or temporarily produce alterations in brain structure and function (Breedlove & Hampson, 2002). These are called the organizational and activational effects of sex steroids, and we will now describe the historical origins of this highly influential concept in further detail.

Origins of the Organizational–Activational Hypothesis

The organizational hypothesis traces its roots to several important landmark studies published from around 1960 to the early 1980s. A publication on the guinea pig by Phoenix, Goy, Gerall, and Young 44

Sex Differences in Cognition

(1959) was a major turning point in the study of sex differences and provided the conceptual framework for two major distinct classes of steroid effects. Briefly, Phoenix and colleagues found that treating female guinea pigs with T prenatally masculinized the pattern of reproductive behaviors they showed as adults—when primed in adulthood with a stimulus dose of hormone to elicit sexual responding, females exposed to T prenatally displayed less lordosis and greater mounting than female controls. In other words, the treated females showed both defeminization and masculinization of their sexual behavior as a consequence of exposure to T during the developmental period when the brain was organizing. The effects that resulted were permanent and could not be duplicated by treatment with T postnatally or in adulthood. Phoenix and his colleagues suggested that T’s effects were “organizational” in that they influenced the differentiation of sites within the CNS: “testosterone or some metabolite acts on those central nervous tissues in which patterns of sexual behaviour are organized” (Phoenix et al., 1959, p. 381). Thus, the concept of “organizational effects” was born. It gave rise to the modern concept that certain regions of the brain are developmentally modified by specific sex steroids if those steroids are available during a particular temporal window in early development. Once induced, the effects are long-lasting (usually permanent). Although we do not always know where to look in the brain to see the molecular or cellular changes that are the substrate for organizational effects apparent at the behavioral level, some type of permanent neural change is caused by the steroid exposure and is a basis for the surface sex difference we see expressed. The organizational hypothesis has evolved over time to add a number of refinements (see Arnold, 2009), such as the fact that T may exert its organizing effects via conversion to a metabolite (estradiol or DHT), the fact that masculinization and defeminization are orthogonal processes, and the new and fascinating prospect that puberty might be another period in development when the CNS becomes sensitive to the organizing effects of steroids. However, the basic idea of long-lasting effects on neuronal differentiation that take place under steroid guidance during prenatal or early postnatal development remains at the core of  the organizational concept. Processes such as neurite outgrowth, the amount of programmed cell death (apoptosis), the numbers and types of synapses formed, synaptic pruning, and possibly cell migration and specification of cell type or epigenetic

changes are among the developmental processes subject to the organizational effects of sex steroids (Arnold, 2009). Phoenix did not identify a particular brain region that might underlie the effects he observed, but other landmark studies soon followed. Sex differences in neural structure were discovered in the 1970s. One of the earliest studies was an electron microscopy study by Raisman and Field (1973), who discovered that female rats had more of a particular synapse type in the preoptic area (POA) of the hypothalamus than male rats did. In accordance with the organizational hypothesis, manipulations of early androgen exposure were able to reverse the sex difference. Sexual dimorphisms in the brain were also discovered to exist at a macroscopic level, such as the sexually dimorphic nucleus of the preoptic area (SDN-POA) of the rat, a prominent dimorphism first identified by Roger Gorski’s group at UCLA (Gorski, Gordon, Shryne, & Southam, 1978). The SDN-POA is now considered a classic model of how organizational effects of steroids can regulate hypothalamic differentiation in the perinatal period (see Breedlove & Hampson, 2002, for a review). The discovery of the spinal nucleus of the bulbocavernosus (Breedlove & Arnold, 1980) showed that sexual dimorphism resulting from ­steroid exposure was not limited to the hypothalamus, and revealed further new insights into the vagaries of organizational processes (Breedlove & Hampson, 2002). Although less explored in the early days of the field than organizational effects, activational effects of steroid hormones were also described by Phoenix et al. (1959). Notably, for sexual behavior to be triggered in adult guinea pigs, it was necessary to treat the adult animals acutely with estradiol benzoate (followed by progesterone) or with testosterone ­propionate to stimulate the expression of mating behavior. Phoenix et al. (1959) recognized that they were looking at a second type of steroid effect, where adult treatment elicited effects that were transient, reversible, with no lasting residues, and whose end result was to “activate” specific neural pathways in order for a particular behavior to be expressed: “Adulthood, when gonadal hormones are being secreted, is a period of activation; neural tissues are the target organs and mating behaviour is brought to expression” (Phoenix et al., 1959, pp. 379–380). Phoenix accordingly dichotomized hormonal effects as either organizational or activational. We now understand that activational effects do not require a

critical window, are reversible, are temporally linked to the presence of steroid in the bloodstream, and involve the modulation of neuronal activity in preexisting target pathways via genomic regulation of neurotransmitter function (McEwen & Alves, 1999), or via dynamic changes in the synaptic contacts between neurons, or via rapid nongenomic signaling mechanisms (McEwen, Akama, Spencer-Segal, Milner, & Waters, 2012). Although the exact details of the physiological processes have been refined and elaborated since Phoenix’s era, the conceptual distinction he proposed between two broad classes of steroid effects in the brain has withstood the test of time. It is a theoretical framework that has been found to explain a large majority of the phenotypic sex differences seen in brain and behavior across many mammalian species. Because of their wide applicability, we might expect the same principles to apply to the human brain.

Do Organizational and Activational Effects Apply to Extra-Hypothalamic Brain Areas?

The earliest tests of the organizational–activational hypothesis were limited to investigating the effects of various endocrine manipulations on behaviors closely related to reproduction and mating, and to the sexual differentiation of the hypothalamus. An important question, therefore, was whether such effects extended to nonreproductive functions and to brain regions outside the hypothalamus. An early clue that organizational and activational effects would not be limited to the hypothalamus came from songbirds such as zebra finches who use learned songs as part of their courtship rituals. In these species, song production is sexually dimorphic, with males using song to attract female mates. Correspondingly, the volumes of several ­regions involved in birdsong display substantial sex differences and contain more neurons in males than females (Nottebohm & Arnold,  1976). Studies of the zebra finch showed that sexual dimorphism in the size of these nuclei arises from exposure to hormones (estradiol) early in development, and that adult singing is then activated by androgens acting upon a male-typical neural substrate that was laid down during the ­organizational period (Gurney & Konishi,  1980). However, studying organizational effects in songbirds has proven to be complex, with some endocrine manipulations inexplicably having little or no effect, leading to suggestions that elements of sexual differentiation in birds may proceed Hampson

45

under the direct guidance of genes (“cell autonomous effects”; Arnold, 2009). Although the learning and production of birdsong can be construed as “cognitive” and clearly involves nonhypothalamic brain regions, the prospect of direct gene effects complicates its interpretation as a straightforward model of organizational effects on cognition. Both organizational and especially activational effects in extra-hypothalamic brain regions have now been confirmed by experimental and observational studies using rodents or nonhuman primates as subjects. Examples of androgen- and estrogendriven effects on brain physiology are so numerous that a 40-page review can now be written on the effects of a single hormone (e.g., estradiol) in a single neurotransmitter system (e.g., s­erotonin; Bethea, Lu, Gundlah, & Streicher, 2002). Reversible effects of estradiol have been found to extend beyond the neurochemical level to include features of brain microanatomy, such as the regional densities of dendritic spines (Woolley & McEwen,  1992) or rates of neurogenesis in the hippocampus of the adult rat (Galea et al.,  2013; Tanapat, Hastings, Reeves, & Gould, 1999). However, it is not always clear from basic studies exactly how the effects reported translate to the functional level (i.e., cognitive or behavioral counterparts). Progress in documenting organizational effects outside of the hypothalamus has been slower, because of the practical challenges of studying ­organizational effects and of knowing where to look. Presumably, functions subject to the effects of steroids would be those that show sex differences, but cognitive sex differences are less well characterized in nonhuman species than in humans, and cognition (especially in rodents) is often conceptualized in a less finely fractionated way. Still, early suggestions of an effect of neonatal gonadectomy on cortical thickness in the rat (Stewart & Kolb, 1988), discovery of high levels of aromatase activity in the association cortex of the developing rhesus monkey (MacLusky, Naftolin, & Goldman-Rakic, 1986), and observations of sexual dimorphism in the dentate gyrus of the mouse (Roof & Havens, 1992; Wimer & Wimer, 1989) all indicated that organizational effects might potentially extend to the differentiation of cortical and limbic regions, implying that effects of steroids on higher order functions, including cognitive functions, might be found. Numerous studies in rats and mice have now demonstrated a wide range of effects on learning and memory. Hippocampally mediated forms of memory have been an especially active focus. Recent 46

Sex Differences in Cognition

work has begun to explore the effects of estrogens in adult animals on increasingly sophisticated tasks, including measures of social cognition, and has implicated the rapid effects of estrogens and not merely their classic genomic mode of action in generating these effects (see Galea et al.,  2017, for an overview). For instance, infusion of 17β-estradiol or the ERα agonist PPT into the dorsal hippocampus of the mouse rapidly improves performance in social recognition, object recognition, and object placement tasks, whereas the ERβ agonist DPN is ineffective (Phan et al., 2015). In recent work, increasing attention is being paid to moderator variables such as the form of estrogen studied; dose dependency of the effects; distinctions between working and reference memory; defining whether hormonal manipulations act at the encoding, consolidation, or retrieval stages of memory processing; and identifying which form of the estrogen receptor mediates any effects observed. Older studies, however, which led to these modern advances, initially focused only generically on spatial memory. Spatial memory became a focal point of early work primarily because several types of spatial tasks (i.e., tasks that emphasize knowledge of object positions, accurate movement through space, or navigation to reach a distant target) commonly used to evaluate learning and memory in rodents elicit robust sex differences (making them candidates for organizational or activational influences or both). The radial-arm maze (RAM) and Morris water maze (see Figure 4.1) were commonly used in this early body of work to assess spatial learning. These two tasks involve, respectively, learning over a series of trials which arms of a maze are baited with food rewards and learning where a hidden platform (which serves as a refuge) is located within a circular enclosure that is filled with opaque water in which a rat is initially placed. Both tasks depend on hippocampal function and both are acquired more rapidly by male than female rats ( Jonasson, 2005). (In c­ ontrast, the reverse pattern, more rapid acquisition by ­females, is seen for certain other forms of learning, such as active avoidance or classical trace eye-blink conditioning; Dalla & Shors, 2009). Thus, the presence and direction of a sex difference on a cognitive task is dependent on the particular cognitive processes required and the set of brain regions brought to bear. Consistent with the idea that activational effects of estrogens can exert a modulatory influence on hippocampal function, the performance of female rats on tests of hippocampally d ­ ependent spatial

Figure 4.1  An illustration of the radial-arm maze (top) and Morris water maze (bottom), commonly used to assess spatial learning and memory in laboratory animals. In the radial maze, food rewards are placed at the ends of the arms. All of the arms may be baited or only a fixed subset, allowing the experimenter to differentiate between working memory errors and reference memory errors. Entries into never-rewarded arms are reference memory errors; re-entering an arm already visited within a particular trial is a working memory error. In the water maze task, the animal is released from different vantage points around the circular periphery of the pool and over a series of spatial training trials learns to navigate directly to the location of the hidden escape platform (dashed line) submerged below the surface of the water, by using visual cues present in the environment to orient toward the platform location (which is fixed). Dependent variables typically measured in the water maze include latency to find the hidden platform and path length. An optional probe trial may be used to evaluate memory, where the refuge is removed following the spatial training. Time spent searching in the (now-empty) platform quadrant is then monitored.

memory appears to be influenced by estradiol concentrations over a rat’s estrous cycle. Some studies, primarily in mice, report improved spatial memory in the standard water maze paradigm under higher estrogen conditions,1 but most rat studies report lower performance in female rats when tested at proestrus (when endogenous production of ovarian hormones is highest) compared with estrus or diestrus (when hormones are low; Frye, 1995; Korol et al., 1994; Pompili, Tomaz, Arnone, Tavares, & Gasbarri, 2010; Warren & Juraska, 1997). Poorer water maze acquisition in high estrogen states is also  seen in meadow voles and deer mice (Galea, Kavaliers, Ossenkopp, & Hampson,  1995; Galea, Kavaliers, Ossenkopp, Innes, & Hargreaves,  1994). Plasma­ ­estradiol correlated positively with latency to locate the hidden platform (Galea et al., 1995), indicating poorer spatial performance. These effects have been confirmed by experiments where estradiol and 1   It should be noted that a small female, not male, advantage is evident in mice performing the water maze task (Jonasson, 2005).

progesterone were administered exogenously to ovariectomized female rats (Chesler & Juraska, 2000; Frye, 1995; Snihur, Hampson, & Cain,  2008) to simulate the hormonal conditions found naturally at proestrus. Some experiments have further suggested that progesterone and not just estradiol might contribute (e.g., Chesler & Juraska,  2000; Johansson, Birzniece, Lindblad, Olsson, & Bäckström, 2002). Some of estrogen’s effects are age dependent (i.e., different in young vs. old rats) or dosage dependent (e.g., Markham, Pych, & Juraska, 2002; Foster, Sharrow, Kumar, & Masse, 2003). The standard Morris water maze procedure emphasizes reference memory, but if the procedure is modified to emphasize the “unlearning” of previously learned platform positions by relocating the hidden platform each test day, then the performance of female rats is better if they are tested at high ­instead of low levels of gonadal steroids (Dohanich, Korol, & Shors, 2009). This modification might increase the recruitment of the frontal cortex during task performance (Sinopoli, Floresco, & Galea, 2006), which could explain the differing results. Higher estradiol also facilitates working memory in the ­ RAM (Bimonte & Denenberg, 1999; Fader et al., 1999; Pompili et al., 2010). Either way, a systematic change in performance under different hormonal conditions implies that an activational effect is present. A recent review concluded that hippocampal function is sensitive to changes in estrogens that occur over the estrous cycle, and with pregnancy, motherhood, or aging in the rat (Duarte-Guterman, Yagi, Chow, & Galea, 2015). These behavioral effects are thought to be mediated by effects of estrogens on hippocampal morphology and physiology. Consistent with this prospect, the rodent hippocampus expresses estrogen receptors, especially in the cornu ammonis (CA) subfield CA1, but also in CA3 and the dentate gyrus. Both types of classic ERs (ERα and ERβ) and the G-proteincoupled membrane estrogen receptor (GPER) are expressed in the hippocampus, making it an important target for estrogens. The hippocampus shows dynamic neuroplastic changes in dendritic spine densities in CA1 as a function of the estrous cycle (the number of spine synapses in CA1 is increased about 30 percent during proestrus in the rat; Woolley & McEwen,  1992), and the dentate is a site of estrogen-dependent rates of neurogenesis in adult brains (Duarte-Guterman et al., 2015). Some of estradiol’s effects on spines are initiated within minutes of estradiol application. These and other observations confirm that activational effects of Hampson

47

estrogens take place in the hippocampus at a cellular level. Modulatory effects of estradiol on hippocampal neurochemistry during place learning also contribute to spatial memory, and formal models have been proposed (Dohanich et al., 2009). Activational effects of androgens with respect to spatial learning have received less study. Early research on learning and memory suggested that removal of the testes in male rats (at age 20, 50, 90, 130, or 170 days) did not affect maze acquisition (Commins, 1932). This suggested that T is inert with respect to spatial memory in an activational sense. However, recent work on androgens has thrown this overly strong conclusion into doubt (Goudsmit, Van De Poll, & Swaab, 1990; Sandstrom, Kim, & Wasserman, 2006; Spritzer, Gill, Weinberg, & Galea, 2008) and suggests that effects may be especially prominent for working memory. Castration also causes a decline in dendritic spine density in hippocampal CA1, which can be reversed by the acute administration of T or DHT in male rats (Leranth, Petnehazy, & MacLusky, 2003). Whether or not adult T concentrations have an influence, perinatal exposure to T appears to be important for sexual differentiation of neural structures that support spatial learning in the rat. Early work involving neonatal hormone manipulations indicated a possible role for sex steroids (Dawson, Cheung, & Lau,  1975; Joseph, Hess, & Birecree, 1978; Stewart, Skvarenina, & Pottier,  1975), but was not followed up until 1990. Organizational effects on learning in the RAM were then elegantly demonstrated by Williams, Barnett, and Meck (1990), who showed that neonatal manipulations of sex steroids could systematically alter and even ­sex-reverse the adult levels of accuracy displayed by male and female rats in the RAM. Adult female rats exposed to estradiol during the first few days of life showed a level of acquisition in the RAM that resembled male controls, whereas male rats castrated on postnatal day 1 resembled female controls. Furthermore, the same manipulations altered rats’ reliance on geometry versus landmark cues when navigating in the RAM (Williams et al., 1990). (Differences in utilizing the two classes of visual cues is another, related, sexually dimorphic facet of spatial performance in the rat). Roof and Havens (1992) showed that rats exposed to T neonatally displayed visible changes in the hippocampal dentate gyrus. The width of the granule cell layer predicted the efficiency of acquisition in the reference memory version of the water maze in adulthood. In the rodent brain, some of these early organizational effects depend upon the 48

Sex Differences in Cognition

aromatization of T to e­ stradiol (Williams et al., 1990; but see Isgor & Sengelaub, 1998), although species differences exist and it is not clear that aromatization to estrogens is necessary for T to organize the primate brain. Compared with rodents, the effects of gonadal steroids on cognition in nonhuman primates are far less studied. There is preliminary support for an activational influence of estrogens on learning and memory (Lacreuse, Mong, & Hara, 2015), at least in aging female animals, but the research base is not extensive. In the first study of its kind, Lacreuse, Chiavetta, Shirai, Meyer, and Grow (2009) found that administration of testosterone enanthate to young male monkeys following gonadal suppression with a gonadotropin-releasing hormone (GnRH) agonist reduced their performance on a delayed nonmatching-to-sample task of recognition memory. This finding might suggest that androgens, not just estrogens, can exert activational effects within the medial temporal lobe. Even fewer studies of organizational effects are available from nonhuman primates. Early work suggested a possible role for T in the developing frontal cortex of the rhesus monkey (Clark & GoldmanRakic, 1989; see also MacLusky et al., 1986). Female monkeys treated with T propionate on postnatal days 1 to 46 showed male-like performance on an object reversal task dependent upon the orbitofrontal cortex when tested at 75 days of age (Clark & Goldman-Rakic, 1989). A temporary male a­ dvantage in object reversal can be seen in young monkeys (Clark & Goldman-Rakic,  1989) and young children (Overman, Bachevalier, Schuhmann, & Ryan, 1996). Although Clark and colleagues believed the sex difference to be only transient, a sex difference on a probabilistic reversal task has recently been observed in human adults (Evans & Hampson, 2015). Clearly, much work remains to be done, but, based on the limited data presently available, effects of reproductive steroids on learning and memory do appear to generalize to the primate brain.

Translating the Organizational–Activational Hypothesis to Humans

A number of reliable sex differences in cognition and behavior are found in humans (for reviews see Halpern, 2012; Maccoby & Jacklin, 1974). Unlike laboratory animals, however, who are raised under standardized controlled conditions, human behavior is more prominently influenced by variation in the social environment (e.g., differences in family upbringing or the experiences typical of girls vs.

boys in a given society). Therefore, even though a  body of data suggests that the organizational–­ activational hypothesis applies to humans too, social variables must always be considered. In principle, a sex difference in cognitive performance could be caused by an activational effect of sex steroids on brain function, an organizational effect of steroids exerted during early brain development, relevant experiential differences that exist between females and males due to cultural expectations, direct sex-linked gene effects, or any combination of these variables. The challenge of dissociating the influence of culture is most acute for researchers who study ­organizational effects. On the other hand, if a difference is systematically accentuated, is reduced, or disappears altogether when it is evaluated under various hormonal conditions in adults, then an activational effect of sex steroids may be implicated. Although sex differences have been reported in a number of human cognitive functions, only a few have been studied with respect to the ­organizational– activational hypothesis. In the following sections, we concentrate on spatial cognition as a core e­ xample, because in its various forms it has received the most study to date. One might at first think the emphasis on spatial cognition grew out of the animal studies of learning and memory discussed previously. In fact, though, in the human literature it grew out of a separate tradition, and its origins preceded all but the earliest animal studies. It might seem narrow to use spatial cognition as a key example, but the term spatial cognition encompasses a number of separable processes and brain regions, not all of which respond to sex steroids in the same way. Although we might wish that a wider range of cognitive processes had been studied, even a single example of a particular domain of human cognition where organizational or activational effects can be demonstrated is “proof of concept” that may encourage researchers to begin inquiring into other areas of the cognitive panorama.

Common Paradigms Used to Study the Organizational Hypothesis in Humans

In humans, it would be unethical to experimentally manipulate hormones prenatally to study the ­resulting effects on brain development. Accordingly, researchers interested in characterizing the effects of androgens on cognitive differentiation have studied clinical conditions in which androgen production is increased or decreased outside the normal range (i.e., “experiments of nature”). Foremost among these is the classical form of congenital adrenal hyperplasia (CAH; Merke & Bornstein, 2005).

CAH is an autosomal recessive condition where the adrenal cortex is deficient in the enzyme 21-hydroxylase. One consequence is elevated production of androgens by the adrenal glands, beginning in the third month of fetal gestation. Although d ­ifficult to detect in males, excess ­production in f­emales causes masculinization of the external genitalia, which varies in severity. The genital ambiguity is usually detected at birth or soon after, at which point the androgen overproduction can be normalized by introducing appropriate hormone replacement therapy. Most females with CAH begin treatment by about 2 weeks of postnatal age. Females with CAH usually undergo surgical correction of the genital ambiguity in infancy and are raised as females despite the increased androgen exposure before birth. Because psychosocial upbringing tends to be female typical (and often even more gender stereotypical than control females; Pasterski et al., 2005), CAH affords a unique opportunity to dissociate the effects of upbringing and biology. Studies of CAH have been supplemented recently by studies that involve other methodologies. One is the study of opposite-sex twins in which diffusion of androgens from the male co-twin is thought to expose the developing female fetus to higher than usual concentrations of androgens prenatally (i.e., higher than a female fetus would normally experience before birth). The source of T is the gonads of the male co-twin, which undergo activation from about weeks 8 to 24 of gestation (Forest, de Peretti, & Bertrand, 1976), a period when genital differentiation occurs in the normally developing male fetus. Although significantly lower than the levels of androgen exposure seen in CAH, the availability of small amounts of testicular-derived T is thought to potentially have an impact on the development of the female’s CNS. It should be noted that this mechanism is still somewhat speculative in humans, but there is a precedent in other species for the masculinization or defeminization of females’ behavior via a similar ­mechanism, when male and female ­fetuses develop in close proximity in utero (Ryan & Vandenbergh, 2002). In the past 20 years, it has been suggested that a  particular sexually dimorphic feature of digit (finger) differentiation, visible postnatally, can serve as a retrospective biological marker of the relative degree of T exposure that a normally developing fetus experienced prenatally (Manning, 2002). The so-called 2D:4D digit ratio is defined as the ratio of the lengths of the second to fourth digits of the Hampson

49

hand. When studied in adults using standardized measurement procedures, the 2D:4D ratio is said to reflect, in a graded fashion, the level of androgen activity during a particular period in utero that is important for differentiation of the hands. The validity of the 2D:4D ratio as an indicator of prenatal T level is controversial. Evidence for an impact of androgen levels on the ratio is based on data from CAH (Hönekopp & Watson,  2010), data from ­individuals with an inherited insensitivity of the androgen receptor (Berenbaum, Bryk, Nowak, ­ Quigley, & Moffat, 2009), and a few experiments in birds or mammals suggesting that the ratio is perceptibly influenced by differences in T availability during early development. However, counterevidence to the androgen theory also exists (e.g., Hampson & Sankar, 2012), and the issue of validity is not settled. Over the past 20 years, the 2D:4D ratio has been used in a significant number of studies as a gauge of prenatal T concentrations, but it must be taken under advisement until its validity (or lack thereof ) can be more conclusively established. Although it is an infrequently used paradigm, a few researchers have directly measured fetal T and/ or estradiol in amniotic fluid collected from expectant mothers, or have measured neonatal T from infant saliva collected after birth. A great advantage of these methods is the ability to measure hormones directly. However, for amniotic fluid the timing of sample collection is dictated by when amniocentesis is performed and typically only a single measurement (sample of fluid) is available for a given pregnancy. Nevertheless, direct sampling of hormone concentrations provides an important opportunity to assess whether or not hormone levels encountered at these time points are predictive of later cognitive abilities.

Evidence for an Organizational Effect of Androgens on Spatial Cognition

Many studies have used measures of mental rotation (i.e., the ability to dynamically manipulate an object in one’s mind, viewing it from different angles or points of view; see Figure 4.2) or other spatial tasks requiring visualization (e.g., transforming in one’s mind the three-dimensional shape of an object) to assess spatial performance, or have used either virtual or real-world tasks that require learning to navigate in an unfamiliar spatial environment. These types of tasks typically exhibit sex differences. On average, adult males as a group show superior performance relative to adult females, although there are very large individual differences within both sexes. Only 50

Sex Differences in Cognition

A

B

C

Figure 4.2  The original Shepard-Metzler stimuli, shown here, are used in several standard mental rotation tasks, including the Vandenberg and Kuse Mental Rotations Test (1978). (A) A “same” pair that differ by an 80-degree angular rotation in the picture plane. (B) A “same” pair that differ by an 80-degree rotation in the depth plane. (C) A “different” pair that cannot be aligned by any rotation. The Vandenberg and Kuse rotations employ similar figures, but the test uses a multiple-choice format. (From Shepard, R. N., & Metzler, J. [1971]. Mental rotation of three-dimensional objects. Science, 171, 701–703. Reprinted with permission from AAAS.)

the last of these carries a significant memory load, because the visual stimuli in the first two types of tasks normally remain in view throughout each experimental trial. Meta-analysis shows that mental rotation and spatial navigation tasks tend to elicit the largest sex differences, with tasks requiring three-dimensional mental rotation or navigating in a three-dimensional virtual environment exhibiting a mean sex difference on the order of d = 0.80 or above (Voyer, Voyer, & Bryden, 1995). Whereas learning to navigate in a new environment is a “spatial memory” task and elicits activity in the hippocampus (e.g., Pintzka, Evensmoen, Lehn, & Håberg, 2016), often accompanied by activation in the retrosplenial cortex (Goodrich-Hunsaker, Livingstone, Skelton, & Hopkins,  2010), mental rotation typically recruits sites in the parietal cortex, not the hippocampus (Zacks, 2008). Therefore, the tasks used in human studies are not always hippocampally dependent and do not necessarily parallel spatial findings obtained from other species. To ­interpret the varying effects of steroids across d ­ ifferent studies, we must take into account the differing brain regions recruited by different sorts of spatial tasks. Resnick, Berenbaum, Gottesman, and Bouchard (1986) were the first to demonstrate superior

­ erformance on mental rotation tasks in females p with the classical form of CAH (who are exposed to elevated androgens prenatally) compared to unaffected female controls. The effect was not seen in males with CAH, who did not differ from male controls. Spatial superiority in females with CAH has been replicated by several studies using mental rotation or other visualization tasks (Berenbaum, Korman Bryk, & Beltz, 2012; Hampson, Rovet, & Altmann, 1998), but the observation remains controversial because not all studies have observed these effects (e.g., Hines et al., 2003). A possible resolution is suggested by recent work from our lab (Hampson & Rovet, 2015); we discovered that superior spatial performance is not as evident in females with the simple-virilizing form of CAH as in patients who have the salt-­ wasting form, where the severity of androgen excess is usually much greater. (These are two phenotypic forms of classical CAH). Most studies do not permit separate analysis by subtype because of the limited sample sizes usually available (classical CAH affects 1 in 30,000 female births; Merke & Bornstein, 2005). Because the proportion of saltwasters varies considerably from one study to another, it might explain the variable findings across studies. In favor of this hypothesis, superior performance was also observed in females with saltwasting but not simple-virilizing CAH in a study that used a virtual Morris water maze to assess spatial learning (Mueller et al., 2008). Although the cognitive demands of the water maze differ from mental rotation and it recruits the hippocampus to a greater degree, both sets of findings support the hypothesis of an organizational effect of androgens on spatial functions. If an organizational effect does occur prenatally, then evidence for an androgen influence ought to be seen using other research methodologies. Although the effect size is much smaller than for CAH, twin studies have found better mental rotation capabilities in female twins who gestated in utero with a  male co-twin compared with those who had a female co-twin (Heil, Kavšek, Rolke, Beste, & Jansen, 2011; Vuoksimaa et al., 2010). It is unlikely that the improved spatial performance is due to psychosocial or experiential effects of having a brother because the effect was not seen in other, nontwin, girls who had brothers of a similar age (Heil et al., 2011). It is seldom possible to quantify prenatal T directly, but a significant correlation between the level of T in second-trimester amniotic fluid and efficiency of mental rotation was found in a sample

of healthy girls (offspring of the sampled pregnancies) who were tested on a children’s mental rotation test at age 7 (Grimshaw, Sitarenios, & Finegan, 1995). This important study has yet to be replicated. However, Auyeung et al. (2012) found a significant correlation between T concentration in second-­ trimester amniotic fluid and another type of spatial test (but not mental rotation) in children who were tested at age 7 to 10 years. Ideally mental rotation would be assessed at older ages, as not all children are yet capable of doing the rotations by age 7. Given the twin findings, it seems likely that the critical period for T to act in the CNS is prenatal. Yet studies employing the 2D:4D ratio as an indicator of the level of prenatal androgen exposure have largely failed to support any association between the digit ratio and mental rotation, in either women or men (e.g., Hampson, Ellis, & Tenk, 2008; see Puts, McDaniel, Jordan, & Breedlove, 2008, for a metaanalysis). The lack of association with 2D:4D in females conceivably could reflect the smaller range of prenatal T found in normally developing female fetuses (consistent with the lack of distinct spatial effects in females who have simple-virilizing CAH, where androgen production is raised but only modestly) or could reflect inaccuracies associated with the digit ratio as a measurement technique, or questions surrounding its validity. Males ordinarily produce a high enough level of T during prenatal development to organize brain pathways that underlie mental rotation. However, correlations between 2D:4D and mental rotation are largely lacking in male samples (see Puts et al., 2008), except where sample sizes are enormous (e.g., Peters, Manning, & Reimers, 2007, n > 134,000). An association between behavioral markers of androgen exposure and mental rotation competency has been reported in samples of men where prenatal exposure is expected to have a larger range than usual (e.g., Rahman & Wilson, 2003). Where found, associations tend to be positive (i.e., higher mental rotation scores are associated with indicators of higher prenatal androgen levels). Paradoxically, reduced spatial function in boys with CAH has been reported by several labs, including our own (Hampson et al., 1998; Hines et al., 2003), but this phenomenon is not well understood. Potentially it may be explained by the dynamics of adrenal androgen production in CAH and its feedback effects on the testicular production of T by the male fetus (Hampson et al., 1998; but see Grimshaw et al.,  1995 for a similar  association observed in a nonclinical sample of boys). Hampson

51

If there are organizational effects of androgens on cognitive function, how widely do they apply? Few other cognitive functions have been studied. At present, candidates include, among others, verbal fluency and phonological processes (e.g., Hampson & Rovet,  2015; Rahman, Abrahams, & Wilson, 2003), but if anything the direction of the associations is reversed, with a greater level of androgen exposure predicting reduced scores (e.g., Helleday, Bartfai, Ritzén, & Forsman,  1994). This agrees with the fact that small sex differences in these functions are found in the general population, but the differences are in favor of females, not males. Accordingly, it may not be surprising to see a reversed association with prenatal androgens. Language-related functions are of special interest in light of an elevated incidence of speech or language problems and developmental dyslexia reported among girls with CAH (Inozemtseva, Matute, & Juárez, 2008; Nass & Baker, 1991), although these issues remain poorly documented in the clinical literature. Effects of prenatal androgens on language processes, if eventually confirmed, would be in keeping with morphological sex differences in the language zones of the left hemisphere reported in a few studies in  normally developing male and female brains (e.g., Harasty, Double, Halliday, Kril, & McRitchie, 1997; Witelson, Glezer, & Kigar,  1995) and the modest cognitive differences reported in these same processes (Hawke, Olson, Willcutt, Wadsworth, & DeFries, 2009; Heister, 1982). Unfortunately, functions other than mental rotation are inadequately studied, and there are simply too few data to permit any conclusions. Far from being exceptional, organizational effects of androgens may in fact turn out to be widespread in the human cortex: anatomical imaging shows that sexual dimorphism is present in numerous cortical regions of the human brain, many of them known to express sex steroid receptors during early brain development in other species (Goldstein et al., 2001). For most of these regional differences, we know little about whether or not there are any correlates at the functional level. As one example, Clark and Goldman-Rakic (1989) reported a possible organizational effect of neonatal androgens in the orbitofrontal cortex of the rhesus monkey, and ­reports exist of volumetric sex differences in this region in humans (Welborn et al., 2009), but it is only recently that human researchers have begun to describe behavioral differences between adult men and women on tasks that rely on the orbitofrontal cortex (Evans & Hampson, 2015; Reavis & Overman, 52

Sex Differences in Cognition

2001; Reber & Tranel, 2017; van den Bos, Homberg, & de Visser, 2013). Thus, human research is beginning to move beyond the conventional boundaries of verbal and spatial functions to the study of more sophisticated and abstract domains of cognition where new sex differences are only now being revealed. As studies begin to venture beyond mental rotation, we must be open to the possibility that not all cognitive functions are organized during the same temporal window in early development. We and others have long pointed to the possible importance of the postnatal infant period (e.g., Collaer & Hines, 1995; Hampson,  1995). Many critical neuro­developmental processes, including migration and cellular pruning, occur in the first few months after birth and are important candidates for intervention by sex steroids given that, in human male infants (and monkeys), there is a second rise in T production by the testes that takes place in the first few months of postnatal life. Following a brief period of activity, the testes are then quiescent until puberty. Hines et al. (2003) speculated that the critical (“sensitive”) period when androgens modify neural processes that support mental rotation might in fact be postnatal (during the first six months of infancy), not prenatal. In retrospect, this is unlikely because the best data currently available from studies of CAH and opposite-sex twins now implicate an earlier time point for sexual differentiation of mental rotation, but it is still possible that the postnatal peak in T production could coincide with the differentiation of other cognitive or behavioral functions, affording an opportunity for androgens to exert ­organizational effects. Such a possibility would be compatible with Goldman-Rakic’s finding in monkeys (Clark & Goldman-Rakic,  1989) and with a handful of reports that imply that T exposure during infancy may be important for the development of certain lateralized functions (Drea, Wallen, Akinbami, & Mann, 1995; Hampson, 2016). The possibility that there may be cognitively relevant processes that undergo a postnatal period of sen­ sitivity to androgens is territory that is virtually uncharted. In humans and other primates, it is typically ­believed that the aromatization of T to estradiol is not essential for sexual differentiation of the CNS to occur. However, this assumption is questionable because it has rarely been tested. In a rare exception, Hines and colleagues studied females exposed in utero to the synthetic estrogen diethylstilbestrol

(DES), used to prevent miscarriage. This compound has masculinizing effects on the behavior of guinea pigs. Among women who were exposed to DES in utero, Hines and Sandberg (1996) found no differences in mental rotation or verbal fluency between the DES-exposed women and their unexposed sisters. However, because aromatization is an important avenue to sexual differentiation in the rodent brain, we should not be too hasty to dismiss it as a potential route to masculinization of the CNS in humans too, including, potentially, cognitive functions. Unfortunately, opportunities to experimentally test the aromatization hypothesis in humans are exceedingly limited.

Activational Effects of Sex Steroids on Cognition

In other species, sex differences can be caused by sex steroids acting on the differentiation of the CNS during an early window in fetal or neonatal development (i.e., organizational effects of sex steroids). However, another class of steroid effect that can bring about sex differences is activational effects, which are  exerted in the adult brain. Activational effects modify activity within existing neural pathways, but the effects are impermanent and reversible. They wax and wane as a function of the concentration of the active steroid that is available in the bloodstream and dissipate when the steroid milieu changes. Sex differences can be generated by either class of effect acting on its own (earlier termed Type II [activational only] and Type III [­organizational only] behaviors; Goy & McEwen, 1980), but also by activational effects that occur within brain regions previously subject to organization by steroids in early development (Type I). Mental rotation is an example of a cognitive function that may be subject to both the ­organizational and activational effects of steroids.

Paradigms Used to Study the Activational Hypothesis in Humans

During the past two decades, many cognitive investigations have focused on the activational roles of estrogens in cognitive function. One of the most common paradigms used to explore the effects of ovarian hormones is the menstrual cycle. Typically, cognitive testing is prospectively timed to coincide with target phases of a woman’s cycle that differ significantly in the levels of circulating estradiol (and/or progesterone) that are present, allowing a researcher to test for small differences in performance that are evident under high vs. low hormone conditions. Individual differences in the characteristic

length of the cycle must be carefully taken into account along with other subject variables that affect the pattern, timing, or magnitude of endocrine changes (Hampson & Young,  2008). Serum or saliva assays are typically used to confirm that the expected endocrine profile was in fact present when the cognitive testing was performed and, because concentrations vary across women within each phase of the cycle, assays can be used to search for graded correlations between cognitive performance and the levels of a particular hormone. Menstrual cycle paradigms allow for statistically powerful within-subject designs to be used and have the virtue of being naturalistic—the identity of the hormones studied, the timing of their changes in the bloodstream, and the concentration ranges seen are all completely physiological. Although activational effects are easier to study than organizational effects, there are still ethical constraints on the types of research designs that may be used. Randomized placebo-controlled designs are not normally feasible except as part of a medical trial where a hormonal treatment is being tested to  alleviate a pre-existing medical condition (e.g., hypogonadism). Randomized designs have been used fruitfully in the context of menopause, for example. It is the adult reproductive years, however, where we would expect activational effects of sex steroids to be maximally evident, and fewer opportunities exist to exogenously deplete or else deplete then restore sex steroid levels in young adults. Risks that must be navigated in studies involving hormonal interventions include the potential for manipu­ lations to disrupt the menstrual cycle and thus ­increase the risk of an unplanned pregnancy or, for androgens, the potential to cause lasting virilization (i.e., development of male physical characteristics, such as deepening of the voice) if not managed carefully. Rarely, a placebo-controlled design may be permitted for purposes that are purely experimental if only a single administration of active steroid is to be given. Synthetic hormones are typically used medically and are often given at arbitrary doses on a  temporal schedule that does not replicate the ­endogenous conditions under which the natural form of the hormone is secreted. For these reasons, cognitive studies that employ a true experimental design, especially in young adults, are fairly infrequent and not always straightforward to interpret.

Mental Rotation and the Menstrual Cycle

Our own early studies suggested an effect of the  menstrual cycle on certain elements of spatial Hampson

53

function (e.g., the perception of true vertical, Hampson & Kimura,  1988), but mental rotation per se was not evaluated in our earliest work (Hampson, 1990a, 1990c; Hampson & Kimura, 1988). We did observe a significant correlation between serum estradiol concentrations and women’s performance on a spatial visualization test (which required mental rotation but included other spatial elements too; Hampson, 1990a). A correlation was found at the preovulatory phase of the ovarian cycle where estradiol peaks but progesterone is low. This implicated estradiol rather than progesterone, but it did not conclusively demonstrate an effect on the rotational process per se. Although controversial at the time, this work was considered seminal because it was the first indication that circulating estradiol levels may affect cognitive function in women, suggesting that ovarian hormones affect brain function beyond the hypothalamic-­pituitary region. Follow-up studies employed standard psychometric mental rotation tasks, such as the ShepardMetzler figures (1971) or the Vandenberg and Kuse mental rotations test (1978), and confirmed that performance varied with phase of the menstrual cycle. Consistent with the phase-related differences we had observed in our own studies (e.g., Hampson, 1990c), accuracy on rotation tests was found to be best at menses (when estradiol is lowest) and slightly diminished when women were tested during the midluteal phase of the cycle (when estradiol is elevated; Hausmann, Slabbekoorn, Van Goozen, ­ Cohen-Kettenis, & Güntürkün, 2000; Maki, Rich, & Rosenbaum, 2002; Mäntylä, 2013; McCormick & Teillon,  2001; Moody,  1997; Phillips & Silverman, 1997; Silverman & Phillips, 1993; Šimić & Santini, 2012). In a study featuring multiple ­repeated a­ ssessments of the same women, Courvoisier et al. (2013) showed that mental rotation performance was reliably slightly better during menses. These studies reported predictable phase-of-cycle differences. Fewer studies tested for a direct correlation with measured estradiol levels, but studies by Maki et al. (2002) and Hausmann et al. (2000), as well as from our own lab (Hampson, LevyCooperman, & Korman,  2014), ­ independently tested and verified a significant inverse correlation between estradiol and mental rotation using serum or salivary quantification of concomitant estradiol concentrations. Saliva has an advantage over serum in that it directly reflects the bioavailable fraction of  the circulating hormone (i.e., the fraction that is  free to interact with the brain and diffuse into neurons). Importantly, the observed correlations 54

Sex Differences in Cognition

are selective; they are not observed for control tasks that do not exhibit any sex differences and, if anything, the direction of the correlation is reversed for tasks that exhibit a female superiority instead of a male superiority in performance (although less evidence is available for such functions; see, e.g., Maki et al., 2002). Of course, there are also studies that have failed to find a significant menstrual cycle effect, but most can be explained by methodological issues. In a recent study from our lab, we found that the menstrual cycle effect is evident only for large angles of rotation, but not when an overly simple, minimal angular rotation is required to bring two visually presented stimuli into alignment (Hampson et al., 2014). This suggests that some failures to detect the effect might be explained by the use of insensitive mental rotation tasks. We also confirmed that a correlation was present for estradiol, but not for progesterone levels, which also varied across women in our recent study (Hampson et al., 2014). Data presently available suggest that for mental rotation at least, progesterone has weak if any effects on performance (see also Maki et al., 2002). Collectively, the existing menstrual cycle findings suggest that estradiol has an activational effect in one or more brain regions required for mental rotation. However, a recent review concluded that further confirmatory research is needed (Sundström Poromaa & Gingnell, 2014). Correlational evidence is not evidence for causation, so it is important to note that several studies using research designs where estradiol (ethinyl estradiol) was administered exogenously have also found a weak effect on mental rotation scores (e.g., Slabbekoorn, Van Goozen, Megen, Gooren, & Cohen-Kettenis,  1999; Van Goozen, Cohen-Kettenis, Gooren, Frijda, & Van de Poll, 1995; but see Haraldsen, Egeland, Haug, Finset, & Opjordsmoen, 2005). These studies, however, are complex to interpret for several reasons, including the fact that estrogens were given to biological males, not females, whose brains therefore differ in organizational history. Confounds introduced by multiple repeated test administrations and incompletely counterbalanced designs (Schmidt et al., 2013) have posed a significant problem for many studies involving exogenous hormone administration. Recently, support for an activational effect of ­estrogens on mental rotation processes has begun to  arise from a new source: brain imaging studies investigating functional activation of the cortex during mental rotation. Studies using functional magnetic resonance imaging (fMRI), a technique

that enables the direct visualization of changes in brain activation under contrasting conditions, have reported visible alterations over the menstrual cycle in brain regions involved in mental rotation. Women were scanned twice, at contrasting phases of the menstrual cycle, while they performed mental rotation (or control tasks). Dietrich et al. (2001), for example, scanned women during menses (low estradiol) and at the preovulatory peak in estradiol and found that the number of activated voxels in the parietal cortex (BOLD activation, the blood oxygen level–­dependent signal) was markedly greater under high than low estradiol conditions (798 vs. 36 activated voxels; see Figure 4.3). A phase-of-cycle effect was not evident in a control task. In women who performed mental rotation at the midluteal phase, Schöning et al. (2007) found that individual differ-

ences in estradiol (but not progesterone) correlated significantly with the level of BOLD activation in the superior parietal lobe bilaterally and inferior parietal lobe on the right. Because fMRI affords a direct way to visualize the physiological effects of estrogens on task-related neuronal activation, it is a unique and powerful tool to investigate steroid-driven effects on brain function.

Mental Rotation and Oral Contraceptive Use

Coincident with our menstrual cycle studies in the 1980s and early 1990s, we collected data on spatial function in women using combination oral contraceptives (COCs). COCs consist of an estrogen (ethinyl estradiol) combined with 1 of 19 synthetic progestins and are most often used to prevent

number of voxels (p = 0.0001) Control

37

Menses

36 798

Ovulation

Figure 4.3  Regions of significant brain activation during mental rotation in a functional magnetic resonance imaging (fMRI) study (Dietrich et al., 2001). Areas of significant activation (BA 39/37 and BA 7) in the grouped data are depicted in black on a normalized brain viewed from above. Activation in men (upper right) and women tested during menses (low estradiol; lower right) did not differ. When the same women were tested during the late follicular phase (high estradiol; lower left), a large increase was seen in the number of activated voxels during the peak estradiol condition. The number of activated voxels is shown in the upper left. (Reprinted from Dietrich, T., Krings, T., Neulen, J., Willmes, K., Erberich, S., Thron, A., & Sturm, W. [2001]. Effects of blood estrogen level on cortical activation patterns during cognitive activation as measured by functional MRI. NeuroImage, 13, 425–432. Copyright [2001], with permission from Elsevier.)

Hampson

55

c­onception, especially when prescribed in healthy young women. Based on the effects of estradiol seen during the natural menstrual cycle, one might expect ethinyl estradiol to affect mental rotation in COC users, particularly as it has an affinity for the estrogen receptor (ERα) at least equal to, if not slightly greater than, that of endogenous 17β-estradiol (Escande et al., 2006). However, using the COC paradigm to test for an activational effect is more complex than it first appears. Because they are combined into a single tablet, there is no point in the contraceptive cycle where ethinyl estradiol can be varied by a researcher independently of the progestin it is combined with. Although progesterone produced by the corpus luteum over the natural menstrual cycle typically is  uncorrelated with scores on mental rotation tests, the progestins used in COCs often derive from 19-nortestosterone and biologically they exert androgenic and not simply progestogenic ­effects in tissue. This is problematic in light of ­increasing evidence that circulating levels of androgens, not just estrogens, may influence mental rotation in young adults (see later). If the impact of androgens is to improve mental rotation, as some studies ­suggest, then the simultaneous presence of androgenic progestins might counteract any effect of the ethinyl estradiol component of COCs on mental rotation ability. At the time our first COC data were collected in the 1980s, dosages of ethinyl estradiol contained in COCs ranged from 30 to 50 μg/day. Many present-­ day COCs are substantially lower (15 to 30 μg/day). Dosage reduction during the last few decades adds a further layer of complexity because, over the naturally occurring menstrual cycle, diminished mental rotation performance is only visible at phases of the cycle characterized by the highest levels of estradiol production. Thus, the change in COC formulations leaves open the question of whether current dosages are high enough to support an activational effect. Despite these considerations, COC use is one of  the few opportunities to test for an effect on cognition in young women receiving sex steroids exogenously. Note that when COCs are used, the endogenous production of steroids by the ovaries is suppressed, including testosterone. In our earliest report (Hampson, 1990b), we found that the spatial ability of women taking low-dose COCs (defined as 30 to 35 µg/day of ethinyl estradiol) fell intermediate between naturally cycling women tested at menses (lowest estradiol) and matched naturally cycling women tested at the midluteal phase of 56

Sex Differences in Cognition

the cycle (higher estradiol). The effect appeared to be dependent upon ethinyl estradiol dosage in the pills, because we found that women using higher dose COCs (50 to 80 µg/day of ethinyl estradiol) showed lower performance on the same spatial tests compared with women taking low-dose formulations (Hampson & Moffat, 2004). As pointed out earlier, our spatial tests did not include a dedicated measure of mental rotation, but our findings were soon followed by work from other labs using the Vandenberg and Kuse mental rotations test (Vandenberg & Kuse,  1978). Those studies ­appeared to confirm an effect of COCs on mental rotation scores (McCormick & Teillon, 2001; Moody, 1997; Silverman & Phillips, 1993). They included repeated-measures designs (Moody, 1997; Silverman & Phillips,  1993), in which spatial performance during active COC use was contrasted with performance by the same women during the COC washout phase (the seven-day interval each month when COCs are not taken and the uterine lining is shed). Further evidence that the effect is related to ethinyl estradiol was obtained in a recent study where ethinyl estradiol dosage was discovered to be a significant predictor of scores on the Vandenberg and Kuse rotations test in a group of 148 women taking a wide range of commercially available brands of COCs (Beltz, Hampson, & Berenbaum,  2015). Increased ethinyl estradiol dose was associated with decreased mental rotation scores. Where dosage is not explicitly taken into account, several recent studies have failed to find significant effects of COC use on mental rotation (e.g., Mordecai, Rubin, & Maki, 2008). Some reports suggest that the use of COCs also reduces mental rotation performance in male-to-­ female transgender patients being treated with high doses of COCs in the context of sex reassignment (Van Goozen et al.,  1995). Such data are hard to interpret because antiandrogens are typically administered simultaneously as an adjunct part of the treatment regimen. However, in favor of an inhibitory effect exerted by estradiol, Kozaki and Yasukouchi (2008) found that estradiol levels in ordinary men were systematically related to the slope of the reaction time function on a mental rotation test. Based on ordinary women, the findings from studies of oral contraceptive use are consistent with an activational effect of contraceptive steroids on spatial cognition, but the quality of many existing studies is low, and it is hard to attribute the effects unambiguously to ethinyl estradiol given the complications of the COC paradigm.

Spatial Cognition in Postmenopausal Women

Postmenopausal women have been widely used to study the effects of estrogens on cognitive function. However, episodic memory has been the main focus, not mental rotation as in younger adults. The literature on menopause is rich, complex, and extensive (Hogervorst & Bandelow, 2010). Long-term estrogen deprivation of the CNS as a result of undergoing surgical or natural menopause, and the effects of various hormone replacement therapies have been studied. Observational designs and randomized controlled trials have been used. The menopause literature does speak to the possibility of activational effects of estrogens on memory, but mental rotation is seldom studied because it is a cognitive function that is markedly impaired by aging. Floor effects on mental rotation tests are commonly seen by the age of menopause (e.g., Jansen & Heil, 2010), and the decline in mental rotation competency begins even earlier, in young to middle adulthood (Wilson & Vandenberg, 1978). When studying older adults, it is common for researchers to select simpler rotation tests, precede the tests with extended practice, or give them without time limits, making it difficult to compare the results obtained with studies based on young adults. Studies of postmenopausal women have produced mixed findings; in a large sample of women randomized to receive estradiol valerate, performance on a conventional test of mental rotation was negatively correlated with the absolute levels of estradiol posttreatment (measured in serum) and with the magnitude of the increase in estradiol from pre- to posttreatment (Kocoska-Maras et al., 2013). However, some studies have found no effects on mental rotation or even positive effects of treatment with transdermal estradiol (Duka, Tasker, & McGowan, 2000) or conjugated equine estrogens (Kimura,  1995). The variable findings might reflect the types of estrogens used, age of the participants, adaptations of the tests for older people, or other variables. A more interesting (but theoretical) possibility is that the effects of estradiol on mental rotation might be nonlinear (see later for a discussion of androgen effects, where such a phenomenon has been reported). Small increases in estradiol from a hypogonadal state may improve performance, whereas large increases, beyond some optimum point, may diminish it. Over the ordinary menstrual cycle, estradiol is produced in small amounts even at menses where it is lowest, and studies often compare this time point with points of high estradiol production, such as

the midluteal peak, giving rise to the typical finding from menstrual cycle studies that spatial performance is diminished under high estradiol. The possibility of a nonlinear association over a broadened range of concentrations is largely untested, but it would be consistent with several existing observations, including improved mental rotation in postmenopausal women treated with only low amounts of estrogens, the menstrual cycle findings of reduced performance at the highest estradiol levels, and our own early findings of an inverted U-shaped function covering a wider physiological range of serum estradiol levels in young women who were tested on a visualization task that had a prominent mental rotation requirement (Hampson, 1990a).

Do Androgens Have Activational Effects on Spatial Cognition?

Whether or not circulating levels of androgens influence mental rotation is a controversial and unresolved issue. Androgens have received less study in this context than estrogens (the reverse of the situation that presently prevails for the study of ­organizational effects). The earliest studies dedicated to this question were mostly correlational, but clinical studies based on patient populations receiving T (or other androgens) therapeutically, and more recently studies in which circulating levels of androgens are manipulated experimentally in healthy people, allow for a few causal inferences to be made. Beginning in the mid-1980s, several labs independently reported finding significant correlations between endogenous T concentrations in young men and accuracy on spatial tests (e.g., Christiansen & Knussman, 1987; Shute, Pellegrino, Hubert, & Reynolds, 1983), including tests of mental rotation (Hausmann, Schoofs, Rosenthal, & Jordan, 2009; Moffat & Hampson,  1996; Neave, Menaged, & Weightman,  1999; Silverman, Kastuk, Choi, & Phillips, 1999; Yang, Hooven, Boynes, Gray, & Pope, 2007). Correlations were also reported in aging men, including men followed longitudinally over time (Moffat et al., 2002; Yonker, Eriksson, Nilsson, & Herlitz,  2006), although simpler spatial tests were used. Circulating T levels have also been found in a few studies to predict performance on mental rotation tasks in women (e.g., Hausmann et al., 2000; Moffat & Hampson, 1996; but see van Anders & Hampson,  2005), with higher concurrent T predicting superior performance. However, studies of male samples have been complex to interpret, as a number have indicated that a nonlinear relationship may be present in which the correlation reverses and Hampson

57

is negative at the very highest levels of T in men (Moffat & Hampson,  1996; Neave et al.,  1999; Shute et al., 1983). If this observation is correct, it makes it less likely that simple associations will be identified in studies of men, especially when statistics are used that assume linearity. Studies where T is given exogenously support the possibility of a dose-dependent effect, in that the administration of T to healthy women has been ­reported to temporarily improve performance on mental rotation tasks (Aleman, Bronk, Kessels, Koppeschaar, & van Honk,  2004; Pintzka et al., 2016), whereas the administration of T to healthy young men (who already have high T) impairs it (O’Connor, Archer, Hair, & Wu,  2001). In aged men, evaluated with simpler mental rotation tests, performance was improved after treatment with T  compared with placebo (Cherrier et al.,  2001; Janowsky, Oviatt, & Orwoll, 1994), consistent with their hypogonadal status before the treatment was initiated. However, a perceptible change in mental rotation performance is not always found in aged samples (Resnick et al., 2017; Wolf et al., 2000). Wolf et al. (2000) brought T into the supra-physiological range, which may interfere with seeing any effect. Cherrier, Aubin, and Higano (2009) found a deterioration in mental rotation in men with prostate cancer when they underwent treatment-related androgen blockade, relative to their pretreatment level of performance. Not all studies have found associations between T and mental rotation, however. This may potentially reflect methodological factors, but the failure to detect any association includes several seemingly well-done studies with large sample sizes (e.g., Puts et al.,  2010; Resnick et al.,  2017). For these and other reasons, the question of whether adult T levels presently in the circulation exert activational effects on spatial cognition remains open. Notably, there is substantial genetically based variation across individual men in the responsiveness of the androgen receptor to T binding (Chamberlain, Driver, & Miesfeld, 1994). Receptor polymorphism is rarely taken into account in cognitive research, but because T must act via its receptors to trigger a tissue response, it could be one factor that contributes to variability in study outcomes. A few researchers have ventured beyond mental rotation to explore the effects of T on spatial learning, spatial orientation, or navigational efficiency when exploring new environments in either realworld surroundings or virtual simulations. A male advantage in acquiring the layout of an unfamiliar virtual environment is often reported (Astur, 58

Sex Differences in Cognition

Ortiz, & Sutherland, 1998; Moffat, Hampson, & Hatzipantelis, 1998), allowing potential scope for an activational influence of T to be observed. However, these studies too have been inconsistent in supporting the possibility of an association ­between concurrent T concentrations and spatial performance, whether defined by rate of acquisition, path integration, spatial memory for the location of an escape platform (in the virtual Morris maze), or other variables. Using virtual adaptations of the Morris water maze, T was associated with adeptness at finding the hidden platform in women only (Burkitt, Widman, & Saucier, 2007), or in men only (Driscoll, Hamilton, Yeo, Brooks, & Sutherland, 2005), or was inversely related to men’s performance in accord with the theory positing an o­ptimal level of T (Nowak, Diamond, Land, & Moffat, 2014). When a single dose of T was given to women in a double-blind placebo-controlled design, Pintzka et al. (2016) found improved knowledge of the directions among target objects in a novel virtual environment among women who received T compared with placebo prior to entering the environment. The women treated with T also achieved significantly higher accuracy scores on a mental rotation test. Correlations between T concentrations and proficiency in virtual spatial environments have been harder to identify in men, and again, structural variation in the androgen receptor might conceivably moderate the correlations seen (Nowak et al., 2014). The possible impact of individual variation at the receptor level requires further study in the cognitive context, especially given its purported impact on other androgen-dependent traits (Zitzmann, 2009). Besides studies formally designed to address the question, indirect support for an activational effect comes from studies where another variable (not T) was the primary focus of investigation, but where T happened to be measured too. A few menstrual cycle studies, for example, have ­measured T in ­addition to estradiol concentrations and have ­reported a negative correlation between estradiol levels and mental rotation, but a positive correlation with current T levels ­measured simultaneously in the same women (e.g., Hausmann et al., 2000). This further emphasizes the independence of the two effects. In women with polycystic ovarian disease (a gynecological condition in which androgen production by the ovaries or adrenals is raised and ovulation becomes less frequent), increased mental rotation scores were limited to a hyperandrogenic subgroup of women, and a positive correlation was observed between serum T and the Vandenberg and Kuse mental rotation

test score (Barry, Parekh, & Hardiman, 2013; but see Schattmann & Sherwin, 2007). Data from women using COCs also support an activational influence—based on the hormonal formulations of standard COCs, one analysis revealed that brands containing progestins with more highly androgenic properties were statistically associated with higher scores on a mental rotation test (Wharton et al., 2008), whereas women using brands of COCs containing progestins with anti-androgenic properties performed significantly more poorly (Griksiene, Monciunskaite, Arnatkeviciute, & Ruksenas, 2018). Although this evidence suggests an effect of androgens, it does not implicate the hormone T specifically, as the findings were based on the androgen potencies of specific synthetic progestins contained in the COCs (defined as their level of activity at the androgen receptor). Of course, an improvement in mental rotation was observed directly by Aleman et  al. (2004) in healthy women who were treated with a single acute dose of T on an experimental basis, in a double-blind placebo-controlled crossover design. Mental rotation improved significantly within four to five hours after treatment with T compared with placebo. This is a rare example of T  being administered directly in healthy young adults, albeit on a one-time basis. Because it was a  single administration, and because a repeated-­ measures design was used, it constitutes strong support for the activational hypothesis.

Not All Spatial Functions Are Equal

Mental rotation was used as an example in this chapter because of the large amount of research devoted to this cognitive process. However, not every spatial function exhibits a sex difference. Of those that do, the direction of the difference does not ­invariably favor males. For example, our lab discovered a sex difference in favor of women (not men) that is seen on certain tests of spatial working memory (Duff & Hampson, 2001). The sex difference was replicated by other labs using memory tests possessing similar features (e.g., Lejbak, Vrbancic, & Crossley, 2009). A recent meta-analysis concluded that the female superiority in spatial working memory is seen reliably on this particular family of tasks and certain other working memory tasks that emphasize the spatial positions of object identifiers, under conditions where active short-term maintenance and/or repeated updating of the spatial positional information held in short-term store is required (Voyer, Voyer, & Saint-Aubin, 2017). Thus, despite the fact that the material to be remembered is spatial (i.e.,

spatial positions), the sex difference is in favor of females, not males. An increasing body of data supports the hypothesis of an activational influence of estrogens on working memory, including spatial working memory, and brain regions that support it. Sources of evidence include postmenopausal women tested on working memory tasks, with or without estrogen treatment in the form of ­ standard replacement ­therapy (Duff & Hampson,  2000; Keenan, Ezzat, Ginsburg, & Moore,  2001), or after a short-term experimental treatment with estradiol versus placebo (three days of estradiol treatment; Krug, Born, & Rasch, 2006). fMRI reveals increased activation in regions of the prefrontal cortex during working memory tasks if aging women are tested under ­estradiol versus placebo (e.g., Smith et al.,  2006). Ovariectomized female rhesus monkeys treated with estradiol cypionate showed superior per­formance on the delayed response task (a classic behavioral task used to evaluate working memory in  nonhuman primates) compared with monkeys treated with vehicle only (Rapp, Morrison, & Roberts, 2003). Data from young women are rare, but our lab found that estradiol levels over the menstrual cycle correlated inversely with the numbers of  working memory errors committed on a test of spatial working memory (Hampson & Morley, 2013). Figure 4.4 shows data on the same test collected from a group of young women using COCs, who were tested either while they were taking their COCs or during the monthly pill-free interval (Hampson, unpublished data). As shown in Figure 4.4, fewer working memory errors were ­observed during the active use of COCs than during the one-week washout period when the women a­ bstained from estrogen use. In all of these studies the direction of the estradiol effect is reversed, compared with the effects that have been reported for mental rotation. These two spatial processes rely on differing networks or pathways in the brain, with regions of special importance being the parietal cortex (vicinity of the intraparietal sulcus) for the rotational process and the dorsolateral prefrontal cortex for the executive components of spatial working memory (Owen, Evans, & Petrides, 1996; Zacks, 2008). The differences in cortical representation and in the direction of the sex difference observed emphasize the importance of conceptual precision when making predictions about any particular cognitive process and its hormonal control by sex steroids. It is a mistake to think that a hormone such as estradiol has one ­singular effect that applies across-the-board to all spatial functions. Hampson

59

SPWM

WM Errors

50 40 30 20 10 0

0

1 OC_On

2 Trial

3

OC_Off

4 Males

Figure 4.4  The Spatial Working Memory task (SPWM; shown in top image) involves participants being asked to find the spatial locations of matching pairs of colored dots hidden beneath the doors of a 4 × 5 array. A working memory error (WM error) is defined as revisiting an already-visited pair of locations (for full description see Duff & Hampson, 2001). Bottom image shows the numbers of WM errors committed on three consecutive trials of the SPWM task by female combination oral contraceptive (COC) users (n = 60) and demographically matched male controls (n = 96). SPWM data from women with naturally occurring menstrual cycles (non-COC users) collected in the same study have been reported elsewhere (Hampson & Morley, 2013). All women were tested blind then retrospectively classified into subgroups based on health information supplied at the end of the test session, after cognitive testing was completed. Bottom figure shows the numbers of WM errors committed by women taking a COC on the day they were tested (OC_On, n = 40) and numbers of WM errors committed by equivalent COC users tested during the monthly COC washout phase when no active pills are ingested (OC_Off, n = 20). Women actively taking a COC (higher levels of ethinyl estradiol) displayed superior working memory compared to OC_Off or male controls. All COC users took COCs containing 20 to 35 µg/day of ethinyl estradiol. Thus, women tested at higher estradiol levels made significantly fewer spatial working memory errors, among both COC users (shown here) and nonusers (Hampson & Morley, 2013). (Reprinted from Hampson, E. [2018]. Regulation of cognitive function by androgens and estrogens. Current Opinion in Behavioral Sciences, 23, 49–57. Copyright [2018], with permission from Elsevier.)

The Organizational–Activational Hypothesis Today—Current Status

Mental rotation is the cognitive process that has ­received the most attention to date from human researchers exploring the organizational–activational hypothesis. In contrast, clinical work on menopause has focused heavily on verbal episodic memory (see, e.g., Hogervorst & Bandelow, 2010). A growing variety of data support the existence of organizational or activational effects that operate on cognitive functions, but the details of these relationships remain to be discovered. Over the past decade, ­research 60

Sex Differences in Cognition

a­ ttention has begun to shift to cognitive processes mediated by the frontal cortex and its input and output pathways to the amygdala and striatum (e.g., Hampson & Morley,  2013; Hermans et al., 2010; Peper, Koolschijn, & Crone, 2013; Stanton, Liening, & Schultheiss, 2011). In parallel, increasing effort is being devoted to investigating whether frontal functions might exhibit sex differences (e.g., Duff & Hampson, 2001; Evans & Hampson, 2015). It is important for researchers to get beyond the traditional boundaries of looking only at mental rotation and episodic memory, as we are only scratching

the surface of the many cognitive and affective ­domains where hormonal regulation of the CNS by sex steroids might exist. Although modern-day neuroscience has begun to acknowledge sex differences in the CNS, the dynamic regulatory effects of sex steroids in the adult limbic system and cerebral cortex are still underappreciated, even though they have wide-ranging implications for understanding brain function that go beyond cognition per se. Establishing these ­effects on firmer ground will lead to new theoretical and applied advances. Already, growing evidence for  steroid effects on cognitive function supports theoretical conceptualizations in which cognitive sex differences are discussed as evolved adaptations (e.g., Anderson & Rutherford,  2012; Sherry & Hampson, 1997), not solely the products of social learning as formerly thought. Greater appreciation by clinicians of the important regulatory roles played by sex steroids in extra-hypothalamic regions might inform new insights into the sex differences in prevalence or symptom severity observed in a number of psychiatric or neurological conditions that afflict humans. For basic scientists, awareness might also transform the way that functional imaging is currently done by highlighting the need to control for relevant endocrine variables when studying human brain function using f MRI. Understanding which regions of the CNS are influenced by steroids, at which stages of the lifespan, and why, is an important future goal for cognitive neuroscience.

Acknowledgments

I would like to thank the following funding sources who financially supported work from my laboratory described in this chapter: the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Ontario Mental Health Foundation. I also gratefully acknowledge the following former students for their contributions to data collection on the cognitive associations of oral contraceptive use: S. Duff and E. Morley.

References

Aleman, A., Bronk, E., Kessels, R. P. C., Koppeschaar, H. P. F., & van Honk, J. (2004). A single administration of testosterone improves visuospatial ability in young women. Psychoneuro­ endocrinology, 29, 612–617. Anderson, M.  V., & Rutherford, M.  D. (2012). Cognitive reorganization during pregnancy and the postpartum period: An evolutionary perspective. Evolutionary Psychology, 10, 659–687. Arnold, A. P. (2009). The organizational-activational hypothesis as the foundation for a unified theory of sexual differentiation of all mammalian tissues. Hormones and Behavior, 55, 570–578. Astur, R.  S., Ortiz, M.  L., & Sutherland, R.  J. (1998). A characterization of performance by men and women in a

virtual Morris water task: A large and reliable sex difference. Behavioural Brain Research, 93, 185–190. Auyeung, B., Knickmeyer, R., Ashwin, E., Taylor, K., Hackett, G., & Baron-Cohen, S. (2012). Effects of fetal testosterone on visuospatial ability. Archives of Sexual Behavior, 41, 571–581. Barry, J. A., Parekh, H. S. K., & Hardiman, P. J. (2013). Visualspatial cognition in women with polycystic ovarian syndrome: The role of androgens. Human Reproduction, 38, 2832–2837. Beltz, A.  M., Hampson, E., & Berenbaum, S.  A. (2015). Oral contraceptives and cognition: A role for ethinyl estradiol. Hormones and Behavior, 74, 209–217. Berenbaum, S. A., Bryk, K. K., Nowak, N., Quigley, C. A., & Moffat, S. (2009). Fingers as a marker of prenatal androgen exposure. Endocrinology, 150, 5119–5124. Berenbaum, S. A., Korman Bryk, K. L., & Beltz, A. M. (2012). Early androgen effects on spatial and mechanical abilities: Evidence from congenital adrenal hyperplasia. Behavioral Neuroscience, 126, 86–96. Bethea, C. L., Lu, N. Z., Gundlah, C., & Streicher, J. M. (2002). Diverse actions of ovarian steroids in the serotonin neural system. Frontiers in Neuroendocrinology, 23, 41–100. Bimonte, H.  A., & Denenberg, V.  H. (1999). Estradiol facil­ itates performance as working memory load increases. Psychoneuroendocrinology, 24, 161–173. Breedlove, S. M., & Arnold, A. P. (1980). Hormone accumulation in a sexually dimorphic motor nucleus of the rat spinal cord. Science, 201, 564–566. Breedlove, S. M., & Hampson, E. (2002). Sexual differentiation of the brain and behavior. In J. B. Becker, S. M. Breedlove, D.  Crews, & M.  M.  McCarthy (Eds.), Behavioral endocrinology (pp. 75–114). Cambridge, MA: MIT Press. Burkitt, J., Widman, D., & Saucier, D. M. (2007). Evidence for the influence of testosterone in the performance of spatial navigation in a virtual water maze in women but not in men. Hormones and Behavior, 51, 649–654. Chamberlain, N. L., Driver, E. D., & Miesfeld, R. L. (1994). The length and location of CAG trinucleotide repeats in the androgen receptor N-terminal domain affect transactivation function. Nucleic Acids Research, 22, 3181–3186. Cherrier, M. M., Asthana, S., Plymate, S., Baker, L., Matsumoto, A.  M., Peskind, E., . . . Craft, S. (2001). Testosterone supplementation improves spatial and verbal memory in healthy older men. Neurology, 57, 80–88. Cherrier, M. M., Aubin, S., & Higano, C. S. (2009). Cognitive and mood changes in men undergoing intermittent combined androgen blockade for non-metastatic prostate cancer. Psychooncology, 18, 237–247. Chesler, E. J., & Juraska, J. M. (2000). Acute administration of estrogen and progesterone impairs the acquisition of the spatial Morris water maze in ovariectomized rats. Hormones and Behavior, 38, 234–242. Christiansen, K., & Knussman, R. (1987). Sex hormones and cognitive functioning in men. Neuropsychobiology, 18, 27–36. Clark, A. S., & Goldman-Rakic, P. S. (1989). Gonadal hormones influence the emergence of cortical function in nonhuman primates. Behavioral Neuroscience, 103, 1287–1295. Collaer, M.  L., & Hines, M. (1995). Human behavioral sex differences: A role for gonadal hormones during early development? Psychological Bulletin, 118, 55–107. Commins, W.  D. (1932). The effect of castration at various ages  upon learning ability of male albino rats. Journal of Comparative Psychology, 14, 29–53.

Hampson

61

Courvoisier, D. S., Renaud, O., Geiser, C., Paschke, K., Gaudy, K., & Jordan, K. (2013). Sex hormones and mental rotation: An intensive longitudinal investigation. Hormones and Behavior, 73, 345–351. Dalla, C., & Shors, T.  J. (2009). Sex differences in learning processes of classical and operant conditioning. Physiology and Behavior, 97, 229–238. Dawson, J.  L.  M., Cheung, Y.  M., & Lau, R.  T.  S. (1975). Developmental effects of neonatal sex hormones on spatial and activity skills in the white rat. Biological Psychology, 3, 213–229. Dietrich, T., Krings, T., Neulen, J., Willmes, K., Erberich, S., Thron, A., & Sturm, W. (2001). Effects of blood estrogen level on cortical activation patterns during cognitive activation as measured by functional MRI. NeuroImage, 13, 425–432. Dohanich, G., Korol, D., & Shors, T. (2009). Steroids, learning and memory. In D. W. Pfaff, A. P. Arnold, S. E. Fahrbach, A.  M.  Etgen, & R.  T.  Rubin (Eds.), Hormones, brain and behavior (pp. 539–578). San Diego, CA: Academic Press. Drea, C.  M., Wallen, K., Akinbami, M.  A., & Mann, D.  R. (1995). Neonatal testosterone and handedness in yearling rhesus monkeys (Macaca mulatta). Physiology and Behavior, 58, 1257–1262. Driscoll, I., Hamilton, D.  A., Yeo, R.  A., Brooks, W.  M., & Sutherland, R. J. (2005). Virtual navigation in humans: The impact of age, sex, and hormones on place learning. Hormones and Behavior, 47, 326–335. Duarte-Guterman, P., Yagi, S., Chow  C., & Galea, L.  A.  M. (2015). Hippocampal learning, memory, and neurogenesis: Effects of sex and estrogens across the lifespan in adults. Hormones and Behavior, 74, 37–52. Duff, S. J., & Hampson, E. (2001). A sex difference on a novel spatial working memory task in humans. Brain and Cognition, 47, 470–493. Duff, S. J., & Hampson, E. (2000). A beneficial effect of estrogen on working memory in postmenopausal women taking hormone replacement therapy. Hormones and Behavior, 38, 262–276. Duka, T., Tasker, R., & McGowan, J. F. (2000). The effects of 3-week estrogen hormone replacement on cognition in elderly healthy females. Psychopharmacology, 149, 129–139. Escande, A., Pillon, A., Servant, N., Cravedi, J.-P., Larrea, F., Muhn, P.,  .  .  .  Balaguer, P. (2006). Evaluation of ligand selectivity using reporter cell lines stably expressing estrogen receptor alpha or beta. Biochemical Pharmacology, 71, 1459–1469. Evans, K.  L., & Hampson, E. (2015). Sex differences on prefrontally-dependent cognitive tasks. Brain and Cognition, 93, 42–53. Fader, A. J., Johnson, P. E., & Dohanich, G. P. (1999). Estrogen improves working but not reference memory and prevents amnestic effects of scopolamine of a radial-arm maze. Pharmacology Biochemistry & Behavior, 62, 711–717. Forest, M. G., de Peretti, E., & Bertrand, J. (1976). Hypothalamicpituitary-gonadal relationships in man, from birth to puberty. Clinical Endocrinology, 5, 551–569. Foster, T. C., Sharrow, K. M., Kumar, A., & Masse, J. (2003). Interaction of age and chronic estradiol replacement on memory and markers of brain aging. Neurobiology of Aging, 24, 839–852. Frye, C. A. (1995). Estrus-associated decrements in a water maze task are limited to acquisition. Physiology & Behavior, 57, 5–14.

62

Sex Differences in Cognition

Galea, L.  A.  M., Frick, K.  M., Hampson, E., Sohrabji, F., & Choleris, E. (2017). Why estrogens matter for behavior and brain health. Neuroscience & Biobehavioral Reviews, 76, 363–379. Galea, L. A. M., Kavaliers, M., Ossenkopp, K.-P., & Hampson, E. (1995). Gonadal hormone levels and spatial learning performance in the Morris water maze in male and female meadow voles, Microtus pennsylvanicus. Hormones and Behavior, 29, 106–125. Galea, L. A. M., Kavaliers, M., Ossenkopp, K.-P., Innes, D., & Hargreaves, E. L. (1994). Sexually dimorphic spatial learning varies seasonally in two populations of deer mice. Brain Research, 635, 18–26. Galea, L.  A.  M., Wainwright, S.  R., Roes, M.  M., DuarteGuterman, P., Chow, C., & Hamson, D.  K. (2013). Sex, hormones and neurogenesis in the hippocampus: Hormonal modulation of neurogenesis and potential functional implications. Journal of Neuroendocrinology, 25, 1039–1061. Goldstein, J.  M., Seidman, L.  J., Horton, N.  J., Makris, N., Kennedy, D. N., Caviness, V. S. Jr., . . . Tsuang, M. T. (2001). Normal sexual dimorphism of the adult human brain assessed by in vivo magnetic resonance imaging. Cerebral Cortex, 11, 490–497. Goodrich-Hunsaker, N. J., Livingstone, S. A., Skelton, R. W., & Hopkins, R.  O. (2010). Spatial deficits in a virtual water maze in amnesic participants with hippocampal damage. Hippocampus, 20, 481–4891. Gorski, R. A., Gordon, J. H., Shryne, J. E., & Southam, A. M. (1978). Evidence for a morphological sex difference within the medial preoptic area of the rat brain. Brain Research, 148, 333–346. Goudsmit, E., Van De Poll, N.  E., & Swaab, D.  F. (1990). Testosterone fails to reverse spatial memory decline in aged rats and impairs retention in young and middle-aged animals. Behavioral and Neural Biology, 53, 6–20. Goy, R. W., & McEwen, B. S. (1980). Sexual differentiation of the brain. Cambridge, MA: MIT Press. Griksiene, R., Monciunskaite, R., Arnatkeviciute, A., & Ruksenas, O. (2018). Does the use of hormonal contraceptives affect the mental rotation performance? Hormones and Behavior, 100, 29–38. Grimshaw, G.  M., Sitarenios, G., & Finegan, J.  K. (1995). Mental rotation at 7 years: Relations with prenatal testosterone levels and spatial play experiences. Brain and Cognition, 29, 85–100. Gurney, M. E., & Konishi, M. (1980). Hormone-induced sexual differentiation of brain and behavior in zebra finches. Science, 208(4450), 1380–1383. Halpern, D.  F. (2012). Sex differences in cognitive abilities (4th ed.). New York, NY: Taylor & Francis. Hampson, E. (1990a). Estrogen-related variations in human spatial and articulatory-motor skills. Psycho-neuroendocrinology, 15, 97–111. Hampson, E. (1990b). Influence of gonadal hormones on cognitive function in women. Clinical Neuropharmacology, 13(Suppl. 2), 522–523. Hampson, E. (1990c). Variations in sex-related cognitive abilities across the menstrual cycle. Brain and Cognition, 14, 26–43. Hampson, E. (1995). Spatial cognition in humans: Possible modulation by androgens and estrogens. Journal of Psychiatry and Neuroscience, 20, 397–404.

Hampson, E. (2016). The development of hand preference and dichotic language lateralization in males and females with congenital adrenal hyperplasia. Laterality: Asymmetries of Body, Brain, and Cognition, 21, 415–432. Hampson, E., Ellis, C.  L., & Tenk, C.  M. (2008). On the relation between 2D:4D and sex-dimorphic personality traits. Archives of Sexual Behavior, 37, 133–144. Hampson, E., & Kimura, D. (1988). Reciprocal effects of hormonal fluctuations on human motor and perceptualspatial skills. Behavioral Neuroscience, 102, 456–459. Hampson, E., Levy-Cooperman, N., & Korman, J. M. (2014). Estradiol and mental rotation: Relation to dimensionality, difficulty, or angular disparity? Hormones and Behavior, 65, 238–248. Hampson, E., & Moffat, S.  D. (2004). The psychobiology of gender: Cognitive effects of reproductive hormones in the adult nervous system. In A.  H.  Eagly, A.  E.  Beall, & R. J. Sternberg (Eds.), The psychology of gender (pp. 38–64). New York, NY: Guilford Press. Hampson, E., & Morley, E. E. (2013). Estradiol concentrations and working memory performance in women of reproductive age. Psychoneuroendocrinology, 38, 2897–2904. Hampson, E., & Rovet, J.  F. (2015). Spatial function in adolescents and young adults with congenital adrenal hyperplasia: Clinical phenotype and implications for the androgen hypothesis. Psychoneuroendocrinology, 54, 60–70. Hampson, E., Rovet, J. F., & Altmann, D. (1998). Spatial reasoning in children with congenital adrenal hyperplasia due to 21-hydroxylase deficiency. Developmental Neuropsychology, 14, 299–320. Hampson, E., & Sankar, J. S. (2012). Re-examining the Manning hypothesis: Androgen receptor polymorphism and the 2D:4D digit ratio. Evolution and Human Behavior, 33, 557–561. Hampson, E., & Young, E. A. (2008). Methodological issues in the study of hormone-behavior relations in humans: Understanding and monitoring the menstrual cycle. In J.  B.  Becker, K.  J.  Berkley, N.  Geary, E.  Hampson, J.  P.  Herman, & E.  A.  Young (Eds.), Sex differences in the grain: From genes to behavior (pp. 63–78). New York, NY: Oxford University Press. Haraldsen, I.  R., Egeland, T., Haug, E., Finset, A., & Opjordsmoen, S. (2005). Cross-sex hormone treatment does not change sex-sensitive cognitive performance in gender identity disorder patients. Psychiatry Research, 137, 161–174. Harasty, J., Double, K.  L., Halliday, G.  M., Kril, J.  J., & McRitchie, D.  A. (1997). Language-associated cortical regions are proportionally larger in the female brain. Archives of Neurology, 54, 171–176. Hausmann, M., Schoofs, D., Rosenthal, H. E. S., & Jordan, K. (2009). Interactive effects of sex hormones and gender stereotypes on cognitive sex differences—A psychobiosocial approach. Psychoneuroendocrinology, 34, 389–401. Hausmann, M., Slabbekoorn, D., Van Goozen, S.  H.  M., Cohen-Kettenis, P.  T., & Güntürkün, O. (2000). Sex hormones affect spatial abilities during the menstrual cycle. Behavioral Neuroscience, 114, 1245–1250. Hawke, J. L., Olson, R. K., Willcutt, E. G., Wadsworth, S. J., & DeFries, J. C. (2009). Gender ratios for reading difficulties. Dyslexia, 15, 239–242. Heil, M., Kavšek, M., Rolke, B., Beste, C., & Jansen, P. (2011). Mental rotation in female fraternal twins: Evidence for intrauterine hormone transfer? Biological Psychology, 86, 90–93.

Heister, G. (1982). Sex differences in verbal fluency: A short note. Current Psychological Research, 2, 257–260. Helleday, J., Bartfai, A., Ritzén, E. M., & Forsman, M. (1994). General intelligence and cognitive profile in women with congenital adrenal hyperplasia. Psychoneuroendocrinology, 19, 343–356. Hermans, E.  J., Bos, P.  A., Ossewaarde, L., Ramsey, N.  F., Fernández, G., & van Honk, J. (2010). Effects of exogenous testosterone on the ventral striatal BOLD response during reward anticipation in healthy women. NeuroImage, 52, 277–283. Hines, M., Fane, B. A., Pasterski, V. L., Mathews, G. A., Conway, G. S., & Brook, C. (2003). Spatial abilities following prenatal androgen abnormality: Targeting and mental rotations performance in individuals with congenital adrenal hyperplasia. Psychoneuroendocrinology, 28, 1010–1026. Hines, M., & Sandberg, E. C. (1996). Sexual differentiation of cognitive abilities in women exposed to diethylstilbestrol (DES) prenatally. Hormones and Behavior, 30, 354–363. Hogervorst, E., & Bandelow, S. (2010). Sex steroids to maintain cognitive function in women after the menopause: A metaanalysis of treatment trials. Maturitas, 66, 56–71. Hönekopp, J., & Watson, S. (2010). Meta-analysis of digit ratio 2D:4D shows greater sex difference in the right hand. American Journal of Human Biology, 22, 619–630. Inozemtseva, O., Matute, E., & Juárez, J. (2008). Learning disabilities spectrum and sexual dimorphic abilities in girls with congenital adrenal hyperplasia. Journal of Child Neurology, 23, 862–869. Isgor, C., & Sengelaub, D. R. (1998). Prenatal gonadal steroids affect adult spatial behavior, CA1 and CA3 pyramidal cell morphology in rats. Hormones and Behavior, 34, 183–198. Janowsky, J. S., Oviatt, S. K., & Orwoll, E. S. (1994). Testosterone influences spatial cognition in older men. Behavioral Neuro­ science, 108, 325–332. Jansen, P., & Heil, M. (2010). Gender differences in mental rotation across adulthood. Experimental Aging Research, 36, 94–104. Johansson, I.  M., Birzniece, V., Lindblad, C., Olsson, T., & Bäckström, T. (2002). Allopregnanolone inhibits learning in the Morris water maze. Brain Research, 934, 125–131. Jonasson, Z. (2005). Meta-analysis of sex differences in rodent models of learning and memory: A review of behavioral and biological data. Neuroscience and Biobehavioral Reviews, 28, 811–825. Joseph, R., Hess, S., & Birecree, E. (1978). Effects of hormone manipulations and exploration on sex differences in maze learning. Behavioral Biology, 24, 364–377. Keenan, P.  A., Ezzat, W.  H., Ginsburg, K., & Moore, G.  J. (2001). Prefrontal cortex as the site of estrogen’s effect on cognition. Psychoneuroendocrinology, 36, 577–590. Kimura, D. (1995). Estrogen replacement therapy may protect against intellectual decline in postmenopausal women. Hormones and Behavior, 29, 312–321. Kocoska-Maras, L., Rådestad, A. F., Carlström, K., Backström, T., von Schoultz, B., & Hirschberg, A. L. (2013). Cognitive function in association with sex hormones in postmenopausal women. Gynecological Endocrinology, 29, 59–62. Korol, D. L., Unick, K., Goosens, K., Crane, C., Gold, P. E., & Foster, T. C. (1994). Estrogen effects on spatial performance and hippocampal physiology in female rats. Society for Neuroscience Abstracts, 20, 586.9.

Hampson

63

Kozaki, T., & Yasukouchi, A. (2008). Relationships between salivary estradiol and components of mental rotation in young men. Journal of Physiological Anthropology, 27, 19–24. Krug, R., Born, J., & Rasch, B. (2006). A 3-day estrogen treatment improves prefrontal cortex-dependent cognitive function in postmenopausal women. Psychoneuroendocrinology, 31, 965–975. Lacreuse, A., Chiavetta, M. R., Shirai, A.-A. C., Meyer, J. S., & Grow, D. R. (2009). Effects of testosterone on cognition in young adult male rhesus monkeys. Physiology & Behavior, 98, 524–531. Lacreuse, A., Mong, J.  A., & Hara, Y. (2015). Neurocognitive effects of estrogens across the adult lifespan in nonhuman primates: State of knowledge and new perspectives. Hormones and Behavior, 74, 157–166. Lejbak, L., Vrbancic, M., & Crossley, M. (2009). The female advantage in object location memory is robust to verbalizability and mode of presentation of test stimuli. Brain and Cognition, 36, 148–153. Leranth, C., Petnehazy, O., & MacLusky, N. J. (2003). Gonadal hormones affect spine synaptic density in the CA1 hippocampal subfield of male rats. Journal of Neuroscience, 23, 1588–1592. Maccoby, E.  E., & Jacklin, C.  N. (1974). The psychology of sex differences. Stanford, CA: Stanford University Press. MacLusky, N. J., Naftolin, F., & Goldman-Rakic, P. S. (1986). Estrogen formation and binding in the cerebral cortex of the developing rhesus monkey. Proceedings of the National Academy of Sciences of the United States of America, 83, 513–516. Maki, P. M., Rich, J. B., & Rosenbaum, R. S. (2002). Implicit memory varies across the menstrual cycle: Estrogen effects in young women. Neuropsychologia, 40, 518–529. Manning, J. T. (2002). Digit ratio: A pointer to fertility, behavior, and health. New Brunswick, NJ: Rutgers University Press. Mäntylä, T. (2013). Gender differences in multitasking reflect spatial ability. Psychological Science, 24, 514–520. Markham, J. A., Pych, J. C., & Juraska, J. M. (2002). Ovarian hormone replacement to aged ovariectomzied female rats benefits acquisition of the Morris water maze. Hormones and Behavior, 42, 284–293. McCormick, C.  M., & Teillon, S.  M. (2001). Menstrual cycle variation in spatial ability: Relation to salivary cortisol levels. Hormones and Behavior, 39, 29–38. McEwen, B. S., Akama, K. T., Spencer-Segal, J. L., Milner, T. A., & Waters, E.  M. (2012). Estrogen effects on the brain: Actions beyond the hypothalamus via novel mechanisms. Behavioral Neuroscience, 126, 4–16. McEwen, B. S., & Alves, S. E. (1999). Estrogen actions in the central nervous system. Endocrine Reviews, 20, 279–307. Merke, D.  P., & Bornstein, S.  R. (2005). Congenital adrenal hyperplasia. Lancet, 365, 2125–2136. Moffat, S. D., & Hampson, E. (1996). A curvilinear relationship between testosterone and spatial cognition in humans: Possible influence of hand preference. Psychoneuro­endocrinology, 21, 323–337. Moffat, S.  D., Hampson, E., & Hatzipantelis, M. (1998). Navigation in a “virtual” maze: Sex differences and correlation with psychometric measures of spatial ability in humans. Evolution and Human Behavior, 19, 73–87. Moffat, S.  D., Zonderman, A.  B., Metter, E.  J., Blackman, M.  R., Harman, S.  M., & Resnick, S.  M. (2002). Longitudinal assessment of serum free testosterone concentration predicts memory performance and cognitive

64

Sex Differences in Cognition

status in elderly men. Journal of Clinical Endocrinology & Metabolism, 87, 5001–5007. Moody, M. S. (1997). Changes in scores on the Mental Rotations Test during the menstrual cycle. Perceptual and Motor Skills, 84, 955–961. Mordecai, K. L., Rubin, L. H., & Maki, P. M. (2008). Effects of menstrual cycle phase and oral contraceptive use on verbal memory. Hormones and Behavior, 54, 286–293. Mueller, S. C., Temple, V., Oh, E., VanRyzin, C., Williams, A., Cornwell, B.,  .  .  .  Merke, D.  P. (2008). Early androgen exposure modulates spatial cognition in congenital adrenal hyperplasia (CAH). Psychoneuroendocrinology, 33, 973–980. Nass, R., & Baker, S. (1991). Learning disabilities in children with congenital adrenal hyperplasia. Journal of Child Neurology, 6, 306–312. Neave, N., Menaged, M., & Weightman, D.  R. (1999). Sex differences in cognition: The role of testosterone and sexual orientation. Brain and Cognition, 41, 245–262. Nottebohm, F., & Arnold, A. P. (1976). Sexual dimorphism in vocal control areas of the songbird brain. Science, 194, 211–213. Nowak, N.  T., Diamond, M.  P., Land, S.  J., & Moffat, S.  D. (2014). Contributions of sex, testosterone, and androgen receptor CAG repeat number to virtual Morris water maze performance. Psychoneuroendocrinology, 41, 13–22. O’Connor, D.  B., Archer, J., Hair, W.  M., & Wu, F.  C.  W. (2001). Activational effects of testosterone on cognitive function in men. Neuropsychologia, 39, 1385–1394. Overman, W.  H., Bachevalier, J., Schuhmann, E., & Ryan, P. (1996). Cognitive gender differences in very young children parallel biologically based cognitive gender differences in monkeys. Behavioral Neuroscience, 110, 673–684. Owen, A. M., Evans, A. C., & Petrides, M. (1996). Evidence for a two-stage model of spatial working memory processing within the lateral frontal cortex: A positron emission tomography study. Cerebral Cortex, 6, 31–38. Pasterski, V. L., Geffner, M. E., Brain, C., Hindmarsh, P., Brook, C., & Hines, M. (2005). Prenatal hormones and postnatal socialization by parents as determinants of male-typical toy play in girls with congenital adrenal hyperplasia. Child Development, 76, 264–278. Peper, J.  S., Koolschijn, C.  M.  P., & Crone, E.  A. (2013). Development of risk taking: Contributions from adolescent testosterone and the orbito-frontal cortex. Journal of Cognitive Neuroscience, 25, 2141–2150. Peters, M., Manning, J. T., & Reimers, S. (2007). The effects of sex, sexual orientation, and digit ratio (2D:4D) on mental rotation performance. Archives of Sexual Behavior, 36, 251–260. Phan, A., Suschkov, S., Molinaro, L., Reynolds, K., Lymer, J. M., Bailey, C.  D.  C., . . . Choleris, E. (2015). Rapid increases in immature synapses parallel estrogen-induced hippocampal learning enhancements. Proceedings of the National Academy of Sciences of the United States of America, 112, 16018–16023. Phillips, K., & Silverman, I. (1997). Differences in the relationship of menstrual cycle phase to spatial performance on two- and three-dimensional tasks. Hormones and Behavior, 32, 167–175. Phoenix, C. H., Goy, R. W., Gerall, A. A., & Young, W. C. (1959). Organizing action of prenatally administered testosterone propionate on the tissues mediating mating behavior in the female guinea pig. Endocrinology, 65, 369–382. Pintzka, C.  W.  S., Evensmoen, H.  R., Lehn, H., & Håberg, A. K. (2016). Changes in spatial cognition and brain activity

after a dose of testosterone in healthy women. Behavioural Brain Research, 298, 78–90. Pompili, A., Tomaz, C., Arnone, B., Tavares, M. C., & Gasbarri, A. (2010). Working and reference memory across the estrous cycle of the rat: A long-term study in gonadally intact females. Behavioural Brain Research, 213, 10–18. Puts, D. A., Cárdenas, R. A., Bailey, D. H., Burriss, R. P., Jordan, C. L., & Breedlove, S. M. (2010). Salivary testosterone does not predict mental rotation performance in men or women. Hormones and Behavior, 58, 282–289. Puts, D. A., McDaniel, M. A., Jordan, C. L., & Breedlove, S. M. (2008). Spatial ability and prenatal androgens: Meta-analyses of CAH and digit ratio (2D:4D) studies. Archives of Sexual Behavior, 37, 100–111. Rahman, Q., Abrahams, S., & Wilson, G.  D. (2003). Sexualorientation-related differences in verbal fluency. Neuropsychology, 17, 240–246. Rahman, Q., & Wilson, G. D. (2003). Large sexual-orientationrelated differences in performance on mental rotation and judgment of line orientation tasks. Neuropsychology, 17, 25–31. Raisman, G., & Field, P. M. (1973). Sexual dimorphism in the neuropil of the preoptic area of the rat and its dependence on neonatal androgen. Brain Research, 54, 1–29. Rapp, P.  R., Morrison, J.  H., & Roberts, J.  A. (2003). Cyclic estrogen replacement improves cognitive function in aged ovariectomized rhesus monkeys. Journal of Neuroscience, 23, 5708–5714. Reavis, R., & Overman, W. H. (2001). Adult sex differences on a decision-making task previously shown to depend on the orbital prefrontal cortex. Behavioral Neuroscience, 115, 196–206. Reber, J., & Tranel, D. (2017). Sex differences in the functional lateralization of emotion and decision making in the human brain. Journal of Neuroscience Research, 95, 270–278. Resnick, S. M., Berenbaum, S. A., Gottesman, I. I., & Bouchard, T.  J. (1986). Early hormonal influences on cognitive functioning in congenital adrenal hyperplasia. Developmental Psychology, 22, 191–198. Resnick, S.  M., Matsumoto, A.  M., Stephens-Shields, A.  J., Ellenberg, S. S., Gill, T. M., Shumaker, S. A., . . . Snyder, P. J. (2017). Testosterone treatment and cognitive function in older men with low testosterone and age-associated memory impairment. JAMA, 317, 717–727. Roof, R.  L., & Havens, M.  D. (1992). Testosterone improves maze performance and induces development of a male hippocampus in females. Brain Research, 572, 310–313. Ryan, B. C., & Vandenbergh, J. G. (2002). Intrauterine position effects. Neuroscience & Biobehavioral Reviews, 26, 665–678. Sandstrom, N.  J., Kim, J.  H., & Wasserman, M.  A. (2006). Testosterone modulates performance on a spatial working memory task in male rats. Hormones and Behavior, 50, 18–26. Schattmann, L., & Sherwin, B.  B. (2007). Testosterone levels and cognitive functioning in women with polycystic ovary syndrome and in healthy young women. Hormones and Behavior, 51, 587–596. Schmidt, P.  J., Keenan, P.  A., Schenkel, L.  A., Berlin, K., Gibson, C., & Rubinow, D. R. (2013). Cognitive performance in healthy women during induced hypogonadism and ovarian steroid addback. Archives of Women’s Mental Health, 16, 47–58. Schöning, S., Engelien, A., Kugel, H., Schäfer, S., Schiffbauer, H., Zwitserlood, P., . . . Konrad, C. (2007). Functional anatomy of visuo-spatial working memory during mental rotation is influenced by sex, menstrual cycle, and sex steroid hormones. Neuropsychologia, 45, 3203–3214.

Schumacher, M., Mattern, C., Ghoumari, A., Oudinet, J.  P., Liere, P., Labombarda, F., . . . Guennoun, R. (2014). Revisiting the role of progesterone and allopregnanolone in the nervous system: Resurgence of the progesterone receptors. Progress in Neurobiology, 113, 6–39. Shepard, R. N., & Metzler, J. (1971). Mental rotation of threedimensional objects. Science, 171, 701–703. Sherry, D.  F., & Hampson, E. (1997). Evolution and the hormonal control of sexually-dimorphic spatial abilities in humans. Trends in Cognitive Sciences, 1, 50–56. Shute, V.  J., Pellegrino, J.  W., Hubert, L., & Reynolds, R.  W. (1983). The relationship between androgen levels and human spatial abilities. Bulletin of the Psychonomic Society, 21, 465–468. Silverman, I., Kastuk, D., Choi, J., & Phillips, K. (1999). Testosterone levels and spatial ability in men. Psychoneuro­ endocrinology, 24, 813–822. Silverman, I., & Phillips, K. (1993). Effects of estrogen changes during the menstrual cycle on spatial performance. Ethology and Sociobiology, 14, 257–270. Šimić, N., & Santini, M. (2012). Verbal and spatial functions during different phases of the menstrual cycle. Psychiatria Danubina, 24, 73–79. Sinopoli, K.  J., Floresco, S.  B., & Galea, L.  A.  M. (2006). Systemic and local administration of estradiol into the prefrontal cortex or hippocampus differentially alters working memory. Neurobiology of Learning and Memory, 86, 293–304. Slabbekoorn, D., Van Goozen, S.  H.  M., Megen, J., Gooren, L. J. G., & Cohen-Kettenis, P. T. (1999). Activating effects of cross-sex hormones on cognitive functioning: A study of short-term and long-term hormone effects in transsexuals. Psychoneuroendocrinology, 24, 423–447. Smith, Y. R., Love, T., Persad, C. C., Tkaczyk, A., Nichols, T. E., & Zubieta, J. K. (2006). Impact of combined estradiol and norethindrone therapy on visuospatial working memory assessed by functional magnetic resonance imaging. Journal of Clinical Endocrinology & Metabolism, 91, 4476–4481. Snihur, A. W. K., Hampson, E., & Cain, D. P. (2008). Estradiol and corticosterone independently impair spatial navigation in the Morris water maze in adult female rats. Behavioural Brain Research, 187, 56–66. Spritzer, M.  D., Gill, M., Weinberg, A., & Galea, L.  A.  M. (2008). Castration differentially affects spatial working and reference memory in male rats. Archives of Sexual Behavior, 37, 19–29. Stanton, S.  J., Liening, S.  H., & Schultheiss, O.  C. (2011). Testosterone is positively associated with risk taking in the Iowa Gambling Task. Hormones and Behavior, 59, 252–256. Stewart, J., & Kolb, B. (1988). The effects of neonatal gonadec­ tomy and prenatal stress on cortical thickness and asymmetry in rats. Behavioral and Neural Biology, 49, 344–360. Stewart, J., Skvarenina, A., & Pottier, J. (1975). Effects of neonatal androgens on open-field behavior and maze learning in the prepubescent and adult rat. Physiology & Behavior, 14, 291–295. Sundström Poromaa, I., & Gingnell, M. (2014). Menstrual cycle influence on cognitive function and emotion processing— from a reproductive perspective. Frontiers in Neuroscience, 8(380). doi:10.3389/fnins.2014.00380 Tanapat, P., Hastings, N. B., Reeves, A. J., & Gould, E. (1999). Estrogen stimulates a transient increase in the number of new neurons in the dentate gyrus of the adult female rat. Journal of Neuroscience, 19, 5792–5801.

Hampson

65

Van Anders, S. M., & Hampson, E. (2005). Testing the prenatal androgen hypothesis: Measuring digit ratios, sexual orientation, and spatial abilities in adults. Hormones and Behavior, 47, 92–98. van den Bos, R., Homberg, J., & de Visser, L. (2013). A critical review of sex differences in decision-making tasks: Focus on the Iowa Gambling Task. Behavioural Brain Research, 238, 95–108. Van Goozen, S. H. M., Cohen-Kettenis, P. T., Gooren, L. J. G., Frijda N. H., & Van de Poll, N. E. (1995). Gender differences in behaviour: Activating effects of cross-sex hormones. Psychoneuroendocrinology, 20, 343–363. Vandenberg, S.  G., & Kuse, A.  R. (1978). Mental rotations: A  group test of three-dimensional spatial visualization. Perceptual and Motor Skills, 47, 599–601. Voyer, D., Voyer, S., & Bryden, M.  P. (1995). Magnitude of sex differences in spatial abilities: A meta-analysis and consideration of critical variables. Psychological Bulletin, 117, 250–270. Voyer, D., Voyer, S. D., & Saint-Aubin, J. (2017). Sex differences in visual-spatial working memory: A meta-analysis. Psychonomic Bulletin and Review, 24, 307–334. Vuoksimaa, E., Kaprio, J., Kremen, W.  S., Hokkanen, L., Viken,  R.  J., Tuulio-Henriksson, A., & Rose, R.  J. (2010). Having a male co-twin masculinizes mental rotation performance in females. Psychological Science, 21, 1069–1071. Warren, S. G., & Juraska, J. M. (1997). Spatial and nonspatial learning across the rat estrous cycle. Behavioral Neuroscience, 111, 259–266. Welborn, B.  L., Papademetris, X., Reis, D.  L., Rajeevan, N., Bloise, S. M., & Gray, J. R. (2009). Variation in orbitofrontal cortex volume: Relation to sex, emotion regulation and affect. Social Cognitive and Affective Neuroscience, 4, 328–339. Wharton, W., Hirshman, E., Merritt, P., Doyle, L., Paris, S., & Gleason, C. (2008). Oral contraceptives and androgenicity: Influences on visuospatial task performance in younger indivi­duals. Experimental and Clinical Psychopharmacology, 16, 156–164.

66

Sex Differences in Cognition

Williams, C.  L., Barnett, A.  M., & Meck, W.  H. (1990). Organizational effects of early gonadal secretions on sexual differentiation in spatial memory. Behavioral Neuroscience, 104, 84–97. Wilson, J.  R., & Vandenberg, S.  G. (1978). Sex differences in cognition: Evidence from the Hawaii Family Study. In T. E. McGill, D. A. Dewsbury, & B. D. Sachs (Eds.), Sex and behavior: Status and prospectus (pp. 317–335). New York, NY: Plenum Press. Wimer, C. C., & Wimer, R. E. (1989). On the sources of strain and sex differences in granule cell number in the dentate area of house mice. Developmental Brain Research, 48, 167–176. Witelson, S. F., Glezer, I. I., & Kigar, D. L. (1995). Women have greater density of neurons in posterior temporal cortex. Journal of Neuroscience, 15, 3418–3428. Wolf, O.  T., Preut, R., Hellhammer, D.  H., Kudielka, B.  M., Schürmeyer, T. H., & Kirschbaum, C. (2000). Testosterone and cognition in elderly men: A single testosterone injection blocks the practice effect in verbal fluency, but has no effect on spatial verbal memory. Biological Psychiatry, 47, 650–654. Woolley, C., & McEwen, B.  S. (1992). Estradiol mediates fluctuation in hippocampal synapse density during the estrous cycle in the adult rat. Journal of Neuroscience, 12, 2549–2554. Yang, C.-F. J., Hooven, C. K., Boynes, M., Gray, P. B., & Pope Jr., H.  G. (2007). Testosterone levels and mental rotation performance in Chinese men. Hormones and Behavior, 51, 373–378. Yonker, J. E., Eriksson, E., Nilsson, L.-G., & Herlitz, A. (2006). Negative association of testosterone on spatial visualization in 35 to 80 year old men. Cortex, 42, 376–386. Zacks, J. M. (2008). Neuroimaging studies of mental rotation: A meta-analysis and review. Journal of Cognitive Neuroscience, 20, 1–19. Zitzmann, M. (2009). The role of the CAG repeat androgen receptor polymorphism in andrology. Frontiers in Hormone Research, 37, 52–61.

CH A PT E R

5

Involvement of the Sex Hormones in Learning and Memory

Kelsy S. J. Ervin and Elena Choleris

Abstract Learning and memory can be defined as the processes by which we acquire information about our environment and experiences, to be used later in similar situations. These cognitive processes are important for behaviors crucial to survival, such as finding food and shelter, assessing risk, and behaving appropriately in social contexts. Research provides insights into how hormones influence animal and human cognition. Here, we focus on the sex hormones, which influence both learning or acquisition and memory consolidation through two mechanisms: (i) the genomic pathways, acting longer-term to modulate gene transcription, protein expression, and structural changes implicated in establishing memory traces; and (ii) rapid pathways in which receptor activation initiates cell signaling cascades and more immediate neural responses. Investigating the roles of sex hormones in learning and memory enhances our understanding of neural networks involved in different types of learning, in ways that may be applied to human health and cognition. Keywords:  Learning, memory, sex steroids, encoding, retrieval, acquisition

The sex hormones are steroid molecules synthesized via multiple steroidogenic pathways from cholesterol (see Nelson,  2017, for an introduction to sex ­hormones). Steroidogenesis has been described in the gonads and other organs, including the brain. All three classes of sex hormones (i.e., androgens, estrogens, and progestogens) are produced in both biological sexes and act on multiple physiological functions, including behavior and cognition. Progesterone is the most abundant of the progestogens; it binds to the progesterone receptor (PR), and it can be converted to a number of metabolites that act on other neurochemical systems (reviewed in Compagnone & Mellon, 2000). For example, allopregnanolone (5α-tetrahydroprogesterone) is an allosteric modulator of the GABAA receptor. Thus, progestogens can affect biological systems and behavior either directly, via the PR, or indirectly, through their metabolites. With a few synthesizing steps, progesterone is also a precursor to the other two classes of sex steroids and other steroid hormones, including aldosterone,

cortisol, and corticosterone (Compagnone & Mellon, 2000). Androgens, such as testosterone (T), act on the androgen receptor (AR). Testosterone can be converted to the potent androgen dihydrotestosterone (DHT), GABAA active metabolites, and the three estrogens 17β-estradiol, estrone, and estriol. The conversion of testosterone to estrogens is catalyzed by the aromatase enzyme and is very relevant to our understanding of testosterone effects on brain and cognition (reviewed in Atwi, McMahon, Scharfman, & MacLusky,  2016) because testosterone affects behavior via both androgenic (AR-mediated) and estrogenic (estrogen receptor– mediated after aromatization) mechanisms (reviewed in Atwi et al.,  2016). In humans, 17β-estradiol, also referred to simply as estradiol, is the most biologically active of the estrogens and it is also the most abundant estrogen during the reproductive years. Postmenopausally, 17β-estradiol levels greatly decline and estrone ­becomes the most abundant estrogen (Rannevik, Carlström, Jeppsson, Bjerre, & Svanberg, 67

1986; reviewed in Galea, Frick, Hampson, Sohrabji, & Choleris,  2017). Estrogens act via the estrogen receptors (ERs), the better described being ERα, ERβ, and the G protein–coupled ER 1 (GPER1). Other ERs have also been described, including ER-X and the Gq-coupled membrane ER (Gq-mER), but their functions are less characterized (reviewed in Galea et al., 2017; Srivastava & Evans, 2013). The steroid hormones have long been known for their genomic actions. PR, AR, ERα, and ERβ are all intracellular ligand-regulated transcription factors (reviewed in Gustafsson, 2016; although some ligand-independent activity also exists, reviewed in Heldring et al.,  2007) that, in combination with multiple coactivators, can drive the activity of thousands of genes (Carroll et al., 2006). These genomic effects are delayed in their onset and are long-lasting (hours to days). Much more rapid (minutes) effects of the steroid hormones have also been described and recently have been receiving much research attention, especially in investigations on estrogen effects. The rapid effects of estrogens are being found to be mediated by the classical transcription factor receptors and the membrane-bound GPER1 and Gq-mER, and depend on membrane-associated actions of intracellular ERs and/or membranebound ERs triggering intracellular signaling cascades (reviewed in Galea et al., 2017; Srivastava, Woolfrey, & Penzes, 2013). Three main cellular pathways have been identified that are rapidly triggered by estrogens and affect cell function: (1) rapid activation of actin signaling cascades (e.g., Rap, RhoA); (2) rapid initiation of second messenger systems (i.e., signaling molecules released by the cell to trigger physio­ logical changes) such as protein kinase C (PKC), protein kinase A (PKA), the Akt signaling pathway, extracellular signal-regulated kinase (ERK) 1 and 3, and the c-Jun N-terminal kinase (JNK) pathway; and (3) rapid modulation of local protein synthesis mechanisms such as 4E-BP1 (for recent reviews, see Frick, Kim, Tuscher, & Fortress,  2015; Srivastava et al., 2013). Sex hormones act in the brain to regulate or modulate multiple functions, including learning and memory, and dysfunctions of these actions have been linked to a number of psychiatric disorders such as anxiety, depression, and various types of cognitive impairments including autism spectrum disorder and Alzheimer’s disease (reviewed in Gobinath, Choleris, & Galea, 2017). This chapter focuses on the involvement of the sex hormones in learning and memory and describes effects that are mediated via delayed long-term and rapid mechanisms. 68

Learning, the acquisition of novel information, involves the formation of new memories through various stages including acquisition (initial encoding of the memory), consolidation, and recall (retrieval of the consolidated memory). In more recent years, a fourth stage has been added: reconsolidation, which is triggered by the reactivation of consolidated memories at recall (Nader & Hardt, 2009). During consolidation and reconsolidation, memories are labile to both interfering and enhancing effects (reviewed in Kandel, Dudai, & Mayford, 2014). Two types of memory consolidation have been described: cellular and system consolidation. Cellular consolidation occurs within the first one to two hours postacquisition and requires DNA transcription and protein synthesis. Following the initial cellular consolidation, there is evidence that system consolidation occurs over multiple days/weeks and involves the reorganization of brain structures and the transfer of the new memories to long-term storage (Clark, Broadbent, Zola, & Squire, 2002; Dudai, 2004; Frankland & Bontempi, 2005). Consolidated memories are characterized by slow decay and resistance to disruption (reviewed in Kandel et al.,  2014). Experimental manipulations can target the various phases of memory to elucidate their underlying mechanisms and to determine the influence of specific chemicals. For example, effects of a treatment that was administered before the acquisition phase can be attributed to effects on the initial encoding and/or consolidation (depending on how long that treatment lingers in the system) phases of memory formation. Instead, effects observed shortly after acquisition can elucidate action on the consolidation phase. When treatments are administered after a certain retention delay after acquisition and consolidation, any effects can be attributed to mechanisms underlying the retrieval or reconsolidation of a memory. Finally, treatment effects postretrieval can identify the processes underlying memory reconsolidation (e.g., Winters, Tucci, & DaCosta-Furtado, 2009). Researchers have been using this approach to determine the effects of sex steroids on learning and memory. Long-term potentiation (LTP) and long-term depression (LTD) are key cellular processes of synaptic plasticity that are believed to be critical for memory acquisition/encoding and cellular consolidation (reviewed in Kandel et al., 2014). LTP is the enhanced responsiveness of a synapse following intense stimulation and was initially described in the dentate gyrus region of the hippocampus (Bliss & Lømo, 1973) and subsequently found in the CA1

Involvement of the Sex Hormones in Learning and Memory

and CA3 hippocampal regions and other brain areas, with similar and different characteristics (reviewed in Kandel et al., 2014). In addition to synaptic plasticity, structural plasticity—changes in dendritic spine and spine synapse density, patterns of dendritic arborization, and neuronal survival—in the hippocampus and other regions have been implicated in memory. Dendritic spines are the site of most excitatory synapses, and an increase in their number and changes in their structural and molecular characteristics are associated with LTP and the formation and consolidation of new memories (reviewed in Bailey, Kandel, & Harris,  2015). Neurogenesis (i.e., the proliferation, survival, migration, and functional integration of new neurons) in the hippocampal circuit has also been linked to new and consolidated memories in humans and other animals (reviewed in Gonçalves, Schafer, & Gage, 2016; Ihunwo, Tembo, & Dzamalala, 2016). All three classes of sex hormones affect synaptic and structural plasticity in the brain. LTP and synaptic plasticity are enhanced by estrogens (reviewed in Srivastava et al., 2013) and inhibited by testosterone (Skucas et al., 2013). Progesterone was found to enhance, inhibit, or not affect LTP (reviewed in Hansberg-Pastor, González-Arenas, Piña-Medina, & Camacho-Arroyo, 2015). Dendritic spines are also regulated by sex hormones. In females, dendritic spine density peaks when endogenous estrogen levels are high and is increased with both chronic and within 20-40 minutes of acute estrogen treatments (reviewed in Frankfurt & Luine,  2015; Phan, Lancaster, Armstrong, MacLusky, & Choleris, 2011; Phan et al., 2015; Srivastava et al.,  2013). Progesterone also enhances dendritic spines after chronic treatments (reviewed in Hansberg-Pastor e­ t al., 2015). In males, dendritic spines fluctuate with androgen levels, and this appears to be an androgen- and not estrogenmediated effect. Conversely, in females, testosteroneenhancing effects on dendritic spines are mediated by estrogenic mechanisms (reviewed in Atwi et al., 2016). Sex hormones also affect various aspects of neurogenesis. In adult females, endogenous hormonal fluctuations are correlated with hippocampal neurogenesis in a manner consistent with enhancing effects of ovarian hormones (reviewed in Mahmoud, Wainwright, & Galea, 2016). Several investigations have highlighted a role for estrogens in cell proliferation and survival even though these effects seem to be modulated by age and hormonal environment at the time of treatment (reviewed in Mahmoud et al., 2016). Limited investigations into the effects of progesterone on neurogenesis point at a modulatory

role on the effects of estrogens in the female ­hippocampus (reviewed in Mahmoud et al., 2016). In adult males, long-term exposure to androgens enhances the survival of new neurons in the hippocampus, whereas estrogens seem to enhance cell survival only with short-term exposure during axon growth (reviewed in Atwi et al., 2016; Mahmoud et al., 2016). In summary, estrogens have been shown to enhance LTP, dendritic spines, dendritic arborization, and neurogenesis. The effects of progesterone and testosterone are less investigated (especially progesterone effects) but are often similar to those of estrogens, even though results tend to be more mixed. Collectively, these effects of sex hormones on brain plasticity may explain why these hormones also have profound effects on learning and memory, as reviewed next.

Sex Hormones and Spatial Memory

Spatial memory involves memory for spatial positioning and/or for environmental or contextual cues. In tests of spatial memory, spatial cues or context may be learned passively or as the result of predicting an outcome such as a reward or an aversive experience (in the laboratory the latter is often a mild electric shock). In both humans and animals, the hippocampus is necessary for encoding of spatial cues. Spatial memory can be tested in animals using mazes (e.g., Morris water maze, radial arm maze), memory for context (e.g., contextual fear conditioning), or spatial recognition tasks (e.g., object placement tasks; Luine & Frankfurt,  2012; Tuscher, Fortress, Kim, & Frick, 2015; Figure 5.1). Similar tasks are performed in humans with the advantage that humans can report what they remember (Squire, 1992).

The Study of Spatial Memory in Laboratory Rodents

In spatial versions of the radial arm maze and Morris water maze, animals are provided with extra-maze cues with which they can learn and locate the position of an escape platform (Morris water maze) or in which arm they receive a food reward (radial arm maze). Another test of spatial memory in rodents is the object placement task, a spatial recognition task in which an animal is habituated to objects in a testing area (e.g., open field, home cage) and, at test, one object is moved to a new location (Figure 5.1A). Rodents naturally investigate novel environmental stimuli; therefore, preferential investigation of the displaced object indicates that the animal identifies the object’s position as new and/or recognizes that the other object has not been Ervin and Choleris

69

moved from the old position (Ervin, Phan, Gabor, & Choleris, 2013; Koss & Frick, 2017; Luine, 2015; Tuscher et al., 2015). Less common in the literature on sex hormones and behavior are hippocampusdependent contextual fear conditioning and inhibitory avoidance tasks. In inhibitory avoidance, the animal learns to avoid a location it normally prefers (e.g., a nonelevated location or a dark chamber) when it is associated with footshock, and instead to inhibit leaving a nonpreferred location (e.g., an elevated platform or a brightly lit area; Ervin et al., 2013; Luine, 2014; Luine & Frankfurt, 2012; Figure 5.1B).

Endogenous Hormones and Spatial Memory

In humans, only subtle sex differences exist in spatial memory abilities. No consistent sex differences emerge in tests using a human version of a radial arm maze, and men only outperform women on an adapted Morris water maze with no landmarks available, suggesting that even though performance is comparable in the sexes, men and women use different strategies to encode spatial information (Koss & (A)

Frick, 2017). Sex differences in mental rotation tasks are often cited as differences in spatial abilities; however, there is evidence that mental rotation depends on object working memory, rather than spatial working memory (Hyun & Luck, 2007; Kaufman, 2007). In this section on spatial memory, we will focus on hippocampus-dependent spatial memory. Sex differences in the spatial memory abilities of laboratory rodents further support a role for the gonadal hormones in learning and memory for spatial environmental cues. For example, male rats typically outperform females on the radial arm maze (Luine,  2014,  2015). Studies of female rodents across the stages of the estrous cycle and during pregnancy further implicate the sex hormones in regulation of spatial memory. Female rats and mice in the proestrus phase (high circulating hormones) of the estrous cycle tend to perform better than those in diestrus (low circulating hormones) on object placement, inhibitory avoidance, and the Morris water maze (Frick, 2009; Luine & Frankfurt, 2012; Tuscher et al., 2015). Similarly, pregnant, postpartum, and lactating rats also outperform rats that are (B)

Figure 5.1  Spatial memory tasks. Both object placement (A) and inhibitory avoidance (B) are frequently used to test hippocampaldependent spatial memory in rodents. In object placement (A), an animal is presented with and allowed to investigate one or more objects either in an open field or in the home cage. At test, one object is moved to a new location and the animal shows intact spatial memory if it preferentially investigates the object in the new location (e.g., the cylinder in the diagram). In the inhibitory avoidance task (B), an animal is placed in a nonpreferred location such as a brightly lit chamber or elevated platform. When the animal steps down off the platform or through to a darkened chamber, a normally preferred location, it receives a mild foot shock. If the animal remembers the context in which it receives the foot shock, it will inhibit or delay entering the darkened chamber or stepping down off the platform. Other spatial memory tasks include the Morris water maze and radial-arm mazes, in which animals navigate the maze to find an escape platform or food reward, respectively, using extra-maze visual cues.

70

Involvement of the Sex Hormones in Learning and Memory

not pregnant or are in early pregnancy, analogous to the first trimester in humans. Although estrogens and progesterone are low during lactation, it is possible that effects on hippocampal plasticity by steroid hormones are sustained by the effects of elevated prenatal hormones and/or that other hormones are involved in enhancing spatial memory (Tuscher et al., 2015). Multiparous rats perform better on object placement than primiparous rats, suggesting that these sex hormones have a sustained effect on memory (Tuscher et al., 2015). Complementary to these findings, ovariectomy impairs spatial memory, and as female rats and mice age, circulating sex hormones decrease and reproductive function declines, and spatial memory on object placement and in the Y- and T-maze, radial arm maze, and Morris water maze is impaired (Frick,  2009; Luine & Frankfurt,  2012; Tuscher et  al., 2015). Collectively, these results point at a role for gonadal hormones in spatial learning.

Estrogenic Enhancement of Spatial Memory in Female Rodents

While aged and ovariectomized female rodents are impaired in spatial memory, chronic treatment with estrogens typically restores or ameliorates performance. Chronic 17β-estradiol or estradiol benzoate treatment ameliorates performance on object placement, delayed matching to sample, the T-maze, the radial arm maze, and the Morris water maze in ovariectomized rats and mice (Fortress & Frick,  2014; Luine & Frankfurt,  2012; Tuscher et al.,  2015). Similarly, spatial memory can be improved in aged rats and mice with estradiol treatments, although effectiveness varies depending on the timing and dose (Frick, 2009; Luine & Frankfurt, 2012). Studies using acute treatments of estrogens provide further evidence for specific estrogenic regulation of spatial memory. Preacquisition treatment with 17β-estradiol improved object placement in ovariectomized rats when given in two doses over two days (Jacome et al., 2010) or in ovariectomized mice when tested within 40 minutes of one acute treatment (Phan et al., 2012). In the latter study, the short duration of the test likely rules out classical genomic effects of estradiol, and the learning improvement was likely due to rapid mechanisms of action through activation of intracellular signaling cascades (Ervin et al., 2013; Luine & Frankfurt, 2012; Phan et al., 2012; Tuscher et al., 2015). Postacquisition treatments with estradiol show a similar pattern of effects. Estradiol administered immediately postacquisition to ovariectomized female mice improved object placement memory when tested 24 hours after

acquisition. Similar treatment enhanced memory in object placement, the Morris water maze, and inhibitory avoidance in ovariectomized female rats, and in the Morris water maze in castrated male rats (Luine,  2015; Luine & Frankfurt,  2012; Tuscher et al., 2015). Estradiol treatment is only effective at enhancing object recognition memory if it is given immediately postacquisition; if given after a delay of even one hour, it is ineffective (Ervin et al., 2013; Frick et al., 2015; Luine, 2015; Luine & Frankfurt, 2012; Tuscher et al., 2015). Animals in these studies were tested at least four hours after training, at a time when memory is consolidated via DNA transcription-dependent processes (Bourtchuladze et al., 1994; Da Silva et al., 2008; Nguyen, Abel, & Kandel, 1994). However, there is an accumulating literature supporting the idea that estradiol’s action through rapid nonnuclear cell signaling pathways is necessary for the memory-enhancing effects on object recognition (Fortress & Frick, 2014; Luine, 2014; Tuscher et al., 2015). Thus, estrogens enhance spatial memories, likely through both genomic and nongenomic mechanisms, the latter likely via cell signaling processes that are also involved in nonspatial memory.

Estrogen Action in the Hippocampus

There is growing evidence identifying the hippocampus as a site of estrogenic improvements in spatial memory, as well as in other types of learning and memory. Recently, it has become clear that estrogens are synthesized de novo in the hippocampus. Although estrogens in the hippocampus correlate with fluctuating levels of circulating hormones over the estrous cycle, hippocampal levels of estradiol are about 40 times higher than circulating levels in female and male rats (Luine,  2014; Srivastava & Penzes, 2011). Furthermore, induced hippocampal neural activity increases local estradiol synthesis in rats and songbirds, and aromatase has been detected in the human brain (Luine,  2014). This suggests that effects of hippocampal estrogens in these model species may generalize to humans. Estradiol treatment also affects dendritic morphology in the hippocampus. Estradiol administered systemically or to hippocampal slices ex vivo increases the density of dendritic spines, the postsynaptic sites of excitatory neural transmission, and ovariectomy reduces hippocampal dendritic spine density (Fortress & Frick, 2014; Koss & Frick, 2017; Luine, 2014; Phan et al., 2015). Increases in spine density occur as early as 30 minutes after systemic treatment with estradiol; this and further manipulations of downstream cell signaling targets of estradiol Ervin and Choleris

71

indicate that estradiol likely acts through rapid mechanisms to influence hippocampal dendritic morphology (Srivastava et al., 2013). These actions of estrogens are thought to enhance memory by providing the substrate upon which new memories may be formed (Ervin et al., 2013; Fortress & Frick, 2014; Koss & Frick, 2017; Luine, 2014; Phan et al., 2015). With regard to spatial memory specifically, 17β-estradiol infused into the dorsal hippocampus immediately postacquisition enhanced object placement memory in ovariectomized rats and mice (Kim, Szinte, Boulware, & Frick, 2016; Luine, 2015; Luine & Frankfurt, 2012; Tuscher et al., 2015) and improved Morris water maze performance in both ovariectomized female and castrated male rats (Packard, 1998). Preacquisition infusion of estradiol similarly facilitated object placement in ovariectomized mice tested within 40 minutes of treatment (Phan et al., 2015). While the specific intracellular mechanisms of the estrogenic effects on spatial memory are unclear, studies of their effects on object memory suggest involvement of the estrogen-­ mediated activation of ERK, phosphatidylinositol 3-kinase (PI3K), mammalian target of rapamycin (mTOR) pathways, and/or epigenetic changes to histone acetylation or DNA methylation (Fortress & Frick,  2014; Kim et al.,  2016; Luine,  2014; Tuscher et al., 2015).

Roles of Estrogen Receptors in Estrogenic Enhancement of Spatial Memory

Research has investigated whether ERα, ERβ, or GPER1 plays a role in estrogenic enhancement of spatial memory. Long-term manipulations of the receptors suggest that ERβ plays a prominent role. Mice with ERβ gene knockout (ERβKO) exhibit impaired Morris water maze performance, and estradiol was only effective in improving performance on object placement and in a spatial Y-maze task in ERαKO and wild-type control mice (with intact ERβ expression), but not in ERβKO mice (Luine & Frankfurt, 2012; Tuscher et al., 2015). Treatment with the ERβ agonist diarylpropionitrile (DPN) also improved object placement in wild-type but not ERβKO mice and performance on delayed matching to sample and the radial arm maze in female rats (Luine, 2014; Luine & Frankfurt, 2012; Tuscher et al., 2015). The ERα agonist propyl pyrazole triol (PPT) was ineffective at improving spatial memory via delayed, long-term mechanisms, in which animals received PPT treatments 48 hours or more prior to testing, except in one study in which chronic PPT treatment enhanced delayed matching to sample (Hammond, 72

Mauk, Ninaci, Nelson, & Gibbs, 2009). Thus, ERα activation may facilitate spatial memory but is not necessary for estrogenic enhancement of performance on these tasks. Alternatively, chronic treatment may lead to a loss of selectivity of the agonist for the receptor (Luine & Frankfurt, 2012). ER-specific agonists also allow the investigation of the contributions of rapid estrogenic effects on spatial memory. Preacquisition and postacquisition treatments with PPT, both systemically and in the hippocampus, enhanced object placement in female mice (Frye, Duffy, & Walf, 2007; Phan et al., 2011, 2015). However, postacquisition PPT in female rats had no  effect on object placement or Morris water maze  performance (Jacome et al.,  2010; Rhodes & Frye, 2006). Differences in PPT effects on spatial memory could be due to differences in ER expression in rats and mice, or to dosage and timing of treatments with respect to the phases of memory, acquisition, or consolidation (Ervin et al., 2013; Luine & Frankfurt, 2012; Tuscher et al., 2015). Consistent with ERKO studies, preacquisition treatment with the ERβ agonist DPN enhanced object placement in female mice, and postacquisition DPN enhanced object placement and Morris water maze performance in female rats (Ervin et al., 2013; Luine & Frankfurt, 2012), confirming a prominent role for ERβ in spatial memory. Less is known about the role of the GPER1, but chronic treatment with the agonist G1 enhanced spatial delayed matching to  position in ovariectomized female rats, and acute systemic and intrahippocampal preacquisition treatment rapidly facilitated object placement in ovariectomized female mice (Ervin et al., 2013; Gabor, Lymer, Phan, & Choleris, 2015; Hammond, Nelson, Kline, & Gibbs, 2012; Hawley et al., 2013; Kim et al.,  2016; Lymer, Robinson, Winters, & Choleris,  2017). GPER1-mediated enhancements of spatial memory are dependent on the JNK pathway, as inhibition of c-Jun phosphorylation blocks the enhancing effects of the agonist G1, but ERK inhibition has no effect (reviewed in Frick, 2015). This suggests a mechanism of GPER1’s role in spatial memory that is distinct from ERα and ERβ.

Progesterone Effects on Spatial Memory

Less research has focused on the role of progesterone in spatial learning and memory. In young adult rats and mice, postacquisition progesterone enhanced object placement memory (Frye et al., 2007; Frye, Koonce, & Walf, 2013; but see Frye & Walf, 2008). Similarly, chronic and acute postacquisition progesterone improved performance in

Involvement of the Sex Hormones in Learning and Memory

the object placement task in aged mice (Frye & Walf,  2008,  2010). When administered chronically, progesterone impaired inhibitory avoidance, but acute treatment given immediately posttraining improved performance (Farr et al.,  1995; Frye & Lacey, 2000). These findings suggest either that progesterone impairs memory only through genomic mechanisms or that rather than impairing memory, chronic treatment with progesterone may have affected inhibitory avoidance performance through the analgesic and anxiolytic effects of its metabolites, such as allopregnanolone (Bitran, Hilvers, & Kellogg, 1991; Frye & Duncan,  1994). Intrahippocampal progesterone treatment had no effect on Morris water maze performance (Tuscher et al.,  2015). Overall, these findings suggest that the role of ­progesterone in spatial learning may be less robust than that of estrogens, but further investigation is warranted.

Androgen-Mediated Effects on Spatial Memory

Although some androgen effects on spatial memory are attributable to testosterone conversion to estrogens via the aromatase enzyme, there is likely a nonestrogenic mechanism that also influences performance. In castrated male rats that are impaired in spatial memory tasks, chronic testosterone propionate e­nhances object placement memory and performance on Y-maze, T-maze, and radial arm maze tasks (Gibbs,  2005; Spritzer et al.,  2011; Hawley et al., 2014). In the test of object placement, similar improvements in memory could not be replicated with chronic treatment with estradiol benzoate, suggesting that in males, androgen receptor–­mediated mechanisms contribute to spatial memory (Luine, 2015).

Conclusions

Sex differences in spatial memory capabilities seem to involve the action of gonadal steroid hormones in the brain. Estrogens play a major role in the acquisition and consolidation of spatial information, likely through their action in the hippocampus. In females, the ERβ seems to mediate these effects, both rapidly and longterm, though ERα and GPER1 also influence spatial memory through rapid effects (Ervin, Lymer, et al., 2015; Tuscher et al., 2015). In males, testosterone has similar enhancing effects on spatial learning and memory, some of which occur through aromatization to estradiol and action on the ERs. However, androgen-specific effects on spatial memory have been found (Luine, 2015). The hippocampus is a prominent site of neurosteroid action

with regard to spatial information processing, and there are sex differences in the structure and function of the rodent hippocampus. In addition to, or potentially because of, these differences, male and female rodent hippocampal circuits are affected differently by gonadectomy. Notably, gonadectomy generally decreases CA1 dendritic spine density in females, but not in males. Castrated males also have enhanced LTP, neuronal excitability, and synaptic transmission in the hippocampus, while the opposite is true for ovariectomized females. Castrated male rats also have longer, more branching dendrites in the CA3 than intact controls, and have more mossy fiber projections to the CA3, both important for establishing LTP (Atwi et al., 2016; Mendell et al., 2017). Rather than being mediated through estrogenic pathways from testosterone aromatized to estradiol, it seems these effects in males are mediated by androgenic interaction with brain-derived neurotrophic factor (BDNF) and/or by other metabolites such as dihydrotestosterone and 5α-androstane-3α17β-diol (Atwi et al., 2016; Mendell et al., 2017). These structural and functional sex differences in the hippocampus suggest organizational as well as activational effects of neurosteroids and highlight a need to understand their effects through development. Spatial memory is important in animals for a variety of essential processes, such as finding food, identifying a dangerous environment, and establishing a territory (Fagan et al., 2013; Sherry, 1998). Territoriality in particular is important for both male and female mate selection, and this could be one reason that sex hormones influence cognitive aspects of territorial behaviors. The benefits of sex hormones, notably estrogens, on hippocampaldependent memory could be specific to spatial memory. However, estrogens’ effects on hippocampal dendritic plasticity, neurogenesis, and epigenetic changes could support learning and memory more generally.

Sex Hormones and Declarative Object Memory

Object memory is a type of declarative memory well characterized in humans and other animals. Humans can be asked to recall whether they recognize an object or image. Object memory in rats and mice is typically tested using an object recognition task. Similar to object placement, animals are presented with one, two, or more objects to freely explore during a training or acquisition phase (Figure 5.2A). After one or more training exposures, the animal is tested for recognition by swapping one now-familiar Ervin and Choleris

73

seen with spatial memory, female rats and mice in the proestrus phase (high circulating hormones) of the estrous cycle tend to perform better than those in diestrus (low circulating hormones), pregnant rats perform better than rats that are not pregnant, and aged female rats with reduced reproductive function are impaired on object recognition (Luine & Frankfurt,  2012; Tuscher et al.,  2015). These findings suggest an important role for the sex hormones in object recognition.

(A)

Estradiol Effects on Object Recognition Memory (B)

Figure 5.2  Object memory task. Object memory in rodents is commonly tested using the object recognition task. The animal is first presented with one or more objects in an open field or in the home cage (A). After one or more of these habituation sessions, the animal is given a test in which one object is swapped for a novel object (B). If the animal preferentially investigates the novel object (e.g., the pyramid in the diagram), it is interpreted as having recognized or shown intact object memory for the object it investigated during the previous habituation sessions (e.g., the cube).

object for a new object. If the animal recognizes the previously encountered object, it will spend more time exploring the new object and less time with the familiar objects from the training phases (e.g., Ervin et al., 2013; Luine, 2015; Tuscher et al., 2015; Winters, Saksida, & Bussey,  2008; Figure  5.2B). Object memory uses the ventral visual stream, and the perirhinal cortex has been shown to be particularly important (Winters et al., 2008).

Endogenous Hormones and Rodent Object Recognition

Male and female animals with high levels of circulating sex hormones display enhanced object recognition, whereas gonadectomy impairs it in both sexes (Frick et al., 2015; Luine, 2015). Similar to patterns 74

Although estrogens, progesterone, and androgens have all been implicated in modulating object recognition, most of the research has focused on estrogens. Chronic 17β-estradiol or estradiol benzoate preserved object recognition abilities in ovariectomized female rats and mice (Luine, 2015; Luine & Frankfurt, 2012; Tuscher et al., 2015). Similarly, acute treatments with 17β-estradiol, estradiol benzoate, or 17α-estradiol administered before the acquisition phase enhanced object recognition (Luine & Frankfurt, 2012; Tuscher et al., 2015). Treatments in these studies were administered at least four hours before the test or recognition phase, therefore the memory-enhancing effects may be mediated by either delayed long-term or rapid effects of estrogens, or both. Acute, preacquisition treatments in which the animal was tested within 40 minutes provide evidence that rapid effects of estrogens play a critical role in object recognition (Phan et al., 2012, 2015). In many other studies, estradiol treatment was administered immediately postacquisition, ruling out estradiol’s effects on acquisition, or learning about the objects, which allows for assessment of its effects on memory consolidation specifically. As with spatial memory tasks, estradiol only enhances object recognition memory if administered immediately after the acquisition phase (Ervin et al.,  2013; Frick et al., 2015; Luine, 2015; Luine & Frankfurt, 2012; Tuscher et al., 2015). These changes depend on ERK signaling, as inhibition of ERK phosphorylation and its downstream targets prevents estradiol-mediated improvements of object recognition (Fernandez et al., 2008; Fortress, Fan, Orr, Zhao, & Frick, 2013; Lewis, Kerr, Orr, & Frick, 2008). Other research also implicates the PI3K, mTOR, and PKA pathways; inhibition of any of these pathways abolishes the improving effects of estradiol on object recognition memory. Estradiol interactions with the N-methyl-D-aspartate (NMDA) and metabotropic glutamate receptors are also

Involvement of the Sex Hormones in Learning and Memory

necessary for its improvements of object memory (Fortress & Frick, 2014; Luine, 2014). Longer term actions of estrogens on object ­recognition may be mediated through epigenetic mechanisms, likely both through classical genomic mechanisms involving estrogen response elements on DNA and by downstream effects of rapid initiation of cellular signaling cascades (Fortress & Frick, 2014; Galea et al.,  2017; Gustafsson,  2016). Estrogens likely influence DNA histone wrapping through acetylation of the H3 domain, dependent on estrogenic activation of the ERK pathway. However, estradiol also reduces expression of histone deacetylase 2 (HDAC2) and HDAC3 in the dorsal hippocampus, thus reducing the impairing effects of HDAC2 and HDAC3 on hippocampaldependent memory (Fortress & Frick, 2014). Less is known about specific estrogenic actions influencing DNA methylation; however, methylation does play a role in estrogen-mediated object memory consolidation, as DNA methyltransferase (DNMT) inhibitors block the enhancing effects of estradiol on object recognition. Estradiol seems to specifically increase de novo methylation of DNA by increasing expression of DNA methyltransferases DNMT3a and DNMT3b, suggesting a possible mechanism for estrogenic enhancement of long-term memory consolidation (Fortress & Frick, 2014).

In several studies, the ERβ agonist DPN enhanced memory when given pre- and postacquisition in mice and rats, and estradiol was effective at enhancing object recognition in ERα (but not in ERβ) knockout mice (Luine & Frankfurt,  2012; Tuscher et al.,  2015). However, the results are mixed and some studies find no enhancing effect of DPN on object recognition when administered systemically or intrahippocampally (Pereira, Bastos, de Souza, Ribeiro, & Pereira, 2014; Phan et al., 2011,  2015). Although the gene knockout study suggests that ERα may not be necessary for object recognition, the agonist PPT enhanced performance in rats and mice when administered systemically or in the dorsal hippocampus (Ervin et al., 2013; Luine & Frankfurt,  2012; Phan et al., 2011, 2015; Tuscher et al., 2015; but see Jacome et al.,  2010). In addition, studies investigating the role of GPER1 have found that it too rapidly enhanced object recognition in mice when given systemically or intrahippocampally preacquisition or immediately postacquisition (Gabor et al., 2015; Kim et al.,  2016; Lymer et al., 2017). In conclusion, there is no one estrogen receptor that is solely responsible for the estrogenic effects on object recognition, and it is possible that the three main receptors for estrogens act synergistically to enhance object memory.

Estrogenic Action in the Hippocampus for Object Recognition Memory

Progesterone and Object Recognition

Despite the well-known role of the hippocampus in spatial memory, this brain region is also a site of estrogenic improvements of object recognition memory. Preacquisition infusion of 17β-estradiol into the hippocampus improved object recognition memory within 40 minutes of treatment, as did postacquisition treatment, suggesting enhancement of memory consolidation and likely also acquisition or learning mechanisms (Phan et al., 2015; Tuscher et al., 2015). Learning enhancements with preacquisition treatment with estradiol seem to occur concurrently with estrogen-induced rapid changes in dendritic spine density, specifically an increase in immature “silent spines” with low α-amino-3-hydroxy-5-methyl-4isoxazolepropionic acid (AMPA) receptor sensitivity (Luine & Frankfurt, 2012; Phan et al., 2015).

Roles of Estrogen Receptors in Object Recognition

Much evidence on specific estrogen receptors points to a role for ERβ in mediating the memoryenhancing effects of estradiol on object recognition.

Progesterone’s effects on object recognition have been far less studied, but a few investigations suggest it also plays a role. When administered postacquisition, progesterone or combined estradiol and progesterone treatment improved object recognition memory in ovariectomized female mice and rats. Similar to effects of estrogenic manipulations, these treatments were only effective if given immediately postacquisition (Tuscher et al., 2015). Although estrogens administered to aged female animals do not always enhance object recognition, acute and chronic progesterone treatments improved performance in middle-aged and aged mice (Tuscher et al., 2015). It is unclear whether these memory enhancements are due to progesterone’s action at progesterone receptors or to its metabolites such as androgens, estrogens, and/or allopregnanolone. The effects of acute postacquisition treatments suggest that, like estrogens, progesterone acts through rapid cell signaling mechanisms to enhance memory. While the mechanisms are currently unclear, one study has shown that when progesterone was infused into the dorsal hippocampus of mice, levels of phosphorylated ERK increased Ervin and Choleris

75

five minutes after treatment, and blockade of the ERK and mTOR pathways prevented progesterone’s enhancing effects on object recognition memory (Orr, Rubin, Fan, Kent, & Frick, 2012). Thus, progesterone, like estradiol, seems to rapidly activate cell signaling cascades that support object memory consolidation (Tuscher et al., 2015).

Conclusions

Object memory, like spatial memory, is affected by the hormonal status of the animal. In general, gonadectomy impairs object memory, and hormone replacement with either estrogens or progesterone tends to improve or restore it, though the effects of estrogens have received more research attention. Estrogens seem to influence object memory through a variety of mechanisms including intracellular signaling cascades (e.g., PI3K, mTOR, and PKA), by interacting with ionotropic and metabotropic glutamate receptors, through changes in dendritic spine plasticity, and through changes to the epigenome, primarily through histone acetylation, though there is evidence for estrogens’ effects on DNA methyltransferases (Fernandez et al., 2008; Fortress & Frick, 2014; Lewis et al.,  2008; Luine,  2014; Luine & Frankfurt, 2012; Phan et al., 2015). There is no one ER that is clearly responsible for estrogens’ effects on object memory, as ERβ, ERα, and GPER1 all seem to play a role, and may act synergistically (Gabor et al., 2015; Jacome et al., 2010; Kim et al., 2016; Luine & Frankfurt, 2012; Lymer et al., 2017; Pereira et al., 2014; Phan et al., 2011; 2015; Tuscher et  al., 2015). Many of the effects of estrogens on object memory have been observed in the hippocampus, though it is unclear if estrogens in the hippocampus are required for object memory encoding itself or for encoding other relevant contextual information. Object memory has adaptive relevance for a few different reasons. It can be involved in processes such as recognizing predators or food sources, risk appraisal, and/or associating objects in the environment with contingent fearful or rewarding stimuli and situations (Dere, Kart-Teke, Huston, & De Souza Silva, 2006; Morris, 2001). Object memory in more general terms can be classified as a type of declarative memory. If object memory is generalizable to declarative memory, it has clear adaptive advantages for identifying relevant stimuli in the environment or even for remembering specific events (Dere et al., 2006; Morris, 2001). Enhancement of object memory by estrogens and other neurosteroids may be indicative of their general role in memory enhancement 76

or of a specific role for encoding information about stimuli in the environment.

Sex Hormones and Memories Within the Social Context

Social memory seems to involve neural pathways that can both overlap and be distinct from those underlying memory for objects or emotional stimuli (e.g., fear). Social memory involves social recognition, or recognizing others within a social group based on either the unique identity of specific individuals or features shared by a class of individuals (e.g., kinship, position in a dominance hierarchy, familiarity) (Choleris, Clipperton-Allen, Phan, & Kavaliers, 2009; Ervin, Lymer, et al., 2015). In most tests of social recognition, the animal is tested for familiarity recognition using a similar test to object recognition. An animal is presented with, and habituates to, one or more conspecific stimulus animals (Figure 5.3A). At test, one stimulus animal is swapped for an unfamiliar animal. As in tests of object recognition, if the animal recognizes the animal it encountered during habituation sessions, it will preferentially investigate the novel conspecific (Ervin, Lymer, et al.,  2015; Ervin et al., 2013; Figure 5.3B). Social learning is a distinct process from social recognition, in which an animal acquires new information about its environment from another individual, rather than learning about that individual. Humans and other animals use social learning for many types of adaptively relevant information, such as how and/or where to find food, how to avoid predators, and with whom to mate (Ervin, Lymer, et al.,  2015). The neurobiology of social learning has been most thoroughly studied in birdsong and rodents. In passerine bird species, males must learn species-specific songs by listening to an adult tutor (Bolhuis & Moorman,  2015). In rodents, social learning is often investigated using the social transmission of food preferences phenomenon, in which an “observer” animal given a choice of novel flavored foods preferentially eats a food it previously smelled on the breath of another “demonstrator” animal (Ervin, Lymer, et al., 2015; Figure 5.4). In humans, women tend to perform better when evaluating emotional faces, and seem to attend more to social stimuli than men (reviewed in Little, 2013; Sherwin, 2012). Physiological measurements support these findings: when presented with social stimuli, women show higher activity (measured with electroencephalography or functional magnetic resonance imaging) in cortical areas of the brain associated with facial recognition (e.g., fusiform face area)

Involvement of the Sex Hormones in Learning and Memory

(A)

(B)

Endogenous Hormones in Rodent Social Recognition and Social Learning

In tests of social recognition, female mice in proestrus during acquisition had enhanced memory for a familiar conspecific. Similarly, memory for a socially acquired food preference was facilitated among proestrus or postpartum mice (compared to diestrus mice) that interacted with a “demonstrator” conspecific that was previously fed a novel food. Postpartum rats and pregnant gerbils also showed an enhanced memory for a socially acquired food preference. Together, these studies suggest a prominent role for sex hormones in the acquisition of and/or memory for social information (reviewed in Choleris et al., 2009; Ervin, Lymer, et al., 2015).

Estrogens and Social Recognition

Figure 5.3  Memory for familiar conspecifics. Social recognition is commonly tested using a familiarity recognition test similar to that of object recognition tasks. The animal is first presented with one or more conspecifics (A). At test, one conspecific is swapped for a novel conspecific (B). If the animal recognizes the conspecific it encountered previously in habituation sessions (e.g., the gray mouse in the diagram), it will preferentially investigate the novel conspecific (e.g., the black mouse).

and with mirror neuron systems (in which brain areas associated with a movement are activated while watching another person perform the movement; Luine, 2014; Proverbio, 2017). While these findings suggest neurobiological underpinnings of a sex difference in social information processing, it is difficult in humans to disentangle biological from cultural influences on the ability to attend to and use social information by men and women. Girls may show an early preference for social interactions and emotional stimuli, and then develop an enhanced competency at interpreting these stimuli by virtue of increased exposure and “practice.” Furthermore, little is known about how changes in hormone levels over the menstrual cycle and/or with use of hormonal contraceptives affect performance on these tasks (Proverbio,  2017). Animal studies may therefore provide useful insight into possible mechanisms of social cognition that could generalize to aspects of human social behavior.

As with other tests of recognition memory, estrogens administered exogenously to ovariectomized animals tend to enhance memory for a familiar conspecific. These effects occurred when estradiol benzoate or estradiol dipropionate was administered days before acquisition and when 17β-estradiol was administered within 40 minutes of acquisition and testing, suggesting that estrogens can enhance social recognition through both rapid and long-term mechanisms (Ervin, Lymer, et al., 2015). The ERα seems to play a major role in the longterm and rapid estrogenic modulation of social recognition, unlike other recognition tasks in which ERβ may play a more significant role, at least in studies that do not focus on rapid effects (e.g., Ervin, Lymer, et al., 2015). ERα gene knockout abolished social recognition abilities in female mice and reduced memory retention for a familiar conspecific in male mice. Correspondingly, the ERα agonist PPT improved social recognition when administered 48 hours before acquisition in both ovariectomized and gonadally intact mice. Treatment with the ERβ agonist DPN 48 hours prior to acquisition also enhanced social recognition in female mice. However, social recognition was impaired, but not blocked, in female ERβKO mice. These findings suggest that ERα is required for social recognition in female mice, whereas ERβ helps modulate but is not necessary for social recognition (reviewed in Ervin, Lymer, et al., 2015; Gabor, Phan, Clipperton-Allen, Kavaliers, & Choleris, 2012). Investigation of the rapid effects of estrogens through specific ERs confirms a prominent role for ERα in the control of social recognition. Acute, systemic treatment with the agonist PPT, but not the ERβ agonist DPN, improved social recognition in Ervin and Choleris

77

(A)

(B)

(C)

Figure 5.4  Memory for socially acquired information. Social learning in rodents can be assessed using the social transmission of food preferences. First, a “demonstrator” animal eats a novel-flavored food (A). In a subsequent social interaction, an “observer” animal, naïve to that flavor, can smell the food the demonstrator ate on its breath (B). When the observer is later tested for a food preference and offered two or more novel-flavored foods, it prefers to eat the food it smelled on the demonstrator’s breath (C). Social learning is observed when the observer eats more of the demonstrated food than the other novel food(s). Social learning is impaired when the observer prefers neither food and eats roughly equal quantities of each.

ovariectomized mice when it was administered just 15 minutes before acquisition and mice were tested within 40 minutes of treatment. This time scale rules out most classical genomic effects and provides evidence for rapid estrogenic enhancements of social recognition specifically through ERα (Phan, Lancaster, Armstrong, MacLusky, & Choleris, 2011). There has been less research on the involvement of the GPER, but one study showed a rapid facilitation of social recognition when administered within 40 minutes of testing (Gabor et al.,  2015). Whether the ERα and GPER1 act independently or synergistically to improve social recognition is yet to be determined, and further research is also required to clarify the role of the different ERs in male mice.

Brain Regions Involved in Estrogenic Facilitation of Social Recognition

Both the hippocampus and medial amygdala have been implicated in the estrogenic control of social recognition. Infusion of 17β-estradiol, the ERα agonist PPT, or the GPER1 agonist G1 into the dorsal hippocampus of ovariectomized female mice enhanced social recognition within 40 minutes, suggesting that estrogens in the hippocampus act through ERα to rapidly facilitate social recognition 78

(Ervin, Lymer, et al.,  2015; Lymer et al.,  2017; Phan et al.,  2015). However, these experiments involved testing social recognition in the home cage. When tested in a Y-maze designed to limit spatial cues, 17β-estradiol administered systemically still enhanced social recognition, but not when infused into the dorsal hippocampus. Estrogens in the hippocampus may therefore facilitate social recognition indirectly by encoding spatial cues associated with the stimulus mice. The fact that ­systemic estradiol still enhanced social recognition in the Y-maze suggests that other brain regions are involved in the rapid effects of estrogens on social recognition (Ervin et al., 2013). The medial amygdala seems to play a major role in the rapid and long-term effects of estrogens on social recognition. ERα gene silencing with short hairpin RNA in the medial amygdala prevented enhancement of social recognition with estradiol and progesterone treatment in ovariectomized rats (Spiteri et al.,  2010). Furthermore, infusions of 17β-estradiol or the GPER1 agonist G1 directly into the medial amygdala improved social recognition in ovariectomized mice within the rapid 40-minute timeframe. In contrast to findings with systemic and intrahippocampal investigations, the ERβ agonist

Involvement of the Sex Hormones in Learning and Memory

DPN also rapidly enhanced social recognition when infused into the medial amygdala, whereas a much higher dose of the ERα agonist was required (Lymer et al., 2018). Estrogens likely influence social recognition both through genomic action in the medial amygdala through the control of oxytocin synthesis and receptor expression (Choleris et al., 2003), and through rapid, cell signaling mechanisms (Ervin, Lymer, et al., 2015). Overall, both the medial amygdala and the hippocampus seem to be involved in estrogenic effects on social recognition, with the hippocampus playing a lesser role, possibly by associating the conspecific with environmental cues when they are available.

Progesterone and Social Recognition

Little research has investigated the influence of progesterone in social recognition. Overall, it seems that progesterone may impair social recognition in male rats via long-term mechanisms, whereas it could facilitate social recognition in female animals. As previously noted, female mice in proestrus perform better than mice in diestrus on social recognition, a stage during which both estradiol and progesterone levels are elevated (Choleris et al.,  2009; Ervin, Lymer, et al., 2015). In one study, ovariectomized rats that received a combined treatment of estradiol and progesterone also had enhanced social recognition compared to vehicle-treated control rats (Spiteri & Ågmo, 2009). Conversely, male rats that received three progesterone treatments over three days prior to social recognition acquisition and testing showed impaired recognition (Bychowski & Augur, 2012). This impairment was reversed by postacquisition infusion of vasopressin into the lateral septum, suggesting that progesterone likely acts through longterm genomic mechanisms to disrupt vasopressin signaling required for social recognition in the male rats (Bychowski, Mena, & Augur, 2013). In agreement with these findings, male rats that received prenatal exposure to alcohol had impaired social recognition in adulthood, as well as elevated circulating progesterone, and higher levels of its precursor pregnenolone and the metabolite allopregnanolone in the cortex and hippocampus (Barbaccia et al., 2007). These findings are in contrast with the improving effects found in female mice and rats when given postacquisition, possibly due to rapid mechanisms of action (reviewed in Tuscher et al., 2015). These differences could indicate a sex difference in the role of progesterone in social recognition. It could also be that chronic elevation or repeated treatments of progesterone activated long-term mechanisms that

impaired social recognition, and it is yet unknown how rapid mechanisms of progesterone action or how postacquisition treatment affect social recognition in either males or females.

Androgens and Social Recognition in Male Rodents

Compared with tests of spatial and object memory, social recognition in male mice and rats seems to depend more heavily on testosterone conversion to estrogens via aromatase, and less on androgen receptor–mediated mechanisms. The effect of gonadectomy on social recognition in males is less clear than in females. Some studies show that castrated male rats show prolonged habituation to a previously encountered individual compared with gonadally intact males, suggesting prolonged social recognition memory (Bluthé, Gheusi, & Dantzer,  1993; Thor, 1980). In contrast, another study found that castrated male rats habituated only to repeated exposures to a live conspecific, whereas gonadally intact males habituated to exposures to a conspecific or exposures to odor cues from the conspecific animal, suggesting an impairment of social recognition (Sawyer, Hengehold, & Perez, 1984). Other studies found similarly mixed effects in which the social recognition abilities of castrated male rats seemed to depend on the timing of exposures to the social stimuli (reviewed in Choleris et al., 2009). Studies using prenatal manipulations of androgen receptors or gene knockout models support a primary role for estrogens in social recognition. Prenatal treatment with an androgen receptor antagonist did not affect social recognition in male rats (Axelson, Smith, & Duarte, 1999). Male aromatase knockout (ArKO) mice are impaired in social recognition. Because aromatase is necessary to convert testosterone into estrogens, ArKO mice have negligible levels of estrogens and higher levels of testosterone. Thus, disruptions in social recognition in ArKO mice could be due to lack of estrogens and/or impairing effects of testosterone (Fisher, Graves, Parlow, & Simpson,  1998; Pierman et al.,  2008). Treatment with the testosterone metabolites estradiol and dihydrotestosterone restored social recognition in ArKO mice (Pierman et al., 2008). Together, these studies suggest that androgens do not have major developmental effects on mechanisms involved in social recognition, but rather that estrogens are critical (Choleris et al., 2009). The effects of castration and ArKO in male rats and mice on social recognition are likely mediated by estrogenic mechanisms and downstream regulation of arginine v­asopressin. Ervin and Choleris

79

Castrated male rats and male ArKO rats have reduced vasopressin levels in limbic structures associated with social recognition, including the medial amygdala. In castrated rats, these effects were reversed with testosterone or estradiol treatment, but not dihydrotestosterone, a nonaromatizable androgen (Brot, De Vries, & Dorsa, 1993; De Vries, Wang, Bullock, & Numan,  1994). Although arginine vasopressin synthesis requires both androgenic and estrogenic mechanisms, it seems that estrogenic mechanisms play a predominant role in social recognition, even in male animals (Choleris et al., 2009).

Neurohormonal Control of Social Learning

Compared with social recognition, little is known about the influence of gonadal hormones on social learning. However, studies with ovariectomized mice suggest a prominent role for estrogens in the modulation of social learning in rodents. Similarly, a broader range of research in birdsong learning also supports a role for estrogens and androgens in social learning. Social learning in rodents is commonly tested using the social transmission of food preference. This test, like spontaneous recognition tasks, takes advantage of a natural behavior in rodents. In brief, an “observer” animal investigates the nose and mouth of a “demonstrator” animal that previously ate a novel food (Figure  5.4A, B). When the observer subsequently encounters this food, smelled on the breath of the demonstrator, and another novel food, it will typically prefer, or  consume more of, the demonstrated food (Figure 5.4C). The social transmission of food preference is thought to be a strategy by which rodents can overcome neophobia and expand their diet to novel foods that are likely to be safe (Ervin, Lymer, et al., 2015; Ervin et al., 2013). As mentioned previously, mice in the proestrus phase tend to show an enhanced preference for the demonstrated food (Choleris, Clipperton-Allen, Gray, Diaz-Gonzalez, & Welsman, 2011). Ovariectomized female mice administered estrogens exogenously days before or within 45 minutes of testing similarly show enhanced social learning, suggesting that estrogens may facilitate this learning strategy through both long-term and rapid mechanisms (Ervin, Lymer, et al., 2015). Unlike social recognition, which seems to strongly involve the ERα, social learning seems to be more influenced by estrogenic action through ERβ and the GPER1. A study of the longterm effects of estrogens on social learning, in which the ERα agonist PPT and the ERβ agonist WAY-200070 were administered 48 hours and 80

72 hours, respectively, prior to testing, found that activation of ERβ prolonged the duration for which observers preferred the demonstrated food, whereas ERα activation blocked the socially acquired food preference (Clipperton, Spinato, Chernets, Pfaff, & Choleris, 2008). On a shorter time scale, in which mice were treated with ER agonists and tested for a food preference within 45 minutes, only the GPER1 agonist G1 facilitated social learning in a way similar to 17β-estradiol. ERα and ERβ agonists instead shortened the duration of the socially acquired food preference, suggesting that activation of these ERs may inhibit social learning (Ervin, Lymer, et al., 2015; Ervin, Mulvale, Gallagher, Roussel, & Choleris, 2015). In conclusion, it seems that ERβ may play a role in enhancing social learning, but only through long-term mechanisms, having no effect or perhaps impairing social learning through rapid mechanisms. Similarly, ERα has no effect, or a slight impairing effect, on social learning through rapid mechanisms and impairs it through long-term mechanisms. These findings are in contrast with the roles of these ERs  on social recognition, and other recognition learning tasks, and perhaps set social learning apart as a ­distinct learning system. The GPER1, however, facilitated social learning, consistent with its rapid facilitatory effects on social recognition and other types of memory. The long-term enhancing effect of the ERβ agonist on social learning co-occurred with an increase in submissive behaviors displayed by the observer mouse to the demonstrator, which may have facilitated acquisition of the socially acquired food preference. This would be in agreement with other studies involving socially learned avoidance of biting flies in which observers learned better from a more dominant than a subordinate demonstrator (Kavaliers, Colwell, & Choleris, 2005). Thus, ERβ activation may indirectly facilitate social learning by changing the nature of the social interaction (i.e., when observers acquire information), rather than acting directly on social learning mechanisms (Choleris et  al., 2009; Ervin, Lymer, et al.,  2015; Ervin et  al., 2013). However, further research is required to determine the specific mechanisms by which the different ERs influence social learning. Social learning in the social transmission of food preference is well known to involve acetylcholine and dopamine, and a recent study showed a sex difference in manipulations of the hippocampal dopaminergic system, in which female mice were less sensitive to a dopamine D1-type receptor blockade in the hippocampus (Matta, Tiessen, & Choleris, 2017). Estrogens’ interactions

Involvement of the Sex Hormones in Learning and Memory

with dopamine and/or their aforementioned effects on cell signaling cascades and epigenetic mechanisms could potentially underlie the estrogenic effects on social learning. Research on hormonal modulation of birdsong learning can provide additional insight on the role of the sex hormones on social learning. There is a high density of androgen receptors throughout the neural pathways involved in song learning and production, and testosterone has been found to ­ drive development of a mature, crystallized song in young male songbirds (Bolhuis & Moorman, 2015; Brainard & Doupe,  2002). These effects seem to arise through downstream effects on the vasotocin system (evolutionarily homologous to vasopressin in mammals) and likely contribute to the control of a suite of t­erritorial behaviors (Goodson & Bass, 2001). In addition to testosterone action through androgen receptors, there is also evidence of its influence through estrogenic mechanisms after aromatization to estradiol. Song learning is disrupted by aromatase inhibition or blockade of ERs. Estradiol also modulates song preferences of female songbirds for male songs (Bailey & Saldanha, 2015). Whether the evidence from birdsong learning suggesting involvement of androgens and v­ asopressin translates to rodent social learning about food requires further investigation.

Conclusions

Estrogens play a stronger role in the modulation of social recognition learning than other types of learning in both male and female animals (Choleris et al., 2009; Ervin, Lymer, et al., 2015). ERα plays a prominent role in social recognition through both rapid and long-term mechanisms, while ERβ can support social recognition but is not necessary (Ervin, Lymer, et al., 2015). Progesterone also seems to facilitate social recognition in female animals but disrupts it in male rodents, likely through downstream inhibition of vasopressin signaling (Choleris et al., 2009; Bychowski & Augur, 2012; Bychowski et al., 2013; Tuscher et al., 2015). In contrast with the findings from social recognition, ERα impairs social learning as observed with the social transmission of food preferences. ERβ is involved to some extent through long-term mechanisms, either by direct enhancement of learning or indirectly by changing the dominance relationship between the observer and demonstrator (Clipperton et al., 2008; Ervin, Lymer, et al., 2015). Unlike the classical ERs, GPER1 has been found to enhance not only social recognition and social learning but also nonsocial

learning such as object recognition and object placement (Ervin, Lymer, et al., 2015; Gabor et al., 2015; Lymer et al., 2017). Thus, GPER1 may have a general facilitatory effect on learning. Conversely, ERα and ERβ may have more specific roles depending on the type of social or nonsocial information. ERα-mediated enhancements of social recognition may play a larger role in a suite of sexual and territorial behaviors. ERα also plays a prominent role in the estrogenic control of aggression and sexual behaviors; social recognition is crucial in both of these contexts for recognizing intruders versus familiar individuals, and for recognizing potential mates. Combined with the fact that ERα also mediates anorexic effects of estradiol, it could be that ERα-activated mechanisms prioritize reproductively motivated behaviors over foraging and/or behaviors related to social learning (reviewed in Ervin, Lymer, et al., 2015). While estrogens clearly play a role in social learning in rodents, the mechanisms and ERs involved are unclear. The involvement of estrogens and androgens and their downstream effect on ­vasopressin in another form of social learning, birdsong learning, may provide some clues about the neurohormonal underpinnings of social learning in rodents (Goodson & Bass,  2001; Bailey & Saldanha,  2015). However, one major difference between the role of sex hormones in birdsong learning and the social transmission of food preference is the probable motivational differences in these two types of behaviors. As testosterone promotes aggression and territorial behaviors and sexual behavior, it may be no surprise that it also supports song learning and production in male songbirds for whom songs serve to attract mates and/or establish territories. This is in contrast to the social transmission of food preference in rodents, in which aggression can impede acquisition of a socially learned food preference (Choleris et al., 1998; Ervin, Lymer, et al., 2015). Nonetheless, the convergence of the songbird and rodent literature supporting a role for estrogens provides strong evidence for their role in social learning and other social behaviors.

General Conclusion

We have summarized the literature on the results of laboratory studies examining the involvement of sex steroids in learning and memory. The results of these investigations paint a complex picture—the type of learning examined, hormone, receptor, brain region, timing of administration, and sex can all affect learning and memory. Unsurprisingly, then, the results of these investigations are mixed. Some hormones, Ervin and Choleris

81

estrogens in particular, seem to promote social and nonsocial cognition more than others. But even in this case, there exist sex differences in their effects and receptor-, mechanism- (long-term versus rapid), and learning type–specific results. This complex picture should not be surprising; after all, these hormones are highly conserved in animal evolution and play critical roles in multiple functions that may synergistically or competitively affect survival and reproduction. The complexity of these actions should not deter, but rather should promote, further research. The stakes are high. Millions of men and women develop diseases that are hormone dependent and may be treated with either hormones or hormone antagonists (e.g., Gonzalez et al., 2015; Schilder & Schagen, 2007). Understanding the consequences of these treatments on cognition and the complex facets of the biological underpinnings of these effects will be critical in informing medical practices and the expanding field of individualized or personalized medicine. It is likely—and desirable—that hormonetargeting treatments be specifically tailored for men and women, different age groups, and conditions in the future. This has the potential for enhanced effectiveness with reduced unwanted side effects (recently discussed in Choleris, Galea, Sohrabji, & Frick, 2018).

References

Atwi, S., McMahon, D., Scharfman, H., & MacLusky, N.  J. (2016). Androgen modulation of hippocampal structure and function. Neuroscientist, 22, 46–60. Axelson, J.  F., Smith, M., & Duarte, M. (1999). Prenatal flutamide treatment eliminates the adult male rat’s dependency upon vasopressin when forming social–olfactory memories. Hormones and Behavior, 36, 109–118. Bailey, C. H., Kandel, E. R., & Harris, K. M. (2015). Structural components of synaptic plasticity and memory consolidation. Cold Spring Harbor Perspectives in Biology, 7, a021758. Bailey, D. J., & Saldanha, C. J. (2015). The importance of neural aromatization in the acquisition, recall, and integration of song and spatial memories in passerines. Hormones and Behavior, 74, 116–124. Barbaccia, M. L., Scaccianoce, S., Del Bianco, P., Campolongo, P., Trezza, V., Tattoli, M., . . . Steardo, L. (2007). Cognitive impairment and increased brain neurosteroids in adult rats perinatally exposed to low millimolar blood alcohol concentrations. Psychoneuroendocrinology, 32, 931–942. Bitran, D., Hilvers, R.  J., & Kellogg, C.  K. (1991). Anxiolytic effects of 3 alpha-hydroxy-5alpha[beta]-pregnan-20-one: Endogenous metabolites of progesterone that are active at the GABAA receptor. Brain Research, 561, 157–161. Bluthé, R.  M., Gheusi, G., & Dantzer, R. (1993). Gonadalsteroids influence the involvement of arginine–vasopressin in social recognition in mice. Psychoneuroendocrinology, 18, 323–335. Bliss, T. V. P., & Lømo, T. (1973). Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized

82

rabbit following stimulation of the perforant path. Journal of Physiology, 232, 331–356. Bolhuis, J. J., & Moorman, S. (2015). Birdsong memory and the brain: In search of the template. Neuroscience and Biobehavioral Reviews, 50, 41–55. Bourtchuladze, R., Frenguelli, B., Blendy, J., Cioffi, D., Schutz, G., & Silva, A. J. (1994). Deficient long-term-memory in mice with a targeted mutation of the camp-responsive elementbinding protein. Cell, 79, 59–68. Brainard, M. S., & Doupe, A. J. (2002). What songbirds teach us about learning. Nature, 417, 351–358. Brot, M.  D., De Vries, G.  J., & Dorsa, D.  M. (1993). Local implants of testosterone metabolites regulate vasopressin mRNA in sexually dimorphic nuclei of the rat brain. Peptides, 14, 933–940. Bychowski, M. E., & Auger, C. J. (2012). Progesterone impairs social recognition in male rats. Hormones and Behavior, 61, 598–604. Bychowski, M.  E., Mena, J.  D., & Auger, C.  J. (2013). Vasopressin infusion into the lateral septum of adult male rats rescues progesterone-induced impairment in social recognition. Neuroscience, 246, 52–58. Carroll, J. S., Meyer, C. A., Song, J., Li, W., Geistlinger, T. R., Eeckhoute, J., Brodsky, A. S., . . . Brown, M. (2006). Genomewide analysis of estrogen receptor binding sites. Nature Genetics, 38, 1289–1297. Choleris, E., Clipperton-Allen, A.  E., Gray, D.  G., DiazGonzalez, S., & Welsman, R. G. (2011). Differential effects of dopamine receptor D1-type and D2-type antagonists and  phase of the estrous cycle on social learning of food preferences, feeding, and social interactions in mice. Neuropsychopharmacology, 36, 1689–1702. Choleris, E., Clipperton-Allen, A. E., Phan, A., & Kavaliers, M. (2009). Neuroendocrinology of social information processing in rats and mice. Frontiers in Neuroendocrinology, 30, 442–459. Choleris, E., Galea, L. A. M., Sohrabji, F., & Frick, K. M. (2018). Sex differences in the brain: Implications for behavioral and biomedical research. Neuroscience and Biobehavioral Reviews, 85, 126–145. Choleris, E., Gustafsson, J.-Å., Korach, K.  S., Muglia, L.  J., Pfaff, D. W., & Ogawa, S. (2003). An estrogen-dependent four-gene micronet regulating social recognition: A study with oxytocin and estrogen receptor-alpha and -beta knockout mice. Proceedings of the National Academy of Sciences of the United States of America, 100, 6192–6197. Choleris, E., Valsecchi, P., Wang, Y., Ferrari, P., Kavaliers, M., & Mainardi, M. (1998). Social learning of a food preference in male and female Mongolian gerbils is facilitated by the anxiolytic, chlordiazepoxide. Pharmacology Biochemistry and Behavior, 60, 575–584. Clark, R.  E., Broadbent, N.  J., Zola, S.  M., & Squire  L.  R. (2002). Anterograde amnesia and temporally graded retro­ grade amnesia for a nonspatial memory task after lesions of hippocampus and subiculum. Journal of Neuroscience, 22, 4663–4669. Clipperton, A. E., Spinato, J. M., Chernets, C., Pfaff, D. W., & Choleris, E. (2008). Differential effects of estrogen receptor alpha and beta specific agonists on social learning of food preferences in female mice. Neuropsychopharmacology, 33, 2362–2375. Compagnone, N.  A., & Mellon, S.  H. (2000). Neurosteroids: Biosynthesis and function of these novel neuromodulators. Front. Neuroendocrinology, 21, 1–56.

Involvement of the Sex Hormones in Learning and Memory

Da Silva, W. C., Bonini, J. S., Bevilaqua, L. R. M., Medina, J. H., Izquierdo, I., & Cammarota, M. (2008). Inhibition of mRNA synthesis in the hippocampus impairs consolidation and reconsolidation of spatial memory. Hippocampus, 18, 29–39. Dere, E., Kart-Teke, E., Huston, J. P., & De Souza Silva, M. A. (2006). The case for episodic memory in animals. Neuro­ science and Biobehavioral Reviews, 30, 1206–1224. De Vries, G.  J., Wang, Z.  X., Bullock, N.  A., & Numan, S. (1994). Sex-differences in the effects of testosterone and its metabolites on vasopressin messenger-RNA levels in the bed nucleus of the stria terminalis of rats. Journal of Neuroscience, 14, 1789–1794. Dudai, Y. (2004). The neurobiology of consolidations, or, how stable is the engram? Annual Review of Psychology, 55, 51–86. Ervin, K. S. J., Lymer, J. M., Matta, R., Clipperton-Allen, A. E., Kavaliers, M., & Choleris, E. (2015). Estrogen involvement in social behavior in rodents: Rapid and long-term actions. Hormones and Behavior, 74, 53–76. Ervin, K.  S.  J., Mulvale, E., Gallagher, N., Roussel, V., & Choleris, E. (2015). Activation of the G protein-coupled estrogen receptor, but not estrogen receptor α or β, rapidly enhances social learning. Psychoneuroendocrinology, 58, 51–66. Ervin, K. S. J., Phan, A., Gabor, C. S., & Choleris, E. (2013). Rapid oestrogenic regulation of social and nonsocial learning. Journal of Neuroendocrinology, 25, 1116–1132. Fagan, W.  F., Lewis, M.  A., Auger-Méthé, M., Avgar, T., Benhamou, S., Breed, G., . . . Mueller, T. (2013). Spatial memory and animal movement. Ecology Letters, 16, 1316–1329. Farr, S. A., Flood, J. F., Scherrer, J. F., Kaiser, F. E., Taylor, G. T., & Morley, J. E. (1995). Effect of ovarian steroids on footshock avoidance learning and retention in female mice. Physiology and Behavior, 58, 715–723. Fernandez, S.  M., Lewis, M.  C., Pechenino, A.  S., Harburger, L.  L., Orr, P.  T., Gresack, J.  E., . . . Frick, K.  M. (2008). Estradiol-induced enhancement of object memory conso­ lida­ tion involves hippocampal ERK activation and membrane-bound estrogen receptors. Journal of Neuroscience, 28, 8660–8667. Fisher, C.  R., Graves, K.  H., Parlow, A.  F., & Simpson, E.  R. (1998). Characterization of mice deficient in aromatase (ArKO) because of targeted disruption of the cyp19 gene. Proceedings of the National Academy of Sciences of the United States of America, 95, 6865–6970. Fortress, A.  M., Fan, L., Orr, P.  T., Zhao, Z., & Frick, K.  M. (2013). Estradiol-induced object recognition memory consolidation is dependent on activation of mTOR signaling in dorsal hippocampus. Learning and Memory, 20, 147–155. Fortress, A. M., & Frick, K. M. (2014). Epigenetic regulation of estrogen-dependent memory. Frontiers in Neuroendocrinology, 35, 530–549. Frankfurt, M., & Luine, V. (2015). The evolving role of dendritic spines and memory: Interaction(s) with estradiol. Hormones and Behavior, 74, 28–36. Frankland, P. W., & Bontempi, B. (2005). The organization of recent and remote memories. Nature Reviews Neuroscience, 6, 119–130. Frick, K. M. (2009). Estrogens and age-related memory decline in rodents: What have we learned and where do we go from here? Hormones and Behavior, 55, 2–23. Frick, K.  M. (2015). Molecular mechanisms underlying the memory-enhancing effects of estradiol. Hormones and Behavior, 74, 4–18.

Frick, K. M., Kim, J., Tuscher, J. J., & Fortress, A. M. (2015). Sex steroid hormones matter for learning and memory: Estrogenic regulation of hippocampal function in male and female rodents. Learning and Memory, 22, 472–493. Frye, C., Duffy, C., & Walf, A. (2007). Estrogens and progestins enhance spatial learning of intact and ovariectomized rats in the object placement task. Neurobiology of Learning and Memory, 88, 208–216. Frye, C. A., & Duncan, J. E. (1994). Progesterone metabolites, effective at the GABAA receptor complex, attenuate pain sensitivity in rats. Brain Research, 643, 194–203. Frye, C. A., Koonce, C. J., & Walf, A. A. (2013). Progesterone, compared to medroxyprogesterone acetate, to C57BL/6, but  not 5alpha-reductase mutant, mice enhances object recognition and placement memory and is associated with higher BDNF levels in the hippocampus and cortex. Neuroscience Letters, 551, 53–57. Frye, C.  A., & Lacey, E.  H. (2000). Progestins influence performance on cognitive tasks independent of changes in affective behavior. Psychobiology, 28, 550–563. Frye, C. A., & Walf, A. A. (2008). Progesterone to ovariectomized mice enhances cognitive performance in the spontaneous alternation, object recognition, but not placement, water maze, and contextual and cued conditioned fear tasks. Neurobiology of Learning and Memory, 90, 171–177. Frye, C. A., & Walf, A. A. (2010). Progesterone enhances learning and memory of aged wild-type and progestin receptor knockout mice. Neuroscience Letters, 472, 38–42. Gabor, C. S., Lymer, J., Phan, A., & Choleris, E. (2015). Rapid effects of the G-protein coupled oestrogen receptor (GPER) on learning and dorsal hippocampus dendritic spines in female mice. Physiology and Behavior, 149, 53–60. Gabor, C. S., Phan, A., Clipperton-Allen, A. E., Kavaliers, M., & Choleris, E. (2012). Interplay of oxytocin, vasopressin, and sex hormones in the regulation of social recognition. Behavioral Neuroscience, 126, 97–109. Galea, L.  A.  M., Frick, K.  M., Hampson, E., Sohrabji, F., & Choleris, E. (2017). Why estrogens matter for behavior and brain health. Neuroscience and Biobehavioral Reviews, 76, 363–379. Gibbs, R. B. (2005). Testosterone and estradiol produce different effects on cognitive performance in male rats. Hormones and Behavior, 48, 268–277. Gobinath, A., Choleris, E., & Galea, L.  A.  M. (2017). Sex, hormones, and genotype interact to influence psychiatric disease, treatment and behavioral research. Journal of Neuroscience Research, 95, 50–64. Gonçalves, J. T., Schafer, S. T., & Gage, F. H. (2016). Adult neurogenesis in the hippocampus: From stem cells to behavior. Cell, 167, 897–914. Gonzalez, B.  D., Jim, H.  S., Booth-Jones, M., Small, B.  J., Sutton, S.  K., Lin, H.  Y.,  .  .  .  Jacobsen, P.  B. (2015). Course and predictors of cognitive function in patients with prostate cancer receiving androgen-deprivation therapy: A controlled comparison. Journal of Clinical Oncology, 33, 2021–2027. Goodson, J. L., & Bass, A. H. (2001). Social behavior functions and related anatomical characteristics of vasotocin/ vasopressin systems in vertebrates. Brain Research Reviews, 35, 246–265. Gustafsson, J. A. (2016). Historical overview of nuclear receptors. Journal of Steroid Biochemistry and Molecular Biology, 157, 3–6.

Ervin and Choleris

83

Hammond, R., Mauk, R., Ninaci, D., Nelson, D., & Gibbs, R. (2009). Chronic treatment with estrogen receptor agonists restores acquisition of a spatial learning task in young ovariectomized rats. Hormones and Behavior, 56, 309–314. Hammond, R., Nelson, D., Kline, E., & Gibbs, R. B. (2012). Chronic treatment with a GPR30 antagonist impairs acquisition of a spatial learning task in young female rats. Hormones and Behavior, 62, 367–374. Hansberg-Pastor, V., González-Arenas, A., Piña-Medina, A. G., & Camacho-Arroyo, I. (2015). Sex hormones regulate cyto­ skeletal proteins involved in brain plasticity. Frontiers in Psychiatry, 6, 165. Hawley, W. R., Grissom, E. M., Martin, R. C., Halmos, M. B., Bart, C. L., & Dohanich, G. P. (2013). Testosterone modulates spatial recognition memory in male rats. Hormones and Behavior, 63, 559–565. Hawley, W. R., Grissom, E. M., Moody, N. M., Dohanich, G. P., & Vasudevan, N. (2014). Activation of G-protein-coupled receptor 30 is sufficient to enhance spatial recognition memory in ovariectomized rats. Behavioral Brain Research, 262, 68–73. Heldring, N., Pike, A., Andersson, S., Cheng, G., Matthews, J., Hartman, J., . . . Gustafson, J. A. (2007). Estrogen receptors: How do they signal and what are their targets. Physiological Reviews, 87, 905–931. Hyun, J. S., & Luck, S. J. (2007). Visual working memory as the substrate for mental rotation. Psychonomic Bulletin and Review, 14, 154–158. Ihunwo, A.  O., Tembo, L.  H., & Dzamalala, C. (2016). The dynamics of adult neurogenesis in human hippocampus. Neural Regeneration Research, 11, 1869–1883. Jacome, L. F., Gautreaux, C., Inagaki, T., Mohan, G., Alves, S., Lubbers, L.  S., & Luine, V. (2010). Estradiol and ERbeta agonists enhance recognition memory, and DPN, and ERbeta agonist, alters brain monoamines. Neurobiology Learning and Memory, 94, 488–498. Kandel, E.  R., Dudai, Y., & Mayford, M.  R. (2014). The molecular and systems biology of memory. Cell, 157, 163–186. Kaufman, S. B. (2007). Sex differences in mental rotation and spatial visualization ability: Can they be accounted for by differences in working memory capacity? Intelligence, 35, 211–223. Kavaliers, M., Colwell, D. D., & Choleris, E. (2005). Kinship, familiarity and social status modulate social learning about “micropredators” (biting flies) in deer mice. Behavioral Ecology and Sociobiology, 58, 60–71. Kim, J., Szinte, J. S., Boulware, M. I., & Frick, K. M. (2016). 17βEstradiol and agonism of G-protein-coupled estrogen receptor enhance hippocampal memory via different cell-signaling mechanisms. Journal of Neuroscience, 36, 3309–3321. Koss, W. A., & Frick, K. M. (2017). Sex differences in hippocampal function. Journal of Neuroscience Research, 95, 539–562. Lewis, M. C., Kerr, K. M., Orr, P. T., & Frick, K. M. (2008). Estradiol-induced enhancement of object memory consoli­ dation involves NMDA receptors and protein kinase A in the dorsal hippocampus of female C57BL/6 mice. Behavioral Neuroscience, 122, 716–721. Little, A. C. (2013). The influence of steroid sex hormones on the cognitive and emotional processing of visual stimuli in humans. Frontiers in Neuroendocrinology, 34, 315–328. Luine, V. (2014). Estradiol and cognitive function: Past, present and future. Hormones and Behavior, 66, 602–618. Luine, V. (2015). Recognition memory tasks in neuroendocrine research. Behavioral Brain Research, 285, 158–164.

84

Luine, V., & Frankfurt, M. (2012). Estrogens facilitate memory processing through membrane mediated mechanisms and alterations in spine density. Frontiers in Neuroendocrinology, 33, 388–402. Lymer, J.  M., Robinson, A., Winters, B.  D., & Choleris, E. (2017). The rapid effects of the G-protein coupled estrogen receptor in the dorsal hippocampus on learning in female mice. Psychoneuroendocrinology, 77, 131–140. Lymer, J. M., Sheppard, P. A. S., Kuun, T., Blackman, A., Jani, N., Mahbub, S., & Choleris, E. (2018). Estrogens and their receptors in the medial amygdala rapidly facilitate social recognition in female mice. Psychoneuroendocrinology, 89, 30–38. Mahmoud, R., Wainwright, S.  R., & Galea, L.  A. (2016). Sex hormones and adult hippocampal neurogenesis: Regulation, implications, and potential mechanisms. Frontiers in Neuroendocrinology, 41, 129–152. Matta, R., Tiessen, A. N., & Choleris, E. (2017). The role of dorsal hippocampal dopamine D1-type receptors in social learning, social interactions, and food intake in male and female mice. Neuropsychopharmacology 42, 2344–2353. Mendell, A.  L., Atwi, S., Bailey, C.  D.  C., McCloskey, D., Scharfman, H. E., & MacLusky, N. J. (2017). Expansion of mossy fibers and CA3 apical dendritic length accompanies the fall in dendritic spine density after gonadectomy in male, but not female, rats. Brain Structure and Function, 222, 587–601. Morris, R.  G.  M. (2001). Episodic-like memory in animals: Psychological criteria, neural mechanisms and the value of episodic-like tasks to investigate animal models of neuro­ degenerative disease. Philosophical Transactions of the Royal Society B, 356, 1453–1465. Nader, K., & Hardt, O. (2009). A single standard for memory: The case for reconsolidation. Nature Reviews Neuroscience, 10, 224–234. Nelson, R. J. (2017). An introduction to behavioral endocrinology (5th ed.). Sunderland, MA: Sinauer Associates. Nguyen, P. V., Abel, T., & Kandel, E. R. (1994). Requirement of a critical period of transcription for induction of a late-phase of LTP. Science, 265, 1104–1107. Orr, P.  T., Rubin, A.  J., Fan, L., Kent, B.  A., & Frick, K.  M. (2012). The progesterone-induced enhancement of object recognition memory consolidation involves activation of the extracellular signal-related kinase (ERK) and mammalian target of rapamycin (mTOR) pathways in the dorsal hippocampus. Hormones and Behavior, 61, 487–495. Packard, M.  G. (1998). Post-training estrogen and memory modulation. Hormones and Behavior, 34, 126–139. Pereira, L. M., Bastos, C. P., de Souza, J. M., Ribeiro, F. M., & Pereira, G. S. (2014). Estradiol enhances object recognition memory in Swiss female mice by activating hippocampal estrogen receptor alpha. Neurobiology of Learning and Memory, 114, 1–9. Phan, A., Gabor, C. S., Favaro, K. J., Kaschack, S., Armstrong, J. N., MacLusky, N. J., & Choleris, E. (2012). Low doses of 17β-estradiol rapidly improve learning and increase hippocampal dendritic spines. Neuropsychopharmacology, 37, 2299–2309. Phan, A., Lancaster, K. E., Armstrong, J. N., MacLusky, N. J., & Choleris, E. (2011). Rapid effects of estrogen receptor alpha and beta selective agonists on learning and dendritic spines in female mice. Endocrinology, 152, 1492–1502. Phan, A., Suschkov, S., Molinaro, L., Reynolds, K., Lymer, J. M., Bailey, C.  D.  C., . . . Choleris, E. (2015). Rapid increases in

Involvement of the Sex Hormones in Learning and Memory

immature synapses parallel estrogen-induced hippocampal learning enhancements. Proceedings of the National Academy of Sciences of the United States of America, 112, 16018–16023. Pierman, S., Sica, M., Allieri, F., Viglietti-Panzica, C., Panzica, G.  C., & Bakker, J. (2008). Activational effects of estradiol and dihydrotestosterone on social recognition and the arginine–vasopressin immunoreactive system in male mice lacking a functional aromatase gene. Hormones and Behavior, 54, 98–106. Proverbio, A. M. (2017). Sex differences in social cognition: The case of face processing. Journal of Neuroscience Research, 95, 222–234. Rannevik, G., Carlström, K., Jeppsson, S., Bjerre, B., & Svanberg, L. (1986). A prospective long-term study in women from pre-menopause to post-menopause: Changing profiles of gonadotrophins, oestrogens and androgens. Maturitas, 8, 297–307. Rhodes, M., & Frye, C. (2006). ERβ-selective SERMs produce mnemonic-enhancing effects in the inhibitory avoidance and water maze tasks. Neurobiology of Learning and Memory, 85, 183–191. Sawyer, T.  F., Hengehold, A.  K., & Perez, W.  A. (1984). Chemosensory and hormonal mediation of social memory in male rats. Behavioral Neuroscience, 98, 908–913. Schilder, C. M., & Schagen, S. B. (2007). Effects of hormonal therapy on cognitive functioning in breast cancer patients: A review of the literature. Minerva Ginecologica, 59, 387–401. Sherry, D. (1998). The ecology and neurobiology of spatial memory. In R. Dukas (Ed.), Cognitive ecology: The evolutionary ecology of information processing and decision making (pp. 261–296). Chicago, IL: University of Chicago Press. Sherwin, B.  B. (2012). Estrogen and cognitive functioning in women: Lessons we have learned. Behavioral Neuroscience, 126, 123–127. Skucas, V. A., Duffy, A. M., Harte-Hargrove, L. C., MagagnaPoveda, A., Radman, T., & Chakraborty, G. (2013). Testosterone depletion in adult male rats increases mossy fiber transmission, LTP, and sprouting in area CA3 of hippocampus. Journal of Neuroscience, 33, 2338–2355.

Srivastava, D.  P., & Evans, P.  D. (2013). G-protein oestrogen receptor 1: Trials and tribulations of a membrane oestrogen receptor. Journal of Neuroendocrinology, 25, 1219–1230. Srivastava, D. P., & Penzes, P. (2011). Rapid estradiol modulation of neuronal connectivity and its implications for disease. Frontiers in Endocrinology, 2, 1–17. Srivastava, D. P., Woolfrey, K. M., & Penzes, P. (2013). Insights into rapid modulation of neuroplasticity by brain estrogens. Pharmacological Reviews, 65, 1318–1350. Spiteri, T., & Ågmo, A. (2009). Ovarian hormones modulate social recognition in female rats. Physiology and Behavior, 98, 247–250. Spiteri, T., Musatov, S., Ogawa, S., Ribeiro, A., Pfaff, D. W., & Ågmo, A. (2010). The role of the estrogen receptor alpha in the medial amygdala and ventromedial nucleus of the hypothalamus in social recognition, anxiety and aggression. Behavioral Brain Research, 210, 211–220. Spritzer, M.  D., Daviau, E.  D., Coneeny, M.  K., Engelman, S.  M., Prince, W.  T., & Rodriguez-Wisdom, K.  N. (2011). Effects of testosterone on spatial learning and memory in adult male rats. Hormones and Behavior, 59, 484–496. Squire, L. R. (1992). Memory and the hippocampus: A synthesis from findings with rats, monkeys, and humans. Psychological Review, 99, 195–231. Thor, D.  H. (1980). Testosterone and persistence of social investigation in laboratory rats. Journal of Comparative and Physiological Psychology, 94, 970–976. Tuscher, J. J., Fortress, A. M., Kim, J., & Frick, K. M. (2015). Regulation of object recognition and object placement by ovarian sex steroid hormones. Behavioral Brain Research, 285, 140–157. Winters, B. D., Saksida, L. M., & Bussey, T. J. (2008). Object recognition memory: Neurobiological mechanisms of encoding, consolidation and retrieval. Neuroscience and Biobehavioral Reviews, 32, 1055–1070. Winters, B. D., Tucci, M. C., & DaCosta-Furtado, M. (2009). Older and stronger object memories are selectively destabilized by reactivation in the presence of new information. Learning and Memory, 16, 545–553.

Ervin and Choleris

85

CH A PT E R

6

Endocrine Disruptors and Other Environmental Influences on Hormone Action

Laura N. Vandenberg

Abstract Endocrine-disrupting chemicals (EDCs) interfere with hormone action by altering hormone synthesis, secretion, transport in the blood, binding to receptors, metabolism, or excretion. This chapter reviews the history of EDCs and other environmental chemicals, methods used to identify EDCs, and common uses for these chemicals in consumer products. It also describes major principles of endocrinology and how these features influence the actions of EDCs. This chapter will also evaluate controversies in the study and regulation of EDCs, including the concept of “low dose effects,” the question of whether humans are exposed to EDCs at levels that can cause harm, and the determination of “safe” doses of exposure. Finally, this chapter reviews other environmental factors that can influence the health of laboratory animals and interfere with the study of EDCs. Keywords:  endocrine disruptor, adverse effect, potency, low-dose effect, threshold, nonmonotonic

The endocrine system plays an important role in mediating the interaction of individuals with their environments. Hormones are responsible for the maintenance of body temperature and body fat; they allow creatures to utilize sugars and store energy in the body; they regulate fertility and the response of mothers to their young; they mediate social and sexual behaviors; and they regulate water intake and blood pressure. Hormones also regulate development of many of the body’s organs and play an ­important role in cell differentiation and organogenesis. Moreover, hormones allow the organs, ­tissues, and cells of the body to respond to environmental changes. Only recently has it begun to be understood that some environmental influences can interfere with these responses. One aspect of the ­environment that has received significant attention over the last decade is the discovery and understanding of endocrine-disrupting chemicals (EDCs). These compounds affect not only the homeostatic

endocrine machinery that is required to maintain the health of the individual but also the developmental processes and epigenetic regulation of gene expression, inducing effects that can persist for ­generations. This chapter will discuss the effects of EDCs on the function of the endocrine system with a specific focus on hormone action. Knowledge from several scientific fields including environmental health, toxicology, endocrinology, and epidemiology will be combined to describe why EDCs have emerged as an important public health concern. Finally, this chapter will describe other environmental factors that, like EDCs, can interfere with normal biological processes. We will specifically focus on variables that are present in laboratory animal studies but are often ignored; these environmental factors could influence developmental and homeostatic processes, inducing variability in experiments and making it difficult to replicate prior research collected under different environmental conditions.

87

Environmental Chemicals: A Brief Historical Review

Chemicals have been used throughout known human history for the benefit of individuals or groups. Chemicals added to animal hides allowed for the preservation of these materials for use in clothing and other fabrics; chemicals added to food (e.g., salt, sugar) allowed these items to be preserved for later consumption. In the 18th and 19th centuries, chemical manufacturing reached a much larger scale, ­allowing chemicals to be produced by the ton to alter fabrics, foods, and other goods. By the early 20th century, a number of large chemical companies had formed, focused on the production of chemicals for use in detergents, plastics, inorganics, fertilizers, pharmaceuticals, and pesticides. Modern warfare was changed due to the availability of chemicals to protect soldiers from insects—which often were more than a nuisance due to their transmittal of ­infectious diseases—as well as chemicals to be used as weapons. Following World War II, the number of synthetic pesticides soared, allowing humans to “tame” environments that had previously been difficult to farm or inhabit due to insects. Yet, this same use of pesticides that improved the use of lands and the health of many populations at risk from infectious diseases caught the attention of environmental health scientists due to their u ­ nintended effects on nontarget species, including ­exposed humans. Perhaps the most famous example of concern over the widespread and indiscriminate use of pesticides arose from the 1962 publication of Silent Spring, an  environmental science book written by Rachel Carson. Carson explored the unintended consequences of exposures to pesticides, including dichlorodiphenyltrichloroethane (DDT), diel­ drin, endrin, and methoxychlor. With the publication of Silent Spring, experts from a growing number of disciplines, as well as the public at large, became aware that exposures to chemicals in the environment could affect the health of wild animals, with potentially devastating effects that were documented and widely acknowledged. Carson suggested that similar effects could happen to exposed humans, and she documented human exposures to insecticides like DDT, which was found in foods ranging from fruits to breads (Carson, 1987). Lay audiences that recall the work of Carson often point to her “discovery” that DDT affected the health of birds of prey, including American bald eagles. In fact, many researchers were working to understand the effects of DDT on nontarget species. In the 1950s, approximately 1 pound (0.45 kg) 88

was sprayed per acre across more than 1 million acres in Montana; this led to measurable levels of DDT in every fish that was examined (reviewed in Carson, 1987). Carson noted that the fish were not usually killed by DDT, but rather that they experienced higher mortality due to loss of other food sources. Birds of prey, eating these fish, had significantly higher levels of DDT in their bodies, illustrating the bioaccumulation of DDT in body tissues, as well as the biomagnification of this chemical in the food chain (see Figure 6.1). The adverse ­effects that were observed in birds of prey included egg shell thinning (which contributed to decreased ­fecundity, as eggs would often crack when parents sat on their nests to keep them warm), neuromuscular diseases, wasting, and death. Importantly, Carson documented that the effects of DDT and other pesticides were most profound on embryos and hatchlings rather than the exposed adult birds. Further, although observational studies of wild birds were a central focus of her work, Carson also reviewed dozens of studies detailing the effects of controlled experimental studies with a range of species. Although Carson was not alone in her research examining the effects of environmental agents on wildlife populations, Silent Spring is widely credited with raising awareness of multiple scientific issues that have continued to be examined in the field of environmental health. These include (1) acknowledgment that even low doses of chemicals can have profound effects on exposed creatures, many of which were not the intended target of the chemical; (A)

TIME (B)

FOOD CHAIN Figure 6.1  Bioaccumulation and biomagnification of environmental contaminants. (A) Bioaccumulation refers to the increased level of a compound that is stored in the body, typically in fat, after exposures. Bioaccumulation is typically attributed to the physio­chemical properties of a compound (e.g., features of a chemical that increase its lipophilic tendencies). (B) Biomagnification refers to the increased level of exposure observed at higher levels of the food chain.

Endocrine Disruptors and Other Influences on Hormone Action

(2) understanding that chemical mixtures can have compounded effects, inducing more extreme outcomes than are observed after exposures to single chemicals; and (3) the realization that the timing of chemical exposures is critical, indicating that some populations—often developing creatures—are more vulnerable to exposures than others. The chemical industry’s reaction to Carson’s book was extremely negative; although it might be understandable that companies have an interest in protecting their products, the personal vilification of Carson was unprecedented. In spite of this negative attention, Carson’s understanding that many chemicals used as pesticides could alter metabolism and function of the nervous systems of exposed animals continued to draw attention from scientists working in a range of fields. Since the 1960s, many dedicated conservationists and wildlife biologists have described the effects of environmental compounds on wildlife. In the 1980s and 1990s, one ­researcher, Theo Colborn, began to document the effects of chemicals on the health of wildlife in the Great Lakes area in the United States. Collecting papers published by her fellow scientists, Colborn found that adverse health outcomes were often ­reported in birds and fish, with effects seen in a wide range of species. Like Carson, Colborn observed that effects were often seen in animals at the top of the food chain. She also noted, similar to what Carson had documented in Silent Spring, that the worst effects of chemical exposures were often ­measured in the offspring of exposed animals, and not the exposed adults themselves (see Colborn, vom Saal, & Soto, 1993). To try to identify patterns in the data, Colborn created a spreadsheet to tally the effects observed across dozens of species (Krimsky, 2003). With this analysis, she noted that the most common effects were diminished reproduction, thyroid problems, altered behavior, and metabolic changes. For the first time, she noted that the underlying commonality in these effects is the central role of the endocrine system. These findings led Colborn to organize a meeting at the Wingspread Conference Center in Wisconsin. This meeting, held in 1991, brought ­together scientists from a wide range of backgrounds to describe the data each one had collected from their own studies of hormonally active pharmaceuticals and environmental chemicals. The group noted ­effects of these chemicals on cultured cells, laboratory animals, and human populations and p ­ atients. The Wingspread participants further discussed the effects of these chemicals on sexual ­differentiation,

reproductive function, neurobehavioral development, and autoimmune diseases. At the conclusion of the meeting, the scientists collectively wrote a consensus statement that summarized the state of the science on these environmental chemicals. They wrote, “We are certain of the following: A large number of man-made chemicals that have been released into the environment, as well as a few natural ones, have the potential to disrupt the endocrine system of ­animals, including humans” (Colborn et al., 1993 [emphasis added]). The Wingspread meeting marked the first time that the term endocrine disruptor was widely used. Although the meeting participants did not expressly define this term, it stuck; this term continues to be used today. Furthermore, the study of EDCs has moved from a fairly small field with only a few dozen laboratories committed to their study to a field that is acknowledged and addressed by scientists with a wide range of expertise. At the time of Wingspread, the contributions from experts in fields, including ecology, endocrinology, medicine, law, reproductive physiology, toxicology, wildlife management, and cancer biology, came together to advance the study of EDCs. Today, the dogma from each of these fields has significantly affected how public health agencies approach the study and regulation of these compounds.

Defining and Identifying Endocrine-Disrupting Chemicals

Following the Wingspread Conference and publications from its attendees (Colborn et  al.,  1993), regulatory agencies around the world began to pay much closer attention to the issue of EDCs. The first step taken by many of these agencies was to develop a  working definition that could be used for the study, identification, and future regulation of these c­ ompounds. The first agency to define an EDC was the U.S. Environmental Protection Agency (EPA), which offered this definition in 1996: “An exogenous agent that interferes with the production, ­release, transport, metabolism, binding, action, or elimination of natural hormones in the body ­responsible for the maintenance of homeostasis and the regulation of developmental processes” (Kavlock et al., 1996: 716). In 2012, scientists affiliated with the Endocrine Society, a 100-year-old international society composed of basic scientists and physicians dedicated to the study and treatment of conditions and diseases related to hormones, worked to simplify the EPA’s definition. They defined an EDC as “an exogenous chemical, or mixture of chemicals, Vandenberg

89

that interferes with any aspect of hormone action” (Zoeller et al., 2012: 4097). EDCs are a global health issue, and thus other groups, including international agencies, have weighed in on their definition and identification. In 2002, the World Health Organization (WHO) and International Programme on Chemical Safety (IPCS) defined these chemicals as an exogenous substance or mixture that alters function(s) of the endocrine system and consequently causes adverse effects in an intact organism, or its progeny, or (sub)populations. A potential endocrine disruptor is an exogenous substance or mixture that possesses properties that might be expected to lead to endocrine disruption in an intact organism, or its progeny, or (sub)populations. (IPCS, 2002: 1)

There are several reasons that different countries, and different agencies within single countries, have selected different definitions for EDCs. In 2016, a debate within the European Commission over the definition for an EDC highlighted some of these reasons (Bourguignon et  al.,  2016; Kortenkamp et al., 2016). Much of this debate has focused on the political and economic consequences for the wording used within each definition (Solecki et al., 2016): some definitions require that “proof ” of a link ­between a chemical and an adverse effect be established before a compound can be labeled an EDC; this is a high burden to reach, and even determining which endpoints are “adverse” is a complex and controversial issue (Zoeller et  al.,  2014). Others have required that data be available to demonstrate the mechanism by which a chemical causes harm prior to the labeling of a compound as an EDC (Slama et  al.,  2016); again, this is a high burden of  proof, depending on the level of mechanistic ­information that is required. The use of different definitions by different agencies has caused controversy, and likely will cause further political and economic issues, because a chemical could be labeled an EDC according to one definition (in one jurisdiction) but be considered “safe” from this label in another country, or by ­another agency. The EPA’s definition, for example, ­requires only that a compound be demonstrated to affect the action of hormones in the body for it to be considered an EDC, whereas the WHO/IPCS definition requires that such actions be linked to a downstream adverse effect (Beronius & Vandenberg, 2016). The Endocrine Society, having adapted the 90

EPA’s ­definition, argues that any alteration to the action of hormones in the body will induce adverse outcomes, at least in some populations (Zoeller et al., 2012, 2014). Thus, they argue, any compound that is an EDC is also expected to cause harm. Identifying EDCs is, of course, dependent on how these compounds are defined. One definition that was proposed for use in the European Union (and has since been rejected) required that a ­compound be “known to cause harm to humans” before it could be labeled an EDC. This definition would thus require a relatively large body of epidemiological studies before a compound could be identified as an EDC. From the public health perspective, such a requirement is unacceptable; even in relatively lax regulatory environments, requiring that a compound be known to harm humans before it can be regulated is not scientifically or ethically defensible. Starting in the 1990s, the EPA worked with ­scientific experts to develop methods to screen ­environmental chemicals for endocrine-disrupting properties. As a part of the Endocrine Disruptor Screening Program (EDSP), the EPA developed a two-tiered testing system to characterize compounds as EDCs (Borgert et al., 2011). The first tier is composed of 11 assays that allow for the identification of chemicals that have the potential to interact with the endocrine system (Table  6.1). These assays are focused on three hormones (estrogens, androgens, and thyroid hormone) and steroidogenesis pathways. Chemicals that are identified as “positive” in Tier 1 assays are then tested in Tier 2 assays, which examine specific adverse outcomes associated with disruption of the endocrine system (e.g., endpoints that demonstrate developmental and reproductive toxicity). Members of the Endocrine Society and other scientific experts have raised serious concerns about the use of the EDSP assays, suggesting that they are not comprehensive enough to evaluate all aspects of hormone biology; it has been ­suggested—and demonstrated empirically—that the assays will not detect all compounds that interfere with hormone action (Myers, Zoeller, & vom Saal,  2009; Zoeller et  al.,  2012,  2014). Concerns have also been raised about how the data from the EDSP assays are combined and collectively analyzed (Myers et  al.,  2009,  2015; L.  H.  Vandenberg & Bowler,  2014; L.  H.  Vandenberg, Colborn, et  al., 2013). For this reason, alternative testing programs have been proposed (Schug et  al.,  2013) and have begun to be used (Soto, Schaeberle, Maier, Sonnenschein, & Maffini, 2017).

Endocrine Disruptors and Other Influences on Hormone Action

Table 6.1.  Summary of Endocrine Disruptor Screening Program (EDSP) Tier 1 Assays Type of Assay

Assay

Hormone System Evaluated

Cell fraction assay

Aromatase ER binding assay AR binding assay ER transcriptional activation assay Steroidogenesis assay Uterotrophic assay Hershberger assay Pubertal female assay

Enzyme converting testosterone to estradiol Estrogen receptor agonists and antagonists Androgen receptor agonists and antagonists Estrogen receptor agonists and antagonists Hormone synthesis Estrogen receptor agonists (rodent) Androgen receptor antagonists (rodent) Estrogen receptor agonists and antagonists, thyroid hormone receptor antagonists (rodent) Androgen receptor agonists and antagonists, thyroid hormone receptor antagonists (rodent) Estrogen receptor agonists and antagonists (fish) Thyroid hormone receptor agonists and antagonists (frog)

In vitro assays In vivo assays

Pubertal male assay Fish short-term reproductive assay Amphibian metamorphosis assay

For the remainder of this chapter, discussions of EDCs will rely on the Endocrine Society’s definition of an EDC (Zoeller et  al.,  2012), which is a simplification of the definition originally proposed by the EPA (Kavlock et al., 1996).

Sources of Endocrine-Disrupting Chemicals

Estimates from government agencies suggest there  are now more than 80,000 chemicals in ­commerce (Bergman, Heindel, Kasten, et al., 2013). Unfortunately, few of these chemicals have been well evaluated for toxicity, and even fewer tests for endocrine-disrupting activity have been completed. For this reason, it is difficult to determine exactly how many EDCs are in the environment, or to estimate how many EDCs are produced in high volumes. Different groups that have used various criteria to evaluate chemical toxicity data have identified between 800 and 1,200 compounds as EDCs. For example, the Endocrine Disruption Knowledgebase developed by the US Food and Drug Administration (FDA) includes more than 1,000 chemicals based on their ability to bind to just a handful of nuclear hormone receptors in either in vitro or in vivo assays (FDA,  2010; The Endocrine Disruption Exchange [TEDX],  2015). Because most of these chemicals have not been evaluated with Tier 2 assays, the FDA argues that they cannot yet be considered EDCs. Others have argued that Tier 1 assays are sufficient to conclude that a compound is an EDC according to the EPA definition. Compounds that have been identified as EDCs are used in a variety of consumer products, including personal care products, food and beverage p ­ ackaging, detergents, toys and other children’s items, fragrances,

electronics, upholstery and other furnishings, ­medical and sports equipment, and building materials (Bergman, Heindel, Jobling et al., 2013a; Gore et  al.,  2015). Chemicals used in a wide range of ­industrial processes, and compounds that are the byproducts of industrial reactions, have also been identified as EDCs. Some EDCs are chemicals that have been developed specifically to be biologically active. Many chemicals used as pesticides (herbicides, insecticides, fungicides, etc.) were designed to interfere with biological processes and have endocrine-disrupting properties as unintended features. Furthermore, some pharmaceuticals have been ­developed that were designed to mimic or block the actions of hormones (e.g., to treat hormonedependent cancers or as birth control agents). These compounds are used by patients to purposely ­“disrupt” the endocrine system (e.g., by blocking ovulation or preventing the conversion of androgens to estrogens in the case of estrogen-sensitive tumors). However, because these pharmaceuticals are typically administered at high doses, they are ­excreted by patients into the environment, where they disrupt hormone action in wildlife (Bhandari et  al.,  2015; Christen, Hickmann, Rechenberg, & Fent, 2010; Orlando & Ellestad, 2014). Many pharmaceuticals, including contraceptives, have been detected in public drinking water, suggesting that low-level environmental exposures to human populations are also likely. Finally, there are some naturally occurring EDCs, including phytoestrogens (i.e., plant-derived estrogens) and mycoestrogens (i.e., fungal-derived estrogens) and some heavy metals including lead, cadmium, nickel, copper, and ­ ­chromium that have been shown to display hormonal activities. Vandenberg

91

Defining “Hormone Action”

As described previously, the vast majority of compounds identified as actual or putative EDCs by the FDA were discovered to bind to three hormone receptors: estrogen, androgen, and thyroid hormone (FDA, 2010). Because the endocrine system is much more complex than these three nuclear receptors, it is likely that many more compounds with endocrine-disrupting properties have not yet been identified. For example, few chemicals that mimic or block the actions of glucocorticoid or mineralocorticoid hormones have been identified; this is likely due to the absence of high-throughput methods to evaluate compounds that disrupt these endocrine pathways. Although receptor binding (agonism and antagonism) is the most common feature of identified EDCs, hormones have a number of actions; each step in the pathway is subject to possible disruption. In addition to their interactions with receptors, hormones are understood to be synthesized, secreted, transported in blood, metabolized, and eliminated (Diamanti-Kandarakis et al., 2009). Hormones are also known to induce downstream actions, including alterations to gene expression and subsequent changes to tissue morphology and function. EDCs can block any—or all—of these steps; disruption could involve up-regulation or down-regulation of any of these steps as well (see Figure 6.2). Hormones are synthesized and secreted from ­endocrine organs into the bloodstream, where they

are transported in blood and travel to target cells. Within target cells, they bind to receptors located on the cell membrane or within the cell (in the ­cytoplasm or the nucleus). Hormones are metabolized and excreted, allowing serum concentrations to be tightly controlled. By definition, EDCs can interfere with any of these steps, altering hormone action.

Endocrine-Disrupting Chemicals and Endocrine Principles

The endocrine system is complex, with involvement in virtually all aspects of life from conception until death (Diamanti-Kandarakis et  al.,  2009). Thus, although most screening tools and studies of EDCs have focused on the hormones involved in reproduction, dozens of signaling molecules work together to coordinate the organs and systems of the body (Gore, Heindel, & Zoeller,  2006; Myers et  al., 2009). Furthermore, the narrow view of ­hormones in the context of “homeostasis” is an inappropriate understanding of signaling molecules that have important—and vastly different—roles at ­different stages of development (Bergman, Heindel, Jobling, et  al.,  2013b; Diamanti-Kandarakis et  al.,  2009; Heindel & Vandenberg, 2015). A number of prominent endocrinologists have suggested that studies evaluating compounds that interfere with one or more aspect of hormone action would be better informed if they understood the principles of endocrinology (Beausoleil et al., 2013; Gore et al., 2006; Myers et al., 2009;

target cells

1 2

end ocr

ine

org a

n

4a

3 6 4b

1 - synthesis 2 - secretion 3 - transport 4 - binding (a) membrane (b) cytoplasm (c) nucleus 5 - metabolism 6 - excretion

bloo

d ve

ssel

5 4c target cells

Figure 6.2  Hormone actions and disruption by endocrine-disrupting chemicals (EDCs).

92

Endocrine Disruptors and Other Influences on Hormone Action

L.  H.  Vandenberg, Colborn, et  al.,  2013; Zoeller et al., 2012). Five major principles dictate the study of hormones: Principle 1. Hormones are responsible for coordinating the tissues of the body from conception until death (Diamanti-Kandarakis et al., 2009). Hormones have essential roles at all life stages including embryogenesis, fetal development, birth, puberty, pregnancy, parturition, and aging. Because hormones integrate the tissues of the body at all stages of life, disrupting the endocrine system will produce diseases and other types of dysfunction that are specific to the period of disruption, even if the diseases take years or decades to manifest (Heindel, 2005, 2008). For example, exposure to synthetic estrogens such as the potent pharmaceutical estrogen diethylstilbestrol (DES) while in the womb induces clear cell adenocarcinoma of the vagina that begins to manifest in puberty and breast cancer that begins to manifest in  the fourth decade of life (Soto, Vandenberg, Maffini, & Sonnenschein,  2008; Trichopoulos, 1990). Mothers exposed to DES during pregnancy had modest increases in their own breast cancer risk, but no increase in vaginal cancer (McLachlan, Newbold, Burow, & Li, 2001). Although most EDCs that have been identified and well studied disrupt the estrogen, androgen (testosterone), and thyroid hormone signaling ­pathways, the number of hormones operating in the bodies of vertebrates, including humans, is much greater. Thus, the signaling pathways controlled by  these hormones are similarly large in number. There is no reason to expect that EDCs will not affect these different pathways as well, with likely implications on the reproductive tract, brain, cardiovascular system, bone, immune system, metabolic machinery, and others (Heindel,  2008; Schug, Janesick, Blumberg, & Heindel, 2011). In fact, highthroughput screening assays for some of these “novel” pathways are beginning to reveal novel endocrinedisrupting properties (Janesick et  al.,  2016). The concept of “obesogens” and metabolism-disrupting chemicals emerged, in fact, because of compounds that could increase adipocyte number and size, alter function of the liver and pancreas, and induce diseases, including type 2 diabetes and nonalcoholic fatty liver disease (Grun & Blumberg,  2009; Heindel, Blumberg, et al., 2017; Heindel, vom Saal, et al., 2015). Principle 2. Hormones act via highly specific binding to receptors. The activity of hormones is ­mediated by receptors, and hormones have highly specific interactions with one or more receptors

(Norman & Henry,  2015). Only cells that express a  hormone’s receptor will be responsive to that ­molecule. Altering the responses of cells, tissues, and organs to hormones can be mediated by changing the number of receptors or by altering the concentration of the hormone itself, either in circulation or in the target tissue. There are a number of additional features about the interactions between hormones and their receptors that guide how hormones, and EDCs, are often evaluated. One is the concept of binding affinity, which is a way to characterize the relationship between the concentration of the ligand that is required to maximally occupy the ligand-binding site of the receptor. Some hormones have more than one receptor (estrogens bind to estrogen receptor α and β, for example), and some hormone receptors have multiple isoforms (e.g., progesterone receptor). If a hormone has a different binding affinity for the receptors or isoforms, it will influence the likelihood of binding to each receptor, and thus influence subsequent downstream effects, including gene expression. Binding affinity is a constant— hormones have a specific binding affinity for each receptor. With regard to EDCs, there is an expectation that binding affinity can be used to determine which compounds are “weak” or “strong” agonists. Using this approach, a binding affinity similar to the natural hormone is interpreted to be a “strong” agonist. Unfortunately, this approach conflates binding affinity with potency, defined as the concentration of a compound that is required to produce a  biological effect at a given intensity (Bergman et  al.,  2015; Zoeller et  al.,  2014). Potency is thus endpoint specific, and may not necessarily be predicted by binding affinities. In fact, there are ­numerous examples where compounds could have binding affinities that are quite divergent but potencies that are similar, at least for some endpoints. In these cases, compounds that bind to the receptor with much lower affinity compared to the endogenous hormone can have surprisingly “strong” ­effects—which closely mimic the actions of the natural hormone. For this reason, broad-stroke labeling of EDCs as “weak” estrogens, for example, should be avoided. Furthermore, such labels often fail to consider life stage, another important feature of hormones (see Principle 4). Principle 3. Hormones act at exceptionally low doses. Most hormones circulate in the blood at ­part-per-billion (ppb) or part-per-trillion (ppt) concentrations (L.  N.  Vandenberg,  2014; Vandenberg Vandenberg

93

94

% receptors bound

(A)

10 8 6 4 2 0

(B) biological effect

et al., 2012). Specifically, circulating concentrations of hormones are typically in the ppt range for estradiol, testosterone (in adult females), and thyroid hormone; other hormones that circulate in the ppb range include progesterone, testosterone (in adult males), growth hormone, and prolactin (Jones, 1996). These concentrations change depending on age and physiological status (e.g., puberty, pregnancy). For example, testosterone levels in human male neonates are typically in the range of 0.2 to 0.5 ng/ml. These levels increase to 4 to 6 ng/ml in most adult men, which is about a 10-fold increase. Similarly, estradiol concentrations are typically in the range of 50 to 100 pg/ml in women during the early follicular phase of the menstrual cycle, but raise to 600 pg/ ml during the luteal phase, which is again about a 10-fold increase. The specific interactions (described in Principle 2) between a hormone and its receptor(s) are responsible for the ability of hormones to have potent effects at such low concentrations. There are several specific factors responsible for these low-dose effects: the ligand’s binding affinity for the receptor, the number of receptors, the concentration of the hormone, and the developmental stage when exposures occur (Barouki, Gluckman, Grandjean, Hanson, & Heindel, 2012; Charlton, 2009; Lees, Cunningham, & Elliott, 2004; Welshons et al., 2003). There is a nonlinear relationship between the concentration of the hormone and the number of bound receptors, and a nonlinear relationship between the number of bound receptors and the induced biological effect (Vandenberg et al., 2012). What this means is that, at low concentrations, even small increases in hormone concentration will have large biological effects; in contrast, at high concentrations, the same increase in hormone level will have little biological effect (see Figure  6.3). Most hormone receptors are actually unbound—a situation previously termed “the spare receptor hypothesis” (May, Moran, Kimelberg, & Triggle, 1967; Zhu, 1996); less than 10 percent of hormone receptors are typically bound at any time. Thus, small changes in hormone concentration translate to relatively large changes in the total number of bound receptors (Welshons et  al.,  2003). In essence, the system is “primed” to respond to low doses of ­hormone and to small changes in hormone concentration. In the context of EDCs, many of these c­ ompounds circulate in the ppb and ppt concentration range. Thus, it is not surprising that they can have  biological ­effects at these low doses. Some EDCs are found in  tissues in the part-per-million concentration

1

2 3 4 5 6 7 Hormone concentration (arbitrary units)

10 8 6 4 2 0

2

4 6 8 % receptors bound

10

Figure 6.3  Nonlinear relationships between hormone concentration, bound receptors, and biological effects. (A) A nonlinear relationship is observed between hormone concentration and the percentage of receptors that is bound by the hormone. This means that the greatest increase in percent bound receptors is seen in the low-dose range. (B) A nonlinear relationship is observed between percent bound receptors and biological effects. This means that the greatest change in biological effect is seen when relatively few receptors are bound.

range, which is not “low” compared to hormone concentrations. Another important feature that distinguishes some hormones from EDCs is their transport in the blood. When steroid hormones are found in the blood, they are distributed into three phases: (1) free, representing the unconjugated, unbound form; (2) bioavailable, representing hormones bound to low-affinity carrier proteins such as albumin; and (3) inactive, representing the form that is bound to high-affinity binding proteins such as sex hormone– binding globulin (SHBG) or α-fetoprotein (Welshons et al., 2003). For a hormone to be physiologically active and thus capable of inducing biological effects, it must be found in either the free or bioavailable phases. In contrast, many EDCs do not bind to these same circulating proteins; therefore, the entire concentration of EDC found in the blood is ­available and capable of inducing biological effects (Dechaud, Ravard, Claustrat, de la Perriere, & Pugeat, 1999; Milligan, Khan, & Nash, 1998; Nagel, vom Saal, & Welshons, 1999; Sheehan & Young, 1979). Although there is sufficient understanding from multiple disciplines that hormones and other

Endocrine Disruptors and Other Influences on Hormone Action

s­ignaling molecules can have potent effects at low doses, there remains significant controversy about “low-dose effects” in the study of EDCs. This is ­discussed in detail later in this chapter. Principle 4. The effects of hormones depend on life stage (Heindel & Vandenberg, 2015; Wallen, 2009). In over 50 years of study, endocrinologists have come to understand how the same hormone can have different effects depending on the period of exposure. Typically, exposures have been broken down into two categories: exposures occurring during adulthood and those occurring during development. Exposures occurring in adulthood induce “activational” responses because the individual will be activated to respond only as long as the hormone is present. An  example of this is the use of pharmaceutical contraceptives, where women do not ­ovulate when exposed to synthetic estrogens like 17α-ethinyl estradiol (or estrogen plus progesterone combinations), but regain their fertility once they cease using these pharmaceuticals. In contrast, exposures occurring during development (often characterized as embryonic or fetal exposures) induce “organizational” responses because they promote changes to the differentiation of cells and organization of tissues. These organizational effects of hormones can permanently alter the development of cells, tissues, and organs. Only recently have endocrinologists and developmental biologists realized that organizational effects of hormones can extend beyond embryonic and fetal development to other periods of life. For some organs such as the brain, it is well established that development can continue for decades after birth (Heindel, Balbus, et al., 2015). Other studies have revealed that some organs (e.g., the breast) have critical windows of development at several distinct phases of life (embryonic, pubertal, pregnancy, lactation), and thus organizational effects of hormones occur during these critical periods (Macon & Fenton, 2013). EDCs have shown similar time-dependent ­effects. For example, studies in rodents have shown that relatively large (mg/kg) doses of bisphenol A (BPA), an estrogenic EDC, are needed to induce a uterotrophic response (an increase in the wet weight of the uterus) when exposures occur in adults (Kanno et al., 2003). Further, when exposures cease, the uterotrophic response is similarly lost. Yet, much lower doses of BPA during early d ­ evelopment (ng/kg) will shift the phenotype of uterine cells, altering tissue organization permanently (Markey, Luque, Munoz De Toro, Sonnenschein, & Soto,  2001; Markey, Michaelson, Veson, Sonnenschein, & Soto, 2001).

In humans, DES, the pharmaceutical estrogen that was administered to pregnant women in the 1940s to 1970s, provides another example of the organizational versus activational effects of exposure (Bern, 1992). Mothers exposed to DES during pregnancy did not develop reproductive cancers or malformations, whereas their daughters, exposed during gestation, had a significantly increased risk of clear cell adenocarcinoma of the vagina and other reproductive malformations. Further study revealed that there were specific times during gestation when DES induced the most harm in exposed daughters; those that were exposed to the pharmaceutical during the first 15 weeks of gestation had the highest incidence of clear cell adenocarcinomas of the vagina and other vaginal and cervical pathologies (Mittendorf, 1995). These results indicate not only that prenatal development is uniquely sensitive to estrogenic compounds like DES but also that for some health endpoints, the critical window of exposure is even smaller. Principle 5. Hormones display nonlinear and even nonmonotonic relationships between dose and effect. As described earlier, the relationship between the dose/concentration of a hormone and the percentage of receptors that are bound is nonlinear, such that greater increases in bound receptors are observed in the low-dose range (Welshons et al., 2003). A similar nonlinear relationship has been demonstrated between the percentage of bound receptors and the biological effect that is induced, with a greater relative increase in biological effects observed in the low-dose range. For this reason, linear relationships between hormone dose/concentration and any biological outcome are unexpected if a ­ receptor mediates the effect that is being observed (see Figure 6.3). Many dose–response relationships that are observed are nonlinear but are still monotonic; this means that with increasing dose, the effect is always either increasing (see Figure  6.4A) or decreasing (see Figure 6.4B), and although the slope of the line may change, the sign of the slope (either positive or  negative) does not. Yet, there are also many ­examples from the literature where the relationship between dose and effect are nonmonotonic, defined as a dose–response curve where the slope of the line ­representing the dose–response curve changes sign  from negative to positive (or vice versa; see Figure 6.4C; Kohn & Melnick, 2002). These kinds of curves are often referred to as biphasic, U-shaped, or inverted U-shaped. Nonmonotonic curves defy traditional ways of thinking of toxicants, where Vandenberg

95

(B)

10 8

8

6

6

4 2 0

(C)

1

2

3

8

4 Dose

5

6

0

7 (D)

Effect

4 2 2

3

4 Dose

1

2

5

6

7

3

4 Dose

5

6

7

5

6

7

inhibitor of cell proliferation

6 4 cell proliferation (promoter)

2

U-shaped 1

10 8

inverted U-shaped

6

0

4 2

10

Effect

10

Effect

Effect

(A)

0

1

2

3

4 Dose

Figure 6.4  Nonmonotonic relationships are common for hormones and endocrine-disrupting chemicals (EDCs). (A) Nonlinear, monotonic relationship between dose and effect with a positive slope. (B) Nonlinear, monotonic relationship between dose and effect with a negative slope. (C) Examples of nonmonotonic relationships between dose and effect. (D) Nonmonotonic dose responses can manifest due to two overlapping monotonic effects that act on a common endpoint.

more of an exposure is anticipated to lead to more of an effect; consider the expected relationship between poison and death, or consumption of sweets and body weight. Yet, the literature contains many examples that illustrate nonmonotonic relationships; these are observed for essential nutrients and vitamins, where both too much and too little of these substances lead to dysfunction or death, and optimal levels of exposure are in the moderate range. Similarly, the relationship between water consumption and health is nonmonotonic; both too much and too little can lead to death. With regard to hormones and EDCs, the ­mechanisms by which nonmonotonic dose responses can be produced are now well understood. One mechanism is similar to what has been observed for vitamins and other essential nutrients: when levels are too low, cellular function is disrupted, but high levels (often achieved via exogenous hormone exposures) can be overtly toxic. Other mechanisms that generate nonmonotonic dose responses include the activation of negative- and positive-feedback loops within the endocrine system, the overlapping effects of hormones acting on two monotonic responses that influence the same endpoint (e.g., estrogens stimulate both cell proliferation and inhibit cell proliferation; see Figure 6.4D), the down-regulation of receptors 96

at  higher doses, and desensitization of receptors (Vandenberg et  al.,  2012; Zoeller & Vandenberg, 2015). In the case of EDCs, nonmonotonic dose responses can also be observed when a compound binds to more than one hormone receptor and ­induces competing effects; some EDCs bind to one receptor at low doses and a second receptor at higher doses (Watson, Bulayeva, Wozniak, & Alyea, 2007; Watson, Jeng, & Guptarak, 2011). There are hundreds of examples of nonmonotonic dose responses that have been observed for both hormones and EDCs (Vandenberg et al., 2012). These nonmonotonic effects occur in cultured cells and in laboratory animals following controlled administration of compounds. Importantly, they have also been observed in human epidemiology studies examining environmentally exposed groups.

Controversies in the Study of Endocrine-Disrupting Chemicals

A number of scientific statements and position papers from endocrinologists have been published, indicating that (1) a number of chemicals are known to disrupt the action of hormones; (2) humans and wildlife populations are exposed to these compounds; (3) controlled exposures to these compounds induce

Endocrine Disruptors and Other Influences on Hormone Action

a range of effects, including effects associated with disease, in laboratory animals; and (4) observational studies from wildlife and humans also demonstrate associations between chemical exposures and adverse health outcomes (Diamanti-Kandarakis et al., 2009; Gore et al., 2015; Zoeller et al., 2012). Similar conclusions were drawn by other experts. One report was published in February 2013 by the United Nations Environment Programme (UNEP) and World Health Organization (WHO), entitled “State of the Science of Endocrine Disrupting Chemicals – 2012” (WHO, 2013). Similar to the scientific statements written by members of the Endocrine Society, the 2013 UNEP/WHO report concluded the following: (1) Data from controlled laboratory studies confirm that chemicals can ­contribute to endocrine disorders. Many of these diseases have been observed in humans and wildlife populations. (2) Wildlife populations are affected by EDC exposures, with negative effects specifically observed on growth and reproductive endpoints. (3) The methods that are widely used to identify and evaluate the safety of EDCs examine only limited endpoints, missing a large fraction of the known spectrum of endocrine-disrupting effects. As a result, the methods used by decision makers, including risk assessors, to draw conclusions about chemical safety are likely underestimating the harmful effects of EDCs. Finally, (4) the increased risk of diseases due to EDC exposures may be underestimated by regulatory agencies around the world. Other groups, including many researchers and advocates associated with the chemical industry, have denied these conclusions (Lamb et  al., 2014, 2015; Nohynek, Borgert, Dietrich, & Rozman, 2013; Rhomberg, Goodman, Foster, Borgert, & Van Der Kraak,  2012). Importantly, these individuals and the groups they represent often have financial interests in delaying action on EDCs (Bergman et al., 2015; Zoeller et al., 2014). However, the debate over public health concerns about EDCs continues in scientific communities and among government decision makers. Next, we will discuss some of the controversies that have arisen in the identification, study, and regulation of EDCs.

Can Endocrine-Disrupting Chemicals Have Effects at Low Doses?

Studies evaluating the role of hormones in fetal development have confirmed how even small ­ ­differences in hormone concentrations can influence physiological, morphological, and behavioral endpoints. Perhaps the best studied system for

e­valuating the consequences of natural variability in hormone concentrations comes from the study of intrauterine position in animals with litters, ­including mice and rats. In mice and rats, male fetuses produce testosterone in midgestation (starting around gestational day 13/14 in mice), and some of this hormone is transferred passively to neighboring fetuses via the maternal blood supply (J. G. Vandenbergh, 2004). Thus, pups (both male and female) positioned between two males are ­exposed to higher concentrations of testosterone compared to pups positioned between two females. Differences as low as 1 ng/ml testosterone in females and 25 pg/ml estradiol in males have serious effects on a large number of phenotypes, including ­prostate development, mammary gland morphology, responsiveness to testosterone, aggressive behaviors, steroid metabolism, age of pubertal onset, and ­estrous cycle length, among others (Ryan & Vandenbergh, 2002). Studies from human twins have also revealed that exceptionally low differences in hormone exposures during fetal development can influence health endpoints including reproductive success and risk of adult diseases like breast cancer (Vandenberg et al., 2012). Thus, intrauterine position and intrauterine exposures to endogenous hormones provide robust examples of low-dose effects for natural hormones. The low-dose hypothesis, originally articulated in the 1990s, postulated that EDCs have effects on exposed animals at doses that are thought to be safe for humans, and that humans are also affected by these same environmentally relevant doses (Davis et  al.,  1993; Sharpe & Skakkebaek,  1993). With regard to EDCs and other environmental chemicals, the term low dose actually has a number of meanings (Melnick et al., 2002): (1) doses below those used in traditional toxicology studies (i.e., doses below those that induce overt signs of toxicity); (2) doses in the range of what humans are typically exposed to (e.g., environmental, not occupational, exposures); and (3) doses that produce blood levels in experimental animals that mimic the levels m ­ easured in humans, accounting for differences in how species metabolize chemicals. Many studies use the term low dose without specifically defining which of the cutoffs detailed previously was used to determine that the dose is low. These differing definitions for what constitutes a “low dose” produce quite different cutoffs for the same environmental chemical. For example, the ­estrogenic chemical BPA has three different cutoff doses based on these three definitions. For doses Vandenberg

97

below those tested in traditional toxicology studies, the cutoff dose would be set by the no observed adverse effect level (NOAEL) of 50 mg/kg/day ­ (Welshons et al., 2003). For human exposure levels, consumer product assessments estimate that exposures are in the range of ng/kg or low µg/kg daily (Lakind & Naiman,  2008). Finally, for doses that produce blood concentrations found in the general population, doses of 100 to 400 µg/kg/day would be set as the cutoff dose, although there is also controversy about whether this compound can be accurately measured in blood (Vandenberg, Hunt, Myers, & Vom Saal,  2013). Importantly, exposure levels, urine/blood concentrations, and studies that carefully determine how compounds are metabolized are not available for the vast majority of chemicals on the market. For this reason, establishing evidence-based cutoffs to distinguish “low dose” studies will be difficult and will require commitments to produce these data by either chemical manufacturers or regulatory bodies. Considering these limitations, there are several EDCs that have been well studied at low doses. In 2002, the National Toxicology Program assembled an expert panel to evaluate the published ­literature and found evidence for low-dose effects for four compounds (Melnick et  al.,  2002): (1) DES, the pharmaceutical estrogen, induced effects on prostate weight after gestational exposures to low doses; (2) genistein, a phytoestrogen found in soy, produced effects on the volume of sexually dimorphic brain nuclei in male rats, the morphology of the mammary gland in male rats, and the proliferation of immune cells after exposures to low doses; (3) nonylphenol, a surfactant and industrial chemical with estrogenic properties, altered sexually dimorphic nuclei in the brain, thymus weight, estrus cyclicity, and proliferation of immune cells in rats exposed to  low doses during perinatal development; and (4)  methoxychlor, an insecticide with estrogenic properties, altered the immune system of rats exposed to low doses during perinatal development. At that time, the expert panel also noted that both biological features and factors related to study design may impact the ability to detect low-dose ­effects. For example, differences in animal species and strains can influence responses to hormones and EDCs. Furthermore, the use of varying diets with different concentrations of phytoestrogens, and leaching of hormonally active compounds from caging, water bottles, bedding material, and other environmental media, could interfere with the ­ detection of effects from low-dose EDCs. These 98

f­actors are discussed further in the last section of this chapter. In 2012 and 2013, a group of 12 experts from a range of scientific disciplines re-examined the lowdose literature, concluding that low-dose effects have been demonstrated for a larger number of ­environmental chemicals (Vandenberg et al., 2012; Vandenberg, Colborn, et  al.,  2013). Four compounds were evaluated in depth: (1) low doses of BPA on prostate weight and development, mammary gland development, and the response of the mammary gland to secondary environmental challenges including stimulation with carcinogens; (2) atrazine and sexual development and differentiation in amphibians; (3) dioxin and spermatogenesis in exposed rodents and humans; and (4) perchlorate and thyroid function in humans. Approximately two dozen additional compounds were found to have consistent low-dose effects on at least one endpoint (Vandenberg, Colborn, et al., 2013). In 2017, the National Academy of Sciences conducted a systematic review of two chemical families characterized as EDCs, phthalates and brominated flame retardants, and concluded that there was sufficient evidence from the literature for low-dose ­effects (National Academies of Sciences & Medicine, 2017). Specifically, phthalates were shown to alter anogenital distance and male hormone levels at low doses, and brominated flame retardants were shown to affect learning, memory, and IQ at low doses. The National Academy selected just two case studies to evaluate the use of systematic review protocols (discussed briefly later in this chapter) and as a proof of principle to demonstrate that low-dose ­effects are reproducible for these compounds. In looking at the kinds of conditions that have been observed to be associated with low-dose effects, the endpoints include alterations in the size of nuclei in the brain, alterations to the weight of ­endocrine organs (i.e., the prostate, uterus, and ovaries), metabolic disruptions (i.e., increased body weights, alterations to adipose differentiation, and nonalcoholic fatty liver disease), reduced sperm count and abnormal sperm quality, altered morphology of the mammary gland and increased susceptibility to carcinogen challenges, expression of atypical behaviors (i.e., anxiety-like behaviors, hyperactivity, and disruptions to learning and ­ memory), altered release or circulating concentrations of hormones (i.e., testosterone, corticosterone, and thyroid-stimulating hormone), abnormal tissue organization in hormone-sensitive organs (e.g., the prostate, thyroid, mammary gland, and uterus), and

Endocrine Disruptors and Other Influences on Hormone Action

others (reviewed in Vandenberg et al., 2012). These endpoints are considered adverse to endocrinologists, developmental biologists, and physicians, and therefore there is concern about low-dose exposures to these compounds. Based on the number of examples that have now been evaluated by different groups, some scientists (see Vandenberg et al., 2012) have argued that the low-dose hypothesis should not be considered a hypothesis any longer. A significant body of evidence indicates that EDCs can induce adverse effects, including diseases that are relevant to human populations, at doses in the range of human exposures.

Are Humans Exposed to EndocrineDisrupting Chemicals at Doses That Have Effects on Health?

Hundreds of compounds have been measured in human blood or urine, with concentrations that are typically in the ppb or ppt range, although some are measured in the ppm range (e.g., Braun et al., 2014; Kato, Wong, Jia, Kuklenyik, & Calafat, 2011; Woodruff, Zota, & Schwartz, 2011; Ye, Zhou, Wong, & Calafat,  2012). Many of these compounds are characterized as EDCs. Examination specifically of pregnant women shows that these individuals have dozens of chemicals, and specifically EDCs, in their bodies at any one time. These findings are particularly concerning because they provide unequivocal evidence that the fetus is exposed to large numbers of chemicals during gestation (Arbuckle, 2010). But the mere presence of these compounds in human tissues and fluids is not sufficient to conclude that they are also causing harm. Certainly, controlled studies of EDCs in laboratory animals provide evidence that effects on humans are anticipated. Yet, human epidemiology studies also provide support for the conclusion that EDC

exposures are affecting human health. Hundreds of environmental epidemiology studies suggest associations between EDCs and a range of health outcomes including obesity, infertility, attention deficit hyperactivity disorder, cancers, anxiety disorders, thyroid dysfunction, IQ, and others (reviewed in Gore et al., 2015; Vandenberg et al., 2012; Zoeller et al., 2012). Importantly, these studies do not involve controlled or randomized exposures, and thus it can be challenging to draw conclusions about the causal relationships between exposures and effects. Furthermore, many environmental epidemiology studies use cross-sectional study designs, where ­exposures and health outcomes are evaluated at the same time, making it impossible to determine the temporal relationship between the two (e.g., does exposure cause the disease, or does having the disease influence exposure?). In the last decade, epidemiologists have focused on building prospective cohorts, allowing exposures to be measured (often during vulnerable periods of development) and individuals to be followed over time to determine if they develop diseases. These longitudinal studies, although limited in number, have provided stronger evidence for causal relationships between EDCs and disease outcomes (Bergman, Heindel, Kasten, et al., 2013; Heindel, Skalla, Joubert, Dilworth, & Gray, 2017).

Are There Safe Doses of EndocrineDisrupting Chemicals?

The traditional approach to the evaluation of toxic compounds is to determine which (high) doses induce morbidity and mortality, and then find a  (lower) dose where no such effects exist (see Figure  6.5). This point, below which adverse outcomes are not expected following exposures, is often referred to as a “threshold” (Bergman, Andersson, et al., 2013; Bergman et al., 2015). The absence of

10

biological effect

8 6

no-threshold

threshold

4 2 0

1

2

3

4

5

6

7

–2 –4

dose

Figure 6.5  No-threshold and threshold models for environmental chemicals.

Vandenberg

99

biological effects below this threshold dose is rarely demonstrated. This approach, where high-dose ­toxicity is extrapolated to low-dose safety, does not appear to apply to hormonally active substances. This is because the endocrine system is already functioning within organisms, and therefore substances that alter receptor binding (by mimicking or blocking the actions of endogenous hormones) or other aspects of hormone action (synthesis, secretion, metabolism, etc.) are either adding or subtracting from processes that are already ongoing; rather than a substance activating a system that is dormant (and might be considered to have a biological threshold for activation), the endocrine system is active and can readily be disrupted by small amounts of chemicals that activate or antagonize it. In the threshold model, there is a dose below which effects are not observed. Thresholds are assumed to exist for many environmental chemicals, although they are rarely evaluated empirically. On this graph, a threshold model is represented by a dotted line, and the threshold is indicated by a black circle. For EDCs, thresholds are difficult if not impossible to demonstrate because these compounds alter active biological processes. Thus, the no-threshold model can be used, where even minute quantities of EDCs are anticipated to have biological effects, even if they cannot be measured experimentally. The no-threshold model is represented by a solid black line. Some examples from endocrinology illustrate the concept of an active endocrine system and how it can be disrupted by EDCs that either “add” to or “subtract” from those active processes. The first ­example comes from a relatively large literature demonstrating that gestational exposures to estrogens— endogenous and exogenous—can contribute to increased breast cancer risk (Soto et al., 2008). The embryonic mammary tissue expresses estrogen receptors, yet the fetal ovaries produce little if any estrogen (depending on the species; Soto et  al., 2008). Pharmaceutical estrogens like DES and estrogenic environmental chemicals like BPA can disrupt development of the fetal mammary gland, altering gene expression in the epithelium and ­ composition of the stroma. These changes increase the sensitivity of the gland to hormones and carcinogens, contributing to increased risk of precancerous lesions and carcinomas in adulthood. A second example is illustrated by the effect of antiandrogenic compounds like many phthalates on anogenital distance in both rodents and humans. 100

Androgens produced by the fetal testis are responsible for development of the male reproductive tract and increase the size of the tissue between the anus and the genitals (typically measured by either the base of the scrotum or the penis). For centuries, farmers have used anogenital distance as a simple way to distinguish male and female animals at birth because it is a reliable and visible way to determine the sex of animals, even at an age where other sexually dimorphic features (e.g., the presence/absence of testes) are difficult to discern. Antiandrogenic compounds decrease anogenital distance in males, even to the point of making it problematic to distinguish the sexes (Gray et  al.,  2001). This effect is i­mportant not just because it makes it difficult to distinguish male and female animals, but because in humans anogenital distance is predictive of a number of health outcomes in men, including fertility, semen parameters, and serum reproductive hormone concentrations, among others (Mendiola, Stahlhut, Jorgensen, Liu, & Swan, 2011). If there are no “thresholds” for EDCs, the logical conclusion is that there are no doses that would not have at least some effect. The question is whether the tools that are being used to evaluate those effects are sensitive enough, and whether the sample size or population being examined is large enough to provide sufficient power to detect these effects. Rodent studies evaluating carcinogens have found that thousands of mice are needed per treatment group to determine which dose increases cancer incidence by just 1 percent (Littlefield, Farmer, Gaylor, & Sheldon, 1980). These so-called mega-mouse studies suggest that even low doses of compounds could have effects with significant implications for public health, but the likelihood of detecting these effects in traditional studies (with sample sizes of 20 to 50 per treatment group) is low. Another issue to consider when determining the safety of EDCs is the types of tests and the endpoints included in those tests that are used during hazard and risk assessment. As described earlier in this chapter, standardized tests such as those used in the EDSP Tier 1 and Tier 2 evaluate endpoints that are globally recognized as toxicologically relevant. Tier 2 assays and other tests that follow prescribed methodologies (described as test guidelines) examine endpoints that are described as “apical,” including body weight and food consumption in the dam, time to pregnancy and length of gestation, number of pups and uterine implantation sites, sex ratios, gross morphology of exposed pups, weight of organs

Endocrine Disruptors and Other Influences on Hormone Action

in the exposed pups, and histopathology of reproductive organs in exposed pups. Are these the best endpoints to evaluate EDCs, and low-dose effects more specifically? Although these endpoints are likely appropriate for assessing toxicity, they are not sufficient to evaluate the broad range of endocrine diseases and dysfunctions that affect human populations. Suggestions have been made that traditional methods for evaluating chemical safety should incorporate more sensitive endpoints relevant to endocrine diseases, such as mammary gland morphology in whole mount preparations, sensitive behavioral assays that test for sexually dimorphic behaviors, assays that evaluate cardiovascular and immune function, metabolic assays that evaluate pancreatic function, tests that examine endpoints relevant to thyroid hormone action rather than simple measurements of serum hormones, and others (Gore et  al.,  2006; Myers et al., 2009).

Beyond Endocrine-Disrupting Chemicals: Other Environmental Influences

The environment plays an important role in disease risk, severity, latency, and other features of high relevance to human and wildlife health. Understanding how much the environment, and which specific ­aspects of the environment, influences disease is an important part of modern medicine and public health. Many additional environmental factors can influence human health, including alcohol consumption, use of pharmaceuticals, exposure to other pollutants, UV and other radiation exposures, ­vibrations, and noise, among others. Acknowledging how environmental influences can affect controlled laboratory studies is also important, and deserves ­attention from scientists working in a broad range of fields. Differences in environmental factors in animal facilities and other laboratory environments may contribute to the “replication crisis” that has recently received attention in the scientific community; many studies cannot be replicated, either in other labs or in the original labs that conducted the initial studies (Iqbal, Wallach, Khoury, Schully, & Ioannidis, 2016). Changes to environmental conditions in laboratories cannot be discounted (Kolla, Pokharel, & Vandenberg, 2017). Next, other environmental features that can interfere with laboratory studies are briefly described.

Stress

Laboratory animals experience stress from numerous sources, many of which can be controlled in the

research setting. Shipping of animals, or movement within or between animal facilities, is stressful to animals and should be avoided during critical windows of development or periods of time relevant to the exposure or evaluation. Handling of animals is also stress inducing, even when care is given to provide pain-free handling. Furthermore, differences in how experimenters handle animals can influence the induction of stress. A recent study revealed that male experimenters induce greater stress responses in rodents compared to female experimenters, and these heightened stress reactions can diminish the animals’ responses to painful stimulants (Sorge et al., 2014). In the context of EDCs, attention has been given to how animals are exposed to environmental compounds. In many cases, experimenters attempt to recapitulate human-relevant routes of exposure, but the common means of oral exposure involves gavage, a method where a needle is inserted into the animal’s mouth and passed through the esophagus into the stomach, where the test compound is released. The process of gavage is stressful, does not truly replicate human oral exposures because it bypasses the oral mucosa, and can confound the ­evaluation of environmental chemicals (Vandenberg, Welshons, Vom Saal, Toutain, & Myers, 2014).

Food and Water

Researchers often do not pay attention to the type of feed that is provided to their animals. Yet, numerous studies have shown that rodent chow can have varying concentrations of chemical contaminants, including pesticides and heavy metals that can interfere with evaluations of EDCs, or can themselves induce biological effects (Mesnage, Defarge, Rocque, Spiroux de Vendomois, & Seralini, 2015). Similarly, the composition of the feed, including the amount of phytoestrogens, the fat content, and the presence/ absence of other essential nutrients, is an important consideration for laboratory animal studies (Ruhlen et al., 2011). Most animal researchers provide water ad libitum, and otherwise do not think about how this resource will affect their experiments. Not only should the vessel used to deliver water be considered (e.g., glass bottles, plastic bottles, unseen ­plastic piping delivering water directly to cages), but also the source of water is important. Pollutants and natural variation in minerals in water can interfere with experiments, leading some researchers to go to what may be seen as extremes (e.g., the use Vandenberg

101

of reverse-osmosis purification systems) to treat their water.

Housing

Other features of housing can contribute to stress and/or interfere with laboratory experiments. Caging is clearly an important aspect of housing (Koehler et al., 2003), as is the material used for bedding and any enrichment activities provided to animals (e.g., cotton nestlets, wooden dowels, running wheels). Researchers often provide enrichment materials only to animals that appear stressed, but this can lead to uneven chemical exposures if the enrichment activity includes an unknown chemical contaminant. Crowding is often controlled by institutional guidelines for animal housing, but solitary housing can also be stressful for animals that are social. For rodents, long periods of isolation from other conspecifics should be avoided, although this can be challenging when treatments (or age, or reproductive status) induce aggression; this is common in adult male rodents.

Culling and Fostering

Many researchers cull litters to maintain feasible experimental protocols and decrease animal care ­ costs. Yet, culling can introduce (or remove) biological variability, depending on the outcome to be evaluated. Culling has been shown to lead to catchup growth due to the restrictive nutrients experienced in utero with large litters and the sudden availability of nutrients available via nursing when the litter is smaller, altering body weight and metabolic endpoints (Suvorov & Vandenberg, 2016). Culling litters to change the overall sex ratio can also influence behavioral outcomes in pups, as can rearing in single-sex litters, which occasionally is observed without the interference of culling. ­ Researchers also often foster or cross-foster pups ­between litters to equalize litter size, optimize the use of lactating females, or evaluate specific hypotheses. Although most strains of mice and rats will accept and care for fostered pups, this care does deviate somewhat from the care that is provided to a dam’s own pups (Francis, Diorio, Liu, & Meaney, 1999; Maccari et al., 1995).

Emerging Issues for Endocrine-Disrupting Chemicals in Public Health

A number of relatively new steps have been taken in the field of environmental health sciences to better 102

understand and characterize not only the effects of EDCs but also their impact on societies. The first is the use of systematic review methodologies to better synthesize data derived from in vitro, laboratory animal, wildlife, and epidemiology studies (Vandenberg et al., 2016). Systematic reviews are common in the medical field and are important for drawing broader conclusions from multiple studies with approaches arising from disparate fields. Systematic review methods allow researchers to evaluate the quality of published studies and determine the strength of the evidence for specific, directed questions (e.g., does chemical X cause disease Y?). Methods are being developed to use systematic review methods to address much larger questions that are relevant to the study of EDCs (Beronius & Vandenberg,  2016), such as “What is the strength of the evidence that EDCs affect male fertility?” A final approach that has received attention uses tools from health economics to measure the cost of EDC exposures to human populations and society. Using conservative estimates of the contribution of  EDCs to specific diseases, and examining only a  small number of chemicals, researchers have ­assessed the annual health care cost of EDCs in the European Union at more than €150B (Trasande et al., 2015) and in the United States at more than $200B (Attina et al., 2016). Thus, ignoring EDCs and their effects on endocrine diseases is a costly choice, both to human health and to the health care economy.

Conclusion

In humans, evidence is mounting that secular trends for many endocrine-related diseases and disorders are increasing in prevalence (Bergman, Heindel, Jobling et  al.,  2013a). Some of these conditions include infertility (male, female, and idiopathic); neurobehavioral disorders including anxiety disorders, attention deficit hyperactivity disorder, and autism; asthma and autoimmune disorders; birth defects involving the male genital tract; endocrine-related cancers of the breast, ovary, prostate, thyroid, testes, and uterus; early puberty in females; and metabolic syndrome including obesity, type 2 diabetes, and nonalcoholic fatty liver disease, among others. Although improved detection and changes in diagnostic criteria may account for some of these increases, evidence suggests that the absolute risk for many of these conditions is increasing. These changes are occurring not only in the developed

Endocrine Disruptors and Other Influences on Hormone Action

world but also in developing countries where ­noncommunicable diseases recently became more common causes of death than communicable ­diseases (Balbus et al., 2013). Because many of these changes in disease incidence have been observed over short periods of time—some as much as five decades, others as little as two decades—changes in human genetics cannot be responsible. Thus, environmental factors and gene–environment interactions have received significant attention. Societal changes at the level of nutrition and exercise patterns are likely to be involved in some of these diseases, as are changes to occupations and the pollution involved in urbanization. Yet, experimental evidence linking low-level chemical exposure to disease outcomes in rodents has drawn necessary attention to the possible role for EDCs in some, if not all, of these diseases. Furthermore, epidemiology studies of environmentally exposed populations continue to suggest associations between EDCs and disease outcomes. Dismissing the likelihood that EDCs play at least some role in the conditions that increasingly affect human populations requires large bodies of ­literature—thousands of studies—to be ignored.

References

Arbuckle, T.  E. (2010). Maternal-infant biomonitoring of environmental chemicals: The epidemiologic challenges. Birth Defects Research Part A: Clinical and Molecular Teratology, 88(10), 931–937. doi:10.1002/bdra.20694 Attina, T.  M., Hauser, R., Sathyanarayana, S., Hunt, P.  A., Bourguignon, J.  P., Myers, J.  P., . . . Trasande, L. (2016). Exposure to endocrine-disrupting chemicals in the USA: A population-based disease burden and cost analysis. Lancet Diabetes & Endocrinology, 4(12), 996–1003. doi:10.1016/ s2213-8587(16)30275-3 Balbus, J.  M., Barouki, R., Birnbaum, L.  S., Etzel, R.  A., Gluckman, P. D., Grandjean, P., . . . Tang, K.-C. (2013). Earlylife prevention of non-communicable diseases. Lancet, 381(9860), 3–4. Barouki, R., Gluckman, P. D., Grandjean, P., Hanson, M., & Heindel, J.  J. (2012). Developmental origins of noncommunicable disease: Implications for research and public health. Environmental Health, 11, 42. doi:1476069X-11-42 Beausoleil, C., Ormsby, J.  N., Gies, A., Hass, U., Heindel, J. J., Holmer, M. L., . . . Schoenfelder, G. (2013). Low dose effects and non-monotonic dose responses for endocrine active chemicals: Science to practice workshop: Workshop summary. Chemosphere, 93(6), 847–856. doi:10.1016/j. chemosphere.2013.06.043 Bergman, A., Andersson, A. M., Becher, G., van den Berg, M., Blumberg, B., Bjerregaard, P.,  .  .  .  Zoeller, R.  T. (2013). Science and policy on endocrine disrupters must not be mixed: A reply to a “common sense” intervention by

toxicology journal editors. Environmental Health, 12, 69. doi:10.1186/1476-069X-12-69 Bergman, A., Becher, G., Blumberg, B., Bjerregaard, P., Bornman, R., Brandt, I., . . . Zoeller, R. T. (2015). Manufacturing doubt about endocrine disrupter science—A rebuttal of industrysponsored critical comments on the UNEP/WHO report “State of the Science of Endocrine Disrupting Chemicals 2012.” Regulatory Toxicology and Pharmacology, 73(3), 1007– 1017. doi:10.1016/j.yrtph.2015.07.026 Bergman, Å., Heindel, J.  J., Jobling, S., Kidd, K.  A., & Zoeller, R.  T. (Eds.) (2013a). The state-of-the-science of endocrine disrupting chemicals—2012. Geneva, Switzerland: United National Environment Programme and World Health Organization. Retrieved from http://www.who.int/ iris/bitstream/10665/78101/1/9789241505031_eng.pdf Bergman, A., Heindel, J. J., Jobling, S., Kidd, K. A., & Zoeller, R.  T. (2013b). State of the science of endocrine disrupting chemicals 2012. Summary for decision makers. Geneva, Switzerland: United National Environment Programme and World Health Organization. Bergman, A., Heindel, J. J., Kasten, T., Kidd, K. A., Jobling, S., Neira, M., . . . Woodruff, T. J. (2013). The impact of endocrine disruption: A consensus statement on the state of the science. Environmental Health Perspectives, 121(4), A104– A106. doi:10.1289/ehp.1205448 Bern, H. A. (1992). The fragile fetus. In T. Colborn & C. Clement (Eds.), Chemically-induced alterations in sexual and functional development: The wildlife/human connection (pp. 9–15). Princeton, NJ: Princeton Scientific Publishing. Beronius, A., & Vandenberg, L.  N. (2016). Using systematic reviews for hazard and risk assessment of endocrine disrupting chemicals. Reviews in Endocrine and Metabolic Disorders, 16(4), 273–287. doi:10.1007/s11154-016-9334-7 Bhandari, R. K., Deem, S. L., Holliday, D. K., Jandegian, C. M., Kassotis, C.  D., Nagel, S.  C., . . . Rosenfeld, C.  S. (2015). Effects of the environmental estrogenic contaminants bisphenol A and 17alpha-ethinyl estradiol on sexual development and adult behaviors in aquatic wildlife species. General and Comparative Endocrinology, 214, 195– 219. doi:10.1016/j.ygcen.2014.09.014 Borgert, C. J., Mihaich, E. M., Quill, T. F., Marty, M. S., Levine, S.  L., & Becker, R.  A. (2011). Evaluation of EPA’s Tier 1 Endocrine Screening Battery and recommendations for improving the interpretation of screening results. Regulatory Toxicology and Pharmacology, 59(3), 397–411. doi:10.1016/j. yrtph.2011.01.003 Bourguignon, J. P., Slama, R., Bergman, A., Demeneix, B., Ivell, R., Kortenkamp, A., . . . Zoeller, R. T. (2016). Science-based regulation of endocrine disrupting chemicals in Europe: Which approach? Lancet Diabetes & Endocrinology, 4(8), 643–646. doi:10.1016/s2213-8587(16)30121-8 Braun, J. M., Just, A. C., Williams, P. L., Smith, K. W., Calafat, A. M., & Hauser, R. (2014). Personal care product use and urinary phthalate metabolite and paraben concentrations during pregnancy among women from a fertility clinic. Journal of Exposure Science and Environmental Epidemiology, 24(5), 459–466. doi:10.1038/jes.2013.69 Carson, R. (1987). Silent spring: 25th anniversary edition. New York, NY: Houghton Mifflin. Charlton, S.  J. (2009). Agonist efficacy and receptor desensitization: From partial truths to a fuller picture. British Journal of Pharmacology, 158(1), 165–168. doi:10.1111/j.1476– 5381.2009.00352.x

Vandenberg

103

Christen, V., Hickmann, S., Rechenberg, B., & Fent, K. (2010). Highly active human pharmaceuticals in aquatic systems: A concept for their identification based on their mode of action. Aquatic Toxicology, 96(3), 167–181. doi:10.1016/j. aquatox.2009.11.021 Colborn, T., vom Saal, F. S., & Soto, A. M. (1993). Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environmental Health Perspectives, 101(5), 378–384. Davis, D.  L., Bradlow, H.  L., Wolff, M., Woodruff, T., Hoel, D.  G., & Anton-Culver, H. (1993). Medical hypothesis: Xenoestrogens as preventable causes of breast cancer. Environmental Health Perspectives, 101, 372–377. Dechaud, H., Ravard, C., Claustrat, F., de la Perriere, A. B., & Pugeat, M. (1999). Xenoestrogen interaction with human sex hormone-binding globulin (hSHBG). Steroids, 64(5), 328– 334. doi:S0039128X98001147 Diamanti-Kandarakis, E., Bourguignon, J.  P., Guidice, L.  C., Hauser, R., Prins, G. S., Soto, A. M., . . . Gore, A. C. (2009). Endocrine-disrupting chemical: An Endocrine Society scientific statement. Endocrine Reviews, 30, 293–342. The Endocrine Disruption Exchange (TEDX). (2015). TEDX list of potential endocrine disruptors. Retrieved November 21, 2015, from http://endocrinedisruption.org/endocrinedisruption/tedx-list-of-potential-endocrine-disruptors/ overview Francis, D., Diorio, J., Liu, D., & Meaney, M.  J. (1999). Nongenomic transmission across generations of maternal behavior and stress responses in the rat. Science, 286(5442), 1155–1158. Gore, A. C., Chappell, V. A., Fenton, S. E., Flaws, J. A., Nadal, A., Prins, G.  S., . . . Zoeller, R.  T. (2015). EDC-2: The Endocrine Society’s Second Scientific Statement on Endocrine-Disrupting Chemicals. Endocrine Reviews, 36(6), E1–E150. doi:10.1210/er.2015–1010 Gore, A. C., Heindel, J. J., & Zoeller, R. T. (2006). Endocrine disruption for endocrinologists (and others). Endocrinology, 147(6), S1–S3. Gray, L.  E., Ostby, J., Furr, J., Wolf, C.  J., Lambright, C., Parks, L., . . . Guillette, L. (2001). Effects of environmental antiandrogens on reproductive development in experimental animals. Human Reproduction Update, 7(3), 248–264. Grun, F., & Blumberg, B. (2009). Endocrine disrupters as obesogens. Molecular and Cellular Endocrinology, 304(1–2), 19–29. doi:10.1016/j.mce.2009.02.018 Heindel, J.  J. (2005). The fetal basis of adult disease: Role of environmental exposures—introduction. Birth Defects Research Part A: Clinical and Molecular Teratology, 73(3), 131–132. Heindel, J.  J. (2008). Animal models for probing the developmental basis of disease and dysfunction paradigm. Basic & Clinical Pharmacology & Toxicology, 102(2), 76–81. Heindel, J.  J., Balbus, J., Birnbaum, L., Brune-Drisse, M.  N., Grandjean, P., Gray, K., . . . Hanson, M. (2015). Developmental origins of health and disease: Integrating environmental influences. Endocrinology, 156(10), 3416–3421. doi:10.1210/ en.2015–1394 Heindel, J.  J., Blumberg, B., Cave, M., Machtinger, R., Mantovani, A., Mendez, M.  A., . . . Vom Saal, F. (2017). Metabolism disrupting chemicals and metabolic disorders. Reproductive Toxicology, 68, 3–33. doi:10.1016/j.reprotox. 2016.10.001 Heindel, J. J., Skalla, L. A., Joubert, B. R., Dilworth, C. H., & Gray, K. A. (2017). Review of developmental origins of health

104

and disease publications in environmental epidemiology. Reproductive Toxicology, 68, 34–48. doi:10.1016/j.reprotox. 2016.11.011 Heindel, J.  J., & Vandenberg, L.  N. (2015). Developmental origins of health and disease: A paradigm for understanding disease etiology and prevention. Current Opinion in Pediatrics, 27(2), 248–253. Heindel, J.  J., vom Saal, F.  S., Blumberg, B., Bovolin, P., Calamandrei, G., Ceresini, G., et al. (2015). Parma consensus statement on metabolic disruptors. Environmental Health, 14, 54. doi:10.1186/s12940-015-0042-7 International Programme on Chemical Safety (IPCS). (2002). Global assessment of the state-of-the-science of endocrine disruptors. In T. Damstra, S. Barlow, A. Bergman, R. J. Kavlock, & G.  Van der Kraak (Eds.), World Health Organization. Geneva, Switzerland. Iqbal, S.  A., Wallach, J.  D., Khoury, M.  J., Schully, S.  D., & Ioannidis, J. P. (2016). Reproducible research practices and transparency across the biomedical literature. PLoS Biology, 14(1), e1002333. doi:10.1371/journal.pbio.1002333 Janesick, A.  S., Dimastrogiovanni, G., Vanek, L., Boulos, C., Chamorro-Garcia, R., Tang, W., & Blumberg, B. (2016). On the utility of ToxCast and ToxPi as methods for identifying new obesogens. Environmental Health Perspectives, 124(8), 1214–1226. doi:10.1289/ehp.1510352 Jones, K.  A. (1996). Summation of basic endocrine data. In  G.  H.  Gass & H.  M.  Kaplan (Eds.), Handbook of endocrinology (2nd ed., Vol. I, pp. 1–42). New York, NY: CRC Press. Kanno, J., Onyon, L., Peddada, S., Ashby, J., Jacob, E., & Owens, W. (2003). The OECD program to validate the rat uterotrophic bioassay. Phase 2: Dose-response studies. Environmental Health Perspectives, 111(12), 1530–1549. Kato, K., Wong, L. Y., Jia, L. T., Kuklenyik, Z., & Calafat, A. M. (2011). Trends in exposure to polyfluoroalkyl chemicals in the U.S.  population: 1999–2008. Environmental Science Technology, 45(19), 8037–8045. doi:10.1021/es1043613 Kavlock, R. J., Daston, G. P., DeRosa, C., Fenner-Crisp, P., Gray, L. E., Kaattari, S., . . . Tilson, H. A. (1996). Research needs for the risk assessment of health and environmental effects of endocrine disruptors: A report of the U.S.  EPA-sponsored workshop. Environmental Health Perspectives, 104(4), 715–740. Koehler, K. E., Voigt, R. C., Thomas, S., Lamb, B., Urban, C., Hassold, T., & Hunt, P.  A. (2003). When disaster strikes: Rethinking caging materials. Lab Animal, 32(4), 24–27. Kohn, M. C., & Melnick, R. L. (2002). Biochemical origins of the non-monotonic receptor-mediated dose-response. Journal of Molecular Endocrinology, 29, 113–123. Kolla, S., Pokharel, A., & Vandenberg, L.  N. (2017). The mouse mammary gland as a sentinel organ: Distinguishing “control” populations with diverse environmental histories. Environmental Health, 16(1), 25. doi:10.1186/s12940-0170229-1 Kortenkamp, A., Bourguignon, J.  P., Slama, R., Bergman, A., Demeneix, B., Ivell, R., . . . Zoeller, R.  T. (2016). EU regulation of endocrine disruptors: A missed opportunity. Lancet Diabetes & Endocrinology, 4(8), 649–650. doi:10.1016/ s2213-8587(16)30151-6 Krimsky, S. (2003). Hormonal chaos: The scientific and social origins of the environmental endocrine hypothesis. Baltimore, MD: Johns Hopkins University Press. Lakind, J. S., & Naiman, D. Q. (2008). Bisphenol A (BPA) daily intakes in the United States: Estimates from the 2003–2004

Endocrine Disruptors and Other Influences on Hormone Action

NHANES urinary BPA data. Journal of Exposure Analysis and Environmental Epidemiology, 18(6), 608–615. Lamb, J. C., Boffetta, P., Foster, W. G., Goodman, J. E., Hentz, K. L., Rhomberg, L. R., . . . Williams, A. L. (2014). Critical comments on the WHO-UNEP State of the Science of  Endocrine Disrupting Chemicals—2012. Regulatory Toxicology & Pharmacology, 69(1), 22–40. doi:10.1016/j. yrtph.2014.02.002 Lamb, J. C., Boffetta, P., Foster, W. G., Goodman, J. E., Hentz, K. L., Rhomberg, L. R., . . . Williams, A. L. (2015). Comments on the opinions published by Bergman et al. (2015) on critical comments on the WHO-UNEP State of the Science of Endocrine Disrupting Chemicals (Lamb et al. 2014). Regulatory Toxicology & Pharmacology, 73(3), 754–757. doi:10.1016/j.yrtph.2015.10.029 Lees, P., Cunningham, F. M., & Elliott, J. (2004). Principles of pharmacodynamics and their applications in veterinary pharmacology. Journal of Veterinary Pharmacology and Therapeutics, 27(6), 397–414. doi:10.1111/j.1365–2885. 2004.00620.x Littlefield, N.  A., Farmer, J.  H., Gaylor, D.  W., & Sheldon, W. G. (1980). Effects of dose and time in a long-term, lowdose carcinogenic study. Journal of Environmental Pathology and Toxicology, 3(3), 17–34. Maccari, S., Piazza, P. V., Kabbaj, M., Barbazanges, A., Simon, H., & Le Moal, M. (1995). Adoption reverses the long-term impairment in glucocorticoid feedback induced by prenatal stress. Journal of Neuroscience, 15(1), 110–116. Macon, M. B., & Fenton, S. E. (2013). Endocrine disruptors and the breast: Early life effects and later life disease. Journal of  Mammary Gland Biology and Neoplasia, 18(1), 43–61. doi:10.1007/s10911-013-9275-7 Markey, C.  M., Luque, E.  H., Munoz De Toro, M., Sonnenschein, C., & Soto, A. M. (2001). In utero exposure to bisphenol A alters the development and tissue organization of the mouse mammary gland. Biology of Reproduction, 65(4), 1215–1223. Markey, C. M., Michaelson, C. L., Veson, E. C., Sonnenschein, C., & Soto, A. M. (2001). The mouse uterotrophic assay: A reevaluation of its validity in assessing the estrogenicity of bisphenol A.  Environmental Health Perspectives, 109(1), 55–60. doi:sc271_5_1835 May, M., Moran, J. F., Kimelberg, H., & Triggle, D. J. (1967). Studies on the noradrenaline alpha-receptor. II. Analysis of the “spare-receptor” hypothesis and estimation of the concentration of alpha-receptors in rabbit aorta. Molecular Pharmacology, 3(1), 28–36. McLachlan, J.  A., Newbold, R.  R., Burow, M.  E., & Li, S.  F. (2001). From malformations to molecular mechanisms in the male: Three decades of research on endocrine disrupters. APMIS, 109(4), 263–272. Melnick, R., Lucier, G., Wolfe, M., Hall, R., Stancel, G., Prins, G., . . . Kohn, M. (2002). Summary of the National Toxicology Program’s report of the endocrine disruptors low-dose peer review. Environmental Health Perspectives, 110(4), 427–431. doi:sc271_5_1835 Mendiola, J., Stahlhut, R. W., Jorgensen, N., Liu, F., & Swan, S. H. (2011). Shorter anogenital distance predicts poorer semen quality in young men in Rochester, New York. Environ­ mental Health Perspectives, 119(7), 958–963. doi:10.1289/­ ehp.­1103421 Mesnage, R., Defarge, N., Rocque, L. M., Spiroux de Vendomois, J., & Seralini, G. E. (2015). Laboratory rodent diets contain

toxic levels of environmental contaminants: Implications for regulatory tests. PLoS One, 10(7), e0128429. Milligan, S.  R., Khan, O., & Nash, M. (1998). Competitive binding of xenobiotic oestrogens to rat alpha-fetoprotein and to sex steroid binding proteins in human and rainbow trout (Oncorhynchus mykiss) plasma. General and Comparative Endocrinology, 112(1), 89–95. Mittendorf, R. (1995). Teratogen update: Carcinogenesis and teratogenesis associated with exposure to diethylstilbestrol (DES) in utero. Teratology, 51, 435–445. Myers, J.  P., Antoniou, M.  N., Blumberg, B., Carroll, L., Colborn, T., Everett, L.  G., . . . Benbrook, C.  M. (2015). Concerns over use of glyphosate-based herbicides and risks associated with exposures: A consensus statement. Environ­ mental Health, 15, 19. Myers, J. P., Zoeller, R. T., & vom Saal, F. S. (2009). A clash of old and new scientific concepts in toxicity, with important implications for public health. Environmental Health Perspectives, 117(11), 1652–1655. Nagel, S.  C., vom Saal, F.  S., & Welshons, W.  V. (1999). Developmental effects of estrogenic chemicals are predicted by an in vitro assay incorporating modification of cell uptake by serum. Journal of Steroid Biochemistry and Molecular Biology, 69(1–6), 343–357. National Academies of Sciences, Engineering, & Medicine. (2017). Application of systematic review methods in an overall strategy for evaluating low-dose toxicity from endocrine active chemicals. Washington, DC: National Academies Press. Nohynek, G. J., Borgert, C. J., Dietrich, D., & Rozman, K. K. (2013). Endocrine disruption: Fact or urban legend? Toxicology Letters, 223(3), 295–305. doi:10.1016/j.toxlet.2013. 10.022 Norman, A.  W., & Henry, H.  L. (2015). Hormones (3rd ed.). Academic Press. Orlando, E. F., & Ellestad, L. E. (2014). Sources, concentrations, and exposure effects of environmental gestagens on fish and other aquatic wildlife, with an emphasis on reproduction. General and Comparative Endocrinology, 203, 241–249. doi:10.1016/j.ygcen.2014.03.038 Rhomberg, L. R., Goodman, J. E., Foster, W. G., Borgert, C. J., & Van Der Kraak, G. (2012). A critique of the European Commission document, “State of the Art Assessment of Endocrine Disrupters.” Critical Reviews in Toxicology, 42(6), 465–473. doi:10.3109/10408444.2012.690367 Ruhlen, R. L., Taylor, J. A., Mao, J., Kirkpatrick, J., Welshons, W. V., & vom Saal, F. S. (2011). Choice of animal feed can alter fetal steroid levels and mask developmental effects of endocrine disrupting chemicals. Journal of Developmental Origins of Health and Disease, 2(1), 36–48. Ryan, B.  C., & Vandenbergh, J.  G. (2002). Intrauterine position effects. Neuroscience Biobehavior Reviews, 26, 665–678. Schug, T. T., Abagyan, R., Blumberg, B., Collins, T. J., Crews, D., DeFur, P. L., . . . Myers, J. P. (2013). Designing endocrine disruption out of the next generation of chemicals. Green Chemistry, 15, 181–198. Schug, T. T., Janesick, A., Blumberg, B., & Heindel, J. J. (2011). Endocrine disrupting chemicals and disease susceptibility. Journal of Steroid Biochemistry and Molecular Biology, 127(3–5), 204–215. doi:10.1016/j.jsbmb.2011.08.007 Sharpe, R.  M., & Skakkebaek, N.  E. (1993). Are oestrogens involved in falling sperm counts and disorders of the male reproductive tract? Lancet, 341(8857), 1392–1395.

Vandenberg

105

Sheehan, D.  M., & Young, M. (1979). Diethylstilbestrol and estradiol binding to serum albumin and pregnancy plasma of rat and human. Endocrinology, 104(5), 1442–1446. Slama, R., Bourguignon, J. P., Demeneix, B., Ivell, R., Panzica, G., Kortenkamp, A., & Zoeller, R. T. (2016). Scientific issues relevant to setting regulatory criteria to identify endocrinedisrupting substances in the European Union. Environmental Health Perspectives, 124(10), 1497–1503. doi:10.1289/ehp217 Solecki, R., Kortenkamp, A., Bergman, A., Chahoud, I., Degen, G.  H., Dietrich, D., . . . Boobis, A.  R. (2016). Scientific principles for the identification of endocrine-disrupting chemicals: A consensus statement. Archives of Toxicology, 91(2), 1001–1006. doi:10.1007/s00204-016-1866-9 Sorge, R.  E., Martin, L.  J., Isbester, K.  A., Sotocinal, S.  G., Rosen, S., Tuttle, A.  H., . . . Mogil, J.  S. (2014). Olfactory exposure to males, including men, causes stress and related analgesia in rodents. Nature Methods, 11(6), 629–632. doi:10.1038/nmeth.2935 Soto, A. M., Schaeberle, C., Maier, M. S., Sonnenschein, C., & Maffini, M.  V. (2017). Evidence of absence: Estrogenicity assessment of a new food-contact coating and the bisphenol used in its synthesis. Environmental Science Technology, 51(3), 1718–1726. doi:10.1021/acs.est.6b04704 Soto, A. M., Vandenberg, L. N., Maffini, M. V., & Sonnenschein, C. (2008). Does breast cancer start in the womb? Basic and Clinical Pharmacology and Toxicology, 102(2), 125–133. Suvorov, A., & Vandenberg, L. N. (2016). To cull or not to cull? Considerations for studies of endocrine-disrupting chemicals. Endocrinology, 157(7), 2586–2594. doi:10.1210/ en.2016–1145 Trasande, L., Zoeller, R.  T., Hass, U., Kortenkamp, A., Grandjean, P., Myers, J.  P.,  .  .  .  Heindel, J.  J. (2015). Estimating burden and disease costs of exposure to endocrine-disrupting chemicals in the European Union. Journal of Clinical Endocrinology and Metabolism, 100(4), 1245–1255. doi:10.1210/jc.2014–4324 Trichopoulos, D. (1990). Is breast cancer initiated in utero? Epidemiology, 1, 95–96. U.S. Food and Drug Administration (FDA). (2010). Endocrine disruptor knowledge base. Retrieved August 20, 2012, from http://www.fda.gov/ScienceResearch/BioinformaticsTools/ EndocrineDisruptorKnowledgebase/default.htm Vandenbergh, J.  G. (2004). Animal models and studies of in  utero endocrine disruptor effects. ILAR Journal, 45, 438–442. Vandenberg, L.  N. (2014). Low-dose effects of hormones and endocrine disruptors. Vitamins and Hormones, 94, 129–165. doi:10.1016/B978-0-12-800095-3.00005–5 Vandenberg, L. N., Agerstrand, M., Beronius, A., Beausoleil, C., Bergman, A., Bero, L. A., . . . Ruden, C. (2016). A proposed framework for the systematic review and integrated assessment (SYRINA) of endocrine disrupting chemicals. Environmental Health, 15(1), 74. doi:10.1186/s12940-0160156-6 Vandenberg, L.  N., & Bowler, A.  G. (2014). Non-monotonic dose responses in EDSP Tier 1 guideline assays. Endocrine Disruptors, 2(1), e964530. Vandenberg, L.  N., Colborn, T., Hayes, T.  B., Heindel, J.  J., Jacobs, D.  R., Jr., Lee, D.  H., . . . Myers, J.  P. (2012). Hormones and endocrine-disrupting chemicals: Low-dose effects and nonmonotonic dose responses. Endocrine Reviews, 33(3), 378–455.

106

Vandenberg, L.  N., Colborn, T., Hayes, T.  B., Heindel, J.  J., Jacobs, D. R., Lee, D. H., . . . Zoeller, R. T. (2013). Regulatory decisions on endocrine disrupting chemicals should be based on the principles of endocrinology. Reproductive Toxicology, 38C, 1–15. Vandenberg, L. N., Hunt, P. A., Myers, J. P., & Vom Saal, F. S. (2013). Human exposures to bisphenol A: Mismatches between data and assumptions. Reviews on Environmental Health, 28(1), 37–58. Vandenberg, L. N., Welshons, W. V., Vom Saal, F. S., Toutain, P. L., & Myers, J. P. (2014). Should oral gavage be abandoned in toxicity testing of endocrine disruptors? Environmental Health, 13(1), 46. Wallen, K. (2009). The organizational hypothesis: Reflections on the 50th anniversary of the publication of Phoenix, Goy, Gerall, and Young (1959). Hormones and Behavior, 55(5), 561–565. doi:10.1016/j.yhbeh.2009.03.009 Watson, C. S., Bulayeva, N. N., Wozniak, A. L., & Alyea, R. A. (2007). Xenoestrogens are potent activators of nongenomic estrogenic responses. Steroids, 72, 124–134. Watson, C.  S., Jeng, Y.  J., & Guptarak, J. (2011). Endocrine disruption via estrogen receptors that participate in nonge­ nomic signaling pathways. Journal of Steroid Biochemistry and Molecular Biology, 127(1–2), 44–50. doi:10.1016/j. jsbmb.2011.01.015 Welshons, W.  V., Thayer, K.  A., Judy, B.  M., Taylor, J.  A., Curran, E. M., & vom Saal, F. S. (2003). Large effects from small exposures: I.  Mechanisms for endocrine-disrupting chemicals with estrogenic activity. Environmental Health Perspectives, 111, 994–1006. Woodruff, T.  J., Zota, A.  R., & Schwartz, J.  M. (2011). Environmental chemicals in pregnant women in the United States: NHANES 2003–2004. Environmental Health Perspectives, 119(6), 878–885. doi:10.1289/ehp.1002727 World Health Organization (WHO). (2013). State of the science of endocrine disrupting chemicals—2012. An assessment of the state of the science of endocrine disruptors prepared by a group of experts for the United Nations Environment Programme (UNEP) and WHO. WHO/UNEP. Ye, X., Zhou, X., Wong, L.  Y., & Calafat, A.  M. (2012). Concentrations of bisphenol A and seven other phenols in pooled sera from 3–11 year old children: 2001–2002 National Health and Nutrition Examination Survey. Environmental Science and Technology, 46(22), 12664–12671. doi:10.1021/ es303109c Zhu, B.  T. (1996). Rational design of receptor partial agonists and possible mechanisms of receptor partial activation: A theory. Journal of Theoretical Biology, 181(3), 273–291. Zoeller, R.  T., Bergman, A., Becher, G., Bjerregaard, P., Bornman, R., Brandt, I., . . . Vandenberg, L.  N. (2014). A path forward in the debate over health impacts of endocrine disrupting chemicals. Environmental Health, 13(1), 118. doi:10.1186/1476-069x-13-118 Zoeller, R.  T., Brown, T.  R., Doan, L.  L., Gore, A.  C., Skakkebaek, N.  E., Soto, A.  M., . . . vom Saal, F.  S. (2012). Endocrine-disrupting chemicals and public health protection: A statement of principles from the Endocrine Society. Endocrinology, 153(9), 4097–4110. Zoeller, R.  T., & Vandenberg, L.  N. (2015). Assessing doseresponse relationships for endocrine disrupting chemicals (EDCs): A focus on non-monotonicity. Environmental Health, 14(1), 42. doi:10.1186/s12940-015-0029-4

Endocrine Disruptors and Other Influences on Hormone Action

PA RT

Reproductive Behavior

 

2

CH A PT E R

7

Investigating the Ovulatory Cycle An Overview of Research and Methods

Lisa L. M. Welling and Robert P. Burriss

Abstract An abundant amount of research into the human ovulatory cycle and related adaptive shifts in preferences and behaviors has been published in recent decades. Evidence suggests that fertility in women is accompanied by increased preferences for male traits that putatively signal underlying genetic quality, adaptive increases in sexual motivation, and changes in attractiveness. Yet, these supposed adaptive shifts remain controversial and disputed, while methods across studies have been inconsistent. In this chapter, we review the research on phenotypic variation across the human menstrual cycle, focusing on the related areas of preferences for male traits, sexual behavior and motivation, and cues to ovulation. Next, we consider the various methods currently used by researchers to ascertain conception risk and review recently published recommendations intended to guide future research and facilitate comparison across studies. Keywords:  menstrual cycle, fertility, conception risk, mate preferences, mating behavior, attractiveness

Since the 1990s, the study of phenotypic variation across the human ovulatory cycle has expanded greatly. Researchers have investigated cyclical effects on women’s sexual desires, preferences, mate choice, intrasexual competitiveness, physical appearance, and partner mate retention behavior, among other constructs (reviewed by Gangestad & Thornhill, 2008; Gangestad et al., 2016). However, as recently noted by Gangestad et al. (2016), the methods used by researchers to assess women’s fertility are inconsistent, which hinders comparison across studies and led critics to question the validity and robustness of findings. In this chapter, we review the evidence for phenotypic variation across the human ovulatory cycle in three interconnected areas: preferences for male traits, sexual behavior and motivation, and cues to ovulation. We then consider the various methods currently used by researchers to ascertain fertility status and review recently published recommendations.

Preferences for Male Traits

The ovulatory shift hypothesis predicts systematic variation in androphilic women’s behavior, specifically behaviors related to mating and mate preferences, over the course of the ovulatory cycle (e.g., Gangestad & Thornhill,  1998; Grammer,  1993; Thornhill & Gangestad, 1999). The likelihood of a woman conceiving after a single act of intercourse (termed her conception risk) varies over her ovulatory cycle, reaching a peak in the days preceding ovulation. As a woman’s conception risk peaks, her preferences for male traits that indicate good genes should likewise increase (Gangestad & Thornhill, 1998). Preferences are not expected to remain constant because men who possess good genes do not necessarily offer other potential benefits, such as resources and investment (Perrett et al.,  1998). In fact, men who putatively possess these good genes may be less attractive as long-term mates because they invest fewer resources in their partners (e.g., Penton-Voak & Perrett, 2001;

109

Perrett et al., 1998). Preference shifts are especially pronounced when women judge men for shortterm, sexual relationships, as opposed to long-term, committed relationships (e.g., Gangestad, Simpson, Cousins, Garver-Apgar, & Christensen, 2004; Little & Jones, 2012; Penton-Voak et al., 1999; Puts, 2005), which is consistent with the hypothesis that these patterns reflect evolved strategies for obtaining good genes for offspring (Gangestad & Thornhill, 1998). Alternative functions, such as securing physical protection (Thornhill & Gangestad, 2008) or increasing the probability of conception by copulating with a fertile mate (Puts,  2006), are also consistent with these findings. Many studies have demonstrated fertile phase increases in women’s preferences for men’s traits, including facial masculinity (V. S. Johnston, Hagel, Franklin, Fink, & Grammer, 2001; Jones, Little, et al., 2005; Little & Jones, 2012; Little, Jones, & DeBruine,  2008; Penton-Voak & Perrett,  2000; Penton-Voak et al., 1999; Roney, Simmons, & Gray, 2011; Welling et al.,  2007; but see Jones et al., 2018; Marcinkowska et al., 2016; Marcinkowska, Galbarczyk, & Jasienska,  2018), facial symmetry (Little & Jones, 2012; Little, Jones, Burt, & Perrett, 2007; but see Cárdenas & Harris,  2007; Koehler, Rhodes, Simmons, & Zebrowitz,  2006; Mar­ cinkowska, Galbarczyk, & Jasienska, 2018), the faces of bodily symmetrical men (Thornhill & Gangestad, 2003), darker facial skin color (Frost, 1994), masculine body shape (Little, Jones, & Burriss, 2007; but see Peters, Simmons, & Rhodes,  2009; Jünger, Kordsmeyer, Gerlach, & Penke, 2018; Marcinkowska, Galbarczyk, & Jasienska, 2018), voices with masculine characteristics (lower fundamental frequency and formant dispersion; Feinberg et al., 2006; Puts, 2005; but see Jünger, Motta-Mena, et al.,  2018), greater height (Pawłowski & Jasienska, 2005), and the odor of men who are more dominant, symmetrical, and heterozygous at major histocompatibility complex (MHC) loci (Gangestad & Thornhill, 1998; Havlíček, Roberts, & Flegr,  2005; Rikowski & Grammer, 1999; Thornhill & Gangestad,  1999; Thornhill et al., 2003). Although the ovulatory shift hypothesis predicts that women should be choosier at peak conception risk, some studies have documented that at periovulation women’s ratings of male attractiveness generally increase, including ratings of the face (Danel & Pawłowski,  2006), body (Jünger, Kordsmeyer, et al., 2018), and voice (Jünger, MottaMena, et al., 2018). Furthermore, women pay more attention to attractive men (Anderson et al., 2010), and their accuracy at classifying faces as male is 110

Investigating the Ovul atory Cycle

greatest at peak fertility (Macrae, Alnwick, Milne, & Schloerscheidt,  2002), particularly when those faces are masculine (L. Johnston, Miles, & Macrae, 2008). That said, the sexual relevance of the target face is important, with lesbian women categorizing female faces more accurately at peak fertility (BrinsmeadStockham, Johnston, Miles, & Macrae, 2008). Nevertheless, the validity of cyclic shifts in women’s preferences, particularly relating to masculinity preferences, has been questioned. A recent metaanalysis by Gildersleeve, Haselton, and Fales (2014a) found evidence for robust cyclic shifts specific to women’s preferences for hypothesized cues of male genetic quality, at least when men were evaluated for a short-term (i.e., purely sexual) relationship. However, the high number of studies included in that meta-analysis that produced null findings (60 percent) led others to argue that the few significant findings are research artifacts (i.e., publication bias; Wood, Kressel, Joshi, & Louie, 2014) or the result of “p-hacking,” whereby researchers adjust the parameters of their analysis to produce a desired result (Harris, Pashler, & Mickes, 2014). To address these concerns, Gildersleeve, Haselton, and Fales (2014b) constructed p curves of significant findings from their original meta-analytic data, finding that, consistent with their original report of robust cyclic shifts, the p curves were right-skewed and had a large number of highly significant p values (p < .01), which is characteristic of a nonzero true effect (Simonsohn, Nelson, & Simmons,  2014). More recently, in the largest study to date (n = 584) of cyclic shifts in women’s preferences for masculine male faces, Jones et al. (2018) used a within-subjects design and found no evidence that preferences for facial masculinity are related to cyclic variation in women’s steroid hormone levels. This suggests that menstrual cycle shifts, at least in terms of masculinity preference, are not as robust as previously supposed, but future work of this scale (i.e., large within-subject designs) should be conducted to assess preferences for other physical traits. Preferences for nonphysical traits, such as dominant and intrasexually competitive behavior (Gangestad, Garver-Apgar, Simpson, & Cousins, 2007; Gangestad, Simpson, Cousins, Garver-Apgar, & Christensen, 2004; Lukaszewski & Roney, 2009), courtship language (Rosen & López,  2009), and creativity/intelligence (Haselton & Miller, 2006; but see Gangestad, Garver-Apgar, Simpson, & Cousins, 2007; Prokosch, Coss, Scheib, & Blozis, 2009), are also higher at peak fertility than at times when ­conception risk is lower. Guéguen (2009a, 2009b)

showed in two field studies that women may be more receptive to courtship behavior at peak fertility. In these two studies, women were more likely to agree to a man’s request to dance or to share phone numbers if they were in the late follicular (fertile) phase of their cycle (the phase directly preceding ovulation), compared with the luteal phase (the phase following ovulation, when fertility is lower) and menses. Other cyclic shifts have also been detected at other points in the cycle. During the luteal phase, preferences increase for traits such as apparent health (Jones, Little, et al., 2005; Jones, Perrett, et al., 2005) and self-resemblance (DeBruine, Jones, & Perrett, 2005; Holzleitner et al., 2017), suggesting a motivation to promote affiliation with healthy and related individuals prior to and during pregnancy. Corre­spondingly, increases in progesterone across the cycle are associated with increased sensitivity to social information (Maner & Miller, 2014) and the salience of emotional displays indicating danger (Conway et al., 2007). However, preferences for putative signals of investment have not been shown to increase during this phase (Gangestad et  al., 2007), suggesting that these traits remain consistently attractive.

Sexual Behavior and Motivation

Various cyclical effects are nullified when women judge men for a long-term rather than a short-term relationship (Gangestad et al., 2007; Haselton & Miller,  2006; V.  S.  Johnston et al.,  2001; Jones, Perrett, et al., 2005; Little, Jones, & Burriss, 2007; Pawłowski & Jasienska,  2005; Penton-Voak et al., 1999; Puts, 2005; Rosen & López, 2009). This may be because women trade off good genes against material benefits by securing as long-term partners men who provision, while engaging men whose traits signify good genes as short-term or extra-pair copulation partners (Gangestad & Simpson, 2000). In support of this hypothesis is evidence that women at high conception risk report greater sexual desire (Roney & Simmons,  2013), are more interested in extra-pair men (Gangestad, Thornhill, & Garver, 2002), are less motivated toward sex for the purposes of intimacy (Sheldon, Cooper, Geary, Hoard, & DeSoto, 2006), are more sexually opportunistic (Gangestad, Thornhill, & Garver-Apgar, 2010), selectively flirt more with men who possess markers of genetic fitness (Cantú et al., 2014), are more likely to visit a singles nightclub without their primary partner (Grammer, Jutte, & Fischmann, 1997), and show greater interest in extra-pair men if their partners are less attractive (Haselton & Gangestad, 2006;

Larson, Pillsworth, & Haselton, 2012; Pillsworth & Haselton, 2006; see also Meltzer, 2017, for findings related to masculinity and marital satisfaction), are less symmetrical (Gangestad, Thornhill, & GarverApgar,  2005), and have MHC alleles that do not complement their own (Garver-Apgar, Gangestad, Thornhill, Miller, & Olp, 2006). Although there is some evidence that women’s interest in their partners (as indicated by pupil dilation) is greatest at ovulation, a similar pattern is observed for responses to attractive opposite-sex celebrities (Laeng & Falkenberg, 2007). Furthermore, sexual interest and number of sexual encounters also peak during the late follicular phase in lesbians (Burleson, Trevathan, & Gregory, 2002), suggesting that cyclical effects on motivation and behavior may not be related to fear of pregnancy, proximity to men, or a reaction to male sexual advances. Women may be more competitive at peak fertility (e.g., Pearson & Schipper, 2013). Financially, women are more likely to discount the future by preferring smaller rewards in the present over larger rewards later (Lucas & Koff, 2017), suggesting increased impulsivity, although they do so less after viewing images of attractive males (Kaighobadi & Stevens, 2013). Women are less cooperative and more likely to punish other women when playing an ultimatum game during peak fertility (Eisenbruch & Roney,  2016; Lucas, Koff, & Skeath,  2007), especially if their opponent is attractive (Lucas & Koff, 2013). Perhaps relatedly, competition with same-sex rivals for access to desirable men also seems to increase with conception risk, with women exhibiting two main strategies for competing with same-sex rivals for male attention: derogation of female competitors and appearance enhancement (Buss & Schmitt, 1996; although it should be noted that men also use these intrasexually competitive strategies, see Fisher & Cox, 2010). Despite rating their own attractiveness more highly (Röder, Brewer, & Fink, 2009; Schwarz & Hassebrauck, 2008), women derogate competitors by judging their facial attractiveness more harshly (Fisher, 2004; see also Piccoli, Foroni, & Carnaghi, 2013; Vukovic et al., 2009; Welling et al., 2007) when maximally fertile. Exposure to this derogation in turn causes men to lower their attractiveness judgments of the derogating females’ rivals (Fisher & Cox, 2009), demonstrating the effectiveness of this tactic. Spending patterns also appear to reflect intrasexual competitiveness: women at peak fertility are more willing to spend money on and choose products that enhance appearance (Durante, Griskevicius, Hill, Perilloux, & Li,  2011; Hill & Durante,  2009), Welling and Burriss

111

a­lthough the total amount of money they spend ­remains the same (Röder et al., 2009). Additionally, women’s aversion to behaviors that may be sexually maladaptive or dangerous is greater at peak fertility, with increases in ratings of disgust toward incest (Fessler & Navarrete, 2003) and decreases in behaviors that may increase the risk of rape (Bröder & Hohmann,  2003; but see Fessler, 2003). Attention from unwanted suitors around peak conception risk could be tremendously detrimental because rape at this time would be more likely to lead to conception (i.e., would be more reproductively costly), and so ancestral women may have evolved “rape avoidance” strategies that they use more often during peak fertility than at other times. Indeed, fewer women are raped near ovulation than would be expected by chance and women engage in fewer behaviors that may put them at risk of sexual assault (e.g., walking alone in a dimly lit area, going on a first date, going out to a bar) when they are near ovulation, despite being more physically active in general (see Chavanne & Gallup,  1998). Women are more assertive during times characterized by high estradiol and low progesterone (i.e., the late follicular phase; Blake, Bastian, O’Dean, & Denson,  2017) and perform better on a measure of physical strength after reading a sexual assault scenario near ovulation (Petralia & Gallup, 2002), which could reflect adaptive shifts designed to thwart an attacker when conception is most likely. Furthermore, women show greater bias against out-group men near ovulation (Navarrete, Fessler, Fleischman, & Geyer, 2009), and this bias may depend on the extent to which women associate out-group men with physical formidability (i.e., they report a greater preference for familiar men when fertility is high, particularly if they perceive unfamiliar men as being more physically intimidating; McDonald, Asher, Kerr, & Navarrete, 2011). Correspondingly, female gait has been shown to be more attractive during the luteal phase than during the follicular phase (Provost, Quinsey, & Troje, 2008), which the authors argue may serve to prevent fertility from being widely advertised, as changes in gait are likely to be easier to detect at a distance than are changes in face shape or odor. If gait were to be more attractive at peak fertility, male attention may be drawn indiscriminately, which may bring unwanted male attention. These inhibitory effects on motivation may help to ensure that women less frequently perform behaviors that can lead to disadvantageous pairings or ill-timed/unwanted pregnancies. 112

Investigating the Ovul atory Cycle

Cues to Ovulation

The long-held assumption that human females have lost estrus has been challenged by suggestions that there may be physical and behavioral cues to a woman’s cycle phase (for reviews, see Haselton & Gildersleeve, 2011; Thornhill & Gangestad, 2008). At peak fertility, women are more attractive in terms of facial appearance (Puts et al., 2013; Roberts et al., 2004), odor (Gildersleeve, Haselton, Larson, & Pillsworth, 2012; Havlíček, Dvořáková, Bartoš, & Flegr, 2006; Kuukasjärvi et al., 2004; S. L. Miller & Maner, 2011; Singh & Bronstad, 2001), and voice (Bryant & Haselton, 2009; Pipitone & Gallup, 2008; but see Fischer et al.,  2011). In fact, the scent of an  ovulating woman has a direct endocrinological influence on men: men who smell T-shirts worn by women near ovulation, but not worn by women at other cycle phases or T-shirts not worn by anyone, experience an increase in their circulating testosterone levels (S. L. Miller & Maner, 2010), which is a response associated with sexual arousal (e.g., Stoleru, Ennaji, Cournot, & Spira, 1993). Women also have greater breast symmetry (Manning, Scutt, Whitehouse, Leinster, & Walton, 1996; Scutt & Manning,  1996), are perceived as more attractive by their romantic partners (Cobey, Buunk, Pollet, Kippling, & Roberts,  2013), and modulate their appearance and clothing to enhance attractiveness (Durante, Li, & Haselton, 2008; Haselton, Mortezaie, Pillsworth, Bleske-Rechek, & Frederick, 2007; Hill & Durante, 2009; Röder, Brewer, & Fink, 2009; Schwarz & Hassebrauck,  2008; see also Meltzer, McNulty, Miller, & Baker,  2015, for findings related to women’s weight loss goals) when they are most fertile. G. Miller, Tyber, and Jordan (2007) investigated earnings through tips among female lap dancers as a function of menstrual cycle and contraceptive status. They found that lap dancers receive larger tips at peak fertility compared to when they were in nonfertile phases (menstruation and the luteal phase) of their cycle, whereas contraceptive pill users showed no such variation. It is not clear which trait(s) drive this pattern. One possibility is that men are responding to dynamic cues, such as change in a woman’s scent (Thornhill et al.,  2003), flirtation (Cantú et al., 2014), or increased sexual opportunism (Gangestad et al., 2010), which are known to vary over the cycle (discussed earlier). Further support for the existence of perceivable cues to ovulation comes from studies of the effects of a woman’s cycle phase on other persons. If selection has acted on women to promote late-follicular-phase

copulations with high-quality and compatible men, one may also expect it to have acted on their longterm partners to decrease the likelihood of infidelity and/or poaching. Three studies have suggested that women at peak conception risk perceive their partners to be more proprietary and attentive (Gangestad et al., 2002; Haselton & Gangestad, 2006; Pillsworth & Haselton, 2006), and one has shown that men’s ratings of the dominant appearance of other men increase when their partners are fertile (Burriss & Little,  2006). Moreover, men whose proprietary behaviors increase most during their partners’ fertile phase have partners whose interest in extra-pair copulation increases most during peak fertility (Gangestad et al., 2002; Haselton & Gangestad, 2006). This suggests that women’s attention to extra-pair men during estrus (and perhaps their tendency toward ornamentation, see Haselton et al., 2007) may drive their partner’s increased attention to them. In sum, the variety and volume of studies investigating human sexual behavior as a function of cycle status are substantial, and the importance of collecting accurate, comparable ovulatory cycle information is becoming increasingly evident (Ganges­tad et al., 2016). Next, we review the various methods of ascertaining conception risk, including recently published recommendations for future research (Gangestad et al., 2016; Gonzales & Ferrer, 2016; Jones et al., 2018).

Ascertaining Conception Risk

Studies of ovulatory effects differ substantially in their methods, which may explain some of the disparities between studies’ results (e.g., Little, Jones, Burt, & Perrett, 2007, vs. Cárdenas & Harris, 2007). Similar results might suggest that methodological differences between studies are not important, but differences impede comparison of results and some methods may be more reliable than others (Gangestad et al., 2016). At a minimum, studies investigating ovulatory effects must measure an aspect of the phenotype of a woman (or man, e.g., the woman’s primary partner) and relate it to estimates of the woman’s conception risk or hormonal profile. Estimating conception risk requires determining a woman’s position in her cycle (usually the day of her cycle when she participates in a study, with the day of the onset of menstrual bleeding classed as day one, or in relation to the number of days before/after ovulation), or assessing her hormonal profile (by assaying samples of saliva, blood, or urine) and assigning conception risk values or categorical phases according to actuarial data.

Researchers might then determine whether the ­dependent variables of interest vary as a function of cycle phase (e.g., Penton-Voak et al., 1999) or assess the correlation between these dependent variables and assigned conception risk probabilities (e.g., Gangestad & Thornhill, 1998), or both (e.g., Haselton & Miller, 2006). However, although the various methods used to produce estimates of conception risk may differ considerably in accuracy, to suggest that one method is inherently “better” than any other is to overlook the fact that not all are applicable to every situation. For example, it would not be practical to draw blood for hormonal assay or use transvaginal ultrasonography to check follicle growth at every testing session for diary-, field-, or internet-based studies. Thus, our discussion focuses on practical techniques for maximizing accuracy in conception risk estimation for a variety of methods. We begin by outlining the less commonly used methods of transvaginal ultrasonography, basal body temperature, and cervical mucus qualities, and then review the more popular self-report and hormonal assay methods. Lastly, we discuss conception risk values, which can be used in conjunction with other techniques (e.g., self-report).

Transvaginal Ultrasonography, Basal Body Temperature, and Cervical Mucus Qualities

The most accurate method of identifying the timing of ovulation is daily transvaginal ultrasonography. This method involves the insertion of an ultrasound probe into the vaginal canal to examine the ovaries and other reproductive organs. Transvaginal ultrasonography can be used to precisely track follicular growth and release, and to correlate phenotypic changes to these underlying biological events (Cobey et al., 2012, 2013). Although accurate, this method is invasive and requires medical training and expensive equipment. Therefore, transvaginal ultrasonography is unlikely to be considered appropriate for all researchers when the focus is on cyclic shifts in mate preference or behavior. A simpler method of estimating the timing of ovulation involves tracking a woman’s basal body temperature (BBT), which exhibits a periovulatory increase and remains elevated throughout the luteal phase (Marshall, 1968). The day of ovulation can be approximated using the “three over six rule,” meaning that ovulation typically occurs on the first day of the first time in the cycle when BBT is higher for three consecutive days than the mean of the preceding six days (Marshall, 1968). Exceptions to this rule Welling and Burriss

113

may be permitted if there is a spike in temperature on only a single day among six lower temperature days, or if a rise in BBT can be accounted for by illness or other disturbances (Colombo & Masarotto, 2000). Participants may be given a chart on which to record their daily temperature readings (and menstruation days), or a commercially available temperature computer may be used (e.g., the OvuTherm “Sophia,” Craig Medical Distribution). Temperature computer devices measure and record daily BBT readings and calculate the probable fertile days of the cycle (Freundl et al., 2003). The consistency of cervical mucus can also indicate conception risk. Cervical mucus is typically thin, clear, and ductile during the follicular phase. After ovulation, the mucus becomes more viscous and opaque (Colombo & Masarotto, 2000). In addition, follicular phase mucus crystallizes in a “ferning” pattern, which can be seen under a microscope when a drop of mucus is placed on a glass slide and allowed to dry (Freundl et al.,  2003). Mucus collected from the vulva can be palpated and examined with the naked eye for opacity and consistency. Because salivary mucus shows similar cyclic changes in ferning patterns, commercial kits may allow for examination of either cervical or salivary mucus (Freundl et al.,  2003). Ovulation generally occurs on the last day of the cycle during which cervical mucus is thin, clear, and ductile and exhibits a ferning pattern (Colombo & Masarotto, 2000). To compare the accuracy of various approaches for detecting ovulation, Freundl et al. (2003) tested eight different methods of cycle monitoring, each on a sample of 14 to 16 women. The methods included hormonal measures, BBT, mucus characteristics, or a combination of these methods. The true day of ovulation was determined by daily urinary luteinizing hormone (LH) measurements and transvaginal ultrasonography. Temperature computer devices exhibited the lowest false-negative rate but high rates of false positives, whereas the reverse was true of examining mucus ferning patterns under a ­microscope. A hormone computer device based on LH and an estrogen metabolite showed intermediate false-negative and false-positive rates. The symptothermal method of natural family planning, in which women chart both daily body temperature and mucus characteristics, produced the greatest overall predictive accuracy. Thus, without transvaginal ultrasonography, the most reliable approach is probably to use two methods: one with a low rate of false positives (mucus characteristics), the other with a low rate of false negatives (BBT). However, hormonal 114

Investigating the Ovul atory Cycle

methods offer good overall accuracy (with neither high rates of false positives nor high rates of false ­negatives), are less invasive, and rely less on (potentially inaccurate) reports by participants. Importantly, although the more valid methods of estimating ­ cycle day involve monitoring physiological variables that change cyclically, including hormone levels, BBT, and the appearance or consistency of salivary and cervical mucus, these methods are often impractical when predicting ovulatory status for scheduling testing sessions. For these purposes, researchers often rely on self-report data.

Self-Report

Location in the cycle can be quickly estimated via self-report, although women are often inaccurate when asked to estimate their cycle length (Jukic et al., 2008; Small, Manatunga, & Marcus, 2007) or last menstrual onset (Waller, Spears, Gu, & Cunningham, 2000; Wegienka & Baird, 2005). Self-report questions often target the date of the onset of menstrual bleeding for the previous or current menses. Providing a calendar or asking participants in advance to track their menstruation using one of the several available mobile phone applications may help participants report more accurately their cycle onset date, which can then be used to establish the cycle day on which the task is administered. The date of the onset of menses is usually classed as day one (i.e., a woman whose period began 10 days ago would currently be at day 11). Conception risk values or categorical phases can then be assigned according to cycle day. This method, usually called the “forward counting” method, is the easiest to use because it requires the participant to provide only one date, usually at the time the task is administered. No further contact with the participant is required and so it is frequently employed in between-subject internet-based studies or when using a between-subject design in general (e.g., Little, Jones, & Burriss, 2007; Little, Jones, & DeBruine, 2008). The “backward counting” method (sometimes called “reverse cycle day” or “backward” method; see Gangestad et al., 2016), which predicts the day of ovulation by counting back from the onset of the menses phase subsequent to testing, is considered more reliable because the follicular phase tends to be more variable in duration than the luteal phase (Baird et al., 1995). The cycle day of interest in relation to ovulation is calculated by assuming that ovulation occurs 14 (Beckmann et al., 1998; Knaus, 1929; Ogino, 1930) or 15 (Dixon, Schlesselman, Ory, & Blye, 1980) days prior to the first day of the next menstruation. Trussell, Rodriguez, and Ellertson

(1998) suggest that subtracting 13 days from a woman’s usual cycle length (i.e., 14 days before the next menses begins) provides a less biased estimate. Women vary in the duration of their cycles, so a further optional stage in this process is to transform women’s estimated cycle days into their expected equivalents in a 28-day cycle, such that the cycle begins on day 1 and ends on day 28, and the actual or estimated day on which ovulation occurs is day 15 (Burriss et al., 2015; Puts, 2006). There are two general methods of obtaining data that permit the backward-counting method. The first method involves establishing probable cycle duration (by soliciting reports of average ­duration or having participants specify the dates for more than one previous menstrual onset and calculating the interval). It is then assumed that the next menses will begin one expected cycle length after the previous known date of onset. This method is sometimes referred to as the “forward-backward counting” method and days can be classed relative to estimated ovulation (i.e., number of days before/ after). The second method involves having participants report back to the researcher the date of onset of their first menses after testing has ended. This method has the advantage of greater accuracy and of providing data specific to the cycle during which the participant has been tested. However, this method cannot be used to schedule testing during the cycle of interest, and some participants may fail to provide further information after testing has finished, so it is prudent to use both methods in conjunction (e.g., Fales, Gildersleeve, & Haselton, 2014). Self-report methods have been employed successfully in many studies but are not precise. Ovula­ tion does not follow every menstruation (Wilcox, Weinberg, & Baird, 1995), and the opposite is also true (Check et al., 1989). Moreover, there is evidence that the duration of the luteal phase is not fixed and, as with the duration of the menses and follicular phases, can be highly variable (Stern & McClintock, 1998). Wilcox, Dunson, and Baird (2000) found that only 10 percent of women with 28-day cycles ovulated precisely 14 days before menstruation. The accuracy of self-report methods can be improved by omitting women who report irregular cycles or abnormal cycle lengths. Cutoff points for abnormal cycle length vary across studies, with some authors omitting participants with cycle lengths in excess of 28 days (Guéguen, 2009a, 2009b), in excess of 40 days (DeBruine et al.,  2005; Garver-Apgar, Gangestad, & Thornhill, 2008), in excess of 50 days

(Gangestad, Simpson, Cousins, Garver-Apgar, & Christensen, 2004), or fewer than 21 days (Bryant & Haselton,  2009; Frost,  1994; Gangestad & Thornhill, 1998), or omitting those who are more than two standard deviations from a mean of 29 days (Puts, 2005). Although counting methods, and self-report methods in general, are less accurate than other methods (e.g., transvaginal ultrasonography; Cobey et al., 2012,  2013), their relative ­convenience for both researchers and participants makes their use popular. To address which counting techniques are most accurate, Gangestad et al. (2016) created more than 58,000 simulated cycles using published distributions of the lengths of follicular and luteal phases. Using these data, they assessed the validity of various counting methods and provided suggestions for fu­ture research. Their analyses showed modest validity (d range: 0.40 to 0.55) for most counting meth­ods, and that backward-counting with a confirmed next menstrual onset (d = .70) and daily LH testing (d = .85) is most accurate. These authors recommended the use of within-subject designs to ensure sufficient power for analyses, as betweensubject designs require substantially larger sample sizes for equivalent power (see Gangestad et  al., 2016, for power analyses). Moreover, researchers should use the backward-counting method with confirmed next menstrual onset and kits to detect the LH surge and/or assay reproductive hormones (Gangestad et al., 2016). In line with these suggestions, Jones et al. (2018) outlined four common issues with cycle work: insufficient power, overuse of biased self-report methods, overuse of betweensubject designs, and testing hormones only twice in the cycle (at estimated high and low fertility phases). Taking the aforementioned suggestions and addressing these criticisms of menstrual cycle work may yield important insights into the relationships among conception risk, hormones, and women’s behavior.

Hormonal Assessment

To measure hormone levels across the cycle, ­researchers can have participants visit the laboratory multiple times to provide several samples (i.e., saliva, blood, or urine) for hormonal assay across one or more menstrual cycles (e.g., Jones et al., 2018), or they can use counting methods to estimate the date of ovulation and test participants during the fertile window and other points in the menstrual cycle when fertility is likely to be lower (e.g., Haselton, Mortezaie, Pillsworth, Bleske-Rechek, & Frederick, Welling and Burriss

115

2007). Ovulation test kits provide a relatively inexpensive and effective means of monitoring cycling hormones. Most over-the-counter ovulation test kits respond to levels of LH in urine or to both LH and metabolites of estradiol, both of which peak at midcycle. LH levels spike about one or two days before ovulation (Garcia, Jones, & Wright,  1981; World Health Organization,  1980). Urinary tests for the LH surge agree with results from ultrasonography 97 to 100 percent of the time, making this one of the most accurate tests available (Guermandi et al.,  2001). When researchers use ovulation test kits, it is typically in conjunction with counting methods; participants are generally scheduled for study sessions according to self-report data and then take LH tests on several consecutive days around the time the surge is expected (e.g., Burriss et al., 2015; Haselton et al., 2007). Once researchers have confirmed an LH surge, they can be reasonably confident that their participant is within the fertile window of their cycle. A further advantage of this method is that the results of LH tests are available immediately, which makes it possible for researchers to reactively schedule testing sessions. The preovulation rise in LH is also highly correlated with an increase in the ratio of estrogen to  progesterone levels (Baird, Weinberg, Wilcox, McConnaughey, & Musey, 1991; Baird, Weinberg, Wilcox, McConnaughey, Musey, & Collins, 1991). Consequently, the ratio of the levels of these hormones can also provide a fairly accurate measure of a woman’s proximity to ovulation (Wilcox et al., 1995), even without the use of ovulation test kits to confirm ovulation. Therefore, the experimenter can collect biological samples from participants, usually saliva (e.g., Welling et al.,  2007) or urine (e.g., Venners et al., 2006), at the time(s) of their participation and then store the samples in a noncycling freezer before conducting hormonal assays. In studies of cyclical phenotypic variation in which users and nonusers of hormonal contraceptives have been separately tested, cyclical effects present in nonusers have consistently not been present in users (e.g., Burriss & Little, 2006; Frost, 1994; Gangestad & Thornhill,  1998; Guéguen,  2009b; Laeng & Falkenberg, 2007; Little et al., 2007; Penton-Voak et al., 1999; Puts, 2005). This suggests that these effects are controlled by fluctuations in hormone levels and are not connected to changes due to menstruation independent of ovulation (Gangestad & Thornhill, 1998). In other words, the underlying hormonal mechanisms, and not fertility per se, are likely responsible for changes in behavior associated 116

Investigating the Ovul atory Cycle

with the menstrual cycle. Menstrual cycle effects may be driven by estradiol (Feinberg et al., 2006; GarverApgar et al., 2008; Roney & Simmons, 2008; Roney et al.,  2011; Rosen & López,  2009; Rupp et al., 2009), progesterone (Garver-Apgar et al.,  2008; Jones, Perrett, et al., 2005; Puts, 2006; Rupp et al., 2009), prolactin (Puts, 2006), cortisol (López, Hay, & Conklin, 2009), testosterone (Welling et al., 2007), or a combination of these hormones (e.g., Puts et al., 2013). Also, different hormonal fluctuations can be associated with different phenotypic changes, and it is unlikely that one hormone is singularly responsible for all noted cyclic shifts. Therefore, measuring change in multiple hormones over time— not necessarily in conjunction with high versus low fertility (e.g., across the day)—can yield important information about underlying mechanisms.

Conception Risk Values

A woman’s daily conception risk is an estimate of the average likelihood of her conceiving with a single unprotected copulation. After estimating the day of the cycle on which a participant has been tested, a researcher may determine that the day is inside or outside of the cycle’s fertile window or assign the day a conception risk value taken from one of several studies. Daily conception risk values may be reported by cycle day relative to the onset of the previous menses (e.g., Jochle,  1973; Wilcox, Dunson, Weinberg, Trussell, & Baird, 2001) or relative to the day of ovulation (e.g., Wilcox et  al.,  1995). Conception risk is strongly determined by the timing of copulation in relation to ovulation (Wilcox et al., 1995), and the number of days since the onset of the previous menstruation may be a poor estimate of a woman’s proximity to ovulation. For example, the 14th day after the onset of menstruation may be very near ovulation for a woman with a 28-day cycle, but probably several days away from ovulation for a woman with a 32-day cycle. Thus, greater precision in estimating the fertile window facilitates detection of significant relationships between estimated conception risk and preferences, behaviors, or attitudes. However, this is not to say that a narrower fertile window (e.g., six days) is necessarily preferable to a broader window (e.g., nine days): When a woman’s actual date of ovulation is unknown, it is advisable to assign a broader fertile window so as to increase the likelihood of including genuinely high fertility days (Gildersleeve, Haselton, & Fales, 2014b). Gangestad et al. (2016) assessed the optimal window size and found that broader windows (eight to nine

days) are more likely to be valid, except when the backward-counting method is used and the date of ovulation is known (in which case the chances of classifying a fertile day as a nonfertile day are reduced). Nevertheless, methods that rely on ­continuous measures of conception risk rather than discrete cycle windows are consistently more valid (Gangestad et al., 2016). Estimates of daily conception risk values relative to the day of ovulation are available from several sources (Barrett & Marshall, 1969; Bremme, 1991; Colombo & Masarotto, 2000; Schwartz, MacDonald, & Heuchel, 1980; Schwartz, Mayaux, Martin-Boyce, Czyglik, & David,  1979; Stirnemann, Samson, Bernard, & Thalabard,  2013; Weinberg, Wilcox, Baird, & Gladen, 1998; Wilcox et al., 1995, 2001; Wilcox, Weinberg, & Baird, 1998), although some of these studies report reanalyses of the same data set. The appropriate values for use in menstrual effect studies will depend on the quality of the methods and the sample size used in each conception risk study. The reliability of conception risk values in these studies depends on both the precision in detecting ovulation and the proper attribution of pregnancies to coital acts on the days on which conception occurred. As Colombo and Masarotto (2000) point out, attributing pregnancies to the coital day closest to the presumed day of ovulation (e.g., Bremme, 1991) produces a bias that artificially increases conception risks nearer the day of ovulation. The model of Schwartz et al. (1980) produces the best estimate of the day of conception for cycles in which more than one coital act occurred (Colombo & Masarotto, 2000). Other researchers have used the Schwartz et al. (1980) model and have sufficiently reliable methods for detecting ovulation—using BBT (Schwartz et al., 1980), BBT plus mucus ferning (Colombo & Masarotto,  2000), or the ratio of ­estrogen to progesterone levels (e.g., Wilcox et al., 1998)—that we may consider their daily conception risk estimates reliable. Wilcox et al. (2001) provide conception risk values for each day of the cycle (i.e., the probability of conception following unprotected sex for each day of the cycle) using data from 696 cycles (213 women), and these values are commonly used by researchers (e.g., Roney & Simmons, 2008). More recently, Stirnemann et al. (2013) calculated the probability of being within a five-day fertile window for each day of the cycle based on ultrasound measurements from 5,830 early pregnancies and determined that day 12 (using the forward-counting method) had the highest conception probability. Although Wilcox et al. (2001) and

Stirnemann et al. (2013) defined conception risk differently (i.e., probability of conception on a given day vs. probability of being in the five-day fertile window), validities of estimates using either method are roughly equivalent (Gangestad et al., 2016).

Conclusion

The method researchers use depends largely on their available resources and on their quasi-independent variable of interest. If conception risk is the variable of interest, it makes sense to exclude participants when there is evidence that their cycles are anovulatory. However, if a researcher is interested in underlying hormonal mechanisms, then anovulation matters less because there are still (albeit smaller) hormone fluctuations across anovulatory cycles (e.g., Ellison, Lager, & Caffee, 1987). If a researcher is interested in conception probability or the fertile window in general, it is preferable to estimate the day of a woman’s cycle on which she is tested relative to the day on which she ovulates rather than relative to the day on which her previous menses began. The day of ovulation can most accurately be predicted via ovulation test kits or a combination of BBT and mucus methods, with Gangestad et al. (2016) recommending the backward-counting method with participants’ confirmation of next menstrual onset and use of LH test strips to confirm ovulation. Because underlying hormonal mechanisms, rather than fertility per se, likely drive menstrual shifts in phenotype (e.g., Gangestad et al., 2016; Roney & Simmons, 2013), multiple hormonal assays using within-subject designs with sufficiently large samples are preferred (Gangestad et al., 2016; Jones et al., 2018). Furthermore, researchers relying on counting methods are recommended to exclude participants with irregular cycles or abnormal cycle lengths at the recruitment stage (e.g., using a prescreening questionnaire), rather than after data are collected, to avoid the appearance of p-hacking. Finally, none of the suggestions outlined in this chapter necessarily invalidate findings of previous studies whose methods differ from those ­recommended here. However, the suggestions do underline the need for replication of underpowered studies and those studies using less than ideal methods, as well as for preregistering methods to limit the opportunity for post hoc flexibility (Harris et al., 2014). Our recommendations, like those outlined by Gangestad et al. (2016) and Jones et al. (2018), are intended to serve as guidelines for future research on menstrual effects and on the evolution of human behavior in general. Welling and Burriss

117

As reviewed in this chapter, many menstrual cycle effects have been documented in the literature. These shifts in behavior, preferences, and appearance serve putatively adaptive functions that likely increase the chances of reproducing, especially with optimal partners. Menstrual cycle effects illustrate how underlying mechanisms can inform ultimate explanations for human behavior. Despite recent controversies (e.g., Gildersleeve et al., 2014a, 2014b; Harris et al., 2014; Wood et al.,  2014) and inconsistencies (see Jones et al.,  2018), this line of research remains a fruitful area for further investigation.

References

Anderson, U. S., Perea, E. F., Becker, D. V., Ackerman, J. M., Shapiro, J, R., Neuberg, S. L., & Kenrick, D. T. (2010). I only have eyes for you: Ovulation redirects attention (but not memory) to attractive men. Journal of Experimental Social Psychology, 46(5), 804–808. doi:10.1016/j.jesp.2010.04.015 Baird, D. D., McConnaughey, R., Weinberg, C. R., Musey, P. I., Collins, D. C., Kesner, J. S., . . . Wilcox, A. J. (1995). Application of a method for estimating day of ovulation using urinary estrogen and progesterone metabolites. Epidemiology, 6(5), 547–550. Baird, D. D., Weinberg, C. R., Wilcox, A. J., McConnaughey, D.  R., & Musey, P.  I. (1991). Using the ratio of urinary oestrogen and progesterone metabolites to estimate day of ovulation. Statistics in Medicine, 10(2), 255–266. Baird, D. D., Weinberg, C. R., Wilcox, A. J., McConnaughey, D. R., Musey, P. I., & Collins, D. C. (1991). Hormonal profiles of natural conception cycles ending in early, unrecognized pregnancy loss. Journal of Clinical Endocrinology and Metabolism, 72(4), 793–800. doi:10.1210/jcem-72-4-793 Barrett, J. C., & Marshall, J. (1969). The risk of conception on different days of the menstrual cycle. Population Studies, 23(3), 455–461. doi:10.1080/00324728.1969.10405297 Beckmann, C.  R.  B., Ling, F.  W., Hebert, W.  N.  P., Laube, D. W., Smith, R. P., & Barzansky, B. M. (1998). Obstetrics and gynecology. Baltimore, MD: Williams and Wilkins. Blake, K. R., Bastian, B., O’Dean, S. M., & Denson, T. F. (2017). High estradiol and low progesterone are associated with high assertiveness in women. Psychoneuroendocrinology, 75, 91–99. doi:10.1016/j.psyneuen.2016.10.008 Bremme, J. (1991). Sexual behavior and probability of conception (analysis of a prospective study of natural family planning). Heinrich Heine University of Düsseldorf, Düsseldorf. Brinsmead-Stockham, K., Johnston, L., Miles, L., & Macrae, C.  N. (2008). Female sexual orientation and menstrual influences on person perception. Journal of Experimental Social Psychology, 44(3), 729–734. doi:10.1016/j.jesp.2007.05.003 Bröder, A., & Hohmann, N. (2003). Variations in risk taking behavior over the menstrual cycle: An improved replication. Evolution and Human Behavior, 24, 397–398. doi:10.1016/ S1090-5138(03)00055-2 Bryant, G. A., & Haselton, M. G. (2009). Vocal cues of ovulation in human females. Biology Letters, 5(1), 12–15. doi:10.1098/ rsbl.2008.0507 Burleson, M. H., Trevathan, W. R., & Gregory, W. L. (2002). Sexual behavior in lesbian and heterosexual women:

118

Investigating the Ovul atory Cycle

Relations with menstrual cycle phase and partner availability. Psychoneuroendocrinology, 27(4), 489–503. doi:10.1016/ S0306-4530(01)00066-X Burriss, R. P., & Little, A. C. (2006). Effects of partner conception risk phase on male perception of dominance in faces. Evolution and Human Behavior, 27(4), 297–305. doi:10.1016/ j.evolhumbehav.2006.01.002 Burriss, R.  P., Troscianko, J., Lovell, P.  G., Fulford, A.  J.  C., Stevens, M., Quigley, R., . . . Rowland, H. M. (2015). Changes in women’s facial skin color over the ovulatory cycle are not detectable by the human visual system. PLoS One. 10(7), e0130093. doi:10.1371/journal.pone.0130093 Buss, D. M., & Schmitt, D. P. (1996). Strategic self-promotion and competition derogation: Sex and conflict effects on perceived effectiveness of mate attraction tactics. Journal of Personality and Social Psychology, 70, 1185–1204. Cantú, S. M., Simpson, J. A., Griskevicius, V., Weisberg, Y. J., Durante, K. M., & Beal, D. J. (2014). Fertile and selectively flirty: Women’s behavior toward men changes across the ovulatory cycle. Psychological Science, 25(2), 431–438. doi:10.1177/0956797613508413 Cárdenas, R. A., & Harris, L. J. (2007). Do women’s preferences for symmetry change across the menstrual cycle? Evolution and Human Behavior, 28(2), 96–105. doi:10.1016/j.evolhumbehav. 2006.08.003 Chavanne, T. J., & Gallup, G. G. (1998). Variation in risk taking behavior among female college students as a function of the  menstrual cycle. Evolution and Human Behavior, 19(1), 27–32. doi:10.1016/S1090-5138(98)00016-6 Check, J. H., Shanis, B. S., Stanley, C., Chase, J. S., Nazari, A., & Wu, C. H. (1989). Amenorrhea in an ovulatory woman despite a normal uterine cavity: Case report. American Journal of Obstetrics and Gynecology, 160(3), 598–599. doi:10.1016/ S0002-9378(89)80036-5 Cobey, K.  D., Buunk, A.  P., Pollet, T.  V., Kippling, C., & Roberts, S. C. (2013). Men perceive their female partners, and themselves, as more attractive around ovulation. Biological Psy­ chology, 94(3), 513–516. doi:10.1016/j.biopsycho.2013.09.011 Cobey, K. D., Buunk, A. P., Roberts, S. C., Klipping, C., Appels, N., Zimmerman, Y., . . . Pollet, T. V. (2012). Reported jealousy differs as a function of menstrual cycle stage and contraceptive pill use: A within-subjects investigation. Evolution and Human Behavior, 33(4), 395–401. doi:10.1016/j.evolhumbehav. 2011.12.001 Colombo, B., & Masarotto, G. (2000). Daily fecundability: First results from a new data base. Demographic Research, 3(5). doi:10.4054/DemRes.2000.3.5 Conway, C. A., Jones, B. C., DeBruine, L. M., Welling, L. L. M., Law Smith, M. J., Perrett, D. I., . . . Al-Dujaili, E. A. S. (2007). Salience of emotional displays of danger and contagion in faces is enhanced when progesterone levels are raised. Hormones and Behavior, 51(2), 202–206. doi:10.1016/j.yhbeh.2006.10.002 Danel, D., & Pawłowski, B. (2006). Attractiveness of men’s faces in relation to women’s phase of menstrual cycle. Collegium Antropologicum, 30(2), 315–319. DeBruine, L. M., Jones, B. C., & Perrett, D. I. (2005). Women’s attractiveness judgments of self-resembling faces change across the menstrual cycle. Hormones and Behavior, 47(4), 379–383. doi:10.1016/j.yhbeh.2004.11.006 Dixon, G. W., Schlesselman, J., Ory, H. W., & Blye, R. P. (1980). Ethinyl estradiol and conjugated estrogens as postcoital contraceptives. Journal of the American Medical Association, 244, 1336–1339.

Durante, K. M., Griskevicius, V., Hill, S. E., Perilloux, C., & Li, N.  P. (2011). Ovulation, female competition, and product choice: Hormonal influences on consumer behavior. Journal of Consumer Research, 37(6), 921–934. doi:10.1086/656575 Durante, K. M., Li, N. P., & Haselton, M. G. (2008). Changes in women’s choice of dress across the ovulatory cycle: Naturalistic and laboratory task-based evidence. Personality and Social Psychology Bulletin, 34(11), 1451–1460. doi:10.1177/ 0146167208323103 Eisenbruch, A. B., & Roney, J. R. (2016). Conception risk and the Ultimatum Game: When fertility is high, women demand more. Personality and Individual Differences, 98, 272–274. 10.1016/j.paid.2016.04.047 Ellison, P. T., Lager, C., & Caffee, J. (1987). Low profiles of salivary progesterone in college undergraduate women. Journal of Adolescent Health Care, 8, 204–207. Fales, M.  R., Gildersleeve, K.  A., & Haselton, M.  G. (2014). Exposure to perceived male rivals increases testosterone on fertile days relative to nonfertile days of their partner’s ovulatory cycle. Hormones and Behavior, 65(5), 454–460. doi:10.1016/j. yhbeh.2014.04.002 Feinberg, D. R., Jones, B. C., Law Smith, M. J., Moore, F. R., DeBruine, L. M., Cornwell, R. E., . . . Perrett, D. I. . (2006). Menstrual cycle, trait estrogen level, and masculinity preferences in the human voice. Hormones and Behavior, 49, 215–222. doi:10.1016/j.yhbeh.2005.07.004 Fessler, D.  M.  T. (2003). Rape is not less frequent during the ovulatory phase of the menstrual cycle. Sexualities, Evolution & Gender, 5(3), 127–147. doi: 10.1080/14616660410001662361 Fessler, D. M. T., & Navarrete, C. D. (2003). Domain-specific variation in disgust sensitivity across the menstrual cycle. Evolution and Human Behavior, 24(6), 406–417. doi: 10.1016/ S1090-5138(03)00054-0 Fischer, J., Semple, S., Fickenscher, G., Jürgens, R., Kruse, E., Heistermann, M., & Ofer, A. (2011). Do women’s voices provide cues of the likelihood of ovulation? The importance of sampling regime. PLoS One, 6(9), e24490. doi:10.1371/ journal.pone.0024490 Fisher, M. L. (2004). Female intrasexual competition decreases female facial attractiveness. Proceedings of the Royal Society of  London B-Biology Letters, 271(Suppl.), S283–S285. doi:10.1098/rsbl.2004.0160 Fisher, M. L., & Cox, A. (2009). The influence of female attrac­ tiveness on competitor derogation. Journal of Evolutionary Psychology, 7(2), 141–155. doi:10.1556/JEP.7.2009.2.3 Fisher, M., & Cox, A. (2010). Four strategies used during intrasexual competition for mates. Personal Relationships, 18(1), 20–38. doi:10.1111/j.1475-6811.2010.01307.x Freundl, G., Godehardt, E., Kern, P.  A., Frank-Herrmann, P., Koubenec, H. J., & Gnoth, C. (2003). Estimated maximum failure rates of cycle monitors using daily conception probabilities in the menstrual cycle. Human Reproduction, 18(12), 2628–2633. doi:10.1093/humrep/deg488 Frost, P. (1994). Preference for darker faces in photographs at different phases of the menstrual cycle: Preliminary assessment of evidence for a hormonal relationship. Perceptual and Motor Skills, 79, 507–514. doi:10.2466/pms.1994.79.1.507 Gangestad, S.  W., Garver-Apgar, C.  E., Simpson, J.  A., & Cousins, A. J. (2007). Changes in women’s mate preferences across the cycle. Journal of Personality and Social Psychology, 92(1), 151–163. doi:10.1037/0022-3514.92.1.151 Gangestad, S. W., Haselton, M. G., Welling, L. L. M., Gildersleeve, K., Pillsworth, E. G., Burriss, R. P., . . . Puts, D. A. (2016). How

valid are assessments of conception probability in ovulatory cycle research? Evaluations, recommendations, and theoretical implications. Evolution and Human Behavior, 37(2), 85–96. doi:10.1016/j.evolhumbehav.2015.09.001 Gangestad, S.  W., & Simpson, J.  A. (2000). The evolution of  human mating: Trade-offs and strategic pluralism. Behavioural and Brain Sciences, 23, 573–644. doi:10.1017/ S0140525X0000337X Gangestad, S. W., Simpson, J. A., Cousins, A. J., Garver-Apgar, C. E., & Christensen, P. N. (2004). Women’s preferences for male behavioral displays change across the menstrual cycle. Psychological Science, 15(3), 203–207. doi:10.1111/ j.0956–7976.2004.01503010.x Gangestad, S.  W., & Thornhill, R. (1998). Menstrual cycle variation in women’s preferences for the scent of symmetrical men. Proceedings of the Royal Society B: Biological Sciences, 265(1399), 927–933. doi:10.1098/rspb.1998.0380 Gangestad, S.  W., & Thornhill, R. (2008). Human oestrus. Proceedings of the Royal Society B: Biological Sciences, 275, 991–1000. doi:10.1098/rspb.2007.1425 Gangestad, S.  W., Thornhill, R., & Garver, C.  E. (2002). Changes in women’s sexual interests and their partners’ materetention tactics across the menstrual cycle: Evidence for shifting conflicts of interest. Proceedings of the Royal Society B: Biological Sciences, 269(1494), 975–982. doi:10.1098/ rspb.2001.1952 Gangestad, S. W., Thornhill, R., & Garver-Apgar, C. E. (2005). Women’s sexual interests across the ovulatory cycle depend on primary partner developmental instability. Proceedings of the Royal Society B: Biological Sciences, 272(1576), 2023–2027. doi:10.1098/rspb.2005.3112 Gangestad, S. W., Thornhill, R., & Garver-Apgar, C. E. (2010). Fertility in the cycle predicts women’s interest in attractive bodily features and sexual opportunism. Evolution and Human Behavior, 31(6), 400–411. doi:10.1016/j.evolhumbehav. 2010.05.003 Garcia, J. E., Jones, G. S., & Wright, G. L., Jr. (1981). Prediction of the time of ovulation. Fertility and Sterility, 36(3), 308–315. Garver-Apgar, C. E., Gangestad, S. W., & Thornhill, R. (2008). Hormonal correlates of women’s mid-cycle preference for the scent of symmetry. Evolution and Human Behavior, 29(4), 223–232. doi:10.1016/j.evolhumbehav.2007.12.007 Garver-Apgar, C.  E., Gangestad, S.  W., Thornhill, R., Miller, R. D., & Olp, J. (2006). Major histocompatibility complex alleles, sexual responsivity, and unfaithfulness in romantic couples. Psychological Science, 17(10), 830–835. doi:10.1111/ j.1467-9280.2006.01789.x Gildersleeve, K., Haselton, M. G., & Fales, M. R. (2014a). Do women’s mate preferences change across the ovulatory cycle? A meta-analytic review. Psychological Bulletin, 140(5), 1205–1259. doi:10.1037/a0035438 Gildersleeve, K., Haselton, M. G., & Fales, M. (2014b). Metaanalyses and p-curves support robust cycle shifts in mate preferences: Response to Wood & Carden and Harris, Pashler & Mickes (2014). Psychological Bulletin, 140, 1272–1280. doi:10.1037/a0035438 Gildersleeve, K., Haselton, M. G., Larson, C. M., & Pillsworth, E. G. (2012). Body odor attractiveness as a cue of impending ovulation in women: Evidence from a study using hormoneconfirmed ovulation. Hormones and Behavior, 61(2), 157–161. doi:10.1037/a0035438 Gonzales, J.  E., & Ferrer, E. (2016). Efficacy of methods for ovulation estimation and their effect on the statistical

Welling and Burriss

119

detection of ovulation-linked behavioral fluctuations. Behavior Research Methods, 48(3), 1125–1144. doi:10.3758/ s13428-015-0638-4 Grammer, K. (1993). 5-α-Androst-16en-3α-on: A male pheromone? A brief report. Ethology and Sociobiology, 14(3), 201–208. doi:10.1016/0162-3095(93)90006-4 Grammer, K., Jutte, A., & Fischmann, B. (1997). Der kampf der geschlechter und der krieg der signale. In B. Karnitschneider (Ed.), Liebe, Lust, und Leidenschaft. Sexualitat im Spiegal der Wissenschaft. Stuttgart, Germany: Hirzel. Guéguen, N. (2009a). Menstrual cycle phases and female receptivity to a courtship solicitation: An evaluation in a nightclub. Evolution and Human Behavior, 30(5), 351–355. doi:10.1016/j.evolhumbehav.2009.03.004 Guéguen, N. (2009b). The receptivity of women to courtship solicitation across the menstrual cycle: A field experiment. Biological Psychology, 80(3), 321–324. doi:10.1016/j.biopsycho. 2008.11.004 Guermandi, E., Vegetti, W., Bianchi, M. M., Uglietti, A., Ragni, G., & Crosignani, P. (2001). Reliability of ovulation tests in infertile women. Obstetrics and Gynecology, 97(1), 92–96. Harris, C. R., Pashler, H., & Mickes, L. (2014). Elastic analysis procedures: An incurable (but preventable) problem in the fertility effect literature: Comment on Gildersleeve, Haselton, & Fales (2014). Psychological Bulletin, 140, 1260–1264. doi:10.1037/a0036478 Haselton, M.  G., & Gangestad, S.  W. (2006). Conditional expression of women’s desires and men’s mate guarding across the ovulatory cycle. Hormones and Behavior, 49(4), 509–518. doi:10.1016/j.yhbeh.2005.10.006 Haselton, M.  G., & Gildersleeve, K. (2011). Can men detect ovulation? Current Directions in Psychological Science, 20(2), 87–92. doi:10.1177/0963721411402668 Haselton, M. G., & Miller, G. F. (2006). Women’s fertility across the cycle increases the short-term attractiveness of creative intelligence compared to wealth. Human Nature, 17, 50–73. doi:10.1007/s12110-006-1020-0 Haselton, M.  G., Mortezaie, M., Pillsworth, E.  G., BleskeRechek, A. L., & Frederick, D. A. (2007). Ovulatory shifts in human female ornamentation: Near ovulation, women dress to impress. Hormones and Behavior, 51(1), 40–45. doi:10.1016/ j.yhbeh.2006.07.007 Havlíček, J. Dvořáková, R. Bartoš, L., & Flegr, J. (2006). Nonadvertized does not mean concealed: Body odour changes across the human menstrual cycle. Ethology, 112(1), 81–90. doi:10.1111/j.1439-0310.2006.01125.x Havlíček, J., Roberts, S. C. & Flegr, J. (2005). Women’s preference for dominant male odour: Effects of menstrual cycle and relationship status. Biology Letters, 1, 256–259. doi:10.1098/ rsbl.2005.0332 Hill, S. E., & Durante, K. M. (2009). Do women feel worse to look their best? Testing the relationship between self-esteem and fertility status across the menstrual cycle. Personality and Social Psychology Bulletin, 35(12), 1592–1601. doi:10.1177/ 0146167209346303 Holzleitner, I. J., Hahn, A. C., Fisher, C., Lee, A., O’Shea, K. J., Jones, B. C., & DeBruine, L. M. (2017). Hormonal regulation of women’s prosocial, but not sexual, responses to kinship cues. BioRxiv. doi:10.1101/192054 Jochle, W. (1973). Coitus-induced ovulation. Contraception, 7(6), 523–564. doi:10.1016/0010-7824(73)90023-1 Johnston, L., Miles, L., & Macrae, C.  N. (2008). Was that a man? Sex identification as a function of menstrual cycle and

120

Investigating the Ovul atory Cycle

masculinity. Applied Cognitive Psychology, 22(9), 1185–1194. doi:10.1002/acp.1436 Johnston, V. S., Hagel, R., Franklin, M., Fink, B., & Grammer, K. (2001). Male facial attractiveness: Evidence for hormonemediated adaptive design. Evolution and Human Behavior, 21, 251–267. doi:10.1016/S1090-5138(01)00066-6 Jones, B. C., Hahn, A. C., Fisher, C., Wang, H., Kandrik, M., Han, C., . . . DeBruine, L. M. (2018). No compelling evidence that preferences for facial masculinity track changes in women’s hormonal status. Psychological Science, 29(6), 996–1005. doi:10.1177/0956797618760197 Jones, B.  C., Little, A.  C., Boothroyd, L., DeBruine, L.  M., Feinberg, D. R., Law Smith, M. J., . . . Perrett, D. I. (2005). Commitment to relationships and preferences for femininity and apparent health in faces are strongest on days of the menstrual cycle when progesterone level is high. Hormones and Behavior, 48(3), 283–290. doi:10.1016/j.yhbeh.2005. 03.010 Jones, B. C., Perrett, D. I., Little, A. C., Boothroyd, L., Cornwell, R.  E., Feinberg, D.  R., . . . Moore, F.  R. (2005). Menstrual cycle, pregnancy and oral contraceptive use alter attraction to apparent health in faces. Proceedings of the Royal Society B: Biological Sciences, 272(1561), 347–354. doi:10.1098/rspb. 2004.2962 Jukic, A. M. Z., Weinberg, C. R., Wilcox, A. J., McConnaughey, D.  R., Hornsby, P., & Baird, D.  D. (2008). Accuracy of reporting of menstrual cycle length. American Journal of Epidemiology, 167(1), 25–33. doi:10.1093/aje/kwm265 Jünger, J., Kordsmeyer, T. L., Gerlach, T. M., & Penke, L. (2018). Fertile women evaluate male bodies as more attractive, regardless of masculinity. Evolution and Human Behavior, 39(4), 412-423. doi:10.1016/j.evolhumbehav.2018.03.007 Jünger, J., Motta-Mena, N. V., Cárdenas, R. A., Bailey, D. H., Rosenfeld, K., Schild, C., . . . Puts, D. (2018). Do women's preferences for masculine voices shift across the ovulatory cycle? PsyArXiv. Retrieved from https://psyarxiv.com/k9y7s Kaighobadi, F., & Stevens, J.  R. (2013). Does fertility status influence impulsivity and risk taking in human females? Adaptive influences on intertemporal choice and risky decision making. Evolutionary Psychology, 11(3), 700–717. doi:10.1177/ 147470491301100314 Knaus, H. (1929). Eine neue methode zur bestimmung des ovulationstermines. Zentralblatt für Gynäkologie, 53, 2193. Koehler, N., Rhodes, G., Simmons, L. W., & Zebrowitz, L. A. (2006). Do cyclic changes in women’s face preferences target cues to long-term health? Social Cognition, 24(5), 641–656. doi:10.1521/soco.2006.24.5.641 Kuukasjärvi, S., Eriksson, C.  J.  P., Koskela, E., Mappes, T., Nissinen, K., & Rantala, M.  J. (2004). Attractiveness of women’s body odors over the menstrual cycle: The role of oral contraceptives and receiver sex. Behavioral Ecology, 15(4), 579–584. doi:10.1093/beheco/arh050 Laeng, B., & Falkenberg, L. (2007). Women’s pupillary responses to sexually significant others during the hormonal cycle. Hormones and Behavior, 52, 520–530. doi:10.1016/j.yhbeh. 2007.07.013 Larson, C.  M., Pillsworth, C.  G., & Haselton, M.  G. (2012). Ovulatory shifts in women’s attractions to primary partners and other men: Further evidence of the important of primary partner sexual attractiveness. PLoS One, 7(9), e44456. doi:10.1371/journal.pone.0044456 Little, A. C., & Jones, B. C. (2012). Variation in facial masculinity and symmetry preferences across the menstrual cycle is

moderated by relationship context. Psychoneuroendocrinology, 37(7), 999–1008. doi:10.1016/j.psyneuen.2011.11.007 Little, A. C., Jones, B. C., & Burriss, R. P. (2007). Preferences for masculinity in male bodies change across the menstrual cycle. Hormones and Behavior, 51(5), 633–639. doi:10.1016/j. yhbeh.2007.03.006 Little, A. C., Jones, B. C., Burt, D. M., & Perrett, D. I. (2007). Preferences for symmetry in faces change across the menstrual cycle. Biological Psychology, 76(3), 209–216. doi:10.1016/j. biopsycho.2007.08.003 Little, A. C., Jones, B. C., & DeBruine, L. M. (2008). Preferences for variation in masculinity in real male faces change across the menstrual cycle: Women prefer more masculine faces when they are more fertile. Personality and Individual Differences, 45(6), 478–482. doi:10.1016/j.paid.2008.05.024 López, H. H., Hay, A. C., & Conklin, P. H. (2009). Attractive men induce testosterone and cortisol release in women. Hormones and Behavior, 56(1), 84–92. doi:10.1016/j.yhbeh.2009.03.004 Lucas, M., & Koff, E. (2013). How conception risk affects competition and cooperation with attractive women and men. Evolution and Human Behavior, 34(1), 16–22. doi:10.1016/j. evolhumbehav.2012.08.001 Lucas, M., & Koff, E. (2017). Fertile women discount the future: Conception risk and impulsivity are independently associated with financial decisions. Evolutionary Psychological Science, 3(3), 261–269. doi:10.1007/s40806-017-0094-8 Lucas, M.  M., Koff, E., & Skeath, S. (2007). Pilot study of relationship between fertility risk and bargaining. Psychological Reports, 101(1), 302–310. doi:10.2466/pr0.101.1.302-310 Lukaszewski, A. W., & Roney, J. R. (2009). Estimated hormones predict women’s mate preferences for dominant personality traits. Personality and Individual Differences, 47(3), 191–196. doi:10.1016/j.paid.2009.02.019 Macrae, C. N., Alnwick, K. A., Milne, A. B., & Schloerscheidt, A.  M. (2002). Person perception across the menstrual cycle: Hormonal influences of social-cognitive functioning. Psychological Science, 13, 532–536. Maner, J.  K., & Miller, S.  L. (2014). Hormones and social monitoring: Menstrual cycle shifts in progesterone underlie women’s sensitivity to social information. Evolution and Human Behavior, 35(1), 9–16. doi:10.1016/j.evolhumbehav. 2013.09.001 Manning, J. T., Scutt, D., Whitehouse, G. H., Leinster, S. J., & Walton, J. M. (1996). Asymmetry and the menstrual cycle in women. Ethology and Sociobiology, 17(2), 129–143. doi:10.1016/ 0162-3095(96)00001-5 Marcinkowska, U. M., Ellison, P. T., Galbarczyk, A., Milkowska, K., Pawlowski, B., Thune, I., & Jasienska, G. (2016). Lack of support for relation between woman’s masculinity preference, estradiol level and mating context. Hormones and Behavior, 78, 1–7. doi:10.1016/j.yhbeh.2015.10.012 Marcinkowska, U., Galbarczyk, A., & Jasienska, G. (2018). La donna è mobile? Lack of cyclical shifts in facial symmetry, and facial and body masculinity preferences—A hormone based study. Psychoneuroendocrinology, 88, 47–53. doi:10.1016/j. psyneuen.2017.11.007 Marshall, J. (1968). A field trial of the basal-body-temperature method of regulating births. Lancet, 2(7558), 8–10. doi:10.1016/ S0140-6736(68)92886-9 McDonald, M. M., Asher, B. D., Kerr, N. L., & Navarrete, C. D. (2011). Fertility and intergroup bias in racial and minimalgroup contexts: Evidence for shared architecture. Psychological Science, 22(7), 860–865. doi:10.1177/0956797611410985

Meltzer, A.  L. (2017). Wives with masculine husbands report increased marital satisfaction near peak fertility. Evolutionary Behavioral Sciences, 11(2), 161–172. doi:10.1037/ebs0000083 Meltzer, A.  L., McNulty, J.  K., Miller, S.  L., & Baker, L.  R. (2015). A psychophysiological mechanism underlying women’s weight-management goals: Women desire and strive for greater weight loss near peak fertility. Personality and Social Psychology Bulletin, 41(7), 930–942. doi:10.1177/0146167215585726 Miller, G., Tybur, J. M., & Jordan, B. D. (2007). Ovulatory cycle effects on tip earnings by lap dancers: Economic evidence for human estrus? Evolution and Human Behavior, 28, 375–381. doi:10.1016/j.evolhumbehav.2007.06.002 Miller, S. L., & Maner, J. K. (2010). Scent of a woman: Men’s testosterone responses to olfactory ovulation cues. Psychological Science, 21(2), 276–283. doi:10.1177/0956797609357733 Miller, S. L., & Maner, J. K. (2011). Ovulation as a male mating prime: Subtle signs of women’s fertility influence men’s mating cognition and behavior. Journal of Personality and Social Psychology, 100(2), 295–308. doi:10.1037/a0020930 Navarrete, C. D., Fessler, D. M. T., Fleischman, D. S., & Geyer, J. (2009). Race bias tracks conception risk across the menstrual cycle. Psychological Science, 20(6), 661–665. doi:10.1111/ j.1467-9280.2009.02352.x Ogino, K. (1930). Ovulationstermin und konzeptionstermin. Zentralblatt für Gynäkologie, 54, 464. Pawłowski, B., & Jasienska, G. (2005). Women’s preferences for sexual dimorphism in height depend on menstrual cycle phase and expected duration of relationship. Biological Psychology, 70(1), 38–43. doi:10.1016/j.biopsycho.2005.02.002 Pearson, M., & Schipper, B.  C. (2013). Menstrual cycle and competitive bidding. Games and Economic Behavior, 78, 1–20. doi:10.1016/j.geb.2012.10.008 Penton-Voak, I. S., & Perrett, D. I. (2000). Female preference for male faces changes cyclically—further evidence. Evolution and Human Behavior, 21(1), 39–48. doi:10.1016/S10905138(99)00033-1 Penton-Voak, I.  S., & Perrett, D.  I. (2001). Male facial attractiveness: Perceived personality and shifting female preferences for male traits across the menstrual cycle. Advances in the Study of Behavior, 30, 219–259. Penton-Voak, I. S., Perrett, D. I., Castles, D. L., Kobayashi, T., Burt, D.  M., Murray, L.  K., & Minamisawa, R. (1999). Menstrual cycle alters face preference. Nature, 399, 741–742. doi:10.1038/21557 Perrett, D.  I., Lee, K.  J., Penton-Voak, I.  S., Rowland, D.  R., Yoshikawa, S., Burt, D. M., . . . Akamatsu, S. (1998). Effects of sexual dimorphism on facial attractiveness. Nature, 394, 884–887. doi:10.1038/29772 Peters, M., Simmons, L. W., & Rhodes, G. (2009). Preferences across the menstrual cycle for masculinity and symmetry in photographs of male faces and bodies. PLoS One, 4(1), e4138. doi:10.1371/journal.pone.0004138 Petralia, S. M., & Gallup, G. G., Jr. (2002). Effects of a sexual assault scenario on handgrip strength across the menstrual cycle. Evolution and Human Behavior, 23(1), 3–10. doi:10.1016/ S1090-5138(01)00085-X Piccoli, V., Foroni, F., & Carnaghi, A. (2013). Comparing group dehumanization and intrasexual competition among normally ovulating women and hormonal contraceptive users. Personality and Social Psychology Bulletin, 39(12), 1600–1609. doi:10.1177/0146167213499025 Pillsworth, E.  G., & Haselton, M.  G. (2006). Male sexual attractiveness predicts differential ovulatory shifts in female

Welling and Burriss

121

extra-pair attraction and male mate retention. Evolution and Human Behavior, 27(4), 247–258. doi:10.1016/j. evolhumbehav.2005.10.002 Pipitone, R.  N., & Gallup, G.  G. (2008). Women’s voice attractiveness varies across the menstrual cycle. Evolution and Human Behavior, 29(4), 268–274. doi:10.1016/j. evothumbehav.2008.02.001 Prokosch, M. D., Coss, R. G., Scheib, J. E., & Blozis, S. A. (2009). Intelligence and mate choice: Intelligent men are always appealing. Evolution and Human Behavior, 30(1), 11–20. doi:10.1016/j.evolhumbehav.2008.07.004 Provost, M. P., Quinsey, V. L., & Troje, N. F. (2008). Differences in gait across the menstrual cycle and their attractiveness to men. Archives of Sexual Behavior, 37(4), 598–604. doi:10.1007/ s10508-007-9219-7 Puts, D. A. (2005). Mating context and menstrual phase affect women’s preferences for male voice pitch. Evolution and Human Behavior, 26, 388–397. doi:10.1016/j.evolhumbehav. 2005.03.001 Puts, D. A. (2006). Cyclic variation in women’s preferences for masculine traits: Potential hormonal causes. Human Nature, 17(1), 114–127. doi:10.1007/s12110-006-1023-x Puts, D. A., Bailey, D. H., Cárdenas, R. A., Burriss, R. P., Welling, L.  L.  M., Wheatley, J.  R., & Dawood, K. (2013). Women’s attractiveness changes with estradiol and progesterone across the ovulatory cycle. Hormones and Behavior, 63(1), 13–19. doi:10.1016/j.yhbeh.2012.11.007 Rikowski, A., & Grammer, K. (1999). Human body odour, symmetry and attractiveness. Proceedings of the Royal Society B: Biological Sciences, 266(1422), 869–874. doi:10.1098/ rspb.1999.0717 Roberts, S. C., Havlíček, J., Flegr, J., Hruskova, M., Little, A. C., Jones, B. C., . . . Petrie, M. (2004). Female facial attractiveness increases during the fertile phase of the cycle. Proceedings of the Royal Society of London B: Biology Letters, 271(Suppl.), S270–S272. doi:10.1098/rsbl.2004.0174 Röder, S., Brewer, G., & Fink, B. (2009). Menstrual cycle shifts in women’s self-perception and motivation: A daily report method. Personality and Individual Differences, 47(6), 616–619. doi:10.1016/j.paid.2009.05.019 Roney, J. R., & Simmons, Z. L. (2008). Women’s estradiol predicts preference for facial cues of men’s testosterone. Hormones and Behavior, 53(1), 14–19. doi:10.1016/j.yhbeh.2007.09.008 Roney, J. R., & Simmons, Z. L. (2013). Hormonal predictors of women’s sexual desire in normal menstrual cycles. Hormones and Behavior, 63(4), 636–645. doi:10.1016/j. yhbeh.2013.02.013 Roney, J. R., Simmons, Z. L., & Gray, P. B. (2011). Changes in estradiol predict within-women shifts in attraction to facial cues of men’s testosterone. Psychoneuroendocrinology, 36(5), 742–749. doi:10.1016/j.psyneuen.2010.10.010 Rosen, M.  L., & López, H.  H. (2009). Menstrual cycle shifts in attentional bias for courtship language. Evolution and Human Behavior, 30(2), 131–140. doi:10.1016/j. evolhumbehav.2008.09.007 Rupp, H. A., James, T. W., Ketterson, E. D., Sengelaub, D. R., Janssen, E., & Haiman, J. R. (2009). Neural activation in the orbitofrontal cortex in response to male faces increases during the follicular phase. Hormones and Behavior, 56(1), 66–72. doi:10.1016/j.yhbeh.2009.03.005 Schwartz, D., MacDonald, P.  D.  M., & Heuchel, V. (1980). Fecundability, coital frequency and the viability of ova. Population Studies, 34, 397–400.

122

Investigating the Ovul atory Cycle

Schwartz, D., Mayaux, M. J., Martin-Boyce, A., Czyglik, F., & David, G. (1979). Donor insemination: Conception rate according to cycle day in a series of 821 cycles with a single insemination. Fertility and Sterility, 31(2), 226–229. doi:10.1016/S0015-0282(16)43829-X Schwarz, S., & Hassebrauck, M. (2008). Self-perceived and observed variations in women’s attractiveness throughout the menstrual cycle—a diary study. Evolution and Human Behavior, 29(4), 282–288. doi:10.1016/j.evothumbehav. 2008.02.003 Scutt, D., & Manning, J. T. (1996). Symmetry and ovulation in women. Human Reproduction, 11(11), 2477–2480. doi:10.1093/ oxfordjournals.humrep.a019142 Sheldon, M.  S., Cooper, M.  L., Geary, D.  C., Hoard, M., & DeSoto, M. C. (2006). Fertility cycle patterns in motives for sexual behavior. Personality and Social Psychology Bulletin, 32(12), 1659–1673. doi:10.1177/0146167206292690 Simonsohn, U., Nelson, L. D., & Simmons, J. P. (2014). P-curve: A key to the file-drawer. Journal of Experimental Psychology: General, 143, 534–547. doi:10.1037/a0033242 Singh, D., & Bronstad, P.  M. (2001). Female body odour is a potential cue to ovulation. Proceedings of the Royal Society B: Biological Sciences, 268(1469), 797–801. doi:10.1098/rspb. 2001.1589 Small, C.  M., Manatunga, A.  K., & Marcus, M. (2007). Validity of self-reported menstrual cycle length. Annals of Epidemiology, 17(3), 163–170. doi:10.1016/j.annepidem.2006. 05.005 Stern, K., & McClintock, M. K. (1998). Regulation of ovulation by human pheromones. Nature, 392(6672), 177–179. doi:10.1038/32408 Stirnemann, J. J., Samson, A., Bernard, J.-P., & Thalabard, J.-C. (2013). Day-specific probabilities of conception in fertile cycles resulting in spontaneous pregnancies. Human Reproduction, 28(4), 1110–1116. doi:10.1093/humrep/des449 Stoleru, S. G., Ennaji, A., Cournot, A., & Spira, A. (1993). LH pulsatile secretion and testosterone blood levels are influenced by sexual arousal in human males. Psychoneuroendocrinology, 18(3), 205–218. doi:10.1016/0306-4530(93)90005-6 Thornhill, R., & Gangestad, S. W. (1999). The scent of symmetry: A human sex pheromone that signals fitness? Evolution and  Human Behavior, 20(3), 175–201. doi:10.1016/S10905138(99)00005-7 Thornhill, R., & Gangestad, S.  W. (2003). Do women have evolved adaptation for extra-pair copulation? In E. Voland & K.  Grammer (Eds.), Evolutionary aesthetics (pp. 341–368). Heidelberg, Germany: Springer-Verlag. Thornhill, R., & Gangestad, S.  W. (2008). The evolutionary biology of human female sexuality. New York, NY: Oxford University Press. Thornhill, R., Gangestad, S.  W., Miller, R., Scheyd, G., McCollough, J.  K., & Franklin, M. (2003). Major histocompatibility complex genes, symmetry, and body scent attractiveness in men and women (Homo sapiens). Behavioral Ecology, 14(5), 668–678. doi:10.1093/beheco/arg043 Trussell, J., Rodriguez, G., & Ellertson, C. (1998). New estimates of the effectiveness of the Yuzpe regimen of emergency contraception. Contraception, 57(6), 363–369. doi:10.1016/ S0010-7824(98)00042-0 Venners, S. A., Liu, X., Perry, M. J., Korrick, S. A., Li, Z., Yang, F., . . . Wang, X. (2006). Urinary estrogen and progesterone metabolite concentrations in menstrual cycles of fertile women with non-conception, early pregnancy loss or clinical pregnancy.

Human Reproduction, 21(9), 2272–2280. doi:10.1093/humrep/ del187 Vukovic, J., Jones, B. C., DeBruine, L. M., Little, A. C., Feinberg, D.  R., & Welling, L.  L.  M. (2009). Circum-menopausal effects on women’s judgements of facial attractiveness. Biology Letters, 5(1), 62–64. doi:10.1098/rsbl.2008.0478 Waller, D.  K., Spears, W.  D., Gu, Y., & Cunningham, G.  C. (2000). Assessing number-specific error in the recall of onset of last menstrual period. Paediatric and Perinatal Epidemiology, 14(3), 263–267. doi:10.1046/j.1365-3016.2000.00275.x Wegienka, G., & Baird, D. D. (2005). A comparison of recalled date of last menstrual period with prospectively recorded dates. Journal of Women’s Health, 14(3), 248–252. doi:10.1089/ jwh.2005.14.248 Weinberg, C. R., Wilcox, A. J., Baird, D. D., & Gladen, B. B. (1998). The probability of conception as related to the timing of intercourse around ovulation. Genus, 54(3–4), 129–142. Welling, L.  L.  M., Jones, B.  C., DeBruine, L.  M., Conway, C. A., Law Smith, M. J., Little, A. C., . . . Al-Dujaili, E. A. S. (2007). Raised salivary testosterone in women is associated with increased attraction to masculine faces. Hormones and Behavior, 52(2), 156–161. doi:10.1016/j.yhbeh.2007.01.010 Wilcox, A. J., Dunson, D., & Baird, D. D. (2000). The timing of the “fertile window” in the menstrual cycle: Day specific

estimates from a prospective study. British Medical Journal, 321(7271), 1259–1262. doi:10.1136/bmj.321.7271.1259 Wilcox, A. J., Dunson, D. B., Weinberg, C. R., Trussell, J., & Baird, D. D. (2001). Likelihood of conception with a single act of intercourse: Providing benchmark rates for assessment of post-coital contraceptives. Contraception, 63(4), 211–215. doi:10.1016/S0010-7824(01)00191-3 Wilcox, A. J., Weinberg, C. R., & Baird, B. D. (1995). Timing of sexual intercourse in relation to ovulation: Effects on the probability of conception, survival of the pregnancy, and sex of the baby. New England Journal of Medicine, 333(23), 1517–1521. doi:10.1056/NEJM199512073332301 Wilcox, A.  J., Weinberg, C.  R., & Baird, D.  D. (1998). Postovulatory ageing of the human oocyte and embryo failure. Human Reproduction, 13(2), 394–397. Wood, W., Kressel, L., Joshi, P.  D., & Louie, B. (2014). Metaanalysis of menstrual cycle effects on mate preferences. Emotion Review, 6(3), 229–249. doi:10.1177/1754073914523073 World Health Organization (WHO) (1980). Temporal relationships between ovulation and defined changes in the concentration of plasma estradiol-17β, luteinizing hormone, follicle-stimulating hormone, and progesterone. I.  Probit analysis. American Journal of Obstetrics and Gynecology, 138(4), 383–390.

Welling and Burriss

123

CH A PT E R

8

Reproductive Behavior in the Human Male

Stefan M. M. Goetz, Glenn Weisfeld, and Samuele Zilioli

Abstract Given the pivotal role of differential reproduction to the evolutionary success of ancestral men, evolution has produced a plethora of reproductive strategies aimed at solving the complexities of intramale competition and satisfying and/or thwarting the reproductive desires of women. Life history theory recognizes that an organism has limited resources and must invest energy appropriately. Broadly, reproductive strategies can be dichotomized into short-term (emphasizing mating over parental effort) versus long-term (emphasizing parenting over mating effort) strategies. Increasingly, the neuroendocrine system—especially testosterone—has been recognized as the proximate mechanism orchestrating adoption of one strategy over the other. This chapter reviews behaviors geared toward solving problems associated with both long-term and short-term reproductive strategies and discusses the neuroendocrine correlates. The adoption of one strategy over another is conceptualized as conditional or facultative adaptations in which strategic switching points are tuned over evolutionary time to produce optimal fitness responses to men’s social and physical conditions. Keywords:  reproduction, reproductive behavior, mating, parenting, male strategies, short term, long term, testosterone

Differential reproduction is the driving force behind evolution, altering the relative proportion of alleles in the population. Men, and male mammals in general, exhibit greater variance in reproduction relative to women (i.e., asymmetric reproductive skew; Geary, 1998; Jokela, Rotkirch, Rickard, Pettay, & Lummaa, 2010), so much so that men such as Ismail ibn Sharif (the warrior king) is estimated to have sired upward of 888 children, overshadowing the countless childless bachelors. In contrast, the most children any one woman has ever birthed is 69, in a total of 27 births (Bateman & Bennett,  2006). Although these are extreme examples, they are illustrative of the general pattern. Among modern-day huntergatherers, men’s reproductive variance is about twice that of women’s1 (Betzig,  2012). Furthermore, as 1   This ratio was calculated based on data from Betzig (2012, Table 1), which included five hunter and gatherer societies.

societies become less egalitarian, the men–women difference only intensifies, with men vying for and leveraging status and resources to monopolize a greater proportion of fertile women (Betzig, 1986). Clearly, men in different socioecological and physical conditions face different reproductive challenges and opportunities with significant fitness implications. These challenges drive the evolution of men’s mating strategies. Traditionally, evolutionary psychologists have anchored the study of men’s reproductive strategies along a single continuum ranging from short-term to However, there is some debate as to whether modern-day hunter-gatherers provide a valid model to form inferences about our ancestral past given that modern-day hunter-gatherers primarily inhabit geographically marginal habitats, face pressure from neighboring agricultural societies, and are not free from acculturational influences of state-level societies (Walker, Hill, Flinn, & Ellsworth, 2011).

125

long-term mating strategies, designating short-term strategies with the moniker “cad” (polygynous and uninvolved with childcare) and long-term strategists with the moniker “dad” (monogamous and involved with childcare; e.g., Buss & Schmitt,  1993). Although Western societies do not permit polygamous marriages,2 approximately 20 to 40 percent of men engage in an extramarital affair in their lifetimes (Greeley, 1994). Furthermore, the reproductive consequences of serial monogamy—in which men and women can have several consecutive partners—­resemble those found in polygynous societies. Indeed, in the United States, men who had married and remarried three or more times had 19 percent more children than monogamously married men ( Jokela et al.,  2010). By utilizing more than one woman’s reproductive lifespan, a man can enhance his reproductive success ( Jokela et al., 2010). Following an exclusively monogamous or polygynous mating strategy represents extremes in men’s life history strategies,3 with varying fitness consequences conditioned upon the socioecological and sociocultural environment (Laland & Brown, 2011).

Life History Theory

Life history theory is a branch of evolutionary biology focused on the issue of energy allocation “choices” an organism makes over its lifespan to maximize biological fitness4 (Ellis, Figueredo, Brumbach, &  Schlomer,  2009; Hill,  1993; McNamara & Houston, 1996; Roff, 2002). These choices reflect evolved resource allocation strategies, determining the timing, rate, and extent of somatic growth and investment (e.g., immune function); reproductive maturity; fecundity; mating effort; parental effort; and senescence. All organisms have limited resources in the currency of both time and energy; how they are allocated is balanced as a function of historical selection pressures that optimize the cost–benefit ratio of investing in each (e.g., Hill, 1993). Broadly, energy is allocated into various forms of somatic investment and reproductive investment. Due to the altricial nature of our species—offspring are born immature and require a long period of 2   A survey of world cultures indicates that 80 percent of the 139 societies recorded in the standard cross-cultural sample permit polygynous marriages (Murdock & White, 1980). 3  In evolutionary biology, the term strategy denotes an organism’s phenotype—behavioral, morphological, physiological, and psychological—produced in response to both environmental input and genetic factors. The term does not imply conscious choice. 4   Biological fitness refers to an organism’s reproductive success in passing on genes through the production of offspring or kin.

126

provisioning—the further division of reproductive effort into mating and parental effort is particularly relevant to men’s life history strategies. For instance, in socioecological contexts in which paternal investment is decoupled from offspring reproductive success, men may gain greater fitness returns by divesting from parental effort in favor of mating effort. Indeed, given that parental effort is not an obligate adaptation in men, under some circumstances men can follow mating strategies that require no more than the investment of producing sperm and gaining sexual access to fertile women. However, when paternal investment and offspring reproductive success are closely related, men gain by investing more heavily in parenting effort (Sear & Mace,  2008). These strategies can be viewed as “alternative mating strategies.”

Alternative Mating Strategies

Where life history strategies are discussed in terms of reproduction, evolutionary biologists view them as alternative mating strategies. Parsing the comparative data, three broad categories of alternative mating strategies have been identified: pure alternative, mixed, and conditional strategies (see Gross, 1996). Pure alternative strategies are produced by genetic polymorphic differences in the population. That is, allelic differences underpin static phenotypes that emerge in populations through frequency-dependent selection. For example, negative frequency-­dependent selection occurs when the relative fitness of a trait (e.g., psychopathy) depends on its rarity in the population. Theoretically, frequency-dependent selection can maintain genetic polymorphisms in a population5 (Maynard-Smith, 1982). Mixed strategies, in contrast, are not underpinned by allelic polymorphisms, but rather by organisms enacting each alternative phenotype in some fixed proportion that is invariant at the population level. Phenotypic heterogeneity is the product of each organism following a probabilistic allocation strategy. Finally, the third form is that of conditional strategies. In contrast to pure strategies, conditional strategies are underpinned by a common genetic architecture but differ from mixed strategies, whereby each alternative is not determined probabilistically, but rather as a consistent response 5   Under normal circumstances, the relative fitness conferred by an allele causes that allele to become “fixed” (i.e., all other versions are selected against) within a population. The variance around an optimal trait is due to genetic noise (i.e., random mutations and/or developmental insults).

Reproductive Behavior in the Human Male

mechanism6 to the conditions faced by an organism—­ both external and internal—at several key developmental stages (e.g., Del Giudice, 2009). Across the animal kingdom, few instances of pure genetic polymorphic strategies have been identified.7 Rather, within-species differences in strategies are best explained by conditional and/or facultative adaptions to local conditions. Much like sexual differentiation, within-sex phenotypic variation relies on differential expression of genetic programs; that is, the basis for the strategies is instantiated in a genetic architecture that is conditionally expressed by the varying socioecological challenges faced by an organism (Gross, 1996). Conditional strategies do not necessarily require equal fitness to be maintained within the population; however, fitness is equal across options in contexts where a switching point exists. Stated differently, at a given switch point, one strategy will lead to equal fitness as adopting another strategy (see Gross, 1996). For example, theoretically, there should exist a point at which the socioecological conditions facing, say, a subordinate male chimpanzee provide equal fitness outcomes whether he adopts a possessive mating strategy or a sneaky mating strategy. Beyond that point, however, following one strategy over the other will produce different fitness outcomes. A subordinate male who continues to try to pursue a possessive strategy will leave fewer offspring than a subordinate male who adopts a sneaky mating strategy, ceteris paribus. One oft-cited example can be seen in the various reproductive strategies expressed by male bluegill (Lepomis macrochirus; Gross,  1991). Large males establish and defend a nest, enticing females to lay their eggs where they are then fertilized and defended by the male. However, when the proportion of males following this strategy is high, it creates a niche for smaller males to develop a pseudo-female morph, allowing them to sneak past a larger male’s defense and fertilize the eggs in his nest. Biologists have determined that these morphs are likely not produced through genetic differences between morphs (e.g., in the case of pure alternative strategy), but rather are the result of differential gene expression induced by perception of the male’s social environ6  But there are polymorphic alleles underpinning different “reaction norm” profiles (see Manuck, 2010). 7   For example, the marine isopod Paracerceis sculptta has three mating phenotypes: larger fighter males, medium-sized males that mimic females, and small sneaker males. These three morphs are due to three alleles at a single autosomal locus (Shuster & Wade, 1991).

ment (see Bass,  1996). Each male has the genetic program to produce either morph, with the determining factor activating each morph being the degree of reproductive competition (i.e., frequency of morphs). Looking across nonhuman primate societies, this clearly is the rule. Males in different positions within the hierarchy adopt alternative reproductive strategies (e.g., sneaking copulations or forming temporary consortships away from the group rather than facing reproductive oblivion at the hands of a dominant male). For instance, the sexual strategies observed in a population of chimpanzees from Tanzania were flexible and consisted of three distinct strategies (Tutin, 1979). Males either followed an opportunistic, possessive, or consortship mating strategy, with the latter producing the greatest fitness dividends, save for if the male was the alpha, in which case following the possessive strategy was optimal. Indeed, the strategy adopted was clearly linked to characteristics of the individual (e.g., age, physical condition, and dominance position) and by social factors (e.g., stability of dominance relationships). The discussion of men’s mating strategies that follows takes the perspective that men’s reproductive life history strategies are conditional adaptions, induced by neuroendocrine responses to the environment. Indeed, because these adaptations require that whole sets of mechanisms (both psychological and behavioral) work in concert, the neuroendocrine system represents a viable m ­ echanism for orchestrating these changes. Not only does the endocrine system respond to environmental stimuli, but also its effects are systemic, thus enabling gene expression across diverse neural and somatic tissues (Cox, McGlothlin, & Bonier, 2016; Xu et al., 2012).

Life History Strategies

Traditionally, life history theory focused on ­cross-species comparisons, organizing species and taxa along an r-selected (rapid reproduction) to ­K-selected (slow reproduction) continuum, with scant consideration given to within-species differences (e.g., Pianka, 1970). What attention that was given to within-species differences was viewed as genetic and developmental noise. More recent work has begun to consider the possibility that within-species variation represents adaptive responses—adaptive plasticity (West-Eberhard, 2003)—to the environment, particularly the early developmental environment (e.g., Del Giudice,  2009; Del Giudice, Ellis, & Shirtcliff, 2011). Taking this latter approach, a handful of evolutionarily minded researchers have begun using a fast–slow life history strategy distinction to Goetz, Weisfeld, and Zilioli

127

explain individual differences from an evolutionary perspective (e.g., Belsky, Steinberg, & Draper, 1991; Del Giudice,  2009; Griskevicius, Tybur, Delton, & Robertson, 2011). Slow strategies are characterized by slower rates of development and later onset adrenarche, puberty, and senescence. Behaviorally, slow life history strategies entail later sexual debut, fewer sexual partners, stable reproductive relationships, later ages of reproduction, fewer offspring, and greater parental investment in offspring. Patterns of risk taking also differ, with slow strategists tending to have longer time horizons, lower impulsivity, and greater aversion to risk in general. Essentially, slow strategies involve greater investment in somatic and parental effort. In contrast, fast life history strategies entail less investment in somatic and parental effort, instead favoring investment in mating effort (Griskevicius et al., 2013). Although these differences are in part driven by genetic polymorphisms within human populations (Day et al.,  2016; Plomin, DeFries, Knopik, & Neiderheiser,  2013), much of the current spate of research has focused on adaptive plasticity components (e.g., Del Giudice et al., 2011). Furthermore, behaviors indicative of fast and slow life history strategies can be acutely induced, indicating that life history strategies are not wholly genetically or developmentally determined (Arnocky, Woodruff, & Schmitt, 2016; Griskevicius et al., 2013). Nevertheless, adaptive plasticity of life history strategies stands to explain many key individual differences observed by developmental psychologists and offers a framework for bringing the study of individual differences under the auspices of evolutionary theory.

Adaptive Plasticity in Life History Strategies

Examining the environmental antecedents of these strategies can elucidate the evolutionary pressures that these strategies evolved to address. The factors that appear to be the primary driver of variance in life history strategies are cues to extrinsic risks—that is, risks that are independent of an organism’s behavior, such as mortality due to war (Quinlan, 2007). The translation of parental effort to offspring fitness reaches a saturation point, at which additional effort does not increase offspring fitness. Importantly, the position of this point along the parental effort continuum varies as a function of environmental risk and is lower in high-risk environments (Quinlan, 2007) and in contexts in which paternity certainty is low (Geary, 2005). Thus, parental effort itself can be used as a developmental cue about the predictability of the environment, guiding the development of life 128

history strategies (Del Giudice, 2009). For instance, father absence, indicative of low parental investment, may lead to the development of a fast life history strategy (Draper & Harpending, 1982). Current reproduction is favored over later reproduction, which is achieved by setting sexual maturation earlier and lowering one’s own parental investment (i.e., a quantity-over-quality approach to reproduction; Chisholm, 1993). The experience of an unpredictable environment during development in males engenders what Wilson and Daly (1985) termed the young male syndrome—a pattern of increased competitiveness, risk taking, and violence that represents heavy investment in mating effort at the expense of later reproduction. For instance, sons from father-absent households—which may be indicative of mating low commitment and/or high mortality risk—are more aggressive and display hypermasculine behavior, consistent with heighted male–male competition associated with a fast life history strategy (Draper & Harpending, 1982). Similarly, attachment relationships in infancy and early childhood may serve as a “Socioassay” (Chisholm, 1993), whereby the child gathers information about the danger and predictability of the environment, thus (eventually) balancing trade-offs between current and future reproductive investment and mating and parenting effort (Del Giudice, 2009). Intriguingly, sex differences in attachment emerge during middle childhood (and persist into adulthood) such that boys raised in risky environments tend to develop avoidant attachment styles, whereas girls raised in risky environments tend to develop anxious attachment styles (Corby, 2006; Del Giudice, 2008; Finnegan, Hodges, & Perry, 1996; Granot & Mayseless,  2001; Karavasilis, Doyle, & Markiewicz, 2003; Toth, Szollosi, Danis, Green, & Gervai, 2006). Del Giudice (2009) proposed that an avoidant attachment style may be a compo­nent mechanism of the psychological architecture underpinning short-term mating strategies, promoting a plurality of reproductive relationships.

Manifestations of Mating Effort

Short-term mating can be construed as a zero-sum contest given that, ultimately, fertilization cannot be shared. Furthermore, sexual access to women quickly becomes a limiting resource, especially when considering sex differences in desires for uncommitted mating (Buss & Schmitt, 1993; Symons, 1979). Thus, these factors set the stage for heightened ­intramale contest, particularly in short-term mating domains (Ainsworth & Maner, 2012).

Reproductive Behavior in the Human Male

One natural consequence of zero-sum contests over limited resources is the emergence of dominance hierarchies, which are a ubiquitous feature of primate societies (e.g., Cummins, 2000). As described by the priority-of-access model, through dominance relations, access to mates is regulated such that more dominant individuals generally are granted greater access to fertile mates (Altmann,  1962; Dubuc, Muniz, Heistermann, Engelhardt, & Widdig, 2011). Given the relationship between reproductive success and dominance position (Betzig,  1986; Turke & Betzig, 1985), evolution has produced adaptations that bias men toward gaining/defending status and navigating dominance hierarchies (e.g., Griskevicius et al., 2009). Indeed, this is reflected in our emotional motivation system, in which status itself is a powerful motivational goal (Huberman, Loch, & Önçüler, 2004). Infants as young as 10 months old can extract dominance relationships between two novel “actors” (Thomsen, Frankenhuis, Ingold-Smith, & Carey,  2011), and judgments of other people’s dominance are made within milliseconds and are fundamental to how they are evaluated (Carré, McCormick, & Mondloch,  2009). Furthermore, dominance hierarchies form quickly among unacquainted men8 (Cheng, Tracy, Foulsham, Kingstone, & Henrich, 2013; Rosa & Mazur, 1979). In many social contexts, competition for dominance position is therefore often integral for men to achieve reproductive success (Ainsworth & Maner, 2012; Griskevicius et al., 2009). However, competing for dominance can entail heavy costs. Therefore, the cost–benefit trade-offs of pursing dominance status over, say, somatic effort must be balanced (Wingfield, Lynn, & Soma, 2001). Importantly, under conditions in which these costs are reduced or the benefits increased, mating effort is expected to increase. For instance, when the sex ratio is female biased (i.e., more females than males), male mating effort is expected to be increased because the likelihood of successfully pursuing a short-term mating strategy is greater (Arnkocky, Woodruff, & Schmitt, 2016). Indeed, this is reflected in violent crime rates, an indicator of heightened male mating effort, which tracks sex ratios (Schacht, Tharp, & Smith, 2016). Likewise, under conditions of heightened extrinsic risk (e.g., mortality that cannot be avoided through altering one’s behavior), the benefits of developing 8   In support of the greater importance of dominance among men, women too form dominance hierarchies when first meeting, although dominance hierarchies emerge on longer timescales than they do in men (Anderson, John, Keltner, & Kring, 2001).

traits and behaviors (e.g., risk taking and aggression) that facilitate short-term mating may be favored (Ellis et al., 2012). Testosterone likely functions to balance these risks by allocating resources toward reproductive effort. Specifically, testosterone may increase manifestations of mating effort—intrasexual competitive motivation (Carré & Olmstead,  2015), risk taking (Apicella, Dreber, & Mollerstrom,  2014), and mate-seeking behavior (van der Meij, Almela, Buunk, Fawcett, & Salvador,  2012)—and divert resources away from somatic effort (for review, see Bribiescas, 2001). It should be noted, however, that other androgens have also been implicated in mediating these tradeoffs; for instance, levels of dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulfate (DHEAS) during adolescence may impact later adult attachment styles, putatively in the service of supporting differential life history strategies (Del Giudice, 2009). One stark demonstration of testosterone’s potential role in mediating the trade-off between somatic effort and mating effort is provided by the recent practice of male castration in Korea. Removing the primary source of androgen production in these men appears to result in increased longevity compared to intact controls, potentially indicating that normal levels of testosterone actually lead to shortened lifespans by facilitating traits associated with reproductive effort (e.g., aggression and risk taking; Min, Lee, & Park, 2012). Furthermore, testosterone and immune function (which represents somatic effort) are known to have negative bidirectional effects on each other. For instance, men who received an immune challenge (flu vaccination) show lower testosterone levels than control men (Simmons & Roney,  2009), and conversely, hypogonadal men undergoing testosterone replacement therapy decline in circulating antibodies and cytokines (e.g., Kocar et al., 2000). Given the costs associated with maintaining high levels of testosterone, the endocrine system evolved to adaptively respond to the social environment (Carré, McCormick, & Hariri, 2011; Eisenegger, Haushofer, & Fehr, 2011). The “challenge hypothesis” was proposed to explain fluctuations in testosterone as a facultative adjustment to mating effort (Wingfield, Hegner, Dufty, & Ball, 1990). Although the hypothesis was originally developed to explain testosterone dynamics in birds, it has since found support across numerous taxa, including insects (Tibbetts & Crocker, 2014), fish (Oliveira, 2009), nonhuman primates (Muller & Wrangham, 2004; Goetz, Weisfeld, and Zilioli

129

Sobolewski, Brown, & Mitani, 2013; for a review, see Muller,  2016), and humans (Archer,  2006). For example, chimpanzees form stable linear dominance hierarchies, which reduces aggression and testosterone secretion (Muehlenbein, Watts, & Whitten, 2004; Muller & Wrangham, 2004); however, when a group contains an estrous female— which is advertised by sexual swellings, testosterone levels, along with aggression, have been observed to rise (Muller & Wrangham,  2004; Sobolewski, Brown, & Mitani, 2013). Nulliparous estrous females did not promote testosterone secretion, likely due to the fact that nulliparous females are by definition inexperienced mothers and are thus less attractive to male chimpanzees despite similar copulatory rates to parous females (which indicates that sexual behavior, per se, is not the cause of the observed increase in testosterone).

Risk Taking as a Function of Life History Strategy

As stated previously, the development of a fast life history is partially dependent on cues to environmental risk (see Ellis et al.,  2009). Indeed, if the future is uncertain, investing in the present is likely on average to produce better fitness outcomes than delaying for a future that may never arrive (Ellis et al., 2009). One stark illustration of this principle is the possibility that wartime rape, rather than (solely) an expression of power over the powerless, is a facultative adaptation to the real possibility of imminent death (Henry, Ward, & Hirshberg,  2004). Experimentally priming mortality has been found to increase men’s willingness to engage in risky sexual behaviors, as well as engendering a desire for offspring (Taubman-Ben-Ari, 2004; Wisman & Goldenberg, 2005, respectively). Likewise, cues to life expectancy alter preferences for short-term and long-term mating, such that shorter life expectancy is associated with a preference for short-term mating (Dunkel, Mathes, & Decker,  2010). Risk sensitivity theory frames risk sensitivity in terms of average outcomes, variances around the average, and current need level. If relative reproductive success is in jeopardy, it is to the organism’s advantage to follow a more risky strategy if a lower risk strategy is unlikely to meet an organism’s needs (Mishra & Lalumière, 2010). Recently, Griskevicius and colleagues (2011) showed experimentally that childhood experience of resource scarcity (indexed by socioeconomic status) interacted with mortality cues to induce increased risk taking, consistent with the interpretation that 130

early experience of risk induces the development of fast life history strategies. Facultative adjustment to risk taking has also been observed in mating contexts. In one study, men exposed to pictures of attractive women were less risk averse in a card game and risk taking was positively correlated with their mating motivations, suggesting that risk taking may serve mating effort (Baker & Maner, 2008). Testosterone appears to mediate these effects on behavior, likely by altering the balance between reward and punishment sensitivity (van Honk et al., 2004). Testosterone has been shown to alter the balance between sensitivity for punishment and reward in both animals and humans by reducing punishment sensitivity and increasing reward sensitivity (Boissy & Bouissou, 1994; van Honk et al., 2004). In humans, van Honk and colleagues (2004) demonstrated that administering testosterone to healthy young women shifted their punishment–reward contingencies in the Iowa gambling task toward a risk-prone pattern. Similarly, men with high levels of endogenous testosterone are less risk averse in this task than men with lower levels of endogenous testosterone (Stanton, Liening, & Schultheiss, 2011), suggesting a mechanism by which testosterone amplifies risk taking in the service of mating effort. For instance, Ronay and von Hippel (2010) observed that in the presence of an attractive female, male skateboarders performed more physically risky maneuvers, and this relationship was shown to be partially mediated by increases in testosterone. Likewise, in the domain of financial risk taking, endogenous baseline and responsive testosterone are positively associated with increased risk preferences (Apicella et al.,  2008,  2014). Further­ more, facial masculinity—an indicator of pubertal-testosterone organizational effects (e.g., Hodges-­Simeon, Sobraske, Samore, Gurven, & Gaulin, 2016)—was also found to be positively associated with risk taking (Apicella et al., 2008), and in a more recent study was shown, along with testosterone, to be related to proneness to boredom and sensation seeking (Campbell et al., 2010). With regard to punishment sensitivity, psychopathy—which entails insensitivity to punishment—is positively associated with testosterone in men when under stress (as indicated by heightened cortisol; Welker, Lozoya, Campbell, Neumann, & Carré,  2014). Thus, by amplifying rewards and tamping down sensitivity to punishment, testosterone may serve to enhance short-term mating success by increasing the likelihood of enacting courtship displays and aggressive behaviors, both of which entail some degree of risk.

Reproductive Behavior in the Human Male

Aggression as a Manifestation of Mating Effort

Male mating effort is marked by heighted violence and aggression (Ainsworth & Maner, 2012; Archer, 2006; Griskevicius et al., 2009), particularly in mating markets in which long-term nurturant ­romantic relationships are not pursued (Wilson & Daly,  1985). Young, unmarried men commit the lion’s share of violence (Daly & Wilson, 1988; Wilson & Daly, 1985). Regions with more men have more violent crime (an indicator of increased male–male competition) than regions in which the operational sex ratio9 is female biased (Schacht, Tharp, & Smith, 2016). Marriage has been shown to decrease recidivism (e.g., Horney, Osgood, & Marshall, 1995) and, in turn, marriage is associated with lower testosterone (e.g., Mazur & Michalek, 1998). Finally, the development of men’s sexually dimorphic morphological features (e.g., upper body strength) are mediated by androgens (Hodges-Simeon et al., 2016; Puts, Jones, & DeBruine,  2012) and enhance men’s physical dominance (Puts, 2010; see Table 1 in Sell, Hone, & Pound, 2012, for a list of putative adaptations for combat in men). Taken together, these data strongly support the notion that aggression is  an adaptive expression of heightened male mating effort (i.e., an adaptive product of intramale sexual selection). Aggression, viewed in this context, comports well with the challenge hypothesis (discussed earlier; Wingfield et al., 1990). Heightened mating effort is supported by a facultative adjustment to testosterone levels, enhancing men’s competitive abilities and motivations (Archer, 2004). Under this view, men following a fast life history strategy are expected to show heightened androgen activity (both baseline levels and responsivity to reproductive cues and contexts). Indeed, in a recent study, Zilioli and colleagues (2016) found that men’s testosterone response to erotica was conditioned upon their interest in babies (an indicator of parental effort), such that men who reported less interested in babies secreted more testosterone than men who reported more interest. Furthermore, interest in babies was related to scores on the Mini-K (a self-report measure of life history strategy; Figueredo et al., 2005); however, testosterone reactivity was not related to life history strategy.10 9   Operational sex ratio is the ratio of potentially receptive males to receptive females (Emlen & Oring, 1977). 10   Despite this lack of association, the mini-K has been found to be associated with a short-term mating orientation (Dunkel

Aggression, although detrimental to society and potentially harmful to the individual, nevertheless serves an adaptive function (Archer,  2006; Griskevicius et al., 2009). Men who successfully used aggression in the context of intramale competition to gain sexual access to women and the means toward achieving access (e.g., status) left more descendants than men without these traits. Indeed, in one survey study, when asked to recall the motivating reasons behind their last act of aggression, nearly half of the participants cited status/reputation concerns as the underlying reason (Griskevicius et al., 2009). Moreover, priming mating motives has been shown to increase aggression directed at male competitors (Ainsworth & Maner,  2012; Griskevicius et al., 2009). For instance, men primed with either a courtship or competition motive report a greater willingness to use direct aggression in response to provocation when the target is another man (Griskevicius et al., 2009), and these results generalize beyond self-report aggression (Ainsworth & Maner, 2012). Thus, contexts involving competition for status and mates are often characterized by heightened aggression and testosterone (e.g., Wilson & Daly, 1985). In addition to the challenge hypothesis, Mazur’s “biosocial model of status” provides a functional explanation for the observed effect of competitive outcomes on changes in testosterone (Mazur, 1976, 1985; Mazur & Booth, 1998). The so-called winner/ loser effect, whereby winning a competition tends to increase testosterone whereas losing a competition tends to decrease testosterone, under Mazur’s model is purposed to a­ daptively guide subsequent dominance/deference behaviors; increases in testosterone function to induce behavioral prerogatives of dominance (e.g., aggression directed down the hierarchy and approaching potential mates), whereas decreases serve to dissuade subsequent attempts at gaining dominance over the current competitor. Androgen responses to winning and losing competitions have been observed across numerous taxa. et al., 2010; McDonald, Donnellan, & Navarrete, 2012; Patch & Figueredo, 2016; Strouts, Brase, & Dillon, 2016); in turn, some studies have found that short-term mating orientation is linked with heightened testosterone levels (Edelstein, Chopik, & Kean, 2011; McIntyre et al., 2006; Puts et al., 2015), although several others have failed to find this association (Charles & Alexander, 2011; Farrelly, Owens, Elliott, Walden, & Wetherell, 2015; van Anders et al., 2007). More research directly evaluating the potential relationship between life history strategies and testosterone are warranted. In particular, research into the context in which testosterone responses are induced (e.g., male–male competition) is needed.

Goetz, Weisfeld, and Zilioli

131

In humans, the winner/loser effect has been observed in athletes and their fans, chess players, voters, and the laboratory (for a review of field studies, see Carré & Olmstead, 2015). In the lab, changes in testosterone in response to a competitive outcome have been associated with an increased motivation for future competition, competitive ability, and aggressive behavior, supporting the second half of Mazur’s model (Carré, Campbell, Lozoya, Goetz, & Welker, 2013; Carré & McCormick, 2008; Mehta & Josephs,  2006; Zilioli & Watson, 2014). Mehta and Josephs (2006), using a rigged competition, found that those that lost a competition but still evinced an increase in testosterone were more willing to compete again compared to those that decreased in testosterone. Similarly, Carré and McCormick (2008) showed that men who increased in testosterone in response to behaving aggressively were more likely to choose a competitive task relative to men who did not show an increase in testosterone. In a later study, Carré and colleagues (2013) found that the positive relationship between winning a task and subsequent aggressive behavior was mediated by changes in testosterone, such that increases in testosterone predicted heightened aggression. Subsequent work indicates that these changes may function to increase men’s perceptions of their own dominance (Welling, Moreau, Bird, Hansen, & Carré,  2016). Specifically, in a double-blind, placebo-controlled study, exogenous testosterone was shown to enhance men’s ratings of their own dominance, particularly in men in whom the exogenous testosterone administration produced a large percent increase in testosterone (i.e., those with lower baseline endogenous testosterone levels). Although these studies all involved single-­ interaction episodes, in the environment of evolutionary adaptiveness,11 relationships would involve encountering rivals over multiple episodes. Zilioli and Watson (2014) sought to address this gap by ­designing a study in which men competed across two consecutive days. The usual winner/loser effect was observed, but, importantly, changes in testosterone on the first day were positively associated with competitive task performance on the second day, replicating past murine models (e.g., Fuxjager, Montgomery, & Marler,  2011). Also, consistent   The environment of evolutionary adaptiveness denotes the series of ancestral environments in which selection pressures sculpted an adaptation. It is not a specific time or place, but rather a statistical composite of selection pressures that shaped the adaptive design of an organism over generations (Cosmides & Tooby, 1997). 11

132

with the challenge hypothesis, unstable hierarchies (i.e., pairs of men in which each experienced both victory and defeat) induced the largest increase in testosterone on the second day, in contrast to men who lost on both days, indicating that unstable hierarchies represent a challenge to men’s reproductive success and would hence benefit from increased competitive motivation. Indeed, unstable hierarchies consistently produce heightened aggression and testosterone in nonhuman primates (Holekamp & Strauss, 2016). In sum, aggression and mating effort in men are clearly linked and play a functional role in men’s reproductive lives. Although increased mating effort at the expense of parental and somatic effort has been viewed as indicative of following a fast life history strategy, direct evidence relating individual differences in testosterone and aggression to individual differences in life history strategy is lacking. Future research in this field that incorporates life history measures is warranted.

Mate Seeking, Testosterone, and Life History Strategy

Testosterone and mating success have been observed to be positively associated in humans12 (Peters, Simmons, & Rhodes,  2008; Pollet, van der Meij, Cobey, & Buunk, 2011), indicating that testosterone may function to orient men’s efforts toward courtship behaviors. Indeed, even brief exposures of men to attractive women induce testosterone release (Roney, Mahler, & Maestripieri, 2003), and men’s changes in testosterone during such interactions are associated with women’s ratings of his extraversion and degree of self-disclosure (Roney, Lukaszewski, & Simmons,  2007). Field studies, which achieve greater ecological validity, also find this exposure effect (e.g., Flinn, Ponzi, & Muehlenbein,  2012). Moreover, testosterone appears to facilitate entry into committed romantic relationships. In one study, men with high testosterone at age 21 were more likely to be partnered fathers by age 26 compared to men with low testosterone at age 21 (Gettler, McDade, Feranil, & Kuzawa, 2011). Changes in testosterone in response to the presence of an attractive woman are hypothesized to 12   There may exist a negative feedback loop in which satisfying one’s sexual desire decreases testosterone. When controlling for sociosexual interest, number of sexual partners negatively predicted baseline testosterone levels (Puts et al., 2015), and sexual activity has been found to be associated with lower testosterone levels (Sakaguchi et al., 2007, but see Gettler, McDade, Agustin, Feranil, & Kuzawa, 2013).

Reproductive Behavior in the Human Male

index a positive decision to pursue mating (Roney & Gettler, 2015). Contexts in which mating is precluded or unwanted (e.g., due to sexual satiety) do not induce a rise in testosterone. For instance, subordinate monkeys in the presence of both a receptive female and a dominant male do not produce an increase in testosterone, likely because the dominant male would thwart any attempt at mating (Glick, 1984). Similarly, in mating contexts, the perceived presence of a dominant man can suppress creative displays (e.g., telling funny jokes) of subordinate men (Gambacorta & Ketelaar, 2013). Social context also modulates testosterone responses to potential mates. Flinn and colleagues (2012) documented that men in a Dominican village evinced a testosterone increase when exposed to women, but only if those women were women from another village and not in relationships with kinsmen or friends. Testosterone’s role in promoting mating has been summarized in Roney and Gettler’s (2015) testosterone-relationship cycle model. The testosterone-­ relationship cycle model proposes that testosterone functions to promote mate pursuit, increasing the likelihood of entering into a romantic relationship. However, once in a relationship, mate pursuit may interfere with bond stability and effective parenting. Thus, nurturant relationships are suggested to impose negative feedback on testosterone production and, consequently, mating effort (Roney & Gettler, 2015). For instance, administering testosterone to male house sparrows after pairing prolongs courtship behaviors but reduces offspring survival (Hegner & Wingfield,  1987). In human couples, high testosterone is associated with relationship dissatisfaction and partners of high-testosterone individuals are less satisfied and committed to the relationship (Edelstein, van Anders, Chopik, Goldey, & Wardecker, 2014). Indeed, men’s testosterone positively predicts future divorce (Booth & Dabbs, 1993). This negative feedback will be the focus of the upcoming section on parenting effort. In addition to social constraints, genetic polymorphisms and hypothalamic-pituitary-adrenal axis (HPA) hormones also modulate testosterone responses to mating opportunities. Roney, Simmons, and Lukaszewski (2010) showed that androgen receptor (AR) gene variants, along with cortisol, moderated the effect of exposure to a potential mate on men’s testosterone responses. Specifically, men with fewer CAG repeats (which is associated with greater AR expression; Choong, Kemppainen, Zhou, & Wilson, 1996) and low basal cortisol demonstrated a larger increase in testosterone following a brief

interaction with a young woman than men with more CAG repeats and high basal cortisol. Variations in CAG repeat length predict variables associated with intrasexual competitiveness such as upper body strength and dominance but were not associated with mating strategy, although mating strategy was only assessed via self-reported sociosexuality (Simmons & Roney, 2011).

Coercive Mating Strategies

Sexually coercive mating tactics have been observed across a variety of species and taxa (McKibbin, Shackelford, Goetz, & Starratt, 2008) and are more common in species in which males are more aggressive, eager to mate, sexually assertive, and less discriminating in mating decisions (Thornhill & Palmer, 2000). Although there is clear evidence of adaptations for rape in some species (e.g., scorpionflies have a notal organ designed to restrain females; Thornhill, 1980), evolutionary psychologists have long debated whether sexual coercion in humans represents an adaptation or whether it is a byproduct of other adaptations (i.e., heightened aggression and desire for sexual variety). The adaptationist perspective posits that sexual coercion was selected over evolutionary history as a facultative male reproductive tactic that increased reproductive success by increasing the number of sexual partners (e.g., Thornhill & Thornhill, 1983). Also, when comparing per-incident pregnancy rates between consensual and nonconsensual vaginal sex, nonconsensual pregnancy rates are twice as high (6.42 percent vs. 3.1 percent), and climb to two and a half times higher when adjusted for contraception use (Gottschall & Gottschall,  2003). In addition to increasing partner number and conception rates, sexual coercion within an existing relationship may function as a sperm competition mechanism, an adaptation to the adaptive problems associated with social monogamy, primarily the problem of paternity uncertainty (e.g., Goetz & Shackelford, 2006). In many bird species forming socially monogamous pairs, forced in-pair copulation frequently occurs after extra-pair copulations or situations in which extra-pair copulation may have occurred, paralleling findings in humans and consistent with the interpretation that forced in-pair copulation functions as an anticuckoldry tactic (Goetz & Shackelford, 2006). On the other hand, sexual coercion may be a byproduct of male adaptive desire for sexual variety and male adaptation for heightened aggression (Palmer, 1991; Thornhill & Palmer, 2000). Goetz, Weisfeld, and Zilioli

133

Long-Term Strategies

Marriage is a human universal (Walker, Hill, Flinn, & Ellsworth, 2011). However, ceteris paribus, the fitness benefits of men pursuing short-term mating strategies outweigh the benefits of pursuing longterm mating strategies, which begs the question as to why, then, so many men commit significant portions of their adult lives to a single sexual union (Henrich, Boyd, & Richerson, 2012). The long history and high frequency of monogamous marriage practices suggest that under some conditions, long-term pairbonding proved evolutionarily advantageous (Walker et al., 2011; van Anders, Goldey, & Kuo, 2011). Evolutionary psychologists have proposed several reasons that men might pursue a committed relationship. First, men may pursue a long-term mating strategy because it is imposed on them by women. That is, to gain sexual access to some women, men might have to signal their commitment to the union, a critical evolutionary concern to ancestral women given the cost of bearing offspring (e.g., Schact, Tharp, & Smith, 2016). Second, by conceding to a long-term strategy, men might gain access to more desirable women, provided that what men find desirable reflects women’s fecundity. If women desire a committed mate, a more desirable woman is in a better position to find a man willing to accept the costs of abandoning short-term mating strategies (Perilloux, Cloud, & Buss, 2013). Third, men choosing to pursue a long-term strategy increase paternity certainty. Forming a longer union entails greater sexual access to one mate, which reduces the opportunity for the woman to mate with a different man, and if the woman does engage in sexual intercourse with another man, greater sexual access granted by the long-term pair-bond reduces the likelihood that the interloper will father her offspring (Bethmann & Kvasnicka,  2011). Finally, forming a stable union with a woman provides the father with greater ­opportunity to directly invest in offspring through parenting and provisioning of resources, both physical and social (reviewed in Geary, 2000). Women’s imposition of commitment is reflected in mating market dynamics. Indeed, when the adult sex ratio (ASR; i.e., ratio of sexually mature men to women present in a population) is male biased (i.e., more men than women), the dynamics of the mating market ostensibly push men to cater more to women’s preferred mating desires (i.e., committed relationships;13 e.g., Schacht, Tharp, & Smith, 2016). 13   This is not to say that women do not often desire shortterm mating. For instance, apart from the benefit of possible

134

Conversely, under female-based ratios, uncommitted strategies are more common (Mattingly et al., 2011; Schmitt, 2005). This may seem counterintuitive from a resource scarcity perspective (if sexually accessible women are construed as a resource) in which men are limited in their mating opportunities and are thus expected to compete more vigorously for sexual access to women. However, it appears that the way in which men in industrialized societies compete for women when they are rare is not to increase direct intramale competition, but rather to appear attractive to women by conforming to women’s sexual desires (e.g., Schacht, Tharp, & Smith,  2016). Furthermore, research on the effects of pair-bonding and testosterone—a putative facilitatory mechanism to mating effort—generally indicates that committed partnering decreases testosterone (reviewed in Gray, McHale, & Carré, 2016). However, a recent study using longitudinal data also indicates that lower testosterone levels predict pairing (Dibble, Goldey, & van Anders, 2016). To the extent that lower testosterone indicates a less masculinized pattern of behavior (Manuck et al.,  2010)—in this case, attenuated short-term mating orientation (Buss & Schmitt, 1993; Schmitt, 2005; Schmitt et al., 2012; Symons, 1979)—this result could indicate that entry (in the near term14) into a long-term mating strategy is facilitated by matching women’s preferences. Alternatively, sexual activity may be heightened around the time of relationship initiation and thus could be producing negative feedback on testosterone levels (Puts et al., 2015; Sakaguchi, Oki, Honma, Uehara, & Hasegawa, 2007). Several studies have found that women’s physical attractiveness correlates with fertility (Hill & Hurtado, 1996; Jokela, 2009; Pflüger, Oberzaucher, Katina, Holzleitner, & Grammer, 2012). In the Ache of Paraguay, women with high facial attractiveness were found to have 1.16 times higher fertility than less attractive women (Hill & Hurtado,  1996, immediate material gains or genetic benefits to offspring, shortterm mating can be used as a strategy for defraying the costs of leaving a relationship (e.g., costs of physical threat from a former partner; Buss, Goetz, Duntley, Asao, & Conroy-Beam, 2017). 14   A longitudinal study of marriage and testosterone in the Philippines found the opposite effect; higher testosterone among unmarried men was associated with being married five years later (Gettler et al., 2011). This result has been interpreted as indicating that testosterone benefits men on the mating market, granting them greater success at procuring a mate. However, on shorter intervals (months rather than years), perhaps once intramale competition for a particular woman has been quelled, lower testosterone facilitates signals of commitment (e.g., nurturant behaviors).

Reproductive Behavior in the Human Male

­ . 312). Likewise, in a sample of U.S. women, attracp tive women had 16 percent more children than less attractive women ( Jokela, 2009). Finally, in a small rural community of Austrians, both the total number of pregnancies and the total number of children predicted ratings of facial attractiveness (Pflüger et al., 2012). Thus, attractive women may be able to more easily parlay their heightened mate value into pressuring men to follow their preferred reproductive strategy (i.e., committed relationship). Indeed, more attractive women are better able to actualize their current mating goals (Perilloux, Cloud, & Buss, 2013), which often center on commitment (Buss & Schmitt, 1993; Schmitt et al., 2012). As an altricial species, offspring survival is enhanced by provisioning from the father (Bribiescas, Ellison, & Gray, 2012). However, paternity uncertainty poses an inherent challenge to men providing paternal investment. Before the advent of paternity testing, a man could not be 100 percent certain that his children were in fact his and not the product of an affair. Indeed, some estimates of nonpaternity rates are as high as 20 percent in low-socioeconomic settings, whereas in other cultural contexts they are as low as 1 percent (see Geary, 2005, for a review). Research from cultural anthropology indicates that in cultures in which paternity uncertainty is high, male investment in offspring tends to follow an “avuncular” investment strategy in which men invest not in their own offspring, but rather in their sisters’ offspring (Flinn, 1981; Flinn & Low, 1986). One of the functions of marriage, then, might be an increase in men’s paternity certainty by simultaneously conferring greater sexual access to the husband while reducing extra-dyadic mating opportunities for the wife (Bethmann & Kvasnicka, 2011).

Parenting Effort

As discussed in the mating effort section, testosterone has been linked with increased mating effort (Gray et al., 2016). Indeed, one longitudinal study of men in the Philippines found that higher testosterone positively predicted the likelihood of being married 4.5 years later, suggesting that testosterone enhances mating effort (Gettler et al.,  2011). Conversely, lower testosterone has been shown to be associated with parenting effort. Studies of partnering status (single vs. partnered) have repeatedly found that testosterone is lower among men in monoamorous relationships (Booth & Dabbs, 1993; Burnham et al., 2003; Dibble et al., 2016; Edelstein, Chopik, & Kean, 2011; Gray, Kahlenberg, Barrett, Lipson, & Ellison,  2002; Gray et al.,  2004; Gray, Ellison, &

Campbell, 2007; Sakaguchi et al., 2007; van Anders & Watson, 2006, 2007; van Anders, Hamilton, & Watson, 2007; van Anders & Goldey, 2010). Being in a relationship, however, is not always associated with lower testosterone. Relationship status, per se, does not appear to be the primary cause; rather, how men approach the relationship (relationship orientation) appears to be the key variable along which testosterone varies (e.g., Edelstein et al., 2011; van Anders & Goldey, 2010). For instance, sociosexuality plays a moderating role (Edelstein et al., 2011; van Anders & Goldey, 2010, but see van Anders et al., 2007). Men reporting a desire for uncommitted sexual activity had similar levels of testosterone as single men, indicating that attitudes toward commitment play a key role in whether being in a relationship decreases mating effort (as indicated by testosterone). Indeed, sociosexual desire for uncommitted mating prospectively predicts relationship dissolution (Penke & Asendorpf, 2008), and testosterone increases around the time of divorce but decreases with remarriage (Mazur & Michalek, 1998). Additionally, the transition from a committed to a noncommitted relationship cooccurs with an increase in testosterone (Dibble ­et al., 2016), and testosterone is negatively associated with commitment to and satisfaction with a romantic relationship (Edelstein et al., 2014). Taken together, these data indicate that high testosterone may be detrimental to the maintenance of nurturant relationships. Much of the ethnographic data indicates that norms associated with marriage and childcare moderate the relationship between testosterone and pairing. For instance, in a sample of high-socioeconomic men (university sample) from Beijing, married and unmarried men did not differ in testosterone. However, married men with children did have significantly lower testosterone than unmarried and childless married men, indicating that norms of commitment to a union may be relaxed until a child has been produced. Indeed, among the married nonfathers, many reported having had more than one sex partner during the past year or anticipated having more than one sexual partner in the next five years (Gray, Yang, & Pope, 2006). Also, similar to Western samples (e.g., Edelstein et al.,  2014), testosterone was negatively, albeit not significantly, correlated with relationship satisfaction (Gray et al.,  2006). Similarly, in the Philippines, the relationship between testosterone and marriage was found to be completely mediated by fatherhood, indicating that norms of direct parental care are a driving force Goetz, Weisfeld, and Zilioli

135

mediating the trade-off between mating and parenting effort (Kuzawa, Gettler, Muller, McDade, & Feranil, 2009). In a subsequent study on this population, childcare was shown to further reduce testosterone levels (Gettler et al., 2011). Other cross-cultural research indicates that direct paternal care is associated with decreased testosterone (Alvergne, Faurie, & Raymond, 2009; Gray, 2003; Muller, Marlowe, Bugumba, & Ellison, 2009). Among Senegalese men, a polygynous population but with a high degree of paternal care, fathers had lower testosterone levels than nonfathers. Furthermore, paternal investment in the family (as rated by the mother) was negatively associated with testosterone (Alvergne et al., 2009). Similarly, among a sample of Swahili men, men with younger children had lower testosterone than men with older children, but this relationship was not significant (Gray, 2003). Among the Hadza, a foraging tribe in which fathers invest heavily in direct paternal care, fathers had lower testosterone than nonfathers, and the magnitude of the testosterone decrement was negatively associated with the age of the father’s youngest child. By contrast, fathers and nonfathers among the Datoga pastoralists, who invest little direct paternal care, did not differ in testosterone levels (Muller et al., 2009). Apart from direct investment in offspring, another indicator of commitment that has been found to be associated with testosterone across cultures is whether marriages are monogamous or polygynous; however, the direction of findings has been mixed. For instance, both Swahili and Senegalese men in polygynous marriages were found to have higher testosterone than monogamously married men (Senegalese: Alvergne et al.,  2009; Swahili: Gray, 2003). However, in two separate populations of pastoralists, monogamously compared to polygynously married men did not differ in testosterone levels (Datoga: Muller et al.,  2009; Ariaal: Gray et al., 2007). Intriguingly, Marlowe (2000) has shown that the level of parental care is related to subsistence mode, with the lowest levels found in pastoralists. Furthermore, among pastoralists, wealth and political power inequalities tend to be greater, potentially enabling powerful men to gain more wives through means (e.g., political power) not requiring dispositional mating effort (e.g., aggressive risk taking). Setting aside these differences, Western men pursuing polyamorous relationships have higher testosterone than men in monoamorous relationships (van Anders et al., 2007). Taken together, commitment—whether indicated by paternal investment or monoamory—is 136

c­ onsistently associated with lower levels of testosterone, supporting the notion that testosterone ­mediates the trade-off in reproductive effort between mating and parenting effort (Gray et al., 2016).

Conclusion

Life history theory provides a useful theoretical lens through which to investigate individual differences in many prosocial and antisocial behaviors. Furthermore, the adoption of an adaptive plasticity perspective has provided consilience with a variety of developmental fields that hitherto lacked theoretical depth. By recognizing the trade-offs inherent in allocating resources into any one life history trait, researchers are gaining a deeper understanding of the evolutionary pressures that have shaped individual differences, particularly in the domain of reproductive strategies. Because the cost–benefit ratios associated with these trade-offs are not constant across developmental contexts, differences in allocation schemes, constituting conditional adaptations, maximize reproductive success across conditions. For instance, in socioecological contexts in which paternal investment is closely associated with offspring reproductive success, monogamous mating strategies emerge. A large body of research indicates that the conditional expression of these alternative reproductive strategies is mediated by the neuroendocrine system. In particular, testosterone has been implicated in mediating the trade-off between mating and parental investment. Further incorporation of a life history perspective into the study of social neuroendocrinology will likely prove fruitful in understanding individual differences in neuroendocrine profiles.

References

Ainsworth, S.  E., & Maner, J.  K. (2012). Sex begets violence: Mating motives, social dominance, and physical aggression in men. Journal of Personality and Social Psychology, 103, 819–829. Altmann, S. A. (1962). A field study of the sociobiology of rhesus monkeys, Macaca mulatta. Annals of the New York Academy of Sciences, 102, 338–435. Alvergne, A., Faurie, C., & Raymond, M. (2009). Variation in testosterone levels and male reproductive effort: Insights from a polygynous human population. Hormones and Behavior, 56, 491–497. Anderson, C., John, O. P., Keltner, D., & Kring, A. M. (2001). Who attains social status? Effects of personality and physical attractiveness in social groups. Journal of Personality and Social Psychology, 81, 116–132. Apicella, C. L., Dreber, A., Campbell, B., Gray, P. B., Hoffman, M., & Little, A. C. (2008). Testosterone and financial risk preferences. Evolution and Human Behavior, 29, 384–390.

Reproductive Behavior in the Human Male

Apicella, C.  L., Dreber, A., & Mollerstrom, J. (2014). Salivary testosterone change following monetary wins and losses predicts future financial risk-taking. Psychoneuroendocrinology, 39, 58–64. Archer, J. (2004). Sex differences in aggression in real-world settings: A meta-analytic review. Review of General Psychology, 8, 291–322. Archer, J. (2006). Testosterone and human aggression: An evaluation of the challenge hypothesis. Neuroscience & Biobehavioral Reviews, 30, 319–345. Arnocky, S., Woodruff, N., & Schmitt, D.  P. (2016). Men’s sociosexuality is sensitive to changes in mate availability. Personal Relationships, 23, 172–181. Baker, M.  D., & Maner, J.  K. (2008). Risk-taking as a situationally sensitive male mating strategy. Evolution and Human Behavior, 29, 391–395. Bass, A. H. (1996). Shaping brain sexuality. American Scientist, 84, 352–363. Bateman, P. W., & Bennett, N. C. (2006). The biology of human sexuality: Evolution, ecology and physiology. Verbum et Ecclesia, 27, 245–264. Belsky, J., Steinberg, L., & Draper, P. (1991). Childhood experience, interpersonal development, and reproductive strategy: An evolutionary theory of socialization. Child Development, 62, 647–670. Bethmann, D., & Kvasnicka, M. (2011). The institution of marriage. Journal of Population Economics, 24, 1005–1032. Betzig, L. (1986). Despotism and differential reproduction: A Darwinian view of history. Hawthorne, NY: Aldine. Betzig, L. (2012). Means, variances, and ranges in reproductive success: Comparative evidence. Evolution and Human Behavior, 33, 309–317. Boissy, A., & Bouissou, M.  F. (1994). Effects of androgen treatment on behavioral physiological responses of heifers to fear-eliciting situations. Hormones and Behavior, 28, 66–83. Booth, A., & Dabbs, J.  M., Jr. (1993). Testosterone and men’s marriages. Social Forces, 72, 463–477. Bribiescas, R. G. (2001). Reproductive ecology and life history of the human male. American Journal of Physical Anthropology, 116, 148–176. Bribiescas, R. G., Ellison, P. T., & Gray, P. B. (2012). Male life history, reproductive effort, and the evolution of the genus Homo: New directions and perspectives. Current Anthropology, 53, S424–S435. Burnham, T. C., Chapman, J. F., Gray, P. B., McIntyre, M. H., Lipson, S. F., & Ellison, P. T. (2003). Men in committed romantic relationships have lower testosterone. Hormones and Behavior, 44, 119–122. Buss, D.  M., Goetz, C., Duntley, J.  D., Asao, K., & ConroyBeam, D. (2017). The mate switching hypothesis. Personality and Individual Differences, 104, 143–149. Buss, D. M., & Schmitt, D. P. (1993). Sexual strategies theory: An evolutionary perspective on human mating. Psychological Review, 100, 204–232. Campbell, B. C., Dreber, A., Apicella, C. L., Eisenberg, D. T., Gray, P. B., Little, A. C., . . . Lum, J. K. (2010). Testosterone exposure, dopaminergic reward, and sensation-seeking in young men. Physiology and Behavior, 99, 451–456. Carré, J.  M., Campbell, J.  A., Lozoya, E., Goetz, S.  M., & Welker, K.  M. (2013). Changes in testosterone mediate the  effect of winning on subsequent aggressive behaviour. Psychoneuroendocrinology, 38, 2034–2041.

Carré, J. M., & McCormick, C. M. (2008). Aggressive behavior and change in salivary testosterone concentrations predict willingness to engage in a competitive task. Hormones and Behavior, 54, 403–409. Carré, J. M., McCormick, C. M., & Hariri, A. R. (2011). The social neuroendocrinology of human aggression. Psychoneuroen­ docrinology, 36, 935–944. Carré, J. M., McCormick, C. M., & Mondloch, C. J. (2009). Facial structure is a reliable cue of aggressive behavior. Psychological Science, 20, 1194–1198. Carré, J. M., & Olmstead, N. A. (2015). Social neuroendocrinology of human aggression: Examining the role of competitioninduced testosterone dynamics. Neuroscience, 286, 171–186. Charles, N.  E., & Alexander, G.  M. (2011). The association between 2D:4D ratios and sociosexuality: A failure to replicate. Archives of Sexual Behavior, 40, 587–595. Cheng, J. T., Tracy, J. L., Foulsham, T., Kingstone, A., & Henrich, J. (2013). Two ways to the top: Evidence that dominance and prestige are distinct yet viable avenues to social rank and influence. Journal of Personality and Social Psychology, 104, 103–125. Chisholm, J. S. (1993). Death, hope, and sex: Life-history theory and the development of reproductive strategies. Current Anthropology, 34, 1–24. Choong, C. S., Kemppainen, J. A., Zhou, Z., & Wilson, E. M. (1996). Reduced androgen receptor gene expression with first exon CAG repeat expansion. Molecular Endocrinology, 10, 1527–1535. Corby, B. C. (2006). Attachment and adjustment in preadolescence (Unpublished doctoral dissertation). Ft. Lauderdale: Florida Atlantic University. Cosmides, L., & Tooby, J. (1997). Evolutionary psychology: A primer. University of California at Santa Barbara: Center for Evolutionary Psychology. Cox, R. M., McGlothlin, J. W., & Bonier, F. (2016). Hormones as mediators of phenotypic and genetic integration: An evolutionary genetics approach. Integrative and Comparative Biology, 56, 126–137. Cummins, D. D. (2000). How the social environment shaped the evolution of mind. Synthese, 122, 3–28. Daly, M., & Wilson, M. (1988). Homicide. New York: Aldine de Gruyter. Day, F. R., Helgason, H., Chasman, D. I., Rose, L. M., Loh, P.R., Scott, R.  A., . . . Perry, J.  R.  B. (2016). Physical and neurobehavioral determinants of reproductive onset and success. Nature Genetics, 48, 617–625. Del Giudice, M. (2008). Sex-biased ratio of avoidant/ambivalent attachment in middle childhood. British Journal of Developmental Psychology, 26, 369–379. Del Giudice, M. (2009). Sex, attachment, and the development of reproductive strategies. Behavioral and Brain Sciences, 32, 1–67. Del Giudice, M., Ellis, B.  J., & Shirtcliff, E.  A. (2011). The adaptive calibration model of stress responsivity. Neuroscience and Biobehavioral Reviews, 35, 1562–1592. Dibble, E. R., Goldey, K. L., & van Anders, S. M. (2016). Pair bonding and testosterone in men: Longitudinal evidence for trait and dynamic associations. Adaptive Human Behavior and Physiology, 3, 1–20. Draper, P., & Harpending, H. (1982). Father absence and reproductive strategy: An evolutionary perspective. Journal of Anthropological Research, 38, 255–273. Dubuc, C., Muniz, L., Heistermann, M., Engelhardt, A., & Widdig, A. (2011). Testing the priority-of-access model in a

Goetz, Weisfeld, and Zilioli

137

seasonally breeding primate species. Behavioral Ecology and Sociobiology, 65, 1615–1627. Dunkel, C., Mathes, E., Decker, M. (2010). Behavioral flexibility in life history strategies: The role of life expectancy. Evolutionary Behavioral Sciences, 4, 51–61. Edelstein, R. S., Chopik, W. J., & Kean, E. L. (2011). Sociosexuality moderates the association between testosterone and relationship status in men and women. Hormones and Behavior, 60, 248–255. Edelstein, R. S., van Anders, S. M., Chopik, W. J., Goldey, K. L., & Wardecker, B.  M. (2014). Dyadic associations between testosterone and relationship quality in couples. Hormones and Behavior, 65, 401–407. Eisenegger, C., Haushofer, J., & Fehr, E. (2011). The role of testosterone in social interaction. Trends in Cognitive Sciences, 15, 263–271. Ellis, B.  J., Del Giudice, M., Dishion, T.  J., Figueredo, A.  J., Gray, P., Griskevicius, V., . . . Wilson, D.  S. (2012). The evolutionary basis of risky adolescent behavior: Implications for science, policy, and practice. Developmental Psychology, 48, 598–623. Ellis, B.  J., Figueredo, A.  J., Brumbach, B.  H., & Schlomer, G. L. (2009). Fundamental dimensions of environmental risk. Human Nature, 20, 204–268. Emlen, S. T., & Oring, L. W. (1977). Ecology, sexual selection, and the evolution of mating systems. Science, 197, 215–223. Farrelly, D., Owens, R., Elliott, H.  R., Walden, H.  R., & Wetherell, M.  A. (2015). The effects of being in a “new relationship” on levels of testosterone in men. Evolutionary Psychology, 13, 250–261. doi:10.1177/147470491501300116. Figueredo, A. J., Vásquez, G., Brumbach, B.  H., Sefcek, J.  A., Kirsner, B. R., & Jacobs, W. J. (2005). The K-factor: Individual differences in life history strategy. Personality and Individual Differences, 39, 1349–1360. Finnegan, R. A., Hodges, E., & Perry, D. (1996). Preoccupied and avoidant coping during middle childhood. Child Development, 67, 1318–1328. Flinn, M. V. (1981). Uterine vs. agnatic kinship variability and associated cousin marriage preferences. In R. D. Alexander & D. W. Tinkle (Eds.), Natural selection and social behavior (pp. 439–475). New York, NY: Chiron Press,. Flinn, M., & Low, B. (1986). Resource distribution, social competition, and mating patterns in human societies. In D. I. Rubenstein & R. W. Wrangham (Eds.), Ecological aspects of social evolution (pp. 217–243). Princeton, NJ: Princeton University Press. Flinn, M. V., Ponzi, D., & Muehlenbein, M. P. (2012). Hormonal mechanisms for regulation of aggression in human coalitions. Human Nature, 23, 68–88. Fuxjager, M.  J., Montgomery, J.  L., & Marler, C.  A. (2011). Species differences in winner effect disappear in response to post-victory testosterone manipulations. Proceedings of the Royal Society of London B: Biological Sciences, 278, 3497–3503. doi:10.1098/rspb.2011.0301 Gambacorta, D., & Ketelaar, T. (2013). Dominance and deference: Men inhibit creative displays during mate competition when their competitor is strong. Evolution and Human Behavior, 34, 330–333. Geary, D.  C. (1998). Male, female: The evolution of human sex  differences. Washington, DC: American Psychological Association. Geary, D.  C. (2000). Evolution and proximate expression of human paternal investment. Psychological Bulletin, 126, 55–77.

138

Geary, D.  C. (2005). Evolution of paternal investment. In D. M. Buss (Ed.), The evolutionary psychology handbook (pp. 483–505). Hoboken, NJ: John Wiley & Sons. Gettler, L. T., McDade, T. W., Agustin, S. S., Feranil, A. B., & Kuzawa, C. W. (2013). Do testosterone declines during the transition to marriage and fatherhood relate to men’s sexual behavior? Evidence from the Philippines. Hormones and Behavior, 64, 755–763. Gettler, L. T., McDade, T. W., Feranil, A. B., & Kuzawa, C. W. (2011). Longitudinal evidence that fatherhood decreases testosterone in human males. Proceedings of the National Academy of Sciences of the United States of America, 108, 16194–16199. Glick, B. B. (1984). Male endocrine responses to females: Effects of social cues in cynomolgus macaques. American Journal of Primatology, 6, 229–239. Goetz, A. T., & Shackelford, T. K. (2006). Sexual coercion and forced in-pair copulation as sperm competition tactics in humans. Human Nature, 17, 265–282. Gottschall, J.  A., & Gottschall, T.  A. (2003). Are per-incident rape-pregnancy rates higher than per-incident consensual pregnancy rates? Human Nature, 14, 1–20. Granot, D., & Mayseless, O. (2001). Attachment security and adjustment to school in middle childhood. International Journal of Behavioral Development, 25, 530–541. Gray, P. B. (2003). Marriage, parenting, and testosterone variation among Kenyan Swahili men. American Journal of Physical Anthropology, 122, 279–286. Gray, P. B., Chapman, J. F., Burnham, T. C., McIntyre, M. H., Lipson, S.  F., & Ellisoin, P.  T. (2004). Human male pair bonding and testosterone. Human Nature, 15, 119–131. Gray, P. B., Ellison, P. T., & Campbell, B. C. (2007). Testosterone and marriage among Ariaal men of Northern Kenya. Current Anthropology, 48, 750–755. Gray, P. B., Kahlenberg, S. M., Barrett, E. S., Lipson, S. F., & Ellison, P. T. (2002). Marriage and fatherhood are associated with lower testosterone in males. Evolution and Human Behavior, 23, 193–201. Gray, P. B., McHale, T. S., & Carré, J. M. (2016). A review of human male field studies of hormones and behavioral reproductive effort. Hormones and Behavior, 91, 52–67. Gray, P.  B., Yang, C.  J., Pope, H.  G., Jr. (2006). Fathers have lower salivary testosterone levels than unmarried men and married non-fathers in Beijing, China. Proceedings of the Royal Society B, 273, 333–339. Greeley, A. (1994). Marital infidelity. Society, 31, 9–13. Grevicius, V., Tybur, J.  M., Gangestad, S.  W,. Perea, E.  F, Shapiro, J. R., & Kenrick D. T. (2009). Aggress to impress: Hostility as an evolved context-dependent strategy. Interpersonal Relations and Group Processes, 96, 980–996. Griskevicius, V., Ackerman, J. M., Cantú, S. M., Delton, A. W., Robertson, T.  E., Simpson, J.  A., . . . Tybur, J.  M. (2013). When the economy falters, do people spend or save? Responses to resource scarcity depend on childhood environments. Psychological Science, 24, 197–205. Griskevicius, V., Tybur, J. M., Delton, A. W., & Robertson, T. E. (2011). The influence of mortality and socioeconomic status on risk and delayed rewards: A life history theory approach. Journal of Personality and Social Psychology, 100, 1015–1026. Gross, M.  R. (1991). Evolution of alternative reproductive strategies: Frequency-dependent sexual selection in male bluegill sunfish. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 332, 59–66.

Reproductive Behavior in the Human Male

Gross, M.  R. (1996). Alternative reproductive strategies and tactics: Diversity within sexes. Trends in Ecology & Evolution, 11, 92–98. Hegner, R. E., & Wingfield, J. C. (1987). Effects of experimental manipulation of testosterone levels on parental investment and breeding success in male house sparrows. The Auk, 104, 462–469. Henrich, J., Boyd, R., & Richerson, P. J. (2012). The puzzle of monogamous marriage. Philosophical Transactions of the Royal Society B, 367, 657–669. Henry, N., Ward, T., & Hirshberg, M. (2004). A multifactorial model of wartime rape. Aggression and Violent Behavior, 9, 535–562. Hill, K. (1993). Life history theory and evolutionary anthropology. Evolutionary Anthropology, 2, 78–88. Hill, K., & Hurtado, A. M. (1996). Ache life history: The ecology and demography of a foraging people. New York, NY: Hawthorne. Hodges-Simeon, C. R., Sobraske, K. N. H., Samore, T., Gurven, M., & Gaulin, S.  J. (2016). Facial width-to-height ratio (fWHR) is not associated with adolescent testosterone levels. PLoS One, 11(4), e0153083. Holekamp, K.  E., & Strauss, E.  D. (2016). Aggression and dominance: An interdisciplinary overview. Current Opinion in Behavioral Sciences, 12, 44–51. Horney, J., Osgood, D. W., & Marshall, I. H. (1995). Criminal careers in the short-term: Intra-individual variability in crime and its relation to local life circumstances. American Sociological Review, 60, 655–673. Huberman, B. A., Loch, C. H., & Önçüler, A. (2004). Status as a valued resource. Social Psychology Quarterly, 67, 103–114. Jokela, M. (2009). Physical attractiveness and reproductive success in humans: Evidence from the late 20th century United States. Evolution and Human Behavior, 30, 342–350. Jokela, M., Rotkirch, A., Rickard, I. J., Pettay, J., & Lummaa, V. (2010). Serial monogamy increases reproductive success in men but not in women. Behavioral Ecology, 21, 906–912. Karavasilis, L., Doyle, A.  B., & Markiewicz, D. (2003). Associations between parenting style and attachment to mother in middle childhood and adolescence. International Journal of Behavioral Development, 27, 153–164. Kocar, I. H., Yesilova, Z., Özata, M., Turan, M., Sengül, A., & Özdemir, I. Ç. (2000). The effect of testosterone replacement treatment on immunological features of patients with Klinefelter’s syndrome. Clinical & Experimental Immunology, 121, 448–452. Kuzawa, C. W., Gettler, L. T., Muller, M. N., McDade, T. W., & Feranil, A.  B. (2009). Fatherhood, pairbonding and testosterone in the Philippines. Hormones and Behavior, 56, 429–435. Laland, K.  N., & Brown, G.  R. (2011). Sense and nonsense: Evolutionary perspectives on human behavior. Oxford, UK: Oxford University Press. Manuck, S. B. (2010). The reaction norm in gene x environment interaction. Molecular Psychiatry, 15, 881–882. Manuck, S. B., Marsland, A. L., Flory, J. D., Gorka, A., Ferrell, R.  E., & Hariri, A.  R. (2010). Salivary testosterone and a trinucleotide (CAG) length polymorphism in the androgen receptor gene predict amygdala reactivity in men. Psychoneu­ roen­docrinology, 35, 94–104. Marlowe, F. (2000). Paternal investment and the human mating system. Behavioral Processes, 51, 45–61. Mattingly, B.  A., Clark, E.  M., Weidler, D.  J., Bullock, M., Hackathorn, J., & Blankmever, K. (2011). Sociosexual

orientation, commitment, and infidelity: A mediation analysis. Journal of Social Psychology, 151, 222–226. Maynard-Smith, J. (1982). Evolution and the theory of games. Cambridge, UK: Cambridge University Press. Mazur, A. (1976). Effects of testosterone on status in primate groups. Folia Primatologica, 26, 214–226. Mazur, A. (1985). A biosocial model of status in face-to-face primate groups. Social Forces, 64, 377–402. Mazur, A., & Booth, A. (1998). Testosterone and dominance in men. Behavior and Brain Science, 21, 353–363. Mazur, A., & Michalek, J. (1998). Marriage, divorce, and male testosterone. Social Forces, 77, 315–330. McDonald, M. M., Donnellan, M. B., & Navarrete, C. D. (2012). A life history approach to understanding the Dark Triad. Personality and Individual Differences, 52, 601–605. McIntyre, M.  Gangestad, S.  W., Gray, P.  B., Chapman, J.  F., Burnham, T. C., O’Rourke, M. T., & Thornhill, R. (2006). Romantic involvement often reduces men’s testosterone levels—but not always: The moderating role of extrapair sexual interest. Journal of Personality and Social Psychology, 91, 642–651. McKibbin, W. F., Shackelford, T. K., Goetz, A. T., & Starratt, V. G. (2008). Why do men rape? An evolutionary perspective. Review of General Psychology, 12, 86–97. McNamara, J. M., & Houston, A. I. (1996). State-dependent life histories. Nature, 380, 215–221. Mehta, P. H., & Josephs, R. A. (2006). Testosterone change after losing predicts the decision to compete again. Hormones and Behavior, 50, 684–692. Min, K. J., Lee, C. K., & Park, H. N. (2012). The lifespan of Korean eunuchs. Current Biology, 22, R792–R793. Mishra, S., & Lalumière, M. L. (2010). You can’t always get what you want: The motivational effect of need on risk-sensitive decision-making. Journal of Experimental Social Psychology, 46, 605–611. Muehlenbein, M.  P., Watts, D.  P., & Whitten, P.  L. (2004). Dominance rank and fecal testosterone levels in adult male chimpanzees (Pan troglodytes schweinfurthii) at Ngogo, Kibale National Park, Uganda. American Journal of Primatology, 64, 71–82. Muller, M.  N. (2016). Testosterone and reproductive effort in male primates. Hormones and Behavior, 91, 36–51. Muller, M. N., Marlowe, F. W., Bugumba, R., & Ellison, P. T. (2009). Testosterone and paternal care in East Africa foragers and pastoralists. Proceeding of the Royal Society B, 276, 347– 354. Muller, M.  N., & Wrangham, R.  W. (2004). Dominance, aggression and testosterone in wild chimpanzees: A test of the “challenge hypothesis.” Animal Behavior, 67, 113–123. Murdock, G. P., & White, D. R. (1980). Standard cross-cultural sample. In H. B. A. Schlegel (Ed.), Cross cultural samples and codes (pp. 3–43). Pittsburgh, PA: University of Pittsburgh Press. Oliveira, R.  F. (2009). Social behavior in context: Hormonal modulation of behavioral plasticity and social competence. Integrative & Comparative Biology, 49, 423–440. Palmer, C.  T. (1991). Human rape: Adaptation or by-product? Journal of Sex Research, 28, 365–386. Patch, E.  A., & Figueredo, A.  J. (2016). Childhood stress, life history, psychopathy, and sociosexuality. Personality and Individual Differences. Penke, L., & Asendorpf, J. B. (2008). Beyond global sociosexual orientations: A more differentiated look at sociosexuality and

Goetz, Weisfeld, and Zilioli

139

its effects on courtship and romantic relationships. Journal of Personality and Social Psychology, 95, 1113–1135. Perilloux, C., Cloud, J.  M., & Buss, D.  M. (2013). Women’s physical attractiveness and short-term mating strategies. Personality and Individual Differences, 54, 490–495. Peters, M., Simmons, L. W., & Rhodes, G. (2008). Testosterone is associated with mating success but not attractiveness or masculinity in human males. Animal Behaviour, 76, 297–303. Pflüger, L. S., Oberzaucher, E., Katina, S., Holzleitner, I. J., & Grammer, K. (2012). Cues to fertility: Perceived attractiveness and facial shape predict reproductive success. Evolution and Human Behavior, 33, 708–714. Pianka, E. R. (1970). On r- and K-selection. American Naturalist, 104, 592–596. Plomin, R., DeFries, J.  C., Knopik, V.  S., & Neiderheiser, J. (2013). Behavioral genetics. New York: Palgrave Macmillan. Pollet, T.  V., van der Meij, L., Cobey, K.  D., & Buunk, A.  P. (2011). Testosterone levels and their associations with lifetime number of opposite sex partners and remarriage in a large sample of American elderly men and women. Hormones and Behavior, 60, 72–77. Puts, D. A. (2010). Beauty and the beast: Mechanisms of sexual selection in humans. Evolution and Human Behavior, 31, 157–175. Puts, D.  A., Jones, B.  C., & Debruine, L.  M. (2012). Sexual selection on human faces and voices. Journal of Sex Research, 49, 227–243. Puts, D. A., Pope, L. E., Hill, A. K., Cárdenas, R. A., Welling, L. L., Wheatley, J. R., & Breedlove, S. M. (2015). Fulfilling desire: Evidence for negative feedback between men’s testosterone, sociosexual psychology, and sexual partner number. Hormones and Behavior, 70, 14–21. Quinlan, R. J. (2007). Human parental effort and environmental risk. Proceedings of the Royal Society of London B, 274, 121–125. Roff, D. A. (2002). Life history evolution. Sunderland, MA: Sinauer. Ronay, R., & von Hippel, W. (2010). The presence of an attractive woman elevates testosterone and physical risk taking in young men. Social Psychological and Personality Science, 1, 57–64. Roney, J. R., & Gettler, L. T. (2015). The role of testosterone in human romantic relationships. Current Opinion in Psychology, 1, 81–86. Roney, J.  R., Lukaszewski, A.  W., & Simmons, Z.  L. (2007). Rapid endocrine responses of young men to social interactions with young women. Hormones and Behavior, 52, 326–333. Roney, J. R., Mahler, S. V., & Maestripieri, D. (2003). Behavioral and hormonal responses of men to brief interactions with women. Evolution and Human Behavior, 24, 365–375. Roney, J.  R., Simmons, Z.  L., & Lukaszewski, A.  W. (2010). Androgen receptor gene sequence and basal cortisol concentrations predict men’s hormonal responses to potential mates. Proceedings of the Royal Society of London B: Biological Sciences, 277, 57–63. Rosa, E., & Mazur, A. (1979). Incipient status in small groups. Social Forces, 58, 18–37. Sakaguchi, K., Oki, M., Honma, S., Uehara, H., & Hasegawa, T. (2007). The lower salivary testosterone levels among unmarried and married sexually active men. Journal of Ethology, 25, 223–229. Schacht, R., Tharp, D., & Smith, K. R. (2016). Marriage markets and male mating effort: Violence and crime are elevated where men are rare. Human Nature, 27, 489–500.

140

Schmitt, D.  P. (2005). Sociosexuality from Argentina to Zimbabwe: A 48-nation study of sex, culture, and strategies of human mating. Behavioral and Brain Sciences, 28, 247–311. Schmitt, D.  P., Jonason, P.  K., Byerley, G.  J., Flores, S.  D., Illbeck, B.  E., O’Leary, K.  N., & Qudrat, A. (2012). A reexamination of sex differences in sexuality new studies reveal old truths. Current Directions in Psychological Science, 21, 135–139. Sear, R., & Mace, R. (2008). Who keeps the children alive? A review of the effects of kin on child survival. Evolution and Human Behavior, 29, 1–18. Sell, A., Hone, L.  S., & Pound, N. (2012). The importance of physical strength to human males. Human Nature, 23, 30–44. Shuster, S.  M., & Wade, M.  J. (1991). Equal mating success among male reproductive strategies in a marine isopod. Nature, 350, 608–610. Simmons, Z. L., & Roney, J. R. (2009). Androgens and energy allocation: Quasi-experimental evidence for effects of influenza vaccination on men’s testosterone. American Journal of Human Biology, 21, 133–135. Simmons, Z. L., & Roney, J. R. (2011). Variation in CAG repeat length of the androgen receptor gene predicts variables associated with intrasexual competitiveness in human males. Hormones and Behavior, 60, 306–312. Sobolewski, M. E., Brown, J. L., & Mitani, J. C. (2013). Female parity, male aggression, and the challenge hypothesis. Primates, 54, 81–88. Stanton, S.  J., Liening, S.  H., & Schultheiss, O.  C. (2011). Testosterone is positively associated with risk taking in the Iowa gambling task. Hormones and Behavior, 59, 252–256. Strouts, P. H., Brase, G. L., & Dillon, H. M. (2016). Personality and evolutionary strategies: The relationships between HEXACO traits, mate value, life history strategy, and sociosexuality. Personality and Individual Differences, 115, 128–132. Symons, D. (1979). The evolution of human sexuality. New York, NY: Oxford University Press. Taubman-Ben-Ari, O. (2004). Intimacy and risky sexual behavior. What does it have to do with death? Death Studies, 28, 865–888. Tibbetts, E.  A., & Crocker, K.  C. (2014). The challenge hypothesis across taxa: Social modulation of hormone titres in vertebrates and insects. Animal Behaviour, 92, 281–290. Thomsen, L., Frankenhuis, W. E., Ingold-Smith, M., & Carey, S. (2011). Big and mighty: Preverbal infants mentally represent social dominance. Science, 331, 477–480. Thornhill, R. (1980). Rape in Panorpa scorpionflies and a general rape hypothesis. Animal Behavior, 28, 52–59. Thornhill, R., & Palmer, C. P. (2000). A natural history of rape. Cambridge, MA: MIT Press. Thornhill, R., & Thornhill, N. (1983). Human rape: An evolutionary analysis. Ethology and Sociobiology, 4, 137–173. Toth, I., Szollosi, A., Danis, I., Green, J., & Gervai, J. (2006). Changes of attachment to the mother from infancy to childhood: Girls are less at risk? Infant Mental Health Journal, 27, 343. Turke, P. W., & Betzig, L. L. (1985). Those who can do: Wealth, status, and reproductive success on Ifaluk. Ethology and Sociobiology, 6, 79–87. Tutin, C. E. (1979). Mating patterns and reproductive strategies in a community of wild chimpanzees (Pan troglodytes schweinfurthii). Behavioral Ecology and Sociobiology, 6, 29–38.

Reproductive Behavior in the Human Male

van Anders, S.  M., & Goldey, K.  L. (2010). Testosterone and partnering are linked via relationship status for women and “relationship orientation” for men. Hormones and Behavior, 58, 820–826. van Anders, S.  M., Goldey, K.  L., & Kuo, P.  X. (2011). The steroid/peptide theory of social bonds: Integrating testosterone and peptide responses for classifying social behavioral contexts. Psychoneuroendocrinology, 36, 1265–1275. van Anders, S. M., Hamilton, L. D., & Watson, N. V. (2007). Multiple partners are associated with higher testosterone in North American men and women. Hormones and Behavior, 51, 454–459. van Anders, S. M., & Watson, N. V. (2006) Relationship status and testosterone in North American heterosexual and nonheterosexual men and women: Cross-sectional and longitudinal data. Psychoneuroendocrinology, 31, 715–723. van Anders, S.  M., & Watson, N.  V. (2007). Testosterone levels in women and men who are single, in long-distance relationships, or same-city relationships. Hormones and Behavior, 51, 286–291. van der Meij, L., Almela, M., Buunk, A. P., Fawcett, T. W., & Salvador, A. (2012). Men with elevated testosterone levels show more affiliative behaviours during interactions with women. Proceedings of the Royal Society B, 279, 202–208. van Honk, J., Schutter, D. J. L. G., Hermans, E. J., Putman, P., Tuiten, A., & Koppeschaar, H. (2004). Testosterone shifts the balance between sensitivity for punishment and reward in healthy young women. Psychoneuroendocrinology, 29, 937–943. Walker, R.  S., Hill, K.  R., Flinn, M.  V., & Ellsworth, R.  M. (2011). Evolutionary history of hunter-gatherer marriage practices. PLoS One, 6(4), e19066. Welker, K. M., Lozoya, E., Campbell, J. A., Neumann, C. S., & Carré, J. M. (2014). Testosterone, cortisol, and psychopathic

traits in men and women. Physiology and Behavior, 129, 230–236. Welling, L. L., Moreau, B. J., Bird, B. M., Hansen, S., & Carré, J.  M. (2016). Exogenous testosterone increases men’s perceptions of their own physical dominance. Psychoneuroendo­ crinology, 64, 136–142. West-Eberhard, M.  J. (2003). Developmental plasticity and evolution. Oxford, UK: Oxford University Press. Wilson, M., & Daly, M. (1985). Competitiveness, risk taking, and violence: The young male syndrome. Ethology and Sociobiology, 6, 59–73. Wingfield, J. C., Hegner, R. E., Dufty, A. M., Jr., & Ball, G. F. (1990). The “challenge hypothesis”: Theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. American Naturalist, 136, 829–846. Wingfield, J. C., Lynn, S. E., & Soma, K. K. (2001). Avoid the “costs” of testosterone: Ecological bases of hormonebehavior interactions. Brain, Behavior and Evolution, 57, 239–251. Wisman, A., & Goldenberg, J. L. (2005). From the grave to the cradle: Evidence that mortality salience engenders a desire for offspring. Journal of Personality and Social Psychology, 89, 46–61. Xu, X., Coats, J.  K., Yang, C.  F., Wang, A., Ahmed, O.  M., Alvarado, M., . . . Shah, N. M. (2012). Modular genetic control of sexually dimorphic behaviors. Cell, 148, 596–607. Zilioli, S., Ponzi, D., Henry, A., Kubicki, K., Nickels, N., Wilson, M. C., & Maestripieri, D. (2016). Interest in babies negatively predicts testosterone responses to sexual visual stimuli among heterosexual young men. Psychological Science, 27, 114–118. Zilioli, S., & Watson, N. V. (2014). Testosterone across successive competitions: Evidence for a “winner effect” in humans? Psychoneuroendocrinology, 47, 1–9.

Goetz, Weisfeld, and Zilioli

141

CH A PT E R

9

Mate Preferences Across the Lifespan

Lynda G. Boothroyd and Jovana Vukovic

Abstract Humans show preferential responses to “attractive” individuals from the first hours of life onward. However, these early preferences are subject to later development in terms of increasing agreement on general attractiveness and preferences for specific dimensions of attractiveness. This chapter outlines aspects of mate choice and considers evidence for their hormonal mediation in adults. It then examines preferences for these traits across infancy, puberty, and menopause and considers potential hormonal mediation arguments for the mate choice changes observed during these periods. The chapter finds overall that expression of specific preferences is ambiguous in infancy, but there is clear evidence that preferences become stronger in late childhood and adolescence (albeit subject to disruption around puberty). There is also evidence suggesting a decline in some preferences at menopause in women. Across the developmental and lifespan literature, there is a critical lack of studies directly assessing hormones. The chapter closes with recommendations for future research. Key words:  facial attraction, voices, sexual dimorphism, symmetry, averageness, health, sex steroids, infancy, puberty, menopause

Mate choice lies at the core of evolutionary understandings of human behavior. Given that ­ ­selection of a mate, versus random mating, can yield increased reproductive success even in Drosophila (Edward & Chapman, 2012), it is unsurprising that a long-lived species with high levels of parental investment, such as humans, demonstrates high levels of choosiness with regard to sexual partners. Further­ more, although males may on average invest less than females, high rates of pair-bonding, (serial) monogamy, and male provisioning of offspring lead to the prediction that humans should display mutual mate choice (Brown, Laland & Mulder, 2009; Kokko & Johnstone, 2002; Stewart-Williams & Thomas, 2013), with males and females both seeking to detect and select partners on the basis of traits that will maximize immediate reproductive success and/or long-term inclusive fitness. Although initially most developmental research into attraction assessed the extent to which different

age groups “agreed” in their overall preferences for some faces over others, in the last 10 years, researchers have built a body of data that considers developmental change in preferences for specific traits. Theories around the traits that humans tend to find attractive in potential reproductive partners focus on the immediate benefits to the chooser, the longer term indirect benefits to offspring, or both. In the next section, we outline the principal understandings of the benefits derived from those traits that have been investigated in developmental/lifespan contexts thus far, and the methods used to assess preferences for these traits. Then, we go on to consider critical developmental windows in which hormonal changes may impact preferences: namely, infancy, puberty, and menopause.

Traits Typically Preferred by Adults

Direct indices of current health in a potential partner may include elements of vocal quality (e.g., Orlikoff, 143

1990), levels of carotenoid-linked yellow coloration in the skin (e.g., Stephen, Coetzee, & Perrett, 2011), and the overall visible health of the face and skin quality (e.g., Fink & Matts, 2008; Jones et al., 2001). Selection of a more currently healthy mate yields potential reproductive benefits in terms of increased individual fitness via contagion avoidance, and increased inclusive fitness via (potentially) elevated offspring health (although see Adamo & Spiteri, 2009, for discussion of this point). To date, although the adult literature is increasingly interested in the specific skin and voice properties associated with health, the developmental literature has investigated only global facial health as determined by third-party observers; Boothroyd, Meins, Vukovic, and Burt (2014) constructed pairs of stimuli varying in skin color, texture, and face shape based on composites of individuals rated as appearing very healthy or very unhealthy and found that children preferred the “healthier” faces from 6 to 8 years onward. Relatedly, Kościński (2011, 2013) had child participants rate the attractiveness of individual faces already rated by others for a healthy appearance and found a positive correlation between child-perceived attractiveness and adult-perceived health. Symmetry and averageness are often considered to be indices of underlying quality with significant implications for health (reviewed in, e.g., Stephen & Wei,  2015). Although the literature regarding symmetry presents an overall only modest link to health (Van Dongen & Gangestad, 2011), individuals may yet benefit by selecting more symmetrical partners if the factors that lead to their elevated levels of symmetry result in more viable offspring. Similarly, there is evidence that men with more diverse major histocompatibility complex (MHC) alleles, who may be resistant to more pathogens, have more “average” faces than men with less diverse genetic profiles (Lie, Rhodes, & Simmons, 2008). Thus, averageness may offer comparable potential benefits to offspring as symmetrical traits do. Symmetry can be quantified based on the deviations away from perfect bilateral symmetry (e.g., bodies: Gangestad, Thornhill, & Yeo,  1994; faces: Penton-Voak et al., 2001); it is important, however, to exclude systematic asymmetries, where all individuals tend to show asymmetry in one direction (directional asymmetries, e.g., the heart is situated asymmetrically to the left of midline) or in either direction (antisymmetry, e.g., handedness). The ­remaining “random” asymmetries are known as fluctuating asymmetry. Although some studies of associations between body asymmetries and health 144

or mating outcomes do control for systematic asymmetry (e.g., the body landmarks chosen by Gangestad et al., 1994, are not typically subject to directional asymmetries), most studies of attraction simply look at preferences for stimuli varying in naturally occurring asymmetries without isolating fluctuating asymmetry specifically. For instance, the developmental studies of symmetry preference discussed here all presented pairs of stimuli in which the natural asymmetries of an individual had been reduced/eliminated, left intact, or even exaggerated. Averageness is considered almost exclusively in reference to faces (indeed, we are unaware of any studies looking at other forms of averageness) and refers to the degree to which the proportions of an individual face resemble the population mean for those proportions. Preferences for averageness can be determined by correlating attractiveness ratings of individual faces with composite averageness scores for those faces. The majority of research in developmental contexts, however, tends to manipulate the proportions of individual facial identities to make them closer to population average (usually based on the proportions of 25+ individuals drawn from the same recruitment pool) and to exaggerate the differences between the faces and the average (i.e., to make them more distinctive; see, e.g., Boothroyd et al., 2014; Griffey & Little, 2014). Sexual dimorphism, in the context of attraction research, refers to the degree to which an individual demonstrates sexually dimorphic features—that is, how strongly they show sex-typical shape, color, or size. In terms of body size and shape, there has been a great deal of research into preferences for female waist–hip ratio, which is lower (i.e., more curvy) in women than in men, and for dimensions relating to shoulder or chest width in men (e.g., waist–shoulder and waist–chest ratio, respectively; reviewed in Reeve, Kelly, & Welling, 2016). Facial shape is also sexually dimorphic and is the focus of a vast research literature considering preferences for femininity in female faces and women’s ambiguous attitudes toward masculinity in male faces (reviewed in Little, 2015). Finally, voice pitch is also sexually dimorphic, being deeper in men. Various researchers have linked sexual dimorphism in the female direction (i.e., femininity) with estrogen (e.g., Law Smith et al., 2006; although cf Jones et al.,  2018), whereas dimorphism in the male direction (i.e., masculinity) has been both correlationally (e.g., Pound, Penton-Voak, & Surridge, 2009) and pseudo-experimentally linked (e.g., Verdonck, Gaethofs, Carels, & de Zegher, 1999) to testosterone. Femininity is hypothesized to confer

Mate Preferences Across the Lifespan

advantages in terms of fertility, such that males ­selecting more feminine female partners may be more likely to sire offspring during a given sexual encounter. Masculinity in men is more controversial and has been hypothesized to indicate underlying quality and to yield health benefits to offspring (e.g., Thornhill & Gangestad, 2006; although see Scott, Clark, Boothroyd, & Penton-Voak,  2013, for a counterargument, and Boothroyd et al., 2017, for recent counterevidence). Alternatively, masculinity may be primarily a signal of intrasexual competitiveness (Boothroyd, Jones, Burt, & Perrett, 2007; Puts,  2010). In either case, men who have more feminine faces are thought to be warmer, more cooperative, and more suitable long-term partners (Boothroyd et al., 2007; Perrett et al., 1998), while higher pitched (i.e., more feminine) voices are likewise perceived as less dominant (Puts, Hodges, Cardenas, & Gaulin, 2007). The bulk of the developmental literature into facial preferences has tended to utilize stimuli in which femininity is the preferred direction among the oldest participants, for both male and female images (e.g., Boothroyd et al., 2014; Saxton, DeBruine, Jones, Little, & Roberts, 2009; Saxton et al., 2010). As such, in this context, the question is primarily the extent to which children demonstrate adult-like preferences for femininity. Similarly, where we report on voice preferences, we will focus on whether the preferences exhibited are “adult-like” or not, regardless of direction. The only research to investigate body preferences from a developmental angle focused on children’s emerging preferences for differentiated waist–hip ratio between male and female silhouettes (Connolly, Slaughter, & Mealey, 2004).

Evidence for Hormonal Associations in Adults

To our knowledge, the traits mentioned previously are the only widely recognized specific preferences that have been investigated from a developmental perspective, although Kościński (2010, 2011, 2013) has also considered the extent to which children prefer faces that adults have rated on other nonphysical characteristics (e.g., friendliness, sexy appearance). Preferences for the facial, vocal, and body traits discussed previously have also been the focus of research into endocrine influences on mate choice among adults of reproductive age. Expression of female preferences for multiple forms of male masculinity across the menstrual cycle may be driven by both estrogen (Feinberg et al., 2006; Roney & Simmons, 2008; Roney, Simmons, & Gray, 2011;

Pisanski et al., 2014; although cf Dixson et al., 2018, and Jones et al.,  2018) and testosterone (Bobst, Sauter, Foppa, & Lobmaier,  2014; Thornhill, Chapman, & Gangestad, 2013; Welling et al., 2007; although see Roney et al., 2011), with a potential effect of progesterone in suppressing masculinity preferences (Jones et al.,  2005b; Limoncin et al., 2015; Little, Burriss, Petrie, Jones, & Roberts, 2013; but see Cobey, Little, & Roberts,  2015; see also Welling & Burriss, this volume). Similarly, withinsubject variation in men’s preferences for female femininity is associated with testosterone (Welling et al., 2008; see also Bird et al., 2016). On the other hand, female preference for male healthiness may be driven by progesterone insofar as  women prefer healthiness more in a potential partner during the luteal phase of the cycle than the follicular phase, when taking hormonal contraceptives (which contain artificial progestins) than when naturally cycling, and when pregnant than when not pregnant (Jones, Little, et al., 2005; Jones, Perrett, et al., 2005). There is also a hint that hormones may play a role in symmetry preference; although variation in symmetry preferences across the cycle is subject to limited and conflicting evidence (Gangestad & Thornhill, 2008; Gildersleeve, Haselton, & Fales, 2014), Hadza women prefer more symmetric male faces when pregnant or lactating than when naturally cycling (Little, Apicella, & Marlowe, 2007). Pregnancy and lactation are hormonally distinct periods, however, and the authors do not give their data separately. Furthermore, Cobey et al. (2015) found that women’s masculinity preferences differed between pregnant and postnatal participants, suggesting these periods may also show differences in other trait preferences. As such, it is not clear whether any particular hormone is driving this pattern. Finally, an elevation in symmetry preferences would be surprising during a period in which normal female sex hormones are suppressed, as they are in lactation in particular. Overall, therefore, there is strong reason to ­believe that hormonal factors may drive the expression of mate preferences in adulthood, particularly for masculinity/femininity and health, and may thus have a role to play in developmental change in mate preferences.

Facial Preferences in Infancy

The first step in understanding the development of mate preferences is to consider how early the preferences described previously can be observed. Within the first few hours of life, infants show preferences Boothroyd and Vukovic

145

(as indexed by looking times) for some faces over others, in a manner that is consistent with adult judgments of overall attractiveness. For instance, Slater et al. (1998, 2000) have shown that neonates spend longer looking at female faces that adults have rated as attractive compared to female faces rated as unattractive. Furthermore, reanalyzing all the infant-level data presented by Slater et al. across both papers shows that age in hours (which ranges from 7 to 174 hours) does not relate to the strength of infants’ preferences for the “attractive” faces (r50= 0.195, p = 0.18), which suggests that either the preference is formed within the first six hours of life and remains stable thereafter or that infants are born with a basic representation of attractiveness. Some authors, such as Langlois, Roggman, and Musselman (1994), have argued that averageness is the core feature underlying general attractiveness in adult faces, and Rubenstein, Kalakanis, and Langlois (1999) have suggested that since neonates may see 16+ faces within hours of birth, we might expect an early facial preference for averageness. Specifically, they argued that infants may rapidly construct a prototype of faces within the local population and utilize this prototype in coding individual faces, with faces closer to the prototype (i.e., more average) being easier to p ­ rocess and thus favored. There is evidence for norm-based coding of facial identity and attractiveness in adults’ preferences (e.g., Rhodes, Jeffery, Watson, Clifford, & Nakayama, 2003), and there is also evidence that stimuli that are easier to process trigger more positive affect than those that are harder to process (Winkielman, Halberstadt, Fazendeiro, & Catty, 2006). Indeed, this is one hypothesis behind why infants prefer female faces in general: that they have greater experience with female faces and find them easier to process (for discussion see, e.g., Ramsey-Rennels & Langlois, 2006). However, crucially, de Haan, Johnson, Maurer, and Perrett (2001) found that the ability to form facial prototypes begins at approximately 2 months of age and that early preference must instead have an alternative basis. Although 3-month-old infants treated a composite of previously seen faces as “familiar” (e.g., by attending more to a completely novel face), 1-month-old infants did not show this pattern. These data therefore support the suggestion by Slater et al. that although later experience may revise our internal prototypes, humans must have an innate, simple face representation against which neonates may code novel faces. Research into the specific components of facial attractiveness that infants favor is relatively sparse. Although Rubenstein et al. (1999) found infants 146

preferred a composite (i.e., average) face composed of 32 individual faces to an unattractive face in a paired visual preference test, this is difficult to interpret as we know that composite faces are of at least average attractiveness, and often (but not always) more attractive than the individual faces used (see Perrett, May, & Yoshikawa, 1994, for early evidence regarding attractiveness beyond simple averaging). A more appropriate test of this hypothesis is to compare stimuli in which individual faces have been manipulated to be closer to population average in facial structure (i.e., more average) or further from it (i.e., more distinctive). Using this method, Rhodes, Geddes, Jeffery, Dziurawiec, and Clark (2002) found that 6-month-old infants had a potential preference (indexed by length of “longest look” in each trial, as coded by human observers; there was no difference in overall looking times) for less average/more distinctive faces. Similarly, Griffey and Little (2014) used computer-mounted eye tracking and found that sixty-four 12- to 24-month-olds likewise had a significant visual preference for more distinctive, over more average, faces in terms of overall looking time. Both Rhodes et al. (2002) and Griffey and Little (2014) also investigated preferences for symmetry in faces, and found contrasting results; Griffey and Little’s toddlers showed a significant visual preference for symmetric over asymmetric faces, whereas infants in the Rhodes et al. study showed a longest look fixation that tended toward asymmetric faces. Other research found no evidence for symmetry preferences in infants (Samuels, Butterworth, Roberts, Graupner, & Hole, 1994), although the stimuli used in this research failed to show a symmetry preference in adults either, which is highly unusual in the wider literature (reviewed in Little, 2015) and likely reflects a deliberate decision by the authors to present naturally varying more and less symmetric faces from different individuals in pairs matched to be overall equally attractive. There are two key caveats to be addressed in terms of infant preferences for symmetry and averageness; first, although averageness may be attractive beyond any elements of symmetry (as discussed by, e.g., Rubenstein et al., 1999), it is not clear whether previous research into infant preferences has controlled for the fact that average faces are also more symmetrical than individual/less average faces. Second, most research thus far appears to have concentrated on female faces. Although infants undoubtedly process female faces more efficiently and at a younger age than male faces (see, e.g., Ramsey-Rennels & Langlois, 2006), it is nevertheless important to consider infants’

Mate Preferences Across the Lifespan

preferences for these traits among male faces, not least because approximately half of infants will go on to use male facial features as the basis for choosing a mate. Research into preferences for sex typicality (i.e., masculinity and femininity) is likewise sparse. Early data suggested that infants show preferences for larger eyes (Geldart, Maurer, & Carney, 1999) and more “baby-like” adult features (Kramer, Zebrowitz, San Giovanni, & Sherak,  1995), which are both contributors to a feminine appearance (Boothroyd et al.,  2005). Only two studies have specifically investigated sexual dimorphism. Rennels, Kayl, Langlois, Davis, and Orlewicz (2016) found no visual preference for either high or low masculinity in naturally varying male faces among forty 6- and 12-month-olds. Contrastingly, Griffey and Little (2014) found a significant visual preference among their 12- to 24-month-olds for manipulated femininity in both male and female faces. The advantage to Griffey and Little’s data is again the use of eye tracking rather than human coding of looking times, although the Rennels et al. coders showed high ­interrater reliability. Another difference between the studies is the use of stimuli in which faces were fully masked (i.e., hair and clothing covered; Griffey & Little, 2014) versus covering clothing only (Rennels et al., 2016). Adult data may suggest that preferences for femininity should be stronger in the masked stimuli (DeBruine, Jones, Smith, & Little, 2010), although in pilot data with children, Boothroyd et al. (2014) found a nonsignificant trend whereby facial preferences were stronger using unmasked stimuli. Perhaps most crucially, Rennels et al. (2016) used pairs of faces that varied in masculinity but that were matched in attractiveness (as rated by adult observers), which means direct comparison with objectively manipulated masculinity (as per Griffey & Little,  2014, and a large portion of the adult literature) is challenging. One key finding in the Rennels et al. study, however, is that infants may find highly masculine male faces more challenging to process (as indexed by categorization performance) unless those faces are also high in attractiveness; these findings are ­consistent with the idea that female faces may predominate in infants’ facial experience and should therefore predict a tendency to favor femininity in male faces, which perceptually makes them more similar to the female average (although it does not explain why infants would prefer exaggerated femininity in female faces). Overall, studies of face preferences in infants suggest that there are demonstrable early preferences

for neoteny/femininity, for female faces over male faces, and for distinctive rather than average faces. Preferences for symmetry remain less clear, although the strongest study to date (Griffey & Little, 2014) found a preference for symmetry in late infancy. These data are challenging to integrate. Sex hormones are high in early infancy, but a simple model of hormone levels driving more strongly expressed preferences, as per the adult literature, is clearly inadequate here. Preferences for neoteny/femininity and even symmetry (if such a preference exists) can potentially be explained by visual experience, although preferences for distinctive over average faces defy that explanation. Early sex hormones, however, are often described as “organizational” and thus, once the limitations in infant visual perception are taken into account, may not direct attention to adult-preferred traits. Furthermore, visual preference itself is problematic as a direct proxy for affective preference; visual attention during early development may serve multiple learning purposes without necessarily indicating social preference. All of these early tendencies are also subject to change across childhood and it is in this latter window (and puberty especially) that activational hormonal effects may be more pertinent (the organizational–activational hypothesis is further discussed in Hampson, this volume).

Activation of Preferences Across Childhood and Puberty

Although there is evidence that children continue to broadly agree with adults regarding the faces they consider attractive once they are able to verbally report preferences for naturally varying stimuli (e.g., Boothroyd et al., 2014; Cavior & Lombardi, 1973; Cross & Cross, 1971; Kissler & Bauml, 2000; Saxton, Caryl, & Roberts, 2006), data on specific trait preferences remains sparse for the prepubertal period. To our knowledge, four studies have examined a cross-section of children on specific aspects of traits that contribute to mate preferences in adults. Connolly et al. (2004) found that although there was a general trend for children aged 6 to 15 years to increasingly view male and female body shapes as distinct with age, they only began to significantly distinguish between male and female waist–hip ratios around 10 to 12 years of age. Similarly, Boothroyd et al. (2014) showed children aged 4 to 17 years pairs of faces manipulated in averageness, symmetry, femininity, and healthiness; facial healthiness was significantly preferred over unhealthiness from age 6 to 8 onward, and symmetry and averageness were significantly preferred from 9 years onward. Boothroyd and Vukovic

147

Femininity was not preferred until 17 years (although some authors have found femininity preferences in peripubertal groups; see later in this section). VingilisJaremko and Maurer compared 5-year-olds, 9-yearolds, and adults on preferences for symmetry and averageness; like Boothroyd et al., they did not find evidence for symmetry preferences until 9 years of age (Vingilis-Jaremko & Maurer, 2013a); they did, however, find averageness preferences from 5 years of age onward (Vingilis-Jaremko & Maurer, 2013b). In both cases, preferences became stronger between 5 and 9 years, and between 9 years and adulthood. It is useful to note that Boothroyd et al. used adult stimuli, whereas Vingilis-Jaremko and Maurer used stimuli drawn from each of their three participant age groups and found corresponding patterns of preference in each. The pattern of results observed in infants by Rhodes et al. (2002), and particularly Griffey and Little (2014), strongly suggests that the lack of preferences seen in the 4- to 5-year-olds for all traits in Boothroyd et al., and in the 5-year-olds for symmetry in Vingilis-Jaremko and Maurer’s (2013a), is not due to an inability to perceive the differences between manipulated stimuli. Rather, it seems that from infancy onward, differences between symmetric and asymmetric, and average and distinctive versions of the same faces can be visually distinguished, but preferences for specific characteristics such as these may only emerge (or re-emerge) in later childhood. This point further highlights the fact that maturation of the visual system alone is unlikely to explain the onset of preferences for specific facial characteristics. A larger literature exists examining preferences during puberty. Multiple studies have found evidence for symmetry, averageness, health, and femininity preferences existing from 11 or 12 years onward (Boothroyd et al.,  2014; Kościński,  2011,  2013; Saxton et al., 2009, 2010; Saxton, DeBruine, Jones, Little, & Roberts, 2011), although Little et al. (2010) found a significant preference for masculinized male faces in 11- to 14-year-old girls. Where these studies have utilized adult comparison samples, results have tended to show that although specific preferences have been activated by puberty, they are still not expressed at adult levels until some point after age 14 (e.g., Boothroyd et al., 2014; Vingilis-Jaremko & Maurer, 2013a, 2013b). Similarly, Saxton et al. (2006) have found that the nature of adolescents’ voice and face preferences show more agreement than those of younger groups, but still do not show full agreement with adults’ ratings of the same stimuli. 148

Furthermore, Saxton et al. (2010) found no evidence of 12- to 14-year-old participants showing preferences for masculinized or feminized voice pitch. Documenting and explaining the actual changes that occur during puberty, however, is a more challenging question. The first study to attempt comparison of different peripubertal groups found that 13- to 14-year-olds had stronger facial preferences than 11-year-olds (Saxton et al., 2009), a finding that was replicated in a study comparing 11- to 13- and 14- to 15-year-olds (although there were fewer age differences in preferences for female than male faces; Saxton et al., 2011). However, another study from the same authors suggested that these group differences may have been the result of using different, age-matched stimuli for each age group. When two age groups were tested with matching stimuli, no main effect of age on face preferences was found (Saxton et al., 2010). In fact, although overall means for femininity and symmetry preferences increased with age, preferences for averageness decreased. Correspondingly, Kościński (2010,  2011,  2013) found a similar pattern of means in strength of preference for skin health in girls and boys aged between 11 and 14 years. In the only longitudinal data on facial preferences we are aware of, Saxton et al. (2011) were unable to document any clear changes in adolescents’ facial preferences over a one-year period, although in a later study on voices it was observed that 11- and 13-year-olds showed a drop in voice pitch preference over a 9- to 12-month period (Saxton, DeBruine, Jones, Little, & Roberts, 2013; see also Saxton et al., 2009, for similar cross-sectional patterns in voice pitch preference). These results are concordant with the cross-sectional data of Boothroyd et al. (2014) shown in Figure  9.1. Preferences for health, symmetry, and averageness all emerged in midchildhood but showed a plateau or dip sometime between 10 and 14 years of age before increasing to adult levels by age 17. Likewise, although directional preferences for femininity never exceeded chance levels until age 17, the same pattern of a drop in preferences occurred at 13 to 14 years. There are therefore two issues that need resolving when considering preferences across childhood and adolescence: (1) What causes these preferences to emerge in the first place? (2) Why, when we would expect an explosion in mate-choice-relevant behavior at puberty, do we instead see this emergence disrupted? In explaining the emergence of specific preferences across childhood, one hormone has come under repeated speculation. The earliest stage of puberty is adrenarche, when the adrenal gland

Mate Preferences Across the Lifespan

Proportion of high trait faces chosen

0.8

0.7

0.6

0.5

0.4 4–5yo

6–8yo

9yo

10–12yo 13–14yo

Age Group Averageness

Femininity

17yo

Error bars: 95% Cl Health

Symmetry

Figure 9.1  Summary of the trajectories of specific face preferences with age (data from Boothroyd et al., 2014). Most traits show stronger preferences across mid-childhood, with a dip or plateau around puberty, before increasing to adult levels by 17 years. Children were grouped by putative dehydroepiandrosterone (DHEA) levels/puberty stage; 4–5 DHEA at floor, 6–8 early adrenarche, 9 later adrenarche, 10–12 late adrenarche/early gonadarche, 13–14 late gonadarche, 17 years adult-like hormones. 600 500 400 300 200 100

–1

0

1

3

5

7 Girls

9

11

13

15

Boys

Figure 9.2  DHEA levels (ng/100 ml) across 0–15 years, re-drawn from de Perretti & Forest (1976).

primes the body for full puberty by releasing ­dehydroepiandrosterone (DHEA), a hormone that mainly acts as a precursor to other sex hormones. DHEA release begins around 6 years of age and increases dramatically around 9 to 11 years (de Peretti & Forest,  1976; see Figure  9.2). This surge in DHEA may trigger the first conscious feelings of

sexual attraction in boys and girls (McClintock & Herdt, 1996). Furthermore, levels of DHEA are also high during the postnatal period but drop to floor by age 2. If DHEA plays a role in the development of specific mate preferences, this may explain the apparent U-shaped curve in certain facial preferences across infancy and early childhood; as such, Boothroyd and Vukovic

149

DHEA is an obvious candidate hormone for activation of adult-like preferences for specific ­ facial traits, as has been suggested by multiple ­authors (e.g., Boothroyd et al., 2014; Connolly et al., 2004; Saxton et al., 2009). With respect to facial preferences in particular, it is useful to consider the wider literature on the ­development of face processing. A cognitive approach to facial preferences assumes that preferences at least partially reflect the manner in which faces are processed and stored in the brain. Thus, children’s judgments of attractiveness may become more similar to those of adults as they get older because of increasing expertise in the perceptual and processing skills required to evaluate aspects of attractiveness. Research has not yet investigated links between children’s facial preferences and their processing of other visual stimuli; furthermore, it has been argued that face processing is essentially mature by midchildhood (McKone, Crookes, Jeffery, & Dilks, 2012; McKone, Crookes, & Kanwisher, 2009; but see Susilo, Germine, & Duchaine, 2013). However, much like the pattern of face preferences plateauing during later puberty, a “developmental dip” in other forms of facial processing occurs around the same age. Multiple studies have documented a period between 10 and 14 years of age in which performance on face recognition either dips or plateaus before increasing to adult levels later on (Carey, Diamond, & Woods, 1980; Chung & Thomson, 1995; Lawrence et al., 2008). Recent evidence has shown that the dip is specific to ­processing faces; Johnston et al. (2011) found that both facial identity recognition and facial emotion recognition showed a dip, but that recognition of other complex stimuli, namely, butterflies, did not. These data rule out alternative explanations such as task demands affecting performance and suggest that the “dip” is specific to facial (and perhaps particularly social) stimuli. This developmental dip in face processing has been hypothesized to be part of a broader pubertal reorganization in which the brain is realigned to the social and cognitive challenges of mating. Scherf, Behrmann, and Dahl (2012) suggest that the apparently sex-linked or directly sex-hormone-linked anatomical and functional changes in cortical face processing regions and subcortical bodies (e.g., to the amygdala) during adolescence may demonstrate a period of increasing specialization for adult sexual behavior. They argue that this realignment may result in temporary disruption of social functions and may underlie phenomena such as the developmental 150

dip in face processing. Indeed, Scherf et al. explicitly suggest that attraction may be subject to the same disruption as other aspects of face processing and comment on the current paucity of data assessing this possibility. This suggests that, whereas changes in DHEA may activate the initial onset of adult-like facial preferences, cognitive processing factors may subsequently constrain the expression of preferences until puberty is completed. Attempts to explicitly link pubertal development to changes in mate preferences and facial recognition alike, however, have yielded mixed results. Typically, researchers use standardized self- or parent report measures of puberty to assign puberty stages to participants and assess whether stage or raw scores on pubertal development predict preferences/­ performance (e.g., Boothroyd et al., 2014; Saxton et al., 2010). These puberty measures typically utilize the Tanner classification images or verbal descriptions of physical development that align with the Tanner drawings. Although phenotypic cues to puberty are associated with changes in sex hormones, they remain a blunt instrument in assessing those underlying hormones and are subject to report biases from children and parents alike (see, e.g., Carskadon & Acebo, 1993, for discussion). It is therefore not entirely surprising that Boothroyd et al. (2014) found no link at all between mate preferences and puberty stage, whereas Saxton et al. (2010) found that more advanced puberty was associated with stronger preferences for symmetry, but not for averageness or voice pitch, and that pubertal development had inverse effects on femininity preferences at 12 versus 13 to 14 years. Kościński found little evidence for breast development as a marker of puberty being linked to girls’ preferences for skin health (Kościński, 2011, 2013) and indeed limited evidence for it predicting any other aspects of girls’ preferences (e.g., for “sexy looking” or “friendly” faces). Only Kościński’s (2010) data on 11- and 12-year-old boys showed a strong link between current development and maturity of preferences. To our knowledge, no researchers have assessed pubertal hormones directly when seeking to test links to mate choice. In a pilot study, Boothroyd and colleagues investigated whether salivary DHEAS levels (a diurnally stable metabolite of DHEA) were associated with face preferences in a group of 10-year-old children. The resulting data were largely nonsignificant; however, in part this may be due to 10 years of age being an inappropriate window for assessing these impacts (Boothroyd et al., in preparation). The

Mate Preferences Across the Lifespan

combined weight of the Boothroyd et  al. (2014) cross-sectional data and the Scherf et al. (2012) review, regarding activatory and inhibitory periods of change in mate preferences, would suggest that any activational effects of DHEA may best be observed in the 6-to-9-year window, whereas any inhibitory impact of gonadal hormones may best be observed in the 11-to-14-year window. Future research seeking to test the hypothesis that adrenal and gonadal hormones may alternatively activate mate preferences and then inhibit face and other social ­processing must also concurrently test both mate preferences and facial processing alongside hormone assays to fully explore the interactions between the three. An additional qualification to the mixed results on puberty, however, is that cross-sectional samples may not offer the best determinants of this question. Individual differences in age of puberty are associated with preferences for both faces and voices. Women who experienced first menses at an earlier age prefer more masculinized faces (Cornwell et al., 2006) and deeper voices (Jones, Boothroyd, Feinberg, & DeBruine, 2010) in adulthood. When shown naturally varying male facial stimuli, Hoier’s (2003) participants showed a tendency for earlier first menses to predict rating all faces as more attractive than women with later first menses. These differences may reflect the impacts of early stressors on pubertal development (see Deer, Bernard, & Hostinar, this volume, for further discussion) and mate preferences alike or alternatively reflect differences in sexual and romantic experience, arising from earlier versus later menarche, shaping future preferences. For instance, earlier first menses is associated with earlier first

coitus (e.g., Downing & Bellis, 2009; Udry, 1979), and given the sex difference in age of puberty, those early relationships are likely to be with older males. Related to this, Kościński (2010) found that boys’ sexual experience predicted similar patterns of face preferences as their pubertal development, highlighting the impact experience may have. One key problem with the data on pubertal development and mate choice during puberty may thus be that pubertal hormones may have differing effects within and between individuals. Indeed, although breast development was a weak predictor of preferences among Kościński’s (2013) female participants within 12- and 13-year-old cohorts, girls with earlier menarche had more adult-like preferences for health, facial “sexiness,” and facial cues considered by adults as indicating a good partner. Although Kościński viewed time since menarche as an index of relative pubertal development, the fact that breast development showed no strong patterns suggests these data were picking up on timing of previous developments rather than current hormonal status per se. As such, research into hormonal drivers of pubertal development in mate choice must include longitudinal, within-individual analyses. Indeed, many of the studies that have found hormonal influences on mate preferences in adults have relied on within-individual comparison. For instance, Welling et al. (2008) found no difference between men with overall higher and lower testosterone in terms of their preferences for female femininity; there was, however, a significant difference in mate preferences when comparing between the two sessions in which each man had higher versus lower levels of testosterone.

Box 9.1.  Facial Preferences and Parental Features Despite any potential role of innate representations in early preference as per the neonates in Slater et al. (1998, 2000), later infant preferences are undoubtedly influenced by experience with the faces of conspecifics. For instance, while most studies show a preference among infants for female faces over male faces (see, e.g., Ramsey-Rennels & Langlois., 2006, for a discussion), infants whose primary carer is male show a preference for male faces over female faces (Quinn, Yahr, Kuhn, Slater, & Pascalis, 2002). This strongly suggests that in early life, at least, the primary carer is the major influence on facial prototyping and facial preferences (although see, e.g., Cooper, Geldart, Mondloch, & Maurer, 2006, for evidence on the potential importance of peers in childhood preferences). Parental features are also important in attraction in adulthood. The phenomenon of sexual imprinting (whereby an organism bases its choice of adult sexual partner on the features of its parents) is well documented in some animal species (e.g., Zebra finches: Vos, 1995) and has been shown to operate among humans, with adults showing preferences for parental features in terms of race (Jedlicka, 1980), coloring (Little, Penton-Voak, Burt, & Perrett, 2003), hairiness (Rantala, Pölkki, & Rantala, 2010), age (Perrett et al., 2002), and even personality (Gyuris, Jarai, & Bereczkei, 2010). The ontogeny of this phenomenon, however, remains obscure. (continued )

Boothroyd and Vukovic

151

Box 9.1.  Continued It may be that our facial prototypes are based on the faces we are exposed to, and that although this process is continuous throughout life (e.g., Anzures, Mondloch, & Lackner, 2009; Rhodes et al., 2003; Webster & MacLeod, 2011), extensive exposure to our parents in early life heavily biases our prototypes, and ergo our preferences, toward parental features. However, evidence suggests that a degree of associative learning may also take place, with positive parent–child relationships predicting greater imprinting. For instance, Bereczkei and colleagues (Bereczkei, Gyuris, Koves, & Bernath, 2002; Bereczkei, Gyuris, & Weisfeld, 2004) used a photograph-matching task to show that similarity between an individual’s opposite-sex parent and spouse was significantly greater if the individual reported a warmer relationship with that parent (although see Marcinkowska & Rantala, 2012). Furthermore, Bereczkei et al. (2004) controlled for self-similarity effects and used adoptive fathers and daughters so that genetic explanations can be ruled out. Wiszewska, Pawlowski, and Boothroyd (2007) found that the objective facial proportions of women’s fathers significantly correlated with the proportions of faces they found most attractive out of an array, although only if the daughters reported a positive relationship with their father. Similarly, Kocsor, Saxton, Láng, and Bereczkei (2016) found that participants who reported less rejecting opposite-sex parents had stronger preferences for faces manipulated to resemble that parent. Developmental research into imprinting-like phenomena is particularly rare. Two studies have demonstrated that children may show a preference for parental features prior to puberty. Kocsor, Gyuris, and Bereczkei (2013) used a story completion task with 3- to 6-year-olds and found that boys who completed a story about a young bird in a storm with the bird flying to one of its parents preferred father-like faces more than boys who completed the story with a different ending. There was no such pattern for maternal features or in girls. Using more standardized measures of attachment, Vukovic, Boothroyd, Meins, and Burt (2015) looked at preference for parent-like faces in 9-year-olds drawn from a longitudinal cohort and found that although infant attachment as measured in the strange situation paradigm at 15 months did not predict preferences for parental features in either sex of face, children who reported a currently more accepting relationship with their parents (on the Parental Acceptance and Rejection Questionnaire: Rohner & Khaleque, 2005) favored more parent-like faces. This pattern held for girls looking at both sexes of face, and for boys looking at female faces. For boys looking at male faces, visual exposure predicted preference for paternal features but current relationship did not. Interestingly, although Vukovic et al. (2015) interpreted their data as indicating that parental imprinting may reflect a transient associative impact of positive parent–child relationships on attitudes to parental features, with current experience superseding past experience, Saxton (2016) found that retrospective reports of parent–child relationships had different impacts in different time periods. Specifically, the similarity of women’s ideal/actual partner’s eye color and their father’s eye color was positively predicted by closeness to father in midchildhood and negatively predicted by closeness to father after puberty. This would suggest that puberty may act as a watershed for typical imprinting effects and that incest avoidance mechanisms may come more to the fore once individuals are reproductively mature (although incest avoidance mechanisms regarding siblings typically have a much earlier critical window; reviewed in Rantala & Marcinkowska, 2011).

Facial Preferences and Menopause

Further evidence that hormonal changes across the  lifespan are associated with variation in mate preferences comes from studies that tested pre- and postmenopausal women. As women approach menopause, estrogens, progesterone, and testosterone decrease and fertility declines (Burger, Dudley, Robertson, & Dennersten,  2002; Burger, Hale, Dennerstein, & Robertson, 2008). Just as preferences for overall attractiveness become more c­onsistent within groups as children age (Saxton et al., 2009), so they may become less so across menopause. Kościński (2011) asked peri- and postmenopausal Polish women aged 40 to 62 years to judge the ­attractiveness of stimuli of male facial photographs. The postmenopausal women showed significantly 152

greater variation in their preferences and had a lower intraclass correlation than perimenopausal women of the same age. Studies into specific preferences and menopause have operated on the assumption that the decline in fertility around menopause may trigger a shift away from a mate-oriented mindset toward a community-oriented psychology (Hawkes, O’Connell, Jones, Alvarez, & Charnov, 1998). Thus, postmenopausal women should be less attracted to indices of mate quality. Furthermore, the decline in sex hormones ought to be associated with a decline in face preferences that are strongly modulated by these hormones. Considering that previous research has established a small but robust link between women’s fertility across the menstrual cycle and heightened preferences

Mate Preferences Across the Lifespan

for masculinity in particular (Jones, Little, et al., 2005; Little, Jones, & DeBruine,  2008; Welling et  al., 2007), research around menopause has also tended to focus on this trait. Vukovic et al. (2009) first investigated the association between fertility across the lifespan and women’s masculinity preferences. The authors tested 97 women’s masculinity preferences, 45 of whom were no longer experiencing menstrual cycles due to menopause and all of whom reported that they were not using any hormone replacement therapies or hormonal contraceptives. Participants gave forced-choice preferences for pairs of masculinized and feminized versions of both male and female faces. Postmenopausal women had weaker masculinity preferences than premenopausal women, but the difference between the preand postmenopausal groups was not significant. One possible reason for this null result may have been the small sample. Little et al. (2010) investigated pre- and postmenopausal women’s preferences for facial masculinity in two samples with nearly 200 and nearly 2,000 participants, respectively. In Study 1, participants between the ages of 40 and 65 years judged the attractiveness of 10 trials containing masculinized and feminized versions of men’s faces. In Study 2, women aged 36 to 45 years were compared to women over 45 years (among other groups). In both cases, postmenopausal participants and/or those aged over 45 years preferred less masculine men than did premenopausal/younger women. These ­results suggest that postmenopausal women are more attracted to the direct prosocial benefits signaled by relatively feminine men (see also Fink & PentonVoak, 2002; Gangestad & Simpson, 2000; Jones et al., 2008; Little, Jones, Penton-Voak, Burt, & Perrett, 2002; Rhodes, Simmons, & Peters, 2005). Another factor contributing to the weak results in Vukovic et al. (2009) may be that the stimuli used in the study were of young adults. Therefore, it is possible that circum-menopausal women did not view the young male faces used in the study as faces of potential mates. Accordingly, Jones, Vukovic, Little, Roberts, and DeBruine (2011) replicated the study using face stimuli of individuals closer to the age of a new sample of circum-menopausal participants. Results yielded a significant difference between preand postmenopausal women’s preferences for men’s faces, whereby postmenopausal women preferred the feminized versions of the men’s faces. Collectively, these results provide evidence for hormonally driven variation in women’s masculinity preferences and highlight the importance of testing face preferences using a wide range of stimuli.

It is also worth noting that although Vukovic et al. (2009) and Jones et al. (2011) found an overall preference for femininity, the Little et al. (2010) participants showed a general preference for masculine men’s faces. These results are not surprising considering that other studies have found no consensus as to whether women overall prefer masculine or feminine male faces (see Scott et al., 2014; Fink & Penton-Voak, 2002). It is therefore important for further studies to recruit participants of various backgrounds and ages and from different cultures. Also, although masculine men are perceived to look older (Boothroyd et al., 2005) and older women have been found to prefer older faces (Kościński, 2011), the studies on circum-menopausal women’s masculinity preferences emphasize that masculinity and perceived age are separate, compound facial traits. As such, manipulations of these independently may differentially influence women’s preferences. Moreover, both Vukovic et al. (2009) and Jones et al. (2011) considered both male and female faces. Postmenopausal (i.e., infertile) women preferred more feminine female faces than did ­premenopausal (i.e., fertile) women. The authors controlled for age, suggesting that these results were due to the differences in hormonal profiles between pre- and postmenopausal women. In line with theories of heightened intrasexual competition during periods of high fertility (see Fisher, 2004; Jones, Little, et al., 2005; Welling et al., 2007), this research implies that same-sex competition for potential mates decreases as women’s fertility declines, and that women may become more honest about the attractiveness of other women as they age (i.e., may be less likely to derogate their competitors). Another testosterone-dependent masculine trait in men’s faces, other than male-typical face shape, and which has been studied in relation to menopause is facial hair. To test out women’s preferences for facial hair, Dixson, Tam, and Awasthy (2012) collected stimuli of men at various stages of beard growth: clean shaven, light stubble, heavy stubble, and full beard. Participants were women in New Zealand of various reproductive statuses who were asked to choose the most attractive of the stimuli presented. Interestingly, postmenopausal women found all the stimuli more attractive than did premenopausal women, although cohort effects are possible, suggesting a general shift in baseline thresholds for attractiveness in (young) faces. There was no interaction between menopause and preferences for degree of facial hair in the stimuli, indicating no Boothroyd and Vukovic

153

specific change in attitudes to beards in these groups. Much like in Vukovic et al. (2009), the authors suggested that images of older men should be used in future studies and may yield different results. It is also worth noting that Dixson et al. (2012, 2018) found minimal evidence for differences in fertility status in younger women affecting attitudes to beardedness in the same and a more recent study. As discussed by Dixson et al. (2012), male mate quality is signaled partly by shape-based traits like chins and jaw lines, which are masked by facial hair. As such, beardedness may not show a straightforward relationship with typical predictors of mate preferences, being both a display and a mask of mate quality. Indeed, Dixson, Sulikowski, Gouda-Vossos, Rantala, and Brooks (2016) showed evidence for an interaction between masculinity and beardedness; extremely feminine and masculine men looked more attractive when bearded, but not men in between. Given the evidence that women typically prefer men closer to “average” levels of masculinity (e.g., Scott, Pound, Stephen, Clark, & Penton-Voak,  2010), this suggests beards may operate more as a mask than as a signal. Finally, Kościński (2011) also assessed circummenopausal women’s preferences for mate-relevant aspects of facial appearance, such as health. As in the aforementioned studies, preferences for facial health declined with menopause, as did preferences for faces rated by other women as “sexy” or “marriageable,” although preferences for “friendliness” remained constant. Although health preferences are a reasonably objective feature, it should be noted that “sexiness” and “marriageability” are inherently subjective constructs shaped in part by the preferences of the  perceiver, and as such these declines may simply reflect changing definitions of what constitutes sexiness, for instance, in older women. The results are, however, consistent with those presented earlier for objectively manipulated masculinity and contribute to the overall suggestion that menopause does indeed lead to a decline in those mate preferences commonly ascribed to hormonal factors. Collectively, the aforementioned research suggests that women’s preferences for cues of mate quality in men change across the lifespan. Specifically, these studies show hormonally driven variation in women’s attraction to masculinity, health, and sexiness, whereby postmenopausal women tend to prefer more feminine, less healthy, and less sexy (as rated by younger women) individuals than do fertile women. More research is needed, however, to better understand the link between fertility and 154

mate preferences, and how these preferences influence social interactions within this age cohort. For instance, menopause may not proceed in the same manner for all women (indeed, any study relying on proxies of hormones such as cycle phase is subject to this caveat) and, as in the pubertal period, there is a dearth of direct assessments of hormones in menopausal women. As such, suggestions that these changes are directly the result of circulating hormones rather than, for instance, byproducts of changing self-rated attractiveness with age, or explicit knowledge of changing fertility and thus changing attitudes toward relationships, remain untested. These findings point to a need for more studies using diverse participants, larger sample sizes, and more diverse stimuli of faces of various phenotypes. Additionally, future studies should employ more hormonal assay techniques to more accurately assess participants’ hormonal profiles. Finally, there is also a need to consider age-related hormonal changes in men and the impacts of these changes on preferences. To our knowledge, only one study has examined trajectories in heterosexual male preferences for females in later life. Marcinkowska, Dixson, Kozlov, and Rantala (2015) found a fairly linear decline in men’s preferences for facial femininity in women between 30 and 70 years of age (although even the oldest participants still preferred feminized stimuli more than chance). This is a period in which testosterone may decline initially through lifestyle factors (e.g., marriage and fatherhood; for review see Gray & Campbell,  2009) and later through aging processes (Harman, Metter, Tobin, Pearson, & Blackman, 2001), and thus may lead to decreased interest in indices of female mate quality. However, again, the study was unable to directly assess hormone levels and utilized young female stimuli. Given that feminine faces look younger, it is possible that this decline in part merely reflects an interest in more age-appropriate partners.

Conclusion

There are several key themes that arise when considering these lifespan approaches to mate preferences. First, although there are some aspects of faces about which individuals of all ages from birth through menopause consider attractive, preferences for specific aspects of faces, voices, and bodies that have been linked to mate quality in adults are subject to change over time. When we consider patterns across midchildhood and menopause, it seems likely that pubertal hormones may activate specific mate preferences, whereas menopause and the corresponding

Mate Preferences Across the Lifespan

decline in sex hormones may lead to a decline in mating-relevant preferences. Similarly, although there is apparent disruption of preference development midpuberty, this too is plausibly the result of sex hormones changing social perception in general in line with the further activation of mating-relevant cognition. There remain, however, some key weaknesses in this literature. First, a complete dearth of direct assays of circulating hormones in the participant groups in question means that it is difficult to strongly conclude that hormonal factors are indeed driving these changes. Furthermore, studies rarely—if ever—consider other potential mediators, such as broader face processing during childhood and puberty, or mating versus social attitudes in menopause, in the same samples. Although these mediators may work in tandem with the impacts of sex hormones, rather than as alternative explanations, not assessing them alongside mate preferences limits our understanding of the processes driving change across these periods. This weakness is particularly pertinent when considering the infant literature; although circulating sex hormones are high in early infancy, limitations in face processing likely have a strong impact on expressed face preferences. Thus, although there may be a fruitful narrative to be developed in the future regarding, for instance, organizing effects of hormones in utero and in infancy, or even the possibility that sex hormones may direct attention to materelevant traits even in this early period, without both measuring hormones and controlling for broader face processing, it will be impossible to further consider these possibilities. Likewise, the lack of within-participant comparisons in much of the literature is a serious weakness that urgently needs to be addressed to disentangle within- and between-individual differences in hormones and the differing sources and potential impacts on preferences these may have. Furthermore, the wide degree of variation across studies in terms of the age of the stimuli used, the precise manipulations used, whether or not stimuli were masked, and the way hormonal proxies are assessed (e.g., puberty scales, breast development, reported menstruation, age) also clouds our ability to confidently interpret the overall results for any given trait of interest. The result of these caveats, however, is that we perceive here a literature that is ripe for further research, which may yet yield a truly comprehensive and coherent model of how sex hormones impact on the development and expression of mate preferences across the lifespan.

References

Adamo, S. A., & Spiteri, R. J. (2009). He’s healthy, but will he survive the plague? Possible constraints on mate choice for disease resistance. Animal Behaviour, 77(1), 67–78. Anzures, G., Mondloch, C.  J., & Lackner, C. (2009). Face adaptation and attractiveness aftereffects in 8-year-olds and adults. Child Development, 80, 178–191. Bereczkei, T., Gyuris, P., Koves, P., & Bernath, L. (2002). Homogamy, genetic similarity, and imprinting: Parental influence on mate choice preferences. Personality and Individual Differences, 33, 677–690. Bereczkei, T., Gyuris, P., & Weisfeld, G.  E. (2004). Sexual imprinting in human mate choice. Proceedings of the Royal Society of London Series B: Biological Sciences, 271, 1129–1134. Bird, B. M., Welling, L. L., Ortiz, T. L., Moreau, B. J., Hansen, S., Emond, M., . . . Carré, J. M. (2016). Effects of exogenous testosterone and mating context on men’s preferences for female facial femininity. Hormones and Behavior, 85, 76–85. Bobst, C., Sauter, S., Foppa, A., & Lobmaier, J. S. (2014). Early follicular testosterone level predicts preference for masculinity in male faces–But not for women taking hormonal contraception. Psychoneuroendocrinology, 41, 142–150. Boothroyd, L.  G., Cooper, G., Vukovic, J., Burt, D.  M., & Nettle, D., Meins, E. (in preparation). Adrenal hormone levels and facial preferences in peri-pubertal children. Boothroyd, L.  G., Gray, A.  W., Headland, T., Uehera, R., Waynforth, D., Burt, D. M., & Pound, N. (2017). Male facial appearance and child mortality in two traditional populations. PLoS One, 12, e0169181. Boothroyd, L. G., Jones, B. C., Burt, D. M., Cornwell, R. E., Little, A. C., Tiddeman, B. P., & Perrett, D. I. (2005). Facial masculinity is related to perceived age but not perceived health. Evolution and Human Behavior, 26(5), 417–431. Boothroyd, L.  G., Jones, B.  C., Burt, D.  M., & Perrett, D.  I. (2007). Partner characteristics associated with masculinity, health and maturity in male faces. Personality and Individual Differences, 43(5), 1161–1173. doi:10.1016/j.paid.03.008 Boothroyd, L. G., Meins, E., Vukovic, J., & Burt, D. M. (2014). Developmental changes in children’s facial preferences. Evolution and Human Behavior, 35(5), 376–383. doi:10.1016/ j.evolhumbehav.2014.05.002 Brown, G. R., Laland, K. N., & Mulder, M. B. (2009). Bateman’s principles and human sex roles. Trends in Ecology & Evolution, 24(6), 297–304. doi:10.1016/j.tree.2009.02.005 Burger, H. G., Dudley, E. C., Robertson, D. M., & Dennersten, L. (2002). Hormonal changes in the menopause transition. Recent Progress in Hormone Research, 57, 257–275. Burger, H.  G., Hale, G.  E., Dennerstein, L., & Robertson, D. M. (2008). Cycle and hormone changes during perimeno­ pause: The key role of ovarian function. Menopause, 15(4), 603–612. Carey, S., Diamond, R., & Woods, B. (1980). Development of face recognition: A maturational component. Developmental Psychology, 16(4), 257–269. doi:10.1037/0012-1649.16.4.257 Carskadon, M.  A., & Acebo, C. (1993). A self-administered rating-scale for pubertal development. Journal of Adolescent Health, 14(3), 190–195. doi:10.1016/1054-139x(93)90004-9 Cavior, N., & Lombardi, D. A. (1973). Developmental aspects of judgment of physical attractiveness in children. Developmental Psychology, 8(1), 67–71. Chung, M. S., & Thomson, D. M. (1995). Development of face recognition. British Journal of Psychology, 86, 55–87. Cobey, K. D., Little, A. C., & Roberts, S. C. (2015). Hormonal effects on women’s facial masculinity preferences: The influence

Boothroyd and Vukovic

155

of pregnancy, post-partum, and hormonal contraceptive use. Biological Psychology, 104, 35–40. Connolly, J.  M., Slaughter, V., & Mealey, L. (2004). The development of preferences for specific body shapes. Journal of Sex Research, 41(1), 5–15. Cooper, P. A., Geldart, S. S., Mondloch, C. J., & Maurer, D. (2006). Developmental changes in perceptions of attrac­ tiveness: A role of experience? Developmental Science, 9, 530–543. Cornwell, R. E., Law Smith, M. J., Boothroyd, L. G., Moore, F.  R., Davis, H.  P., Stirrat, M.  R., . . . Perrett, D.  I. (2006). Reproductive strategy, sexual development and attraction to facial characteristics. Proceedings of the Royal Society of London Series B: Biological Sciences, 361(1476), 2143–2154. Cross, J. F., & Cross, J. (1971). Age, sex, race, and perception of facial beauty. Developmental Psychology, 5(3), 433–439. de Haan, M., Johnson, M. H., Maurer, D., & Perrett, D. I. (2001). Recognition of individual faces and average face prototypes by 1-and 3-month-old infants. Cognitive Development, 16(2), 659–678. de Peretti, E., & Forest, M. G. (1976). Unconjugated dehydro­ epiandrosterone plasma levels in normal subjects from birth to adolescence in human: The use of a sensitive radio­ immunoassay. Journal of Clinical Endocrinology and Metabolism, 43(5), 982–991. DeBruine, L.  M., Jones, B.  C., Smith, F.  G., & Little, A.  C. (2010). Are attractive men’s faces masculine or feminine? The importance of controlling confounds in face stimuli. Journal of Experimental Psychology: Human Perception and Performance, 36(3), 751. Dixson, B.  J., Lee, A.  J., Blake, K.  R., Jasienska, G., & Marcinkowska, U.  M. (2018). Women’s preferences for men’s beards show no relation to their ovarian cycle phase and sex hormone levels. Hormones and behavior, 97, 137–144. Dixson, B.  J.  W., Sulikowski, D., Gouda-Vossos, A., Rantala, M. J., & Brooks, R. C. (2016). The masculinity paradox: Facial masculinity and beardedness interact to determine women’s ratings of men’s facial attractiveness. Journal of Evolutionary Biology, 29, 2311–2320. Dixson, B. J., Tam, J. C., & Awasthy, M. (2012). Do women’s preferences for men’s facial hair change with reproductive status? Behavioral Ecology, 24, 708–716. Downing, J., & Bellis, M. A. (2009). Early pubertal onset and its relationship with sexual risk taking, substance use and antisocial behaviour: A preliminary cross-sectional study. BMC Public Health, 9, 446. Edward, D.  A., & Chapman, T. (2012). Measuring the fitness benefits of male mate choice in Drosophila melanogaster. Evolution, 66(8), 2646–2653. doi:10.1111/j.1558–5646.2012. 01648.x Feinberg, D. R., Jones, B. C., Law-Smith, M. J., Moore, F. R., DeBruine, L. M., Cornwell, R. E., . . . Perrett, D. I. (2006). Menstrual cycle, trait estrogen level, and masculinity pre­ ferences in the human voice. Hormones and Behavior, 49(2), 215–222. doi:10.1016/j.yhbeh.2005.07.004 Fink, B., & Matts, P.  J. (2008). The effects of skin colour distribution and topography cues on the perception of female facial age and health. Journal of the European Academy of Dermatology and Venereology, 22(4), 493–498. Fink, B., & Penton-Voak, I. S. (2002). Evolutionary psychology of facial attractiveness. Current Directions in Psychological Science, 11, 154–158.

156

Fisher, M. L. (2004). Female intrasexual competition decreases female facial attractiveness. Proceedings of the Royal Society B, 271, S283–S285. Gangestad, S.  W., & Simpson, J.  A. (2000). The evolution of human mating: Trade-offs and strategic pluralism. Behavioral and Brain Sciences, 23(4), 573–644. Gangestad, S.  W., & Thornhill, R. (2008). Human oestrus. Proceedings of the Royal Society B: Biological Sciences, 275(1638), 991–1000. doi:10.1098/rspb.2007.1425 Gangestad, S.  W., Thornhill, R., & Yeo, R.  A. (1994). Facial attractiveness, developmental stability, and fluctuating asymmetry. Ethology and Sociobiology, 15(2), 73–85. Geldart, S., Maurer, D., & Carney, K. (1999). Effects of eye size on adults’ aesthetic ratings of faces and 5-month-olds’ looking times. Perception, 28(3), 361–374. Gildersleeve, K., Haselton, M. G., & Fales, M. R. (2014). Do women’s mate preferences change across the ovulatory cycle? A meta-analytic review. Psychological Bulletin, 140(5), 1205–1259. doi:10.1037/A0035438 Gray, P. B., & Campbell, B. C. (2009). Human male testosterone, pair bonding and fatherhood. In P. T. Ellison & P. B. Gray (Eds.), Endocrinology of social relationships. Cambridge, MA: Harvard University Press. Griffey, J. A., & Little, A. C. (2014). Infant’s visual preferences for facial traits associated with adult attractiveness judgements: Data from eye-tracking. Infant Behavior and Development, 37(3), 268–275. Gyuris, P., Jarai, R., & Bereczkei, T. (2010). The effect of childhood experiences on mate choice in personality traits: Homogamy and sexual imprinting. Personality and Individual Differences, 49, 467–472. Harman, S.  M., Metter, E.  J., Tobin, J.  D., Pearson, J., & Blackman, M.  R. (2001). Longitudinal effects of aging on serum total and free testosterone levels in healthy men. Journal of Clinical Endocrinology & Metabolism, 86(2), 724–731. Hawkes, K., O’Connell, J. F., Jones, N. B., Alvarez, H., & Charnov, E. L. (1998). Grandmothering, menopause, and the evolution of human life histories. Proceedings of the National Academy of Sciences of the United States of America, 95(3), 1336–1339. Hoier, S. (2003). Father absence and age at menarche. Human Nature, 14(3), 209–233. Jedlicka, D. (1980). A test of the psychoanalytic theory of mate selection. Journal of Social Psychology, 112, 295–299. Johnston, P., Kaufman, J., Bajic, J., Sercombe, A., Michie, P., & Karayanidis, F. (2011). Facial emotion and identity processing development in 5- to 15-year-old children. Frontiers in Psychology, 2, 26. doi:10.3389/fpsyg.2011.00026 Jones, B. C., Boothroyd, L., Feinberg, D. R., & DeBruine, L. M. (2010). Age at menarche predicts individual differences in women’s preferences for masculinized male voices in adulthood. Personality and Individual Differences, 48(7), 860–863. Jones, B.  C., DeBruine, L.  M., Perrett, D.  I., Little, A.  C., Feinberg, D.  R., & Law Smith, M.  J. (2008). Effects of menstrual cycle phase on face preferences. Archives of Sexual Behavior, 37(1), 78–84. Jones, B. C., Hahn, A. C., Fisher, C. I., Wang, H., Kandrik, M., Lao, J., . . . & DeBruine, L. M. (2018). No evidence that more physically attractive women have higher estradiol or progesterone. Psychoneuroendocrinology, 98, 1–5 Jones, B. C., Hahn, A. C., Fisher, C. I., Wang, H., Kandrik, M., Han  C., . . . & DeBruine LM (in press). No compelling evidence that preferences for facial masculinity track changes in women’s hormonal status. Psychological Science.

Mate Preferences Across the Lifespan

Jones, B.  C., Little, A.  C., Boothroyd, L., DeBruine, L.  M., Feinberg, D. R., Law Smith, M. J., . . . Perrett, D. I. (2005). Commitment to relationships and preferences for femininity and apparent health in faces are strongest on days of the menstrual cycle when progesterone level is high. Hormones and Behavior, 48(3), 283–290. doi:10.1016/j.yhbeh.2005.03.010 Jones, B. C., Little, A. C., Penton-Voak, I. S., Tiddeman, B. P., Burt, D. M., & Perrett, D. I. (2001). Facial symmetry and judgements of apparent health: Support for a “good genes” explanation of the attractiveness–symmetry relationship. Evolution and Human Behavior, 22(6), 417–429. Jones, B. C., Perrett, D. I., Little, A. C., Boothroyd, L., Cornwell, R.  E., Feinberg, D.  R., . . . Moore, F.  R. (2005). Menstrual cycle, pregnancy and oral contraceptive use alter attraction to apparent health in faces. Proceedings of the Royal Society B: Biological Sciences, 272(1561), 347–354. doi:10.1098/rspb. 2004.2962 Jones, B.  C., Vukovic, J., Little, A.  C., Roberts, S.  C., & DeBruine, L.  M. (2011). Circum-menopausal changes in women’s preferences for sexually dimorphic shape cues in peer-aged faces. Biological Psychology, 87, 453–455. Kissler, J., & Bauml, K. H. (2000). Effects of the beholder’s age on the perception of facial attractiveness. Acta Psychologica, 104(2), 145–166. Kocsor, F., Gyuris, P., & Bereczkei, T. (2013). The impact of attachment on preschool children’s preference for parentresembling faces—A possible link to sexual imprinting. Journal of Evolutionary Psychology, 11(4), 171–183. Kocsor, F., Saxton, T.  K., Láng, A., & Bereczkei, T.(2016). Preference for faces resembling opposite-sex parents is mod­ erated by emotional closeness in childhood. Personality and Individual Differences, 96, 23–27. Kokko, H., & Johnstone, R.  A. (2002). Why is mutual mate choice not the norm? Operational sex ratios, sex roles and the evolution of sexually dimorphic and monomorphic signalling. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences, 357(1419), 319–330. Kościński, K. (2010). The pattern of facial preferences in boys at early adolescence. Anthropological Review, 73(1), 3–19. Kościński, K. (2011). Life history of female preferences for male faces. Human Nature-An Interdisciplinary Biosocial Perspective, 22(4), 416–438. doi:10.1007/s12110-011-9123-7 Kościński, K. (2013). Facial preferences in early adolescent girls: Pubertal maturity predicts preferences maturity. Collegium Antropologicum, 37(3), 735–743. Kramer, S., Zebrowitz, L. A., San Giovanni, J. P., & Sherak, B. (1995). Infant preferences for attractiveness and babyfaceness. In B.G. Bardy, R.J. Bootsma and Y. Guiard (Eds), Studies in Perception and Action III, Oxford, UK: Psychology Press. Langlois, J. H., Roggman, L. A., & Musselman, L. (1994). What is average and what is not average about attractive faces. Psychological Science, 5(4), 214–220. Law Smith, M., Perrett, D.  I., Jones, B.  C., Cornwell, R.  E., Moore, F.  R., Feinberg, D.  R., . . . Pitman, R.  M. (2006). Facial appearance is a cue to oestrogen levels in women. Proceedings of the Royal Society of London B: Biological Sciences, 273(1583), 135–140. Lawrence, K., Bernstein, D., Pearson, R., Mandy, W., Campbell, R., & Skuse, D. (2008). Changing abilities in recognition of unfamiliar face photographs through childhood and adoles­ cence: Performance on a test of non-verbal immediate memory (Warrington RMF) from 6 to 16 years. [Research Support, Non-U.S. Gov’t]. Journal of Neuropsychology, 2(Pt 1), 27–45.

Lie, H. C., Rhodes, G., & Simmons, L. W. (2008). Genetic diversity revealed in human faces. Evolution, 62(10), 2473–2486. doi:10.1111/j.1558–5646.2008.00478.x Limoncin, E., Ciocca, G., Gravina, G. L., Carosa, E., Mollaioli, D., Cellerino, A., . . . Jannini, E.  A. (2015). Pregnant women’s preferences for men’s faces differ significantly from nonpregnant women. Journal of Sexual Medicine, 12(5), 1142–1151. Little, A. C. (2015). Attraction and human mating. In V. ZeiglerHill, L.M. Welling and T.K. Shackelford (Eds.), Evolutionary perspectives on social psychology (pp. 319–332). New York: Springer International Publishing. Little, A. C., Apicella, C. L., & Marlowe, F. W. (2007). Preferences for symmetry in human faces in two cultures: Data from the UK and the Hadza, an isolated group of hunter-gatherers. Proceedings of the Royal Society of London B: Biological Sciences, 274(1629), 3113–3117. Little, A. C., Burriss, R. P., Petrie, M., Jones, B. C., & Roberts, S.  C. (2013). Oral contraceptive use in women changes preferences for male facial masculinity and is associated with partner facial masculinity. Psychoneuroendocrinology, 38(9), 1777–1785. Little, A. C., Jones, B. C., & DeBruine, L. M. (2008). Preferences for variation in masculinity in real male faces change across the menstrual cycle: Women prefer more masculine faces when they are more fertile. Personality and Individual Differences, 45, 478–482. Little, A. C., Jones, B. C., Penton-Voak, I. S., Burt, D. M., & Perrett, D.  I. (2002). Partnership status and the temporal context of relationships influence human female preferences for sexual dimorphism in male face shape. Proceedings of the Royal Society of London Series B: Biological Sciences, 269(1496), 1095–1100. Little, A. C., Penton-Voak, I. S., Burt, D. M., & Perrett, D. I. (2003). Investigating an imprinting-like phenomenon in humans: Partners and opposite-sex parents have similar hair and eye colour. Evolution and Human Behavior, 24, 43–51. Little, A. C., Saxton, T. K., Roberts, S. C., Jones, B. C., DeBruine, L. M., Vukovic, J., . . . Chenore, T. (2010). Women’s preferences for masculinity in male faces are highest during reproductive age range and lower around puberty and post-menopause. Psychoneuroendocrinology, 35(6), 912–920. doi:10.1016/j. psyneuen.2009.12.006 Marcinkowska, U. M., Dixson, B. J., Kozlov, M. V., & Rantala, M. J. (2015). Men’s preferences for female facial femininity decline with age. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, gbv077. Marcinkowska, U. M., & Rantala, M. J. (2012). Sexual imprinting on facial traits of opposite-sex parents in humans. Evolutionary Psychology, 10, 621–630. McClintock, M. K., & Herdt, G. (1996). Rethinking puberty: The development of sexual attraction. Current Directions in Psychological Science, 5(6), 178–183. McKone, E., Crookes, K., Jeffery, L., & Dilks, D.  D. (2012). A  critical review of the development of face recognition: Experience is less important than previously believed. Cognitive Neuropsychology, 29(1–2), 174–212. McKone, E., Crookes, K., & Kanwisher, N. (2009). The cognitive and neural development of face recognition in humans. Cognitive Neurosciences, 4, 467–482. Orlikoff, R. F. (1990). The relationship of age and cardiovascular health to certain acoustic characteristics of male voices. Journal of Speech, Language, and Hearing Research, 33(3), 450–457.

Boothroyd and Vukovic

157

Penton-Voak, I. S., Jones, B. C., Little, A. C., Baker, S., Tiddeman, B., Burt, D.  M., & Perrett, D.  I. (2001). Symmetry, sexual dimorphism in facial proportions and male facial attractiveness. Proceedings of the Royal Society of London B: Biological Sciences, 268(1476), 1617–1623. Perrett, D.  I., Lee, K.  J., Rowland, D., Yoshikawa, S., Burt, D. M., Henzi, S. P., . . . Akamatsu, S. (1998). Effects of sexual dimorphism on facial attractiveness. Nature, 394, 884–887. Perrett, D. I., May, K. A., & Yoshikawa, S. (1994). Facial shape and judgements of female attractiveness. Nature, 368, 239–242. Perrett, D. I., Penton-Voak, I. S., Little, A. C., Tiddeman, B. P., Burt, D.  M., Schmidt, N., & Barrett, L. (2002). Facial attractiveness judgements reflect learning of parental age characteristics. Proceedings of the Royal Society of London Series B: Biological Sciences, 269, 873–880. Pisanski, K., Hahn, A.  C., Fisher, C.  I., DeBruine, L.  M., Feinberg, D. R., & Jones, B. C. (2014). Changes in salivary estradiol predict changes in women’s preferences for vocal masculinity. Hormones and Behavior, 66(3), 493–497. Pound, N., Penton-Voak, I.  S., & Surridge, A.  K. (2009). Testosterone responses to competition in men are related to facial masculinity. Proceedings of the Royal Society B: Biological Sciences, 276(1654), 153–159. Puts, D. A. (2010). Beauty and the beast: Mechanisms of sexual selection in humans. Evolution and Human Behavior, 31(3), 157–175. doi:10.1016/j.evolhumbehav.2010.02.005 Puts, D. A., Hodges, C. R., Cardenas, R. A., & Gaulin, S. J. C. (2007). Men’s voices as dominance signals: Vocal fundamental and formant frequencies influence dominance attributions among men. Evolution and Human Behavior, 28, 340–344. Quinn, P. C., Yahr, J., Kuhn, A., Slater, A. M., & Pascalis, O. (2002). Representation of the gender of human faces by infants: A preference for female. Perception, 31(9), 1109–1121. Ramsey-Rennels, J.  L., & Langlois, J.  H. (2006). Infants’ differential processing of female and male faces. Current Directions in Psychological Science, 15(2), 59–62. Rantala, M.  J., & Marcinkowska, U.  M. (2011). The role of sexual imprinting and the Westermarck effect in mate choice in humans. Behavioral Ecology and Sociobiology, 65, 859–873. Rantala, M. J., Pölkki, M., & Rantala, L. M. (2010). Preference for human male body hair changes across the menstrual cycle and menopause. Behavioral Ecology, 21, 419–423. Reeve, S. D., Kelly, K. M., & Welling, L. L. (2016). Transitory environmental threat alters sexually dimorphic mate prefe­ rences and sexual strategy. Evolutionary Psychological Science, 2(2), 101–113. Rennels, J.  L., Kayl, A.  J., Langlois, J.  H., Davis, R.  E., & Orlewicz, M. (2016). Asymmetries in infants’ attention toward and categorization of male faces: The potential role of expe­ rience. Journal of Experimental Child Psychology, 142, 137–157. Rhodes, G., Geddes, K., Jeffery, L., Dziurawiec, S., & Clark, A. (2002). Are average and symmetric faces attractive to infants? Discrimination and looking preferences. Perception, 31(3), 315–321. Rhodes, G., Jeffery, L., Watson, T.  L., Clifford, C.  W.  G., & Nakayama, K. (2003). Fitting the mind to the world: Face adaptation and attractiveness aftereffects. Psychological Science, 14(6), 558–566. doi:10.1046/j.0956–7976.2003. psci_1465.x Rhodes, G., Simmons, L. W., & Peters, M. (2005). Attractiveness and sexual behavior: Does attractiveness enhance mating success? Evolution and Human Behavior, 26, 186–201.

158

Rohner, R. P., & Khaleque, A. (2005). Handbook for the Study of Parental Acceptance and Rejection (4th Edition). Storrs, CT: Rohner Research Publications. Roney, J. R., & Simmons, Z. L. (2008). Women’s estradiol predicts preference for facial cues of men’s testosterone. Hormones and Behavior, 53(1), 14–19. doi:10.1016/j.yhbeh.2007.09.008 Roney, J. R., Simmons, Z. L., & Gray, P. B. (2011). Changes in estradiol predict within-women shifts in attraction to facial cues of men’s testosterone. Psychoneuroendocrinology, 36(5), 742–749. doi:10.1016/j.psyneuen.2010.10.010 Rubenstein, A. J., Kalakanis, L., & Langlois, J. H. (1999). Infant preferences for attractive faces: A cognitive explanation. Developmental Psychology, 35(3), 848–855. Samuels, C.  A., Butterworth, G., Roberts, T., Graupner, L., & Hole, G. (1994). Facial aesthetics—Babies prefer attractiveness to symmetry. Perception, 23(7), 823–831. Saxton, T. K. (2016). Experiences during specific developmental stages influence face preferences. Evolution and Human Behavior, 37(1), 21–28. Saxton, T. K., Caryl, P. G., & Roberts, C. S. (2006). Vocal and facial attractiveness judgments of children, adolescents and adults: The ontogeny of mate choice. Ethology, 112(12), 1179–1185. Saxton, T. K., DeBruine, L. M., Jones, B. C., Little, A. C., & Roberts, S. C. (2009). Face and voice attractiveness judgments change during adolescence. Evolution and Human Behavior, 30(6), 398–408. doi:10.1016/j.evolhumbehav.2009.06.004 Saxton, T.  K., DeBruine, L.  M., Jones, B.  C., Little, A.  C., & Roberts, C. S. (2011). A longitudinal study of adolescents’ judgments of the attractiveness of facial symmetry, averageness and sexual dimorphism. Journal of Evolutionary Psychology, 9(1), 43–55. Saxton, T. K., DeBruine, L. M., Jones, B. C., Little, A. C., & Roberts, S. C. (2013). Voice pitch preferences of adolescents: Do changes across time indicate a shift towards potentially adaptive adult-like preferences? Personality and Individual Differences, 55(2), 90–94. Saxton, T.  K., Kohoutova, D., Roberts, S.  C., Jones, B.  C., DeBruine, L.  M., & Havlicek, J. (2010). Age, puberty and  attractiveness judgments in adolescents. Personality and Individual Differences, 49(8), 857–862. doi:10.1016/j.paid.2010. 07.016 Scherf, K.  S., Behrmann, M., & Dahl, R.  E. (2012). Facing changes and changing faces in adolescence: A new model for investigating adolescent-specific interactions between pubertal, brain and behavioral development. Developmental Cognitive Neuroscience, 2(2), 199–219. doi:10.1016/j.dcn. 2011.07.016 Scott, I. M. L., Clark, A. P., Boothroyd, L. G., & Penton-Voak, I. S. (2013). Do men’s faces really signal heritable immuno­ competence? Behavioral Ecology, 24(3), 579–589. doi:10.1093/ beheco/ars092 Scott, I.  M., Clark, A.  P., Josephson, S.  C., Boyette, A.  H., Cuthill, I.  C., Fried, R.  L., . . . Penton-Voak, I.  S. (2014). Human preferences for sexually dimorphic faces may be evolutionarily novel. Proceedings of the National Academy of Sciences of the United States of America, 111(40), 14388–14393. Scott, I.  M.  L., Pound, N., Stephen, I.  D., Clark, A.  P., & Penton-Voak, I.  S. (2010). Does masculinity matter? The contribution of masculine face shape to male attractiveness in humans. PLoS One, 5(10), e13585. doi:10.1371/journal. pone.0013585

Mate Preferences Across the Lifespan

Slater, A., Quinn, P. C., Hayes, R., & Brown, E. (2000). The role of facial orientation in newborn infants’ preference for attractive faces. Developmental Science, 3(2), 181–185. Slater, A., Von der Schulenburg, C., Brown, E., Badenoch, M., Butterworth, G., Parsons, S., & Samuels, C. (1998). Newborn infants prefer attractive faces. Infant Behavior & Development, 21(2), 345–354. Stephen, I. D., Coetzee, V., & Perrett, D. I. (2011). Carotenoid and melanin pigment coloration affect perceived human health. Evolution and Human Behavior, 32(3), 216–227. Stephen, I.  D., & Wei, T.  K. (2015). Healthy body, healthy face? Evolutionary approaches to attractiveness perception. In S. Haque & Sheppard E. (Eds.), Culture and cognition: A collection of critical essays (pp. 45–65). Bern, Switzerland: Peter Lang. Stewart-Williams, S., & Thomas, A.  G. (2013). The ape that thought it was a peacock: Does evolutionary psychology exaggerate human sex differences? Psychological Inquiry, 24(3), 137–168. doi:10.1080/1047840x.2013.804899 Susilo, T., Germine, L., & Duchaine, B. (2013). Face recognition ability matures late: Evidence from individual differences in young adults. Journal of Experimental Psychology: Human Perception and Performance, 39(5), 1212–1217. Thornhill, R., Chapman, J. F., & Gangestad, S. W. (2013). Women’s preferences for men’s scents associated with testosterone and cortisol levels: Patterns across the ovulatory cycle. Evolution and Human Behavior, 34(3), 216–221. doi:10.1016/j. evolhumbehav.2013.01.003 Thornhill, R., & Gangestad, S. W. (2006). Facial sexual dimorphism, developmental stability, and susceptibility to disease in men and women. Evolution and Human Behavior, 27(2), 131–144. Udry, J. R. (1979). Age at menarche, at first intercourse, and at first pregnancy. Journal of Biological Sciences, 11, 411–433. Van Dongen, S., & Gangestad, S. W. (2011). Human fluctuating asymmetry in relation to health and quality: A meta-analysis. Evolution and Human Behavior, 32(6), 380–398. doi:10.1016/ j.evolhumbehav.2011.03.002 Verdonck, A., Gaethofs, M., Carels, C., & de Zegher, F. (1999). Effect of low-dose testosterone treatment on craniofacial

growth in boys with delayed puberty. European Journal of Orthodontics, 21(2), 137–143. Vingilis-Jaremko, L., & Maurer, D. (2013a). The influence of symmetry on children’s judgments of facial attractiveness. Perception, 42(3), 302–320. Vingilis-Jaremko, L., & Maurer, D. (2013b). The influence of averageness on children’s judgments of facial attractiveness. Journal of Experimental Child Psychology, 115(4), 624–639. Vos, D.  R. (1995). The role of sexual imprinting for sex recognition in zebra finches: A difference between males and females. Animal Behaviour, 50(3), 645–653. Vukovic, J., Boothroyd, L. G., Meins, E., & Burt, D. M. (2015). Concurrent parent–child relationship quality is associated with an imprinting-like effect in children’s facial preferences. Evolution and Human Behavior, 36(4), 331–336. Vukovic, J., Jones, B. C., DeBruine, L. M., Little, A. C., Feinberg, D.  R., & Welling, L.  L.  M. (2009). Circum-menopausal effects on women’s judgements of facial attractiveness. Biology Letters, 5(1), 62–64. Webster, M. A., & MacLeod, D. I. (2011). Visual adaptation and face perception. Philosophical Transactions of the Royal Society B: Biological Sciences, 366(1571), 1702–1725. Welling, L.  L.  M., Jones, B.  C., DeBruine, L.  M., Conway, C. A., Smith, M. J. L., Little, A. C., . . . Al-Dujaili, E. A. S. (2007). Raised salivary testosterone in women is associated with increased attraction to masculine faces. Hormones and Behavior, 52(2), 156–161. doi:10.1016/j.yhbeh.2007.01.010 Welling, L. L. M., Jones, B. C., DeBruine, L. M., Smith, F. G., Feinberg, D. R., Little, A. C., & Al-Dujaili, E. A. S. (2008). Men report stronger attraction to femininity in women’s faces when their testosterone levels are high. Hormones and Behavior, 54(5), 703–708. doi:10.1016/j.yhbeh.2008.07.012 Winkielman, P., Halberstadt, J., Fazendeiro, T., & Catty, S. (2006). Prototypes are attractive because they are easy on the mind. Psychological Science, 17(9), 799–806. Wiszewska, A., Pawlowski, B., & Boothroyd, L. G. (2007). Fatherdaughter relationship as a moderator of sexual imprinting: A  facialmetric study. Evolution and Human Behavior, 28, 248–252.

Boothroyd and Vukovic

159

CH A PT E R

10

The Influence of Maternal Stress and Child Maltreatment on Offspring

LillyBelle K. Deer, Kristin Bernard, and Camelia E. Hostinar

Abstract Child maltreatment has significant impacts on developmental trajectories and is one of the most preventable early-life adversities. It shapes the development of many biological systems, with most research concentrating on its influence on stress response systems like the hypothalamicpituitary-adrenal (HPA) axis. This chapter reviews associations between child maltreatment and HPA axis activity across development. One emerging pattern is that of flattened diurnal cortisol slopes in children, adolescents, and adults with childhood maltreatment histories. This effect was moderated by psychiatric diagnosis, maltreatment subtype, and genetic vulnerability in some studies. Effects on cortisol reactivity are more mixed, including reports of higher reactivity, lower reactivity, or no differences compared to nonmaltreated samples. Interventions that focus on enhancing the quality of parent–child relationships early in life may reverse some of the effects of maltreatment. This chapter discusses implications of maltreatment-related alterations in HPA axis function for mental and physical health and concludes with suggested future research directions. Keywords:  maltreatment, HPA axis, cortisol, early-life adversity, maternal stress, intervention

Child maltreatment can have a profound impact on developmental trajectories and is one of the most preventable early-life adversities that children face (Bruce, Gunnar, Pears, & Fisher,  2013; Cicchetti, 2016). Maltreatment during childhood is associated with increased risk of later psychopathology, especially posttraumatic stress disorder (PTSD) and depression (Kaufman & Charney, 2001; Pratchett & Yehuda,  2011; Teicher & Samson,  2013). It also ­forecasts poorer adult physical health, including greater odds of developing cardiovascular disease and diabetes (Ehrlich, Miller, & Chen, 2016; Gilbert et al., 2015), obesity (Danese & Tan, 2014), fibromyalgia, and chronic fatigue syndrome (Borsini, Hepgul, Mondelli, Chalder, & Pariante, 2014; Lee, 2010). It  is theorized that one common pathway leading to the development of these diverse mental and physical health problems involves the dysregulation of the hypothalamic-pituitary-adrenal (HPA)

axis (Burke, Finn, McGuire, & Roche,  2016; Chrousos, 2009; Lee,  2010; Pratchett & Yehuda, 2011). The goal of the present review is to examine the evidence that child maltreatment shapes the ­development of the HPA axis and reveal the nature of the associations between maltreatment and several aspects of HPA functioning: basal activity, reactivity to stressors, and responses to pharmacological challenges. We describe the existing human literature on associations between child maltreatment and indices of HPA activity in childhood, adolescence, and adulthood. We begin by discussing the influence of maternal stress and other risk factors for maltreatment. This is followed by brief overviews of the HPA axis and animal models of maltreatment and its effects on the HPA axis. Then, we summarize the existing human literature on maltreatment and the HPA axis. We conclude by discussing some of the implications of maltreatment-related alterations 161

in HPA functioning for mental and physical health and p ­ roviding some suggested future directions for this area of research.

Child Maltreatment: Definition, Prevalence, and Risk Factors

Although definitions of maltreatment vary based on factors such as culture or purpose (e.g., legal vs. research), maltreatment can be broadly defined as aberrant caregiving behaviors that threaten a child’s well-being and capacity for psychobiological adaptation (Cicchetti, 2016). Maltreatment can be divided into behaviors that reflect abuse (i.e., physical abuse, sexual abuse, emotional abuse) or neglect (i.e., physical neglect, emotional neglect; Barnett, Manly, & Cicchetti, 1993). Physical abuse involves the nonaccidental infliction of physical injury on a child and can range from a temporary injury to a permanent disfigurement. Sexual abuse involves an attempted or actual sexual act between a child and a family member or caretaker for purposes of that person’s sexual enjoyment or financial benefit. Emotional abuse involves threatening behavior that thwarts a  child’s basic emotional needs for psychological safety and security, acceptance and self-­esteem, and age-appropriate autonomy; emotional neglect similarly conveys to a child that he or she is unloved, unwanted, or unworthy but involves the lack of ­appropriate emotional responsiveness, rather than threatening emotional input. Physical neglect can be divided into two different subtypes: failure to provide, which involves the failure to meet the child’s nutritional, medical, or hygiene needs, and lack of supervision, which includes either leaving a child unattended or with an inadequate caregiver. Severity, frequency, and chronicity, as well as the develop­ mental period in which the abuse occurred, are also essential parameters to take into account when assessing child maltreatment (Barnett et al., 1993). The majority of research on the sequelae of child maltreatment has been conducted on samples from the United States and Europe, with comparatively fewer studies reporting results from other regions of the world (Stoltenborgh, Bakermans-Kranenburg, Alink, & van IJzendoorn, 2015). More recent research has begun examining the prevalence and risk factors for child maltreatment in developing countries (Antai, Braithwaite, & Clerk, 2016; Singhi, Saini, & Malhi,  2013). The official prevalence estimate in the United States was of 702,000 abused or neglected children in 2014, or 9.4 victims per 1,000 children, with rates of abuse decreasing with age and neglect being the most common subtype of 162

abuse (U.S.  Department of Health & Human Services, Administration on Children, Youth and Families, Children’s Bureau, 2016). However, statistics based on officially documented cases likely underestimate the overall prevalence of maltreatment, which is 10 times higher according to victim and parent reports (Gilbert et al., 2009). Official records in the United States show that rates of child ­maltreatment have declined since the early 1990s (Jud, Fegert, & Finkelhor,  2016). However, it is unclear whether these statistics represent real change or differences in how these incidents are reported or investigated. For obvious reasons, much less is known about the rate of change over the past two decades among unreported cases. Despite the encouraging indication of improvement over time, child maltreatment remains all too prevalent and is a significant societal concern. Maltreatment is thought to be caused by a ­complex interplay of factors and systems (Garbarino, 1977; Howze & Kotch,  1984; Thompson,  2015). Studies have focused on maternal factors because many of the families where child maltreatment is observed tend to be single-parent families where the mother is the parent, and mothers tend to spend the most time with their children (Taylor, Guterman, Lee, & Rathouz, 2009). The most commonly studied risk factor is maternal stress, which can stem from multiple sources, including intimate partner violence (Antai et  al.,  2016; Taylor et  al.,  2009); having an insecure attachment with their own parents or being maltreated as a child (Cicchetti, Rogosch, & Toth,  2006; De Bellis et  al.,  2001; Widom, Czaja, & DuMont,  2015); having many children (Cicchetti et al., 2006; Kotch et al., 1995); being a single parent or having unstable relationships (Cicchetti et  al.,  2006; Taylor et  al.,  2009); receiving low levels of family support (Cicchetti et al., 2006; Cowen, 2001); suffering from psychopathology such as substance abuse, depression, and PTSD (De Bellis et  al.,  2001; Kotch et  al.,  1995; Taylor et  al.,  2009); having a child with conduct disorder, which can create a vicious cycle of escalating maltreatment and worsening conduct problems (Cowen, 2001; Dodge, Bates, & Pettit, 1990); and certain maternal personality attributes, such as poor impulse control (Cowen,  2001; De Bellis et al., 2001). Low socioeconomic status and being from an area where there is large income inequality are other risk factors (Cicchetti et al., 2006; Cowen, 2001; De Bellis et  al.,  2001; Eckenrode, Smith, McCarthy, & Dineen,  2014; Kotch et  al.,  1995). Residence with a stepparent has also been noted as a

Influence of Maternal Stress and Child Maltreatment

significant epidemiologic risk factor for child abuse, neglect, and murder (Daly & Wilson, 1988, 2005), which predicts outcomes i­ndependently of the other risk factors mentioned previously. Naturalistic studies also note that children living with stepparents exhibit elevated levels of basal cortisol compared to biological children living in the same families, perhaps suggesting greater exposure of ­ stepchildren to adverse events in the home (Flinn, Ward, & Noone, 2005). Evolutionary explanations for these phenomena highlight the reproductive fitness benefits of suppressing violent or conflictual tendencies toward genetically related offspring (Daly & Wilson, 1988). These are some of the predictors that have been linked to higher likelihood of child maltreatment. For the remainder of the chapter we focus on the outcomes of child maltreatment, with a primary emphasis on alterations in the functioning of stress response physiology and possible consequences for mental and physical health. Before proceeding, a brief introduction to the activity of the HPA axis is necessary.

Overview of the Hypothalamic-PituitaryAdrenal Axis

When confronted with physical or psychological challenges that overwhelm the individual’s capacity to cope, the body initiates a number of physiological and behavioral responses through the endocrine, nervous, and immune systems (Gunnar, Doom, & Esposito,  2015). The hypothalamic-pituitary-­adrenal (HPA) axis plays an integral role in these processes by mobilizing energy for coping with stressors and modifying the individual’s responses to similar stressors in the future (Gunnar et al., 2015; Sapolsky, Romero, & Munck, 2000; Smith & Vale, 2006). The activity of the HPA axis can be studied along two basic dimensions: basal functioning and reactivity to stressors (Joëls & Baram, 2009). Basal HPA functioning follows a diurnal rhythm whereby cortisol, the end product of the HPA axis, is secreted in a pulsatile fashion across the day, reaching peak levels in the morning approximately 30 minutes after awakening, and declining gradually across the day to reach minimum levels at night (Joëls & Baram, 2009). Superimposed on this basal rhythm is the reactivity of the HPA axis to physical or psychological threats to well-being (i.e., stressors). Stress-induced cortisol production begins when corticolimbic regions relay threat signals to the paraventricular nucleus (PVN) of the hypothalamus, which releases corticotropin-releasing hormone (CRH) and arginine vasopressin onto the

anterior pituitary gland (Joëls & Baram, 2009; Levy & Tasker, 2012). The pituitary responds by releasing adrenocorticotropic hormone (ACTH) into circulation, which binds to its receptors in the cortex of the adrenal gland. This stimulates the production of the steroid hormone cortisol by the adrenal (Gunnar et al., 2015; Smith & Vale, 2006). Cortisol has pervasive effects across the body (Sapolsky et al., 2000). Acutely, cortisol facilitates the mobilization of energy to the muscles, increases cardiovascular output, sharpens cognition and alertness, and stimulates immune function, while inhibiting other bodily functions that are not as immediately necessary, such as reproductive physiology and appetite (Sapolsky et al., 2000). When cortisol is released into the circulation, it acts upon its receptors throughout the body, of which there are two main types: mineralocorticoid receptors (MRs) and glucocorticoid receptors (GRs; Gunnar et  al.,  2015; Joëls & Baram, 2009). MRs have higher binding affinity for cortisol than GRs and regulate the basal activity of the HPA across the day (Herman et al., 2016). The lower affinity GRs become activated at higher levels of cortisol, including at the peak of the diurnal cycle and during stressors, mediating the effects of these stressors on many organs and systems including the brain (Herman et al., 2016). In the past two decades there has been burgeoning interest in characterizing how early-life experiences shape both the basal activity and the reactivity of the HPA axis to stressors or under pharmacological challenge. Dysregulation of the HPA axis is frequently noted in children and adolescents experiencing psychosocial adversity (Ehlert, 2013; Fisher et  al.,  2016; Tarullo & Gunnar,  2006) and has been increasingly linked to deleterious physical and mental health outcomes (Bruce et  al.,  2013; De Bellis, Spratt, & Hooper, 2011; Ehlert, 2013; Gunnar et al., 2015), as we discuss in more depth in subsequent sections.

Animal Models of Early-Life Maltreatment

Animal models have been critical in substantiating the causal role of maltreatment on neurodevelopmental trajectories (Drury, Sanchez, & Gonzalez, 2016; Sanchez, 2006; Parker & Maestripieri, 2011). Rodent models provided some of the earliest mechanistic data on the effects of low- versus high-quality maternal care on the HPA axis (Meaney & Szyf, 2005; Plotsky & Meaney, 1993). Nonhuman primate models have greatly added to these insights, particularly since there are several important s­ imilarities between them and humans (e.g., the importance Deer, Bernard, and Hostinar

163

of  maternal care for primate development, the prolonged period of postnatal maturation and growth; Sanchez,  2006). As nonhuman primate models are the most similar to human development, we focus on them in this section. Research on early-life adversity in nonhuman primates has primarily focused on maternal stress and its effects on the mother’s HPA system, the offspring’s HPA system, and the dyad’s relationship (Sanchez, 2006; Sanchez, McCormack, & Howell, 2015). Maltreatment tends to occur in 5 to 10 percent of rhesus macaques and other related primate species and occurs primarily in infancy, similar to humans (Maestripieri & Carroll,  1998; Sanchez et al., 2010). Additionally, primates who abuse their offspring tend to have been abused themselves by their mothers, even if they were cross-fostered (i.e., not genetically related to their abusive mothers; Maestripieri, 2005). They also tend to abuse all of their subsequent offspring, irrespective of whether they are adoptive or biological offspring (Maestripieri, Megna, & Jovanovic, 2000). This suggests that the abuse has little to do with the infant and is environmentally transmitted (Maestripieri, Lindell, Ayala, Gold, & Higley, 2005). The variable foraging demand (VFD) model has been used as an ecologically valid way to study the effects of maternal stress on the mother–infant dyad (Andrews & Rosenblum,  1991; Rosenblum & Paully,  1984; Sanchez,  2006). This experimental paradigm disrupts the mothers’ feeding pattern and makes the usual pattern of foraging for food unpredictable. This results in elevated levels of corticotropin-releasing factor (CRF) in the cerebrospinal fluid of the mothers (Coplan et al., 2005) and in more rejecting maternal behavior (e.g., mothers break contact with the infants more often; Rosenblum & Andrews, 1994), compared to mothers maintained under normal feeding conditions. The effects of being randomly assigned to live in a VFD environment are profound on the offspring as well, with infants showing hyperresponsiveness to stressful stimuli and elevated levels of CRF in cerebrospinal fluid (Coplan et al., 2001, 2005). Dysregulation of the HPA system has also been found to persist through the juvenile period, into young adulthood, and in the offspring of the infants raised in VFD conditions (Coplan et  al.,  2001,  2011; Kinnally et al., 2013; Rosenblum, Forger, Noland, Trost, & Coplan, 2001). These experimental studies indicate that induced maternal stress can not only affect the way that dyads interact but also shape the long-term neurobiological development of the offspring. 164

Despite these average trends, there is variability in outcomes, such that some animals are more vulnerable than others to this experimental disruption (Coplan et  al.,  2001). More research is needed to uncover the genetic and experiential bases for these individual differences in reactions to VFD conditions in mothers and their offspring. Other studies have focused on the effect of naturally occurring maltreatment on the offspring’s HPA axis. As in humans, the quality of maternal care alters infants’ HPA axis and their reaction to stress (Drury et al., 2016), as well as their mothers’ ability to buffer their stress response (Sanchez et al., 2015). Maltreated primates exhibit heightened basal cortisol levels in infancy, which appear to normalize at later ages (Howell et al., 2013; Koch, McCormack, Sanchez, & Maestripieri, 2014). Despite this apparent normalization of basal levels, abnormalities in cortisol reactivity persist, including heightened cortisol reactivity to stressors and CRH challenge, and decreased ACTH response to CRH challenge (Drury et al., 2016; Sanchez et al., 2010). In studies of repeated maternal separation, animals similarly show an initial pattern of increased basal cortisol, which is followed by a flattened diurnal cortisol rhythm during the juvenile period, consistent with many studies in humans (Drury et al., 2016). The mechanisms underlying this transition from hypercortisolism early in life to hypocortisolism later in development have yet to be revealed, but downregulation of CRH receptors and epigenetic alterations are thought to be at play (Drury et al., 2016).

Maltreatment and Hypothalamic-PituitaryAdrenal Functioning in Children and Adolescents

The human literature on associations between earlylife maltreatment and indices of HPA axis activity in youth is quite heterogeneous (Tarullo & Gunnar, 2006). As we review in more depth in this section, some studies reveal that maltreated youth exhibit heightened cortisol levels, while others show lower cortisol levels or no differences compared to nonmaltreated groups. There are several reasons for these varying effects. First, these effects depend on the HPA axis indices used (basal levels vs. cortisol reactivity vs. pharmacological challenge). Second, the evidence suggests that effects vary based on features of the maltreatment experience, such as subtype (physical, emotional, or sexual abuse; physical or emotional neglect) or developmental timing. Lastly, there are individual differences based on gender, genetics, and the type of psychopathology

Influence of Maternal Stress and Child Maltreatment

that individuals develop (e.g., internalizing or ­externalizing, posttraumatic stress disorder [PTSD], depression). We discuss findings regarding the ­ ­association between maltreatment and HPA activity in children and adolescents next and highlight how the overall pattern of results differs in light of these factors. Most of the prior studies in youth have examined diurnal cortisol production (e.g., diurnal slope, the typically steep decline in basal cortisol from morning to evening) or cortisol levels at one point during the day. Based on our review of these studies, the emerging pattern was that maltreated children tended to exhibit flattened diurnal cortisol slopes compared to nonmaltreated comparison groups— that is, lower morning cortisol levels and a less steep decline across the day (Bernard, Butzin-Dozier, Rittenhouse, & Dozier, 2010; Bernard, Zwerling, & Dozier, 2015; Dozier et al., 2006; Fisher, Van Ryzin, & Gunnar,  2011). Consistent with this pattern of flatter slopes, some studies found that girls who had experienced sexual abuse had lower levels of morning cortisol than nonmaltreated girls (King, Mandansky, King, Fletcher, & Brewer, 2001), and this effect appears particularly pronounced for girls with PTSD (Keeshin, Strawn, Out, Granger, & Putnam, 2014). Two other studies found that maltreated children were more likely to exhibit flatter cortisol slopes, but in these studies the flatter slopes were more characteristic of maltreated children if they were also depressed (Hart, Gunnar, & Cicchetti, 1996; Kaufman,  1991). Another study found that lower morning cortisol levels and flatter diurnal slopes were more likely among maltreated children exhibiting both internalizing and externalizing symptoms, with a particularly prominent effect in maltreated boys with elevated levels of externalizing symptoms (Cicchetti & Rogosch,  2001b). Both maltreatment subtype and developmental timing appear to matter, as one study found flatter diurnal cortisol slopes among those experiencing early physical or sexual abuse (before age 5), but not for those exposed to abuse later in development or those experiencing other types of maltreatment (Cicchetti, Rogosch, Gunnar, & Toth, 2010). In another study suggesting a role for maltreatment subtype, Bick et al. (2015) reported that adolescents with moderate to severe neglect showed higher levels of afternoon cortisol than adolescents with little to no experience of physical neglect, with no effects of other maltreatment types. Finally, one investigation focusing on the role of genes regulating HPA function among maltreated children reported that flatter slopes

were specific to children with two copies of the TAT haplotype of the CRH receptor 1 (CRHR1) gene, a gene encoding a receptor that binds CRH, a major player in the activation of the HPA axis (Cicchetti, Rogosch, & Oshri, 2011). In conclusion, the basal cortisol literature suggests that maltreated children and adolescents are more likely to evince flatter cortisol slopes across the day compared to nonaffected groups, with some studies finding a main effect of maltreatment and others reporting this effect only in subgroups with psychopathology, greater severity, earlier onset of maltreatment, or a genetic vulnerability. A few studies reported results that at least partially contradict this overall pattern. A recent report showed that girls who had been sexually abused exhibited heightened morning cortisol levels compared to those without a trauma history (Simsek, Yuksel, Kaplan, Uysal, & Alaca, 2015). However, there was a negative correlation between morning cortisol and time since abuse, suggesting decreasing levels of morning cortisol as the time interval from the abuse increased. Furthermore, in this study children who had experienced multiple sexual assaults exhibited lower cortisol levels than the nonmaltreated group. A study with preschool-aged children found that even though children who experienced neglect exhibited the common lowering of morning cortisol compared to a nonmaltreated sample, children who experienced emotional abuse exhibited heightened levels of morning cortisol (Bruce, Fisher, Pears, & Levine, 2009). In a study of school-aged children, those experiencing multiple types of abuse (physical and sexual abuse) showed elevated morning cortisol levels compared to the nonmaltreated group, though children exposed only to physical abuse showed lower levels of morning cortisol than the nonmaltreated group (Cicchetti & Rogosch, 2001a). Finally, a longitudinal study of sexually abused females reported higher morning cortisol in 6- to 16-year-olds recruited within six months of disclosure of the abuse, but this pattern changed across development such that the abused females showed decreasing morning cortisol levels over time and exhibited lower morning cortisol in young adulthood (ages 20 to 32 years) compared to the nonabused control group (Trickett, Noll, Susman, Shenk, & Putnam, 2010). The conclusion emerging from some of these studies (Simsek et  al.,  2015; Trickett et al., 2010) is that morning cortisol levels can be elevated in some instances in the aftermath of the maltreatment experience, but may lower below normative levels over time. This pattern is Deer, Bernard, and Hostinar

165

consistent with meta-analytic findings suggesting that the HPA axis may hypersecrete cortisol soon after an adverse event, but may down-regulate and switch to a pattern of hyposecretion if the stressor becomes chronic or is distant in time (Miller, Chen, & Zhou,  2007). Indeed, naturalistic studies of children in their home environments suggest that cortisol levels are elevated immediately after being reprimanded or punished by a caregiver, then decrease below normal levels in subsequent days and appear below normal in children chronically exposed to stressful situations (for a review, see Flinn et al., 2005). Finally, two studies found no difference in cortisol levels between maltreated children and nonmaltreated comparison groups (Doom, Cicchetti, & Rogosch, 2014; Fisher, Gunnar, Chamberlain, & Reid, 2000). This may be due to small sample sizes (Fisher et al., 2000) or because of greater variability in cortisol levels among the maltreated youth, which can make it difficult to capture group-level mean differences (Doom et al., 2014). In the literature examining cortisol reactivity to a stressor, findings were mixed. Some studies reported that maltreated children and adolescents showed a blunted cortisol response to the stressor compared to nonmaltreated youth (Fisher, Kim, Bruce, & Pears, 2012; Gordis, Granger, Susman, & Trickett, 2008; Hart, Gunnar, & Cicchetti, 1995; MacMillan et  al.,  2009; Peckins, Susman, Negriff, Noll, & Trickett, 2015; Sumner, McLaughlin, Walsh, Sheridan, & Koenen,  2014). Others reported a heightened ­response (Bugental, Martorell, & Barraza,  2003; Harkness, Stewart, & Wynne-Edwards, 2011) or no differences in cortisol reactivity by maltreatment status (Cook, Chaplin, Sinha, Tebes, & Mayes, 2012; Eisen, Goodman, Qin, Davis, & Crayton, 2007). To illustrate some of these findings, one of the earliest studies of cortisol reactivity was on reactions to social conflict in the classroom in preschool boys who had been maltreated, who showed a blunted response to this social stressor (Hart et  al.,  1995). Another widely used paradigm for eliciting a stress response is the Trier Social Stress Test (TSST), a social-evaluative stressor involving public speaking and mental arithmetic components (Kirschbaum, Pirke, & Hellhammer,  1993). In one experiment using the TSST, foster children who had endured physical abuse, and children exposed to physical abuse along with prenatal substance exposure, displayed a blunted cortisol response over time when compared to children who had endured other abuse subtypes and nonmaltreated children (Fisher et  al.,  2012). In another study, maltreated female 166

adolescents showed a blunted cortisol response to the TSST when compared with nonmaltreated ­females (MacMillan et al., 2009). In another analysis of adolescents using the TSST, adolescents with a history of child maltreatment showed a blunted cortisol response to the TSST, and this effect was more pronounced for adolescents who were carriers of one or two G alleles for the rs110402 polymorphism of the CRHR1 gene (Sumner, McLaughlin, Walsh, Sheridan, & Koenen,  2014). As already discussed, the CRHR1 gene plays important roles in regulating HPA function, and this study suggests that variation in the CRHR1 gene may shape the effect of  maltreatment on cortisol reactivity (Sumner et al., 2014). In one of only a few longitudinal studies in this area of research, maltreated adolescents were more likely to have a blunted cortisol response to the TSST than the nonmaltreated comparison group when they were 12 and 13 years old, but no differences emerged at age 18 (Peckins et al., 2015). This was interpreted as a form of adaptive calibration to the environment over time, meaning that the stress response may adapt to meet the changing demands of the environment across development. Adaptation is presumed to be a lengthy process, and thus HPA functioning at any point in time likely reflects longer periods of cumulative life experiences (Peckins et al., 2015). Some studies reported mixed results within the same sample based on maltreatment subtype or psychiatric diagnosis. For instance, adolescents who had experienced physical and/or sexual abuse exhibited a blunted cortisol response to the TSST in comparison to nonmaltreated adolescents, whereas adolescents who had experienced neglect or emotional abuse did not differ from the nonaffected comparison group (Trickett, Gordis, Peckins, & Susman, 2014). Moreover, adolescents who have a history of maltreatment along with mild to moderate levels of depression show a heightened and prolonged cortisol response to the TSST (Harkness et al., 2011), whereas those with moderate or severe depression evince a blunted cortisol response whether they were maltreated or not. Another approach to understanding the effect of maltreatment on the HPA axis is to use pharmacological challenges. In these studies, exogenous CRH, ACTH, or dexamethasone (Dex, a synthetic glucocorticoid) is administered. These studies are quite rare in pediatric populations, and sample sizes are small, which may explain why findings are quite mixed. For instance, one study reported lower ACTH response to CRH challenge in sexually abused youth

Influence of Maternal Stress and Child Maltreatment

(De Bellis et  al.,  1994), whereas another reported higher ACTH response to CRH challenge in depressed abused children currently living in adverse situations (Kaufman et  al.,  1997). Both investigations reported no difference in CRH-stimulated cortisol levels. These findings suggest different abnormalities in pituitary responses to CRH (perhaps due to depressed versus nondepressed status), and possible compensatory mechanisms at the adrenal level allowing similar cortisol levels compared to controls despite differing ACTH levels. Two Dex challenge studies have also reported no differences in postDex cortisol levels, but lower ACTH responses (Bicanic et  al.,  2013; Duval et  al.,  2004). In contrast, one study reported that children who scored higher on the Child Trauma Questionnaire exhibited lower cortisol levels after Dex challenge than children who had lower scores (Lipschitz et  al., 2003), suggesting a more robust negative feedback mechanism among these children. More research is needed to examine the role of psychiatric symptoms and duration and severity of maltreatment in  explaining these somewhat inconsistent findings, as well as the extent to which these presumed alterations in pituitary and adrenal function are long-lasting.

Child Maltreatment and Hypothalamic-Pituitary-Adrenal Functioning in Adulthood

There have been few studies examining diurnal cortisol slopes in adults with maltreatment histories. However, the few extant studies reveal some evidence consistent with the presence of flatter diurnal slopes, similar to the literature on youth. For instance, participants who had been adopted in childhood after experiencing neglect or abuse were more likely to exhibit flatter cortisol slopes in adulthood, especially if they suffered from anxiety and had a history of severe neglect (vs. abuse; van der Vegt, van der Ende, Huizink, Verhulst, & Tiemeier,  2010; van der  Vegt, van der Ende, Kirschbaum, Verhulst, & Tiemeier, 2009). Consistent with the possibility of flatter slopes, adults who were maltreated as children also tended to show lower morning cortisol levels (Power, Thomas, Li, & Hertzman, 2012) and higher afternoon levels (Bremner et al., 2003). Most empirical investigations with adults maltreated as children have focused on cortisol reactivity, but the findings in these studies using experimental reactivity paradigms are mixed. The most widely used paradigm was the TSST, but a few studies employed other psychosocial stressors (e.g., cognitive

challenge: Bremner et al., 2003; conflict role-play: Hagan, Roubinov, Mistler, & Luecken,  2014). In one study on cortisol reactivity using the TSST, adults who retrospectively reported childhood maltreatment showed a blunted cortisol response to the stressor compared to adults without childhood maltreatment exposure (Carpenter et  al.,  2007). ­ Another study found that women who experienced childhood physical abuse exhibited a blunted cortisol response to the TSST when compared to women who had experienced other subtypes of abuse and nonmaltreated women (Carpenter, Shattuck, Tyrka, Geracioti, & Price, 2011). A third study (Buchmann et al., 2014) used the TSST as a stress elicitor to look at the interaction between child maltreatment and the FKBP5 rs1360780 genotype in explaining cortisol reactivity in emerging adults (mean age 19). FKBP5 is a glucocorticoid receptor–regulating cochaperone (i.e., proteins that assist in protein folding and other functions) that has been implicated in the negative feedback inhibition of the HPA axis. Analyses revealed that adults who reported high levels of maltreatment in childhood and were carriers of the rs1360780 CC genotype exhibited a lower cortisol response to the TSST when compared to nonmaltreated adults, whereas carriers of the T allele did not show this difference (Buchmann et al., 2014). In contrast to these reports of lower reactivity, one study found higher cortisol responses to the TSST in women who had experienced child maltreatment if they were also depressed compared to three groups: nondepressed maltreated, depressed nonmaltreated, and nondepressed nonmaltreated women (Heim et al., 2000, 2002). These findings suggest that genetic differences and concurrent ­psychopathology moderate the effect of childhood maltreatment on adult cortisol reactivity. Among the studies using psychosocial stress tests other than the TSST, null results predominated. For instance, a study of adults meeting criteria for PTSD related to child abuse showed no group differences in reaction to a series of cognitive challenges (Bremner et al., 2003), though the study reported higher anticipatory cortisol levels pretask in the PTSD group. Another study with young adults (aged 18 to 22) found a null effect of maltreatment on cortisol reactivity (Hagan et  al.,  2014). This ­experiment used a conflict role-play challenge with a confederate in the lab acting as a peer and found that there was no difference in cortisol reactivity between the maltreated and nonmaltreated group of emerging adults (Hagan et al., 2014). It is difficult to interpret these null results with any confidence. Deer, Bernard, and Hostinar

167

They may be due to the fact that these protocols did not elicit robust enough cortisol responses to reveal individual differences in the samples. Alternatively, lack of main effects at the group level may be the result of averaging across subgroups with opposite profiles (i.e., blunted and heightened reactivity). Indeed, when Hagan et  al. (2014) probed their results further, maltreatment was associated with lower reactivity in those with high levels of externalizing symptoms, and higher reactivity in those with internalizing symptoms. Pharmacological challenge tests have been conducted much more frequently in adults compared to children or adolescents. Two Dex suppression studies revealed enhanced negative feedback (or “super suppression” of the HPA axis) as indicated by lower ACTH or lower cortisol responses in adults with abuse or neglect histories compared to controls, especially if they had a diagnosis of depression or PTSD (Newport, Heim, Bonsall, Miller, & Nemeroff, 2004; Stein, Yehuda, Koverola, & Hanna,  1997; Watson et al., 2007). Clinical studies using a CRH or Dex/CRH challenge have predominantly reported hyporesponsiveness of the HPA axis (lower ACTH and/or lower cortisol) following these challenges in abused patients, especially if they were also suffering from depression or PTSD (Carpenter et al., 2009; Heim, Newport, Bonsall, Miller, & Nemeroff, 2001; Klaassens et  al.,  2009; Rinne et  al.,  2002; Stein et al., 1997). In contrast, chronically abused women with borderline personality disorder (BPD) exhibited a heightened ACTH and cortisol response to a Dex/CRH test (Rinne et al., 2002), and adults with the GG genotype for two single-nucleotide polymorphisms (SNPs) in the CRHR1 gene (rs110402 and rs242924) also exhibited higher cortisol responses to the Dex/CRH challenge than never-maltreated adults and maltreated adults with the other alleles (Tyrka et al., 2009). The pharmacological findings add to the mixed patterns emerging from reactivity studies to underscore the importance of considering psychiatric diagnosis and variations in genes regulating the HPA axis in future studies, given that these two factors may explain the widespread heterogeneity in the associations between maltreatment and the activity of the HPA axis.

Intervention Studies

Most of the human literature on maltreatment and the HPA axis has been correlational, making it ­difficult to infer whether the effects described previously are causal in nature. A growing number of intervention studies have provided important ­evidence 168

that HPA alterations observed in maltreatment can be causal and also reversible. We summarize these studies here. A number of interventions have been developed to improve parent–child relationships in high-risk families (e.g., families referred to Child Protective Services). We review some of the most commonly used interventions in studies that specifically assessed effects on children’s cortisol regulation. The Attachment and Biobehavioral Catch-up (ABC) intervention is a parent-coaching program that aims to enhance the quality of attachment ­between the caregiver and the child and support children’s self-regulatory capabilities. Parent coaches provide in-the-moment feedback in response to parent–child interactions as they occur to promote nurturing, responsive, and nonfrightening care. In randomized controlled trials, ABC has been compared with an active control condition that is meant to enhance motor, cognitive, and language skills (Bernard, Dozier, Bick, & Gordon, 2015; Bernard, Hostinar, & Dozier,  2015; Dozier, Peloso, Lewis, Laurenceau, & Levine, 2008). In a study of infants and toddlers who had been involved with Child Protective Services, infants who received ABC showed more typical diurnal cortisol production than the infants who received the control intervention at a follow-up assessment within several months of the intervention (Bernard, Dozier, et al., 2015). At a preschool follow-up assessment of the same children, at approximately three years postintervention, young children who had received the ABC intervention still showed a more normalized diurnal rhythm than the control group, who showed a flattened diurnal rhythm (Bernard, Hostinar, & Dozier, 2015). In a study of infants in foster care, children randomly assigned to receive ABC showed a lower cortisol response during the Strange Situation, an assessment of attachment that involves brief separations from caregivers, than children in the control intervention (Dozier et al., 2008). Taken together, these studies suggest that the ABC intervention was effective in helping children to regulate both diurnal production and stress reactivity responses of the HPA system. The Multidimensional Treatment Foster Care for Preschoolers (MTFC-P) is a caregiver-based preventative intervention that aims to address the developmental and socioemotional needs of preschool-aged foster children by providing additional support to foster parents, such as parenting skills that promote responsiveness and consistency (Fisher & Stoolmiller, 2008; Fisher, Stoolmiller, Gunnar, & Burraston,

Influence of Maternal Stress and Child Maltreatment

2007; Fisher et al., 2011; Graham et al., 2012). In one study, children who received MTFC-P did not show the increasingly flattened diurnal rhythm over time, whereas foster children in regular foster care did (Fisher et al., 2007). Following foster care placement changes, which were found to be associated with dysregulation in cortisol rhythms among children in regular foster care, children who received MTFC-P continued to show the typical cortisol decline from morning to evening (Fisher et  al., 2011). In a study aimed to identify potential mechanisms leading to changes in cortisol reactivity, Fisher and Stoolmiller (2008) examined changes in foster parent stress as a result of MTFC-P. Indeed, foster parents who received MTFC-P demonstrated lower stress than foster parents in the comparison group. Further, in the regular foster care comparison group, higher foster parent stress was associated with a more blunted diurnal rhythm and lower morning cortisol levels (Fisher & Stoolmiller, 2008). Finally, in a study examining the potentially challenging transition to school, foster children who received the MTFC-P showed a similar response to a community sample of children, whereas children in regular foster care showed elevated morning ­cortisol levels on the fifth day of the transition (Graham et al., 2012). A study that investigated the effects of Promoting First Relationships (PFR), an intervention aimed at improving attachment quality in children who had a recent caregiver change, found no effects on basal morning cortisol levels after the intervention. However, older children who received the intervention exhibited higher cortisol reactivity postintervention compared to children who did not receive the intervention and younger children receiving the intervention (Nelson & Spieker, 2013). This finding was interpreted as a restoration of typical physiological stress reactivity as a result of the intervention, given that most of these at-risk children showed a flat cortisol production pattern during a challenging situation at baseline. Child–Parent Psychotherapy (CPP) is a dyadic intervention that aims to enhance the quality of the relationship between a parent and child by increasing parents’ reflective capacity and sensitivity to child cues; the approach is supportive and nondirective. Over time, maltreated children whose parents received CPP, as well as maltreated children whose parents received a psychoeducational parenting intervention, displayed morning cortisol levels that were similar to nonmaltreated children; in contrast, maltreated children who did not receive an intervention

showed decreasing cortisol levels over time (Cicchetti, Rogosch, Toth, & Sturge-Apple, 2011). Collectively, these results indicate that the effects of child maltreatment on the HPA axis can be prevented or reversed with interventions that ­ focus on enhancing parenting and parent–child relationships, which can inform future intervention strategies and policymaking (Slopen, McLaughlin, & Shonkoff, 2014).

Child Maltreatment and Mental Health

Child maltreatment is associated with increased vulnerability to mental health disorders across the lifespan. This includes higher risk of developing depression, PTSD, anxiety disorders, conduct disorder, substance abuse, and personality disorders (Edwards, Holden, Felitti, & Anda,  2003; Scott, McLaughlin, Smith, & Ellis, 2012; Teicher & Samson, 2013). However, not everyone who experiences maltreatment suffers from mental health issues later in life (Cicchetti, 2016). It has been proposed that alterations in the HPA axis following early trauma may have an effect on whether psychopathology develops, and what form it takes (Heim, Newport, Mletzko, Miller, & Nemeroff, 2008; Susman, 2006). PTSD is one of the most prevalent mental health outcomes linked to child maltreatment. There is a high rate of PTSD among adults who were exposed to maltreatment during childhood (estimated to be as high as 72 to 100 percent in some studies), and there are higher rates of child maltreatment among adult patients diagnosed with PTSD than among those without PTSD (Pratchett & Yehuda,  2011). The current literature on links between childhood maltreatment and adult PTSD offers the hypothesis that one of the mediating pathways for this vulnerability may involve the sensitization of the HPA axis by early maltreatment, which may be a particularly potent risk factor for those exposed to revictimization in adulthood (Brewin, Andrews, & Valentine, 2000; Pratchett & Yehuda, 2011). PTSD that develops after various types of adult trauma has tended to  be associated with lower basal cortisol levels and enhanced negative feedback of the HPA axis (Pratchett & Yehuda,  2011). There are only a few examinations of adults with PTSD secondary to childhood maltreatment, and results are somewhat inconsistent. For instance, one study assessed 24-hour plasma cortisol samples in adult women with a history of child abuse and concurrent PTSD and revealed lower basal cortisol levels in the afternoon in these women compared to abused women without PTSD and women without abuse or PTSD Deer, Bernard, and Hostinar

169

(Bremner, Vermetten, & Kelley,  2007). Another study reported higher levels of 24-hour urinary ­cortisol in women with PTSD related to childhood sexual abuse (Lemieux & Coe,  1995). This latter sample also exhibited a tendency toward obesity, which may explain the elevated cortisol output. Although childhood abuse appears to create a vulnerability for later PTSD especially in those ­ who are re-exposed to trauma in adulthood (Brewin et al., 2000), more research is clearly needed to test the hypothesis that this vulnerability is mediated by dysregulation of the HPA axis. Depression is another prevalent condition in those exposed to childhood maltreatment (Batten, Aslan, Maciejewski, & Mazure, 2004; Heim et al., 2008). Two recent reviews of more than four decades of research on depression and the HPA axis have reported a predominant pattern of hyperactivity of the HPA axis in depression (Pariante & Lightman, 2008; Stetler & Miller, 2011), with the strongest effects being noted in older inpatients with melancholic or psychotic depressive features (Stetler & Miller, 2011). Early-life stress is thought to affect the functioning of glucocorticoid receptors in the brain and the periphery, which results in impaired negative feedback mechanisms and elevated levels of basal or reactive cortisol in many depressed patients (Pariante & Lightman,  2008). Results become less consistent when considering HPA functioning in the context of child maltreatment and concurrent depression, which can manifest with hypo- or hyperreactivity of the HPA axis ­depending on comorbidity with PTSD, duration of early-life stress exposure, and ongoing stress in adulthood (Penza, Heim, & Nemeroff, 2003). Conduct disorder, callous-unemotional behavior, aggression, and externalizing symptoms are also more prevalent among individuals who experience child maltreatment than in the general population (Dackis, Rogosch, & Cicchetti, 2015; Gowin et al., 2013; Maniglio, 2015). A growing literature has reported physiological hypoarousal (e.g., low basal cortisol levels and low sympathetic reactivity) in those with antisocial behavior (Alink et  al.,  2008; Susman, 2006), though effect sizes are smaller than once thought. Hypoarousal of stress systems is thought to be linked to fearlessness, lack of empathy, and seeking stimulation from the environment through disruptive or aggressive behavior. There is increasing evidence that among children exposed to interpersonal violence, blunted cortisol patterns are associated with externalizing behavior (Bernard, Zwerling, & Dozier, 2015; Busso, McLaughlin, & 170

Sheridan,  2016). Some have proposed that the ­effects of early-life trauma on later conduct problems may be, at least in part, mediated by attenuated HPA responses (Susman,  2006), though a ­subgroup of those who exhibit antisocial behavior and HPA axis hypoactivity show these behaviors ­independently of environmental adversity (Hawes, Brennan, & Dadds,  2009). Furthermore, some children show disruptive behavior in the context of HPA hyperreactivity (Hawes et al., 2009). More research is needed to test whether HPA activity plays a causal role in the development of aggressive and antisocial behavior in general and for those exposed to maltreatment in particular. In sum, heterogeneity seems to be the norm rather than the exception when examining patterns of HPA activity linked to psychopathology in those exposed to maltreatment. Several types of psychopathology appear to be associated with hypercortisolism in some studies and hypocortisolism in others. These complex patterns may be, at least in part, explained by a recent meta-analysis of studies on chronic stress and HPA activity (Miller et al., 2007), which revealed that HPA activity increases acutely after stressor onset but reduces over time as stressors become more chronic. Incorporating detailed assessments of recent stressful life events and lifetime patterns of acute and chronic stress exposure in future studies may add clarity to this literature.

Child Maltreatment and Physical Health

Child maltreatment has also been connected to an increased number of hospital visits in adulthood and greater morbidity due to multiple causes, including cardiovascular disease and diabetes (Ehrlich et al., 2016; Gilbert et al., 2015), obesity (Danese & Tan, 2014), fibromyalgia, and chronic fatigue syndrome (Borsini et al., 2014; Lee, 2010). As with the effects of child maltreatment on the development of psychopathology, maltreatment does not always ensure physical health problems, but these conditions are more common in maltreated individuals. Importantly, most theoretical accounts of how maltreatment instantiates these health risks have hypothesized that the activity of the HPA axis plays a central role (Chrousos, 2009). Cortisol has widespread effects throughout the body, including effects on metabolism, immunity, and brain development (Sapolsky et al., 2000). Thus, it is biologically plausible that chronic activation of the HPA axis during abuse or neglect might foster the development of numerous physical health problems. However, few studies have directly tested mediational models

Influence of Maternal Stress and Child Maltreatment

linking child maltreatment to altered HPA functioning and, in turn, physical health symptoms. Accumulating evidence suggests that links ­between HPA activity and inflammation may play a key role in this mediational pathway. Childhood maltreatment is associated with elevated inflammatory biomarkers (Coelho, Viola, Walss-Bass, Brietzke, & Grassi-Oliveira,  2014), which have been implicated in the development of many conditions with inflammatory underpinnings, such as coronary heart disease, diabetes, and obesity (Hotamisligil, 2006). Cortisol is known to play an important role in countering the proinflammatory activity of monocytes and macrophages (Irwin & Cole, 2011), and there is evidence that dysregulated cortisol levels (either abnormally low or chronically high) can impair the control of inflammatory responses (Raison & Miller,  2003; Sapolsky et  al.,  2000). Thus, dysregulation of the HPA axis may be one pathway through which maltreatment may lead to excessive inflammation and chronic diseases of aging precipitated by inflammation (Glaser & Kiecolt-Glaser,  2005). Despite the important role likely played by the HPA axis and inflammation, we must recognize other pathways through which maltreatment might impair health, such as the adoption of health-compromising behaviors that are occasioned or exacerbated by stress, such as smoking, overeating, or sedentary lifestyles (KiecoltGlaser & Glaser,  1988; Raposa, Bower, Hammen, Najman, & Brennan, 2014). In conclusion, there is accumulating evidence to provide piecemeal support for associations between maltreatment and HPA dysregulation, HPA activity and inflammation, and inflammation and multiple disease endpoints. However, testing these full, multistep pathways within the same participants followed longitudinally will be necessary to explicitly examine the mediating role of the HPA axis in the development of many of these health conditions.

Future Directions

Recent literature has brought a greater understanding of the range of possible effects of child maltreatment on the HPA system and its implications for mental and physical health, but many questions remain unanswered. One such question is whether the effects of maltreatment on the HPA axis are transient or persistent across the lifespan, and how they may change with development. A frequently noted gap in the literature has been the scarcity of longitudinal studies to address this question (Bernard et al., 2015; Cicchetti et al., 2011; Heim et al., 2008).

The few existing longitudinal investigations suggest a potential switch from cortisol hypersecretion in the aftermath of trauma to a pattern of hyposecretion later in development (e.g., Trickett et  al., 2010), but more studies are needed to corroborate this pattern. Intervention studies that follow participants over time are especially needed to examine the p ­ ersistence of intervention effects, and to compare interventions to each other (Fisher et al., 2011, 2016; Slopen et al., 2014). There is also a dearth of intervention studies attempting to ameliorate or normalize HPA functioning in adults with childhood trauma exposure. As summarized previously, interventions with children have shown that there is some plasticity in children’s HPA activity, but it is unclear whether this plasticity extends into adulthood. There is also a lack of cortisol reactivity studies in younger maltreated children. The majority of the existing research with these younger participants has focused on diurnal cortisol, whereas studies that incorporated paradigms to test cortisol reactivity have mostly been conducted with preteenagers, teenagers, and adults. This makes it difficult to draw conclusions about developmental changes that may occur across the lifespan, and provides an incomplete characterization of HPA functioning within each life stage. There is some emerging evidence supporting the role of maltreatment subtypes and varying levels of chronicity in shaping HPA outcomes, but this evidence base needs to be expanded (McCrory, De Brito, & Viding,  2010). Furthermore, the role of interactions between genetic variation and maltreatment experiences in influencing psychiatric outcomes is only beginning to be explored. Current results in these areas are relatively mixed, and the number of studies conducted is quite limited. This research could lead to more targeted intervention programs and a greater understanding of the development of individual differences in HPA and mental health outcomes following child maltreatment. More evidence is also needed to test the mediating role of HPA alterations in the relation between child maltreatment and mental or physical health. This research has only recently begun to gain traction (e.g., Bernard, Zwerling, & Dozier,  2015; Hagan et al., 2014), but it is critical for testing many contemporary theories of how early-life stress leads to later psychopathology and chronic disease. There are also a few methodological limitations that have been noted in this literature (McCrory et al., 2010; Pollak, 2015). These limitations include Deer, Bernard, and Hostinar

171

inconsistencies in how HPA function and maltreatment are assessed across studies, the widespread use of retrospective designs in adult studies, and small sample sizes. Cortisol sampling occurs in multiple settings (e.g., laboratory, home, clinic), and there is little consistency in the timing of sample collection. This makes the results incomparable to each other in some respects, and makes it difficult to quantitatively summarize this literature to determine how robust the effects are. Differences in how maltreatment is classified may also lead to heterogeneous patterns of results. Some researchers have used a classification system such as the Maltreatment Classification System (Barnett et  al.,  1993). Other studies use records from Childhood Protective Services and other similar agencies, whereas others collect parental report or retrospective self-report measures. Retrospective and prospective assessments of maltreatment both relate to negative outcomes; however, these different measurement strategies seem to identify only partially overlapping populations, and more research is needed to understand their relationship. Additionally, the prevalence of small sample sizes in this literature (McCrory et al., 2010) is not surprising given the difficulty of recruiting from this vulnerable population, but it can also lower the statistical power to detect effects. Although it may be difficult to address these issues, it is vital to take them into consideration to move this research agenda forward and improve the ­evidence base. In conclusion, it is clear that child maltreatment is associated with altered patterns of HPA activity, which may end up explaining the development of several deleterious mental and physical health outcomes. Intervention studies with children provide hope that these effects can be prevented or mitigated, but more research is needed before concluding that these benefits are long-lasting into adulthood.

Acknowledgments

This manuscript was prepared with support from the Center for Poverty Research at the University of California, Davis, and NSF grant 1327768.

References

Alink, L.  R., van Ijzendoorn, M.  H., Bakermans-Kranenburg, M. J., Mesman, J., Juffer, F., & Koot, H. M. (2008). Cortisol and externalizing behavior in children and adolescents: Mixed meta-analytic evidence for the inverse relation of basal cortisol and cortisol reactivity with externalizing behavior. Developmental Psychobiology, 50(5), 427–450. doi:10.1002/ dev.20300

172

Andrews, M.  W., & Rosenblum, L.  A. (1991). Attachment in  monkey infants raised in variable- and low-demand environments. Child Development, 62(4), 686–693. Antai, D., Braithwaite, P., & Clerk, G. (2016). Social determinants of child abuse: Evidence of factors associated with maternal abuse from the Egypt demographic and health survey. Journal of Injury and Violence Research, 8(1), 25–34. doi:10.5249/jivr.v8i1.630 Barnett, D., Manly, J. T., & Cicchetti, D. (1993). Defining child maltreatment: The interface between policy and research. In  D.  Cicchetti & S.  L.  Toth (Eds.), Child abuse, child development, and social policy (Vol. 8, pp. 2–54). Norwood, NJ: Ablex. Batten, S. V., Aslan, M., Maciejewski, P. K., & Mazure, C. M. (2004). Childhood maltreatment as a risk factor for adult cardiovascular disease and depression. Journal of Clinical Psychiatry, 65(2), 249–254. Bernard, K., Butzin-Dozier, Z., Rittenhouse, J., & Dozier, M. (2010). Cortisol production patterns in young children living with birth parents vs children placed in foster care following involvement of Child Protective Services. Archives of Pediatrics and Adolescent Medicine, 164(5), 438–443. doi:10.1001/archpediatrics.2010.54 Bernard, K., Dozier, M., Bick, J., & Gordon, M.  K. (2015). Intervening to enhance cortisol regulation among children at  risk for neglect: Results of a randomized clinical trial. Development and Psychopathology, 27(3), 829–841. doi:10.1017/ S095457941400073X Bernard, K., Hostinar, C. E., & Dozier, M. (2015). Intervention effects on diurnal cortisol rhythms of Child Protective Servicesreferred infants in early childhood: Preschool follow-up results of a randomized clinical trial. JAMA Pediatrics, 169(2), 112–119. doi:10.1001/jamapediatrics.2014.2369 Bernard, K., Zwerling, J., & Dozier, M. (2015). Effects of early adversity on young children’s diurnal cortisol rhythms and externalizing behavior. Developmental Psychobiology, 57(8), 935–947. doi:10.1002/dev.21324 Bicanic, I. A., Postma, R. M., Sinnema, G., De Roos, C., Olff, M., Van Wesel, F., & Van de Putte, E. M. (2013). Salivary cortisol and dehydroepiandrosterone sulfate in adolescent rape victims with post traumatic stress disorder. Psychon­ euroendocrinology, 38(3), 408–415. doi:10.1016/j.psyneuen. ­2012.06.015 Bick, J., Nguyen, V., Leng, L., Piecychna, M., Crowley, M. J., Bucala, R., . . . Grigorenko, E. L. (2015). Preliminary associations between childhood neglect, MIF, and cortisol: Potential pathways to long-term disease risk. Developmental Psychobiology, 57(1), 131–139. doi:10.1002/dev.21265 Borsini, A., Hepgul, N., Mondelli, V., Chalder, T., & Pariante, C.  M. (2014). Childhood stressors in the development of  fatigue syndromes: A review of the past 20 years of research. Psychological Medicine, 44(9), 1809–1823. doi:10.1017/ S0033291713002468 Bremner, J. D., Vermetten, E., & Kelley, M. E. (2007). Cortisol, dehydroepiandrosterone, and estradiol measured over 24 hours in women with childhood sexual abuse-related posttraumatic stress disorder. Journal of Nervous and Mental Disease, 195, 919–927. Bremner, J. D., Vythilingam, M., Vermetten, E., Adil, J., Khan, S., Nazeer, A., . . . Charney, D. S. (2003). Cortisol response to a cognitive stress challenge in posttraumatic stress disorder (PTSD) related to childhood abuse. Psychoneuroendocrinology, 28(6), 733–750.

Influence of Maternal Stress and Child Maltreatment

Brewin, C.  R., Andrews, B., & Valentine, J.  D. (2000). Meta-analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults. Journal of Consulting and Clinical Psychology, 68(5), 748–766. Bruce, J., Fisher, P.  A., Pears, K.  C., & Levine, S. (2009). Morning cortisol levels in preschool-aged foster children: Differential effects of maltreatment type. Developmental Psychobiology, 51(1), 14–23. doi:10.1002/dev.20333 Bruce, J., Gunnar, M. R., Pears, K. C., & Fisher, P. A. (2013). Early adverse care, stress neurobiology, and prevention science: Lessons learned. Prevention Science, 14(3), 247–256. doi:10.1007/s11121-012-0354-6 Buchmann, A.  F., Holz, N., Boecker, R., Blomeyer, D., Rietschel, M., Witt, S. H., . . . Laucht, M. (2014). Moderating role of FKBP5 genotype in the impact of childhood adver­ sity on cortisol stress response during adulthood. European Neuropsychopharmacology, 24(6), 837–845. doi:10.1016/j. euroneuro.2013.12.001 Bugental, D. B., Martorell, G. A., & Barraza, V. (2003). The hormonal costs of subtle forms of infant maltreatment. Hormones and Behavior, 43(1), 237–244. Burke, N. N., Finn, D. P., McGuire, B. E., & Roche, M. (2016). Psychological stress in early life as a predisposing factor for  the development of chronic pain: Clinical and pre­ clinical evidence and neurobiological mechanisms. Journal of  Neuroscience Research, 95(6), 1257–1270. doi:10.1002/ jnr.23802 Busso, D.  S., McLaughlin, K.  A., & Sheridan, M.  A. (2016). Dimensions of adversity, physiological reactivity, and exter­ nalizing psychopathology in adolescence: Deprivation and threat. Psychosomatic Medicine, 79(2), 162–171. doi:10.1097/ PSY.0000000000000369 Carpenter, L.  L., Carvalho, J.  P., Tyrka, A.  R., Wier, L.  M., Mello, A. F., Mello, M. F., . . . Price, L. H. (2007). Decreased adrenocorticotropic hormone and cortisol responses to stress in healthy adults reporting significant childhood maltreatment. Biological Psychiatry, 62(10), 1080–1087. doi:10.1016/j.biopsych. 2007.05.002 Carpenter, L. L., Shattuck, T. T., Tyrka, A. R., Geracioti, T. D., & Price, L.  H. (2011). Effect of childhood physical abuse on cortisol stress response. Psychopharmacology, 214(1), 367– 375. doi:10.1007/s00213-010-2007-4 Carpenter, L. L., Tyrka, A. R., Ross, N. S., Khoury, L., Anderson, G.  M., & Price, L.  H. (2009). Effect of childhood emo­ tional abuse and age on cortisol responsivity in adulthood. Biological Psychiatry, 66(1), 69–75. Chrousos, G. P. (2009). Stress and disorders of the stress system. Nature Reviews Endocrinology, 5(7), 374–381. doi:10.1038/ nrendo.2009.106 Cicchetti, D. (2016). Socioemotional, personality, and biological development: Illustrations from a multilevel developmental psychopathology perspective on child maltreatment. Annual Review of Psychology, 67, 187–211. doi:10.1146/annurev-psych122414-033259 Cicchetti, D., & Rogosch, F.  A. (2001a). Diverse patterns of neuroendocrine activity in maltreated children. Development and Psychopathology, 13(3), 677–693. Cicchetti, D., & Rogosch, F.  A. (2001b). The impact of child maltreatment and psychopathology on neuroendocrine func­ tioning. Development and Psychopathology, 13(3), 783–804. Cicchetti, D., Rogosch, F.  A., Gunnar, M.  R., & Toth, S.  L. (2010). The differential impacts of early physical and sexual abuse and internalizing problems on daytime cortisol rhythm

in school-aged children. Child Development, 81(1), 252–269. doi:10.1111/j.1467-8624.2009.01393.x Cicchetti, D., Rogosch, F.  A., & Oshri, A. (2011). Interactive effects of corticotropin releasing hormone receptor 1, serotonin transporter linked polymorphic region, and child maltreatment on diurnal cortisol regulation and internalizing symptomatology. Development and Psychopathology, 23(4), 1125–1138. doi:10.1017/S0954579411000599 Cicchetti, D., Rogosch, F. A., & Toth, S. L. (2006). Fostering secure attachment in infants in maltreating families through preventive interventions. Development and Psychopathology, 18(3), 623–649. Cicchetti, D., Rogosch, F. A., Toth, S. L., & Sturge-Apple, M. L. (2011). Normalizing the development of cortisol regulation in maltreated infants through preventive interventions. Development and Psychopathology, 23(3), 789–800. doi:10.1017/ S0954579411000307 Coelho, R., Viola, T.  W., Walss-Bass, C., Brietzke, E., & Grassi-Oliveira, R. (2014). Childhood maltreatment and inflammatory markers: A systematic review. Acta Psychiatrica Scandinavica, 129(3), 180–192. doi:10.1111/acps.12217 Cook, E. C., Chaplin, T. M., Sinha, R., Tebes, J. K., & Mayes, L. C. (2012). The stress response and adolescents’ adjustment: The impact of child maltreatment. Journal of Youth and Adolescence, 41(8), 1067–1077. doi:10.1007/s10964-012-9746-y Coplan, J.  D., Abdallah, C.  G., Kaufman, J., Gelernter, J., Smith, E. L., Perera, T. D., . . . Nemeroff, C. B. (2011). Earlylife stress, corticotropin-releasing factor, and serotonin transporter gene: A pilot study. Psychoneuroendocrinology, 36(2), 289–293. doi:10.1016/j.psyneuen.2010.07.011 Coplan, J. D., Altemus, M., Mathew, S. J., Smith, E. L., Sharf, B., Coplan, P. M., . . . Rosenblum, L. A. (2005). Synchronized maternal-infant elevations of primate CSF CRF concentra­ tions in response to variable foraging demand. CNS Spectrums, 10(7), 530–536. Coplan, J. D., Smith, E. L., Altemus, M., Scharf, B. A., Owens, M. J., Nemeroff, C. B., . . . Rosenblum, L. A. (2001). Variable foraging demand rearing: Sustained elevations in cisternal cerebrospinal fluid corticotropin-releasing factor concentra­ tions in adult primates. Biological Psychiatry, 50(3), 200–204. Cowen, P. S. (2001). Crisis child care: An intervention for at-risk families. Journal of the American Psychiatric Nurses Association, 7(6), 196–204. Dackis, M.  N., Rogosch, F.  A., & Cicchetti, D. (2015). Child maltreatment, callous-unemotional traits, and defensive responding in high-risk children: An investigation of emotion-modulated startle response. Development and Psychopathology, 27(4 Pt 2), 1527–1545. doi:10.1017/ S0954579415000929 Daly, M., & Wilson, M. (1988). Evolutionary social psychology and family homicide. Science, 242, 519–524. doi:10.1126/ science.3175672 Daly, M., & Wilson, M. (2005). The “Cinderella effect” is no fairy tale. Trends in Cognitive Sciences, 9(11), 507–508. doi:10.1016/j.tics.2005.09.007 Danese, A., & Tan, M. (2014). Childhood maltreatment and obesity: Systematic review and meta-analysis. Molecular Psychiatry, 19(5), 544–554. doi:10.1038/mp.2013.54 De Bellis, M. D., Broussard, E. R., Herring, D. J., Wexler, S., Moritz, G., & Benitez, J. G. (2001). Psychiatric co-morbidity in caregivers and children involved in maltreatment: A pilot research study with policy implications. Child Abuse & Neglect, 25(7), 923–944.

Deer, Bernard, and Hostinar

173

De Bellis, M.  D., Chrousos, G.  P., Dorn, L.  D., Burke, L., Helmers, K. Kling, M.  A.,  .  .  .  Putnam, F.  W. (1994). Hypothalamic-Pituitary-Adrenal axis dysregulation in sexually abused girls. Journal of Clinical Endocrinology and Metabolism, 78(2), 249–255. De Bellis, M.  D., Spratt, E.  G., & Hooper, S.  R. (2011). Neurodevelopmental biology associated with childhood sexual abuse. Journal of Child Sexual Abuse, 20(5), 548–587. doi:10.1080/10538712.2011.607753 Dodge, K. A., Bates, J. E., & Pettit, G. S. (1990). Mechanisms in the cycle of violence. Science, 250, 1678–1683. Doom, J. R., Cicchetti, D., & Rogosch, F. A. (2014). Longitudinal patterns of cortisol regulation differ in maltreated and nonmaltreated children. Journal of the American Academy of Child & Adolescent Psychiatry, 53(11), 1206–1215. doi:10.1016/j.jaac.2014.08.006 Dozier, M., Manni, M., Gordon, M.  K., Peloso, E., Gunnar, M.  R., Stovall-McClough, K.  C., . . . Levine, S. (2006). Foster children’s diurnal production of cortisol: An explora­ tory study. Child Maltreatment, 11(2), 189–197. doi:10.1177/ 1077559505285779 Dozier, M., Peloso, E., Lewis, E., Laurenceau, J. P., & Levine, S. (2008). Effects of an attachment-based intervention on the  cortisol production of infants and toddlers in foster care.  Development and Psychopathology, 20(3), 845–859. doi:10.1017/S0954579408000400 Drury, S.  S., Sanchez, M.  M., & Gonzalez, A. (2016). When mothering goes awry: Challenges and opportunities for utilizing evidence across rodent, nonhuman primate and human studies to better define the biological consequences of negative early caregiving. Hormones and Behavior, 77, 182–192. doi:10.1016/j.yhbeh.2015.10.007 Duval, F., Crocq, M.  A., Guillon, M.  S., Mokrani, M.  C., Monreal, J., Bailey, P., & Macher, J.  P. (2004). Increased adrenocorticotropin suppression after dexamethasone admin­ istration in sexually abused adolescents with posttraumatic stress disorder. Annals of the New York Academy of Sciences, 1032, 273–275. doi:10.1196/annals.1314.036 Eckenrode, J., Smith, E. G., McCarthy, M. E., & Dineen, M. (2014). Income inequality and child maltreatment in the United States. Pediatrics, 133(3), 454–461. doi:10.1542/ peds.2013–1707 Edwards, V.  J., Holden, G.  W., Felitti, V.  J., & Anda, R.  F. (2003). Relationship between multiple forms of childhood maltreatment and adult mental health in community respondents: Results from the adverse childhood experiences study. American Journal of Psychiatry, 160(8), 1453–1460. doi:10.1176/appi.ajp.160.8.1453 Ehlert, U. (2013). Enduring psychobiological effects of childhood adversity. Psychoneuroendocrinology, 38(9), 1850– 1857. doi:10.1016/j.psyneuen.2013.06.007 Ehrlich, K.  B., Miller, G.  E., & Chen, E. (2016). Childhood adversity and adult physical health. Development and Psychopa­ thology, 4(1), 1–42. doi:10.1002/9781119125556.devpsy401 Eisen, M. L., Goodman, G. S., Qin, J., Davis, S., & Crayton, J. (2007). Maltreated children’s memory: Accuracy, suggestibility, and psychopathology. Developmental Psychology, 43(6), 1275–1294. doi:10.1037/0012-1649.43.6.1275 Fisher, P.  A., Beauchamp, K.  G., Roos, L.  E., Noll, L.  K., Flannery, J., & Delker, B. C. (2016). The neurobiology of intervention and prevention in early adversity. Annual Review of Clinical Psychology, 12, 331–357. doi:10.1146/annurevclinpsy-032814-112855

174

Fisher, P.  A., Gunnar, M.  R., Chamberlain, P., & Reid, J.  B. (2000). Preventive intervention for maltreated preschool children: Impact on children’s behavior, neuroendocrine activity, and foster parent functioning. Journal of the American Academy of Child and Adolescent Psychiatry, 39(11), 1356–1364. doi:10.1097/00004583-200011000-00009 Fisher, P.  A., Kim, H.  K., Bruce, J., & Pears, K.  C. (2012). Cumulative effects of prenatal substance exposure and early  adversity on foster children’s HPA axis reactivity during  a psychosocial stressor. International Journal of Behavioral Development, 36(1), 29–35. doi:10.1177/ 0165025411406863 Fisher, P. A., & Stoolmiller, M. (2008). Intervention effects on foster parent stress: Associations with child cortisol levels. Development and Psychopathology, 20(3), 1003–1021. doi:10.1017/S0954579408000473 Fisher, P. A., Stoolmiller, M., Gunnar, M. R., & Burraston, B. O. (2007). Effects of a therapeutic intervention for foster preschoolers on diurnal cortisol activity. Psychoneuroendo­ crinology, 32(8–10), 892–905. doi:10.1016/j.psyneuen. 2007.06.008 Fisher, P.  A., Van Ryzin, M.  J., & Gunnar, M.  R. (2011). Mitigating HPA axis dysregulation associated with placement changes in foster care. Psychoneuroendocrinology, 36(4), 531– 539. doi:10.1016/j.psyneuen.2010.08.007 Flinn, M. V., Ward, C. V., & Noone, R. (2005). Hormones and the human family. In D. Buss (Ed.), Handbook of evolutionary psychology (pp. 552–580). New York, NY: Wiley. Garbarino, J. (1977). The human ecology of child maltreatment: A conceptual model for research. Journal of Marriage and Family, 39(4), 721–735. doi:10.2307/350477 Gilbert, L.  K., Breiding, M.  J., Merrick, M.  T., Thompson, W. W., Ford, D. C., Dhingra, S. S., & Parks, S. E. (2015). Childhood adversity and adult chronic disease: An update from ten states and the District of Columbia, 2010. American Journal of Preventive Medicine, 48(3), 345–349. doi:10.1016/j. amepre.2014.09.006 Gilbert, R., Widom, C. S., Browne, K., Fergusson, D., Webb, E., & Janson, S. (2009). Burden and consequences of child maltreatment in high-income countries. Lancet, 373(9657), 68–81. doi:10.1016/S0140-6736(08)61706-7 Glaser, R., & Kiecolt-Glaser, J.  K. (2005). Stress-induced immune dysfunction: Implications for health. Nature Reviews Immunology, 5(3), 243–251. doi:10.1038/nri1571 Gordis, E. B., Granger, D. A., Susman, E. J., & Trickett, P. K. (2008). Salivary alpha amylase-cortisol asymmetry in maltreated youth. Hormones and Behavior, 53(1), 96–103. doi:10.1016/j.yhbeh.2007.09.002 Gowin, J.  L., Green, C.  E., Alcorn, J.  L., 3rd, Swann, A.  C., Moeller, F. G., & Lane, S. D. (2013). The role of cortisol and psychopathy in the cycle of violence. Psychopharmacology, 227(4), 661–672. doi:10.1007/s00213-013-2992-1 Graham, A.  M., Yockelson, M., Kim, H.  K., Bruce, J., Pears, K.  C., & Fisher, P.  A. (2012). Effects of maltreatment and early intervention on diurnal cortisol slope across the start of  school: A pilot study. Child Abuse & Neglect, 36(9), 666–670. doi:10.1016/j.chiabu.2012.07.006 Gunnar, M.  R., Doom, J.  R., & Esposito, E.  A. (2015). Psychoneuroendocrinology of stress: Normative development and individual differences. In M. E. Lamb & R. M. Lerner (Eds.), Handbook of child psychology and developmental science (7th ed., Vol. 3, pp. 106–151). Hoboken, NJ: John Wiley & Sons.

Influence of Maternal Stress and Child Maltreatment

Hagan, M. J., Roubinov, D. S., Mistler, A. K., & Luecken, L. J. (2014). Mental health outcomes in emerging adults exposed to childhood maltreatment: The moderating role of stress  reactivity. Child Maltreatment, 19(3–4), 156–167. doi:10.1177/1077559514539753 Harkness, K. L., Stewart, J. G., & Wynne-Edwards, K. E. (2011). Cortisol reactivity to social stress in adolescents: Role of depression severity and child maltreatment. Psychoneuro­ endocrinology, 36(2), 173–181. doi:10.1016/j.psyneuen. 2010.07.006 Hart, J., Gunnar, M., & Cicchetti, D. (1995). Salivary cortisol in  maltreated children: Evidence of relations between neuroendocrine activity and social competence. Development and Psychopathology, 7, 11–26. Hart, J., Gunnar, M., & Cicchetti, D. (1996). Altered neuroendocrine activity in maltreated children related to symptoms of depression. Development and Psychopathology, 8, 201–214. Hawes, D.  J., Brennan, J., & Dadds, M.  R. (2009). Cortisol, callous-unemotional traits, and pathways to antisocial behavior. Current Opinion in Psychiatry, 22(4), 357–362. doi:10.1097/YCO.0b013e32832bfa6d Heim, C., Newport, D. J., Bonsall, R., Miller, A. H., & Nemeroff, C.  B. (2001). Altered pituitary-adrenal axis responses to provocative challenge tests in adult survivors of childhood abuse. American Journal of Psychiatry, 158(4), 575–581. doi:10.1176/appi.ajp.158.4.575 Heim, C., Newport, D. J., Heit, S., Graham, Y. P., Wilcox, M., Bonsall, R., . . . Nemeroff, C.  B. (2000). Pituitary-adrenal and autonomic responses to stress in women after sexual and physical abuse in childhood. JAMA, 284(5), 592–597. Heim, C., Newport, D.  J., Mletzko, T., Miller, A.  H., & Nemeroff, C. B. (2008). The link between childhood trauma and depression: Insights from HPA axis studies in humans. Psychoneuroendocrinology, 33(6), 693–710. doi:10.1016/j. psyneuen.2008.03.008 Heim, C., Newport, D. J., Wagner, D., Wilcox, M. M., Miller, A. H., & Nemeroff, C. B. (2002). The role of early adverse experience and adulthood stress in the prediction of neuroendocrine stress reactivity in women: A multiple regression analysis. Depression and Anxiety, 15(3), 117–125. doi:10.1002/da.10015 Herman, J. P., McKlveen, J. M., Ghosal, S., Kopp, B., Wulsin, A., Makinson, R., . . . Myers, B. (2016). Regulation of the Hypothalamic-Pituitary-Adrenocortical Stress Response. Comprehensive Physiology, 6, 603–621. Hotamisligil, G. S. (2006). Inflammation and metabolic disorders. Nature, 444(7121), 860–867. doi:10.1038/nature05485 Howell, B. R., McCormack, K. M., Grand, A. P., Sawyer, N. T., Zhang, X., Maestripieri, D., . . . Sanchez, M.  M. (2013). Brain white matter microstructure alterations in adolescent rhesus monkeys exposed to early life stress: Associations with high cortisol during infancy. Biology of Mood & Anxiety Disorders, 3(1), 21. doi:10.1186/2045-5380-3-21 Howze, D. C., & Kotch, J. B. (1984). Disentangling life events, stress and social support: Implications for the primary prevention of child abuse and neglect. Child Abuse & Neglect, 8(4), 401–409. Irwin, M. R., & Cole, S. W. (2011). Reciprocal regulation of the neural and innate immune systems. Nature Reviews Immunology, 11(9), 625–632. doi:10.1038/nri3042 Joëls, M., & Baram, T. Z. (2009). The neuro-symphony of stress. Nature Reviews Neuroscience, 10(6), 459–466. doi:10.1038/ nrn2632

Jud, A., Fegert, J. M., & Finkelhor, D. (2016). On the incidence and prevalence of child maltreatment: A research agenda. Child and Adolescent Psychiatry and Mental Health, 10(17), 1–5. doi:10.1186/s13034-016-0105-8 Kaufman, J. (1991). Depressive disorders in maltreated children.  Journal of the American Academy of Child & Adolescent Psychiatry, 30(2), 257–265. doi:10.1097/00004583199103000-00014 Kaufman, J., Birmaher, B., Perel, J., Dahl, R.  E., Moreci, P., Nelson, B.,  .  .  .  Ryan, N.  D. (1997). The corticotropinreleasing hormone challenge in depressed abused, depressed nonabused, and normal control children. Biological Psychiatry, 42(8), 669–679. Kaufman, J., & Charney, D. (2001). Effects of early stress on brain structure and function: Implications for understanding the relationship between child maltreatment and depression. Development and Psychopathology, 13(3), 451–471. Keeshin, B.  R., Strawn, J.  R., Out, D., Granger, D.  A., & Putnam, F.  W. (2014). Cortisol awakening response in adolescents with acute sexual abuse related posttraumatic stress disorder. Depression and Anxiety, 31(2), 107–114. doi:10.1002/da.22154 Kiecolt-Glaser, J. K., & Glaser, R. (1988). Methodological issues in behavioral immunology research with humans. Brain, Behavior, and Immunity, 2(1), 67–78. King, J.  A., Mandansky, D., King, S., Fletcher, K.  E., & Brewer, J. (2001). Early sexual abuse and low cortisol. Psychiatry and  Clinical Neurosciences, 55(1), 71–74. doi:10.1046/j.1440–1819.2001.00787.x Kinnally, E.  L., Feinberg, C., Kim, D., Ferguson, K., Coplan, J.  D., & Mann, J.  J. (2013). Transgenerational effects of  variable foraging demand stress in female bonnet macaques. American Journal of Primatology, 75(5), 509–517. doi:10.1002/ajp.22130 Kirschbaum, C., Pirke, K.  M., & Hellhammer, D.  H. (1993). The “Trier Social Stress Test”—a tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology, 28(1–2), 76–81. doi:119004 Klaassens, E. R., van Noorden, M. S., Giltay, E. J., van Pelt, J., van Veen, T., & Zitman, F. G. (2009). Effects of childhood trauma on HPA-axis reactivity in women free of lifetime psychopathology. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 33(5), 889–894. doi:10.1016/j. pnpbp.2009.04.011 Koch, H., McCormack, K., Sanchez, M. M., & Maestripieri, D. (2014). The development of the hypothalamic-pituitaryadrenal axis in rhesus monkeys: Effects of age, sex, and early experience. Developmental Psychobiology, 56(1), 86–95. doi:10.1002/dev.21093 Kotch, J.  B., Browne, D.  C., Ringwalt, C.  L., Stewart, P.  W., Ruina, E., Holt, K., . . . Jung, J. W. (1995). Risk of child abuse or neglect in a cohort of low-income children. Child Abuse & Neglect, 19(9), 1115–1130. Lee, Y. R. (2010). Fibromyalgia and childhood abuse: Exploration of stress reactivity as a developmental mediator. Developmental Review, 30(3), 294–307. Lemieux, A. M., & Coe, C. L. (1995). Abuse-related posttraumaticstress-disorder—Evidence for chronic neuroendocrine activa­ tion in women. Psychosomatic Medicine, 57(2), 105–115. Levy, B. H., & Tasker, J. G. (2012). Synaptic regulation of the hypothalamic-pituitary-adrenal axis and its modulation by glucocorticoids and stress. Frontiers in Cellular Neuroscience, 6, 24. doi:10.3389/fncel.2012.00024

Deer, Bernard, and Hostinar

175

Lipschitz, D.  S., Rasmusson, A.  M., Yehuda, R., Wang, S., Anyan, W., Gueoguieva, R., . . . Southwick, S.  M. (2003). Salivary cortisol responses to dexamethasone in adolescents with posttraumatic stress disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 42(11), 1310–1317. doi:10.1097/01.chi.0000084832.67701.0d MacMillan, H.  L., Georgiades, K., Duku, E.  K., Shea, A., Steiner, M., Niec, A., . . . Schmidt, L.  A. (2009). Cortisol response to stress in female youths exposed to childhood maltreatment: Results of the youth mood project. Biological Psychiatry, 66(1), 62–68. doi:10.1016/j.biopsych.2008.12.014 Maestripieri, D. (2005). Early experience affects the intergenera­ tional transmission of infant abuse in rhesus monkeys. Proceedings of the National Academy of Sciences of the United States of America, 102(27), 9726–9729. doi:10.1073/ pnas.0504122102 Maestripieri, D., & Carroll, K.  A. (1998). Child abuse and neglect: Usefulness of the animal data. Psychological Bulletin, 123(3), 211–223. Maestripieri, D., Lindell, S. G., Ayala, A., Gold, P. W., & Higley, J.  D. (2005). Neurobiological characteristics of rhesus macaque abusive mothers and their relation to social and maternal behavior. Neuroscience & Biobehavioral Reviews, 29(1), 51–57. doi:10.1016/j.neubiorev.2004.05.004 Maestripieri, D., Megna, N. L., & Jovanovic, T. (2000). Adoption and maltreatment of foster infants by rhesus macaque abusive mothers. Developmental Science, 3(3), 287–293. Maniglio, R. (2015). Significance, nature, and direction of the association between child sexual abuse and conduct disorder: A systematic review. Trauma, Violence, & Abuse, 16(3), 241– 257. doi:10.1177/1524838014526068 McCrory, E., De Brito, S.  A., & Viding, E. (2010). Research review: The neurobiology and genetics of maltreatment and adversity. Journal of Child Psychology and Psychiatry, 51(10), 1079–1095. doi:10.1111/j.1469–7610.2010.02271.x Meaney, M. J., & Szyf, M. (2005). Maternal care as a model for experience-dependent chromatin plasticity? Trends in Neuroscience, 28(9), 456–463. doi:10.1016/j.tins.2005.07.006 Miller, G. E., Chen, E., & Zhou, E. S. (2007). If it goes up, must it come down? Chronic stress and the hypothalamicpituitary-adrenocortical axis in humans. Psychological Bulletin, 133(1), 25–45. doi:10.1037/0033-2909.133.1.25 Nelson, E. M., & Spieker, S. J. (2013). Intervention effects on morning and stimulated cortisol responses among toddlers in foster care. Infant Mental Health Journal, 34(3), 211–221. doi:10.1002/imhj.21382 Newport, D.  J., Heim, C., Bonsall, R., Miller, A.  H., & Nemeroff, C.  B. (2004). Pituitary-adrenal responses to standard and low-dose dexamethasone suppression tests in adult survivors of child abuse. Biological Psychiatry, 55(1), 10–20. Pariante, C.  M., & Lightman, S.  L. (2008). The HPA axis in major depression: Classical theories and new developments. Trends in Neuroscience, 31(9), 464–468. doi:10.1016/j. tins.2008.06.006 Parker, K. J., & Maestripieri, D. (2011). Identifying key features of early stressful experiences that produce stress vulnerability and resilience in primates. Neuroscience and Biobehavioral Reviews, 35(7), 1466–1483. doi:10.1016/j.neubiorev.2010. 09.003 Peckins, M. K., Susman, E. J., Negriff, S., Noll, J., & Trickett, P. K. (2015). Cortisol profiles: A test for adaptive calibration of the

176

stress response system in maltreated and nonmaltreated youth. Development and Psychopathology, 27(4 Pt 2), 1461–1470. doi:10.1017/S0954579415000875 Penza, K.  M., Heim, C., & Nemeroff, C.  B. (2003). Neurobiological effects of childhood abuse: Implications for the pathophysiology of depression and anxiety. Archives of Women’s Mental Health, 6(1), 15–22. doi:10.1007/s00737002-0159-x Plotsky, P.  M., & Meaney, M.  J. (1993). Early, postnatal experience alters hypothalamic corticotropin-releasing factor (CRF) mRNA, median eminence CRF content and stressinduced release in adult rats. Molecular Brain Research, 18(3), 195–200. Pollak, S.  D. (2015). Multilevel developmental approaches to  under­ s tanding the effects of child maltreatment: Recent  advances and future challenges. Development and  Psychopathology, 27(4 Pt 2), 1387–1397. doi:10.1017/ S0954579415000826 Power, C., Thomas, C., Li, L., & Hertzman, C. (2012). Childhood psychosocial adversity and adult cortisol patterns. British Journal of Psychiatry, 201(3), 199–206. doi:10.1192/ bjp.bp.111.096032 Pratchett, L.  C., & Yehuda, R. (2011). Foundations of posttraumatic stress disorder: Does early life trauma lead to adult posttraumatic stress disorder? Development and Psycho­ pathology, 23(2), 477–491. doi:10.1017/S0954579411000186 Raison, C. L., & Miller, A. H. (2003). When not enough is too much: The role of insufficient glucocorticoid signaling in the  pathophysiology of stress-related disorders. American Journal of Psychiatry, 160(9), 1554–1565. doi:10.1176/appi. ajp.160.9.1554 Raposa, E. B., Bower, J. E., Hammen, C. L., Najman, J. M., & Brennan, P.  A. (2014). A developmental pathway from early life stress to inflammation: The role of negative health behaviors. Psychological Science, 25(6), 1268–1274. doi:10.1177/ 0956797614530570 Rinne, T., de Kloet, E.  R., Wouters, L., Goekoop, J.  G., DeRijk, R.  H., & van den Brink, W. (2002). Hyperre­ sponsiveness of hypothalamic-pituitary-adrenal axis to combined dexamethasone/corticotropin-releasing hormone challenge in female borderline personality disorder subjects with a history of sustained childhood abuse. Biological Psychiatry, 52(11), 1102–1112. Rosenblum, L.  A., & Andrews, M.  W. (1994). Influences of environmental demand on maternal behavior and infant development. Acta Paediatrica, 397(Suppl.), 57–63. Rosenblum, L.  A., Forger, C., Noland, S., Trost, R.  C., & Coplan, J.  D. (2001). Response of adolescent bonnet macaques to an acute fear stimulus as a function of early rearing conditions. Developmental Psychobiology, 39(1), 40–45. Rosenblum, L. A., & Paully, G. S. (1984). The effects of varying environmental demands on maternal and infant behavior. Child Development, 55(1), 305–314. Sanchez, M.  M. (2006). The impact of early adverse care on  HPA axis development: Nonhuman primate models. Hormones and Behavior, 50(4), 623–631. doi:10.1016/j.yhbeh. 2006.06.012 Sanchez, M. M., McCormack, K., Grand, A. P., Fulks, R., Graff, A., & Maestripieri, D. (2010). Effects of sex and early maternal abuse on adrenocorticotropin hormone and cortisol responses to the corticotropin-releasing hormone challenge during the first 3 years of life in group-living rhesus monkeys.

Influence of Maternal Stress and Child Maltreatment

Development and Psychopathology, 22(1), 45–53. doi:10.1017/ S0954579409990253 Sanchez, M. M., McCormack, K. M., & Howell, B. R. (2015). Social buffering of stress responses in nonhuman primates: Maternal regulation of the development of emotional regulatory brain circuits. Social Neuroscience, 10(5), 512–526. doi:10.1080/17470919.2015.1087426 Sapolsky, R. M., Romero, L. M., & Munck, A. U. (2000). How do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocrine Reviews, 21(1), 55–89. doi:10.1210/edrv.21.1.0389 Scott, K. M., McLaughlin, K. A., Smith, D. A., & Ellis, P. M. (2012). Childhood maltreatment and DSM-IV adult mental disorders: Comparison of prospective and retrospective findings. British Journal of Psychiatry, 200(6), 469–475. doi:10.1192/bjp.bp.111.103267 Simsek, S., Yuksel, T., Kaplan, I., Uysal, C., & Alaca, R. (2015). Examining the levels of BDNF and cortisol in children and  adolescent victims of sexual abuse—a preliminary study.  Comprehensive Psychiatry, 61, 23–27. doi:10.1016/j. comppsych.2015.04.013 Singhi, P., Saini, A. G., & Malhi, P. (2013). Child maltreatment in India. Paediatrics and International Child Health, 33(4), 292–300. doi:10.1179/2046905513Y.0000000099 Slopen, N., McLaughlin, K.  A., & Shonkoff, J.  P. (2014). Interventions to improve cortisol regulation in children: A systematic review. Pediatrics, 133(2), 312–326. doi:10.1542/ peds.2013–1632 Smith, S.  M., & Vale, W.  W. (2006). The role of the hypothalamic-pituitary-adrenal axis in neuroendocrine responses to stress. Dialogues in Clinical Neuroscience, 8(4), 383–395. Stein, M.  B., Yehuda, R., Koverola, C., & Hanna, C. (1997). Enhanced dexamethasone suppression of plasma cortisol in  adult women traumatized by childhood sexual abuse. Biological Psychiatry, 42(8), 680–686. Stetler, C., & Miller, G. E. (2011). Depression and hypothalamicpituitary-adrenal activation: A quantitative summary of four decades of research. Psychosomatic Medicine, 73(2), 114–126. doi:10.1097/PSY.0b013e31820ad12b Stoltenborgh, M., Bakermans-Kranenburg, M.  J., Alink, L. R. A., & van IJzendoorn, M. H. (2015). The prevalence of child maltreatment across the globe: Review of a series of meta-analyses. Child Abuse Review, 24(1), 37–50. doi:10.1002/ car.2353 Sumner, J. A., McLaughlin, K. A., Walsh, K., Sheridan, M. A., & Koenen, K.  C. (2014). CRHR1 genotype and history of maltreatment predict cortisol reactivity to stress in adolescents. Psychoneuroendocrinology, 43, 71–80. doi:10.1016/j.psyneuen. 2014.02.002 Susman, E.  J. (2006). Psychobiology of persistent antisocial behavior: Stress, early vulnerabilities and the attenuation hypothesis. Neuroscience & Biobehavioral Reviews, 30(3), 376–389. doi:10.1016/j.neubiorev.2005.08.002

Tarullo, A.  R., & Gunnar, M.  R. (2006). Child maltreatment and the developing HPA axis. Hormones and Behavior, 50(4), 632–639. doi:10.1016/j.yhbeh.2006.06.010 Taylor, C.  A., Guterman, N.  B., Lee, S.  J., & Rathouz, P.  J. (2009). Intimate partner violence, maternal stress, nativity, and risk for maternal maltreatment of young children. American Journal of Public Health, 99(1), 175–183. doi:10.2105/AJPH.2007.126722 Teicher, M. H., & Samson, J. A. (2013). Childhood maltreatment and psychopathology: A case for ecophenotypic variants as clinically and neurobiologically distinct subtypes. American Journal of Psychiatry, 170(10), 1114–1133. doi:10.1176/appi. ajp.2013.12070957 Thompson, R.  A. (2015). Social support and child protection: Lessons learned and learning. Child Abuse & Neglect, 41, 19–29. doi:10.1016/j.chiabu.2014.06.011 Trickett, P.  K., Gordis, E., Peckins, M.  K., & Susman, E.  J. (2014). Stress reactivity in maltreated and comparison male and female young adolescents. Child Maltreatment, 19(1), 27–37. doi:10.1177/1077559513520466 Trickett, P.  K., Noll, J.  G., Susman, E.  J., Shenk, C.  E., & Putnam, F.  W. (2010). Attenuation of cortisol across development for victims of sexual abuse. Development and Psychopathology, 22(1), 165–175. Tyrka, A. R., Price, L. H., Gelernter, J., Schepker, C., Anderson, G. M., & Carpenter, L. L. (2009). Interaction of childhood maltreatment with the corticotropin-releasing hormone receptor gene: Effects on hypothalamic-pituitary-adrenal axis reactivity. Biological Psychiatry, 66(7), 681–685. doi:10.1016/j. biopsych.2009.05.012 U.S. Department of Health & Human Services, Administration on Children, Youth and Families, Children’s Bureau. (2016). Child maltreatment 2014. Retrieved from http://www.acf. hhs.gov/programs/cb/research-data-technology/statisticsresearch/child-maltreatment van der Vegt, E. J., van der Ende, J., Huizink, A. C., Verhulst, F. C., & Tiemeier, H. (2010). Childhood adversity modifies the relationship between anxiety disorders and cortisol secretion. Biological Psychiatry, 68(11), 1048–1054. doi:10.1016/j. biopsych.2010.07.027 van der Vegt, E. J., van der Ende, J., Kirschbaum, C., Verhulst, F. C., & Tiemeier, H. (2009). Early neglect and abuse predict diurnal cortisol patterns in adults: A study of international adoptees. Psychoneuroendocrinology, 34(5), 660–669. doi:10.1016/j.psyneuen.2008.11.004 Watson, S., Owen, B.  M., Gallagher, P., Hearn, A.  J., Young, A. H., & Ferrier, I. N. (2007). Family history, early adversity and the hypothalamic-pituitary-adrenal (HPA) axis: Mediation of the vulnerability to mood disorders. Journal of Neuropsychiatric Disease and Treatment, 3(5), 647–653. Widom, C.  S., Czaja, S.  J., & DuMont, K.  A. (2015). Intergenerational transmission of child abuse and neglect: Real or detection bias? Science, 347, 1480–1485. doi:10.1126/ science.1259917

Deer, Bernard, and Hostinar

177

CH A PT E R

11

Evolution and Human Fatherhood

Adam H. Boyette and Lee T. Gettler

Abstract This chapter reviews research on the evolution of paternal care in humans. It examines human fatherhood within the phylogenetic distribution of paternal care in vertebrates, especially mammals. Phylogenetic comparisons draw out several correlates of paternal care across species, with the most important being social monogamy. Research on the evolution of paternal care in humans has also focused on the relationship between social monogamy and the evolution of paternal care. The chapter reviews this research and major debates around whether male provisioning was a key adaptation in the evolution of pair-bonds and human life history. It discusses evidence for direct male care and summarizes a possible evolutionary sequence of its evolution in humans. It considers studies on the neuroendocrinology of male care in humans, drawing from comparative studies where appropriate. Lastly, this chapter outlines several major biocultural frameworks for understanding population-level patterns of paternal care as integrated, developmental responses to specific socioecological factors. Keywords:  fatherhood, paternal care, testosterone, male care, social monogamy, pair-bonds

Why Do Males Care?

Among mammals, humans are relatively rare in the extent to which fathers commonly cooperate with mothers to raise children. Although the roles of fathers and the range of their investment vary across and within cultures, the unique breadth of ways human males contribute to their offspring’s survival and reproduction opens many questions into the origins of human fatherhood. Evolutionary perspectives are imperative to modeling why humans deviate from the general mammalian pattern of singular maternal care, as well as to explaining the range of existing variation in fathering. In this chapter, we review what we know of the evolution of human paternal care from both ultimate (i.e., phylogeny and ­adaptive value) and proximate (i.e., hormonal and neurobiological mechanisms) perspectives (Tinbergen, 1963). We will finish with a review of approaches that integrate ultimate and proximate perspectives, and discuss multilayered factors that

shape human fatherhood. In this regard, we will highlight how recent advances in the study of the neuroendrocrinology of paternal care have been critical to unpacking how environmental factors are internalized and embodied, serving to moderate care behavior, and how plasticity in response to socioecology has been crucial to human evolution. We should begin with a brief review of taxonomy. Workers in different fields use different terms to refer to ways that males may contribute to their offspring, such that we can speak of paternal involvement, paternal investment, paternal care, and so on (Gray & Anderson, 2010). We will use the specific terms used by those researchers we discuss whenever appropriate. However, for the most part our review will focus on  the evolution of paternal care. Paternal care is typically defined as the set of behaviors performed by an adult male postfertilization that benefit the young and that the male would not perform in the absence of the young (Fernandez-Duque, Valeggia, 179

& Mendoza,  2009; Kleiman & Malcolm,  1981; Woodroffe & Vincent, 1994). For example, defense against conspecifics or predators may benefit the young because they are among the guarded females, but this would not be care. Similarly, as we will discuss more later (see Life History Theory and Human Fatherhood), human hunter-gatherer men’s targeting of large game may contribute to their offspring’s fitness, but because it also benefits potential mates (Hawkes & Bliege Bird, 2002), it cannot be exclusively considered paternal care. Both of these examples, however, are considered paternal investment. Behaviors that are considered paternal care may include feeding, carrying, huddling for thermoregulation, protection of the young (“babysitting”), grooming, teaching, or paying for college expenses (Anderson, Kaplan, & Lancaster,  1999; Woodroffe & Vincent,  1994). These are all types of direct investment, whereas sharing food with a pregnant or nursing mate would be indirect investment.

Phylogenetic Distribution of Paternal Care

Examining the place of human fathers among vertebrates more generally, especially others within the class Mammalia, informs our understanding of the social-ecological conditions that tend to select for paternal care and the evolutionary pathways through which it emerges. Critically, the phylogenetic distribution of paternal care reveals that it is a derived trait among various mammalian lineages. Thus, human fatherhood can be seen as a unique adaptation to specific selective factors faced by our ancestors— although, as we will show later, a similar neuroendocrine biology likely underpins male parenting (or its absence) in all mammals. In general, male care would be expected when the benefits of helping a mate raise offspring outweigh the costs. Costs to a male may be direct, as in the energetic costs of carrying the young (e.g., Achenbach & Snowdon, 2002), or suffered as opportunity costs from the loss of time spent finding additional mates (Clutton-Brock, 1991). Paternal care is relatively rare in mammals, being present in about 3 to 5 percent of species (Clutton-Brock, 1991; Lukas & CluttonBrock, 2013), suggesting the costs of care for the male typically outweigh the fitness benefits of ­deserting his mate. In mammals, because females internally gestate the young until parturition and feed them postpartum for various durations through lactation, males are essentially free of obligations after ejaculation and can pursue further mating, although they are limited by the Fisher condition (see Kokko 180

Evolution and Human Fatherhood

& Jennions,  2008). As a result, male care has not commonly evolved in mammals. This situation can be contrasted with that in birds, where 90 percent of all species exhibit paternal care. Male birds can sit on eggs to aid in incubation, and studies across a variety of species have shown that removing the male has a negative effect on brood size and growth (Clutton-Brock, 1991). As a result, egg guarding is the most common form of paternal care in animals— and one unavailable to mammal fathers (Kleiman & Malcolm, 1981). In both birds and mammals, male care is evolutionarily associated with and likely a result of (at least in most cases) social monogamy (Burley & Johnson, 2002; Lukas & Clutton-Brock, 2013). In mammals but not in birds, however, socially monogamous species also tend to be genetically monogamous (Clutton-Brock & Isvaran, 2006). This suggests that in mammals, paternal care evolved in a context where a stable bond between a breeding pair already benefited males, likely because the benefits of a stable partnership outweighed the costs of finding additional mates (Lukas & Clutton-Brock, 2013). For example, social monogamy is found more commonly in mammalian orders where females live in low densities and rely on food resources of high quality but low abundance. Territoriality and adult–adult intolerance are also common in these groups, as breeding pairs guard resource patches. Subsequent to the evolution of social monogamy, male care evolved across mammalian orders in a variety of ways, resulting in greater fecundity in species with male care (West & Capellini, 2016). According to phylogenetic reconstruction, some monogamous species who gave birth to litters and lived in particularly unpredictable environments went through further evolution toward cooperative breeding (Lukas & Clutton-Brock, 2012, 2013). In cooperative breeding social systems, a group of nonbreeding helpers assist a breeding pair in raising their offspring, and fathers are one of the possible alternative caretakers. Hrdy (1999) has argued that humans are cooperative breeders. Although humans do not give birth to litters, her argument hinges on the fact that human newborns are unusually altricial and require more intensive care for longer than those of our closest genetic relatives, chimpanzees and gorillas. Alloparental/allomaternal care—or care by those other than the mother (Hrdy 1999)—is all but absent among the great apes. In mammals, genera exhibiting paternal care tend to cluster in three groups: social carnivores, rodents, and New World monkeys (Kleiman & Malcolm,

1981). Males in the social carnivore group, such as wolves, coyotes, foxes, and mongooses, play a part in feeding pups and will share food with their mates. In the case of these species, selection has favored paternal investment as a component of risk pooling within cooperative breeding packs. Such species have energetically costly reproduction where females cannot raise litters alone. This situation favors both males forgoing alternative mating opportunities to provide paternal care and reproductive suppression in subordinate females who will forgo breeding to act as helpers to the breeding female. Female reproductive suppression also decreases the available female breeding population, further decreasing a male’s chances of finding alternative mating opportunities (Woodroffe & Vincent, 1994). In addition, game is an unpredictable, high-quality resource that is easily divisible such that it pays for mated pairs to divide the cost of hunting and share captured prey with the young and other nonbreeding pack members (Clutton-Brock,  1991). Male care in social carnivore groups does pay fitness dividends. Phylogenetic analysis has revealed that litters are larger and the duration of lactation is shorter in carnivore groups where males provision the breeding female (West & Capellini,  2016). Shorter lactation times lead to greater reproductive potential for the breeding pair. Several species of rodents also demonstrate paternal care behaviors, particularly grooming, huddling to keep the young warm, and retrieving stray pups (Ribble,  2003; Woodroffe & Vincent,  1994), but the evolutionary context is different than in social carnivore groups. For example, a comparative study of several closely related mice species (Peromyscus spp.) exhibiting the full spectrum of mammalian breeding patterns found that paternal care is associated with delayed reproduction in females, longer reproductive lifespans, smaller litter sizes, and larger offspring (Jašarević et al., 2013). This suggests that paternal care in mice is associated with an overall shift in female life history strategy from investment in offspring quantity to investment in offspring quality (see more in Life History Theory and Human Fatherhood later; see also Vitousek & Schoenle, this volume), and that this shift is related to the greater offspring survivability facilitated by paternal behavior (Jašarević et al.,  2013). Such a pathway to fatherhood contrasts with that of the social carnivores, where larger litters are made possible by males provisioning their mates during shorter lactation periods. These cases represent species-specific advantages of paternal care and clues to its evolution: for mice, direct paternal investment increases the survivability

of fewer, larger offspring, whereas in the larger, slower growing social carnivores, indirect paternal investment increases maternal health and facilitates a higher reproductive rate. However, it does seem that female territoriality and female–female intolerance, which would result in relatively lower female densities, are associated with paternal care in mice, suggesting this factor is still important in initiating the evolution of social monogamy and subsequent male care in this group (Ribble, 2003). It may also be that, for example, in the case of Peromyscus californicus, the larger offspring body size in monogamous species may also be a causal factor in the evolution of paternal care, as male body heat is more crucial to keeping the young warm when litters are smaller (Ribble, 2003). Paternal care in primates is highly diverse, occurring in some form in about 40 percent of all genera (Kleiman & Malcolm,  1981). As is the case for paternal care in mammals in general, its presence is more common in species whose diet is based on foods of low density but high nutritional value, especially fruit (Lukas & Clutton-Brock,  2013). However, nowhere is it more pronounced than among several species of New World monkeys. For example, South American titi monkeys (Callicebus) and owl monkeys (Aotus) exhibit the most extensive and obligate paternal care among primates, where fathers assume the role of primary carrier for infants soon after birth. Females give birth to a single offspring each year, and dependent infants may be carried by their fathers as much as 90 percent of the time, only transferring to their mother for nursing bouts (Fernandez-Duque et al., 2009). In one field study, Spence-Aizenber, Di Fiore, and Fernandez-Duque (2016) observed red titi monkey (Callicebus discolor) fathers to be more involved than mothers in all social interactions with infants except nursing, including grooming, food sharing, inspecting, aggression, and play. Mendoza and Mason (1986) demonstrated in a series of experiments that titi monkey offspring preferentially seek their father over their mother when given the choice, a preference established during the first months of life through frequent rejection by mothers and acceptance by fathers. The role of fathers is also critical among the cooperatively breeding callitrichids. In these monkeys, the marmosets (Callithrix) and tamarins (Saguinus), females frequently give birth to twins whose weight combined might make up as much as 20 percent of a female’s body weight at birth (Digby, Ferrari, & Saltzman  2007). They also have short interbirth intervals as they can conceive during a postpartum Boyet te and Get tler

181

ovulation (Achenbach & Snowdon, 2002). In this reproductive system, contributions from other group members are vital to infant survival, with care involving constant carrying, food sharing, vigilance, defense against predators, and huddling for thermoregulation. The importance of paternal care to male fitness is underscored by the energetic costs of care. For example, Achenback and Snowdon (2002) found that, in a sample of 10 captive adult male cotton-top tamarins (Saguinus oedipus), adult males lost as much as 10 percent of their prebirth body weight during the first 12 months after birth. The weight loss was less the more additional helpers were available in the group. Of the apes besides humans, only among Southeast Asian siamangs (Symphalangus syndactylus), one of the hylobatids, or “lesser apes,” do we see any paternal care. This instance is curious because the other hylobatid group, the gibbons, does not exhibit paternal care, although they share with the siamangs all the other correlates of paternal care—social monogamy, territoriality, and reduced sexual dimorphism (van Schaik & Kappeler, 2003). The great apes, on the other hand, maintain more or less polygynous mating systems with one (gorillas, orangutans) or more (chimpanzees, bonobos) males living with a group of females, who tolerate each other’s presence but raise their own offspring individually. Sexual dimorphism is pronounced in each species. The difference is at its most pronounced among the gorillas, where dominant males can be twice as large as the females in their harem. Males use their added bulk and larger canine teeth to both compete with each other and coerce females to mate. Given this pattern among the other great apes and our own phylogenetic tree, the evolution of paternal care in humans may have been relatively recent and related to other major changes in socioecology and life history. Our lineage separated from the ancestors of the chimpanzees around 7 million years ago, although estimates vary. The clearest evidence in the fossil record that hints at paternal care comes from the level of sexual dimorphism apparent in skeletal remains. As in the other apes, males were larger than females among the earliest bipedal hominins, the australopithecines, who appeared around 4 million years ago. Geary and Flinn (2001) have suggested that australopithecine family groups may have resembled the single-male, multiple-female social structure of gorillas. It remains unclear when the contemporary human pattern of reduced sexual dimorphism evolved, although consensus is that it is an indicator of the evolution of social monogamy 182

Evolution and Human Fatherhood

and the potential for paternal care in our lineage. Some argue that populations of Homo erectus showed relatively modern human-like reduced sexual dimorphism by 2 million years ago (McHenry,  1994; O’Connell, Hawkes, & Blurton Jones, 1999), although more recent finds have called that conclusion into question, demonstrating potentially significant variation in size dimorphism across H.  erectus’ range (Lordkipanidze et al.,  2007; Ruff,  2010; S. W. Simpson et al., 2008). Regardless, it remains doubtful that there was substantial paternal care in ancestral hominins, perhaps even before the appearance of Homo sapiens between 150,000 and 200,000 years ago (Gray & Anderson,  2010). Given this phylogenetic background, we now turn to the major discussion of why paternal care became part of the human evolutionary package.

Life History Theory and Human Fatherhood

Life history theory is a rich resource for understanding both the variability and universal potential for human paternal investment. According to life history theory, all organisms are faced with decisions as to how to allocate their limited somatic resources throughout their lifespan to optimize their inclusive fitness. The nature of these decisions requires that individuals make fundamental trade-offs, between allocating energy to growth versus reproduction, for  example (Roff,  2002; Stearns,  1976). Parental Investment Theory (Trivers, 1972), a branch of life history theory, focuses on the trade-offs inherent to the allocation of reproductive effort. Once an organism has reproduced, it faces the decision of whether to direct its resources toward its current offspring (parenting effort) or toward the production of future offspring (mating effort). As noted earlier, males of most mammals, including the great apes, our closest ­genetic relatives, direct their reproductive effort exclusively toward mating. Hence, paternal care is evolutionarily linked to social monogamy in most species where it occurs, as this entails a reduction in mating effort. As a consequence, debates around the evolution of paternal care as an aspect of human life history have tended to focus on the origin of human pair-bonds (Gray & Crittenden, 2014) and whether or not male contributions helped facilitate the evolution of other unique aspects of the human family and life history (Gettler, 2010; Gurven & Hill, 2009). As is the case with birds, males may be enticed to stay with a current mate if his presence significantly enhances the survivability of the offspring. We wean our offspring before they are nutritionally

independent, which allows relatively short interbirth intervals for our body size (Kaplan, Hill, Lancaster, & Hurtado, 2000). Although potentially increasing our reproductive success by giving birth to multiple dependent offspring in close succession, it dramatically increases the energetic burden on mothers, who must support costly offspring during their long period of development. The lengthy human juvenile period, and the behavioral and nutritional investment required, has been argued to have shaped human cooperation and the family, as mothers required help from others (Hrdy, 1999; Kaplan et al., 2000). In this evolutionary scenario, fathers were one of the individuals who could have subsidized maternal energy budgets by either direct or indirect investment, leading to the evolution of paternal care. The influential “man the hunter” or “man the provisioner” hypothesis of human pair-bonds developed from this line of evidence. The argument is that males formed relationships with females as a result of the benefits to females and their young of male provisioning. Males would acquire high-quality food resources, especially meat, which they would share in the context of a division of labor. The benefits of this relationship to males were reproductive access to a mate and increased offspring survival (Kaplan et al., 2000; Lancaster & Lancaster, 1983; Lovejoy, 1981). Some lines of evidence support this hypothesis; hunting was clearly one critical early human adaptation (Plummer, 2004). There is paleoarcheological evidence for increasingly complex stone tool industries used for hunting by early Homo species beginning as early as 2.6 million years ago. With the subsequent appearance of the large and wide-­ ranging H. erectus, it is argued that females would have required assistance from others to raise multiple, concurrent dependent children; that meat was likely shared; and that it was consumed alongside nutritionally dense plant foods. Early fieldwork among contemporary hunter-gatherers also supports the idea that there was a sexual division of labor in which men typically did most of the hunting and women a majority of the gathering (Lee & De Vore, 1968). This vision of male provisioning—a model based on male parenting effort—was questioned by Kristen Hawkes (1991). She argued that, if the goal of men’s provisioning was to feed their families, they would be better off targeting resources other than the big game typically hunted by hunter-gatherer men. Big game is unpredictable and it is typically shared more widely than other resources among contemporary hunter-gatherers, meaning less is likely to come back to a man’s nuclear family. Rather, she proposed

that men’s big-game hunting does not represent parental effort but serves to advertise men’s quality to women who could become additional wives or extra-pair sexual partners. Data from Hadza hunter-­ gatherer big-game hunting and meat distributions was used to support this hypothesis. Hawkes and her colleagues found that, after meat was shared, a Hadza man’s family did not receive more meat, irrespective of his success or how much time he spent hunting (Hawkes, O’Connell, & Blurton Jones, 2001b), nor did successful hunters’ families receive more meat from others as return payment for their contributions (Hawkes, O’Connell, & Blurton Jones, 2001a). One critique of their interpretation of these data is that if better hunters simply keep supplying the group with meat, what would motivate any one woman to provide mating opportunities to keep the meat coming—temptation to “free ride” on good hunters would disfavor the evolution of such a system (Bliege Bird, Smith, & Bird, 2001). Bliege Bird and colleagues (2001) proposed that men’s hunting must serve as an honest signal of their more general advantageous qualities— that he has “good genes”—such that it pays for any individual woman to mate and produce offspring with him. In general, it does seem that better hunters have greater reproductive success, although it is not clear whether they are better providers (Gurven & Hill, 2009). Although Hawkes and colleagues’ proposition that hunting constitutes showing off, or a “costly signal”, has forced reconsideration of the long-standing notion of men’s care having evolved as provisioning, many are not convinced the empirical evidence supports the position that men’s hunting only reflects mating effort. For example, Marlowe (1999) has shown that, among Hadza hunter-gatherer men, male care is at least partially parenting effort. He found that men with the best hunting reputations are no less likely to provide direct care (carrying, holding, feeding, cleaning, soothing) to their children. Indeed, men who provided more direct care also brought in more food, suggesting they are not trading off direct care for opportunities to show off. Additionally, those men with more biological children versus stepchildren brought in more food, a finding contrary to the expectations of the show-off hypothesis, since relatedness to the young should not matter if provisioning is mating effort. Furthermore, most calories men brought in were not from big-game meat, but rather from other food sources. In another study, Marlowe (2003) found that men increased their productivity during the period Boyet te and Get tler

183

when their wives were nursing and had lower foraging return rates. During this time, men also targeted less widely shared but valuable resources, like honey. As lactation is the most energetically expensive component of reproduction for females, this evidence supports the position that there are fitness advantages to male provisioning within a division of labor and that there was a role for indirect paternal care in the evolution of pair-bonds. The association between male care and lactation has been further substantiated in a cross-cultural study by Quinlan and Quinlan (2008), who found a strong relationship between the stability of pair-bonds in a culture and the mean duration of lactation. Gurven and Hill (2009) also develop several counterarguments to some assumptions of the show-­ off hypothesis. For example, the idea that if men’s goal was to provide for their family they should target more reliable resources is i­nconsistent with their calculations that Ache, Hadza, and Hiwi hunter-­ gatherer men’s average hunting returns are equal to or greater than the returns from gathering more reliable starchy foods. Variability, whether seasonal or daily, they argue, is inherent to foraging and something people adapt to. They also point out that the added nutrient diversity of meat is a valuable component of the division of labor. Furthermore, small game has been a part of human diets throughout our history as a species and, in contrast to widely shared large game, tends to be preferentially shared within families. Moreover, contrary to the assumption that a good hunter’s family would not receive enough of his kill for his hunting to be considered parental effort, they point out that most studies of meat sharing among contemporary hunter-gatherers show the reverse trend, with hunters’ families receiving more. They conclude that it is more reasonable to assume that signaling benefits from hunting would only motivate men to hunt more but are not sufficient to explain why men hunt. Many factors are likely to influence the opportunity costs men might suffer by choosing to provision their families; thus, it is more reasonable to see provisioning as a mix of mating effort and parenting effort that can shift depending on the context. We can conclude that provisioning remains one likely important factor in the evolution of pair-bonds and paternal care. However, an exclusive focus on provisioning is limiting with respect to the potential pathways for men to make significant impacts on their own and their mates’ reproductive success. For one, it ignores the possible evolutionary significance of men’s direct investment in offspring through 184

Evolution and Human Fatherhood

sensitive, nurturant care (Bribiescas, Ellison, & Gray, 2012; Gettler, 2010). For example, Gettler (2010) argues that male carrying of offspring would have significantly reduced energetic burden on mothers, facilitating the evolution of shorter interbirth intervals and higher fertility. Direct investment through socialization behaviors, such as teaching, may also have been critical, and possibly more so after early childhood when carrying and provisioning are most important (Gray & Anderson, 2010). For instance, Kaplan and colleagues (2000) argue that the extraordinarily long human childhood evolved to allow for learning necessary foraging skills. Men’s teaching may have been important in facilitating this critical learning, likely during middle childhood and adolescence (Dira & Hewlett,  2016; Hill, Barton, & Hurtado, 2009; Hrdy, 2009). Scelza (2010) argues that men are especially likely to provide direct investment when it is not easily substitutable, as when they have specific cultural knowledge or status. In support of her argument, she found that, among Martu Aborigines of Australia’s Western Desert, young men whose father was present had their initiation ritual—a gateway to marriage—significantly earlier than those without, and age at initiation predicted the young man’s chance of having a current child and the total number of his current children. In this case, we see direct paternal investment in young adult offspring directly impacting inclusive fitness. Bribiescas, Ellison, and Gray (2012) emphasize that what is truly distinctive about human paternal care is its variability. They note that the diversity of observed male behavior toward offspring varies from carrying, teaching, and provisioning at one extreme to infanticide at the other, depending on  the socioecological context. For instance, Aka hunter-gatherers of the Central African Republic are known for their intimate, high-investing fathering (Hewlett,  1991). However, Hewlett found that the level of direct investment depended on how many brothers a man lived with. Men with more brothers spent more time with other men and less time holding their infants because, Hewlett believes, their investment of time with their social network has indirect payoffs for their children, who will receive shares of  the resources the brothers cooperatively ­acquire. Similarly, among the same group, Meehan (2005) found that men spent less time in direct care when the couple was living matrilocally (i.e., with the wife’s family). In this case, other women in the wife’s family assumed more of the childcare burden. When men were living with their own families, patrilocally, they performed more direct care.

Thus, levels of paternal care vary between and within societies. Moreover, men may divide their resources between direct or indirect parental effort and mating effort differentially across a lifetime or at times both simultaneously. Men’s behavior toward stepchildren illustrates this well. A major factor in predicting male investment is paternity certainty. Thus, all types of male investment are expected to be greater for putatively biological than nonbiological children. Evidence tends to support this claim. In addition to Hadza fathers cited previously (Marlowe, 1999), American men in Albuquerque, New Mexico (Anderson, Kaplan, & Lancaster, 1999), and Xhosa men in Cape Town, South Africa, spent more on their biological children’s education expenses (Anderson, Kaplan, Lam, & Lancaster, 1999), and men in rural Trinidad, where men are typically low investing, spent more time with their genetic children (Flinn, 1988). Most striking is work by Daly and Wilson (1988b) demonstrating that infant and child morbidity and mortality in the United States and in Canada are far greater when the victim lives with one or more stepparents than with both genetic parents. For example, in the United States in 1976, fatal child abuse was 100 times more likely among children living with a stepparent than those living with biological parents (Daly & Wilson, 1988a). At the same time, stepfathers do regularly invest in their current mate’s children. As demonstrated in baboon societies, the evolutionary basis for male investment in this context may be increased chances of mating with their mothers, irrespective of genetic ties to the young (Smuts & Gubernick, 1992). Anderson, Kaplan, and colleagues (1999), for example, found that resident Xhosa stepfathers invested far more time with their stepchildren than did nonresident biological fathers. One interpretation is that such investment in stepchildren can be described as mating effort. The evidence for infanticide by stepfathers highlights another potential evolutionary benefit to the formation of pair-bonds—the prerequisite to paternal care. Although relatively rare in contemporary humans, primate models suggest the threat of infanticide for ancestral hominin females may have been much greater. Sexual dimorphism was greater, suggesting more competition between males over sexual access to females. Since fertility is lower when females have nursing infants, infanticide would have been one potential male strategy to increase female fertility and lay claim to her future offspring. This strategy was famously first observed in use by Hrdy (1977) among Hanuman langurs, but is seen in other

­rimate species as well (Palombit, Seyfarth, & p Cheney, 1997). One solution to the problem of infanticidal males in baboon society is for females to court “friendships” with one or more specific males (Palombit et al., 1997). Females grant these males sexual access and, subsequently, their “friends” have been observed to come to their aid more readily than “nonfriend” males to threats from aggressive or infanticidal males in the group. Thus, these comparative cases suggest that another potential factor in the evolution of pair-bonds in hominins could have been protection from other males. For example, in a scenario favored by Gray and Anderson (2010), mate guarding by groups of low-ranking males would have been an early step in moving from a polygynous social system with a single alpha male monopolizing a group of females to a monogamous social system based on pair-bonds. To conclude this section, the fact that paternal care in humans may at times be mating effort and at times parenting effort suggests that plasticity is central to the evolution of men’s parenting. Male protection and provisioning of females in exchange for sexual access may have been key to the early formation of pair-bonds, a change that could have precipitated many more (Gray & Anderson, 2010). Likely, females had an important role in ensuring (or manipulating) men’s paternity certainty to keep males invested. Continued association with their mates may have facilitated bonding with their offspring, and men may have taken to carrying their young, further increasing their own and their mates’ reproductive success by shortening the interbirth interval. As lives extended, extractive foraging technology became more efficient, networks expanded, and cultures elaborated, men’s teaching and other socialization behaviors would have become more valuable. We must not forget that this scenario most likely occurred within the context of multimale, multifemale groups in which many others had overlapping fitness interests in any single child, and alloparental care was common (Hill et al.,  2009; Hrdy,  1999). Thus, men’s care would have oftentimes been substitutable (Sear, Steele, McGregor, & Mace,  2002)—but at other times perhaps not (Scelza,  2010; Shenk & Scelza, 2012). Next, we review what we know about the evolution of the proximate mechanisms responsible for the diverse array of paternal care we have discussed here. What we know is that underlying the great variation we see in paternal care in humans is ­biology as complex and sensitive to the context of care as is that of mothers. Later, we return to life Boyet te and Get tler

185

history when considering approaches to the evolution of paternal care that integrate the multiple layers of explanation.

Evolution of Hormonal and Neurobiological Mechanisms Related to Human Fathering and Promoting Paternal Care

As one of us (Gettler) has often jokingly remarked during professional talks on this subject, we do not find evidence for H. erectus or early human baby Björn carriers in the archaeological record. Examining contemporary fathers’ neurobiological and hormonal profiles can prove insightful in this regard, however, by helping to fill in some of the potential evidential gaps. The basis for this perspective is grounded in cross-species comparisons. In terms of modeling human paternal neuroendocrinology, cross-species and evolutionary perspectives are salient because the physiological systems that underlie cognition, emotion, and behavior are subject to selective pressure as new demands and strategies arise (such as invested fatherhood in the course of hominin evolution), and, as foreshadowed in the section on “Phylogenetic Distribution of Paternal Care,” those selective pressures and the ecological and social contexts in which paternal care arises and is potentially adaptive may be similar across taxa. Such perspectives and comparisons provide the background to understand what selective factors might have shaped human fathers’ capacity to respond to parenthood physiologically and under what circumstances those social neuroendocrine circuits might be engaged (Gettler,  2010,  2014; Gray & Anderson,  2010; Storey & Ziegler, 2016). Much of the behavioral physiological research that has been conducted on vertebrate fathers was specifically grounded in Wingfield, Hegner, Ball, and Duffy’s (1990) Challenge Hypothesis. Drawing on extensive endocrine research on bird species, this model provided a framework to explain how mating- and parenting-related demands across the reproductive cycle interrelated with male birds’ testosterone (T; Wingfield et al., 1990). For example, in species in which male–male competition was high and paternal care was absent, males were predicted to maintain consistently elevated T, which would help maintain somatic investment in musculature and ornamentation while also facilitating competitive, aggressive, and courtship behaviors. In contrast, among species with biparental care, fathers’ T was expected to decline when they cooperated with mothers to raise young, focusing their attention 186

Evolution and Human Fatherhood

toward their vulnerable offspring and cooperative behavioral dynamics and away from aggression and competition (Wingfield et al., 1990). Building from this framework, research on other vertebrates has found that committed fathers commonly show similar physiological profiles, including some combination of reduced T, elevated prolactin (PRL), and heightened oxytocin (OT) production (Gettler, 2010,  2014; Gray & Anderson,  2010; Storey & Ziegler, 2016; Trumble, Jaeggi, & Gurven, 2015; van Anders, 2013; van Anders, Goldey, & Kuo, 2011). Because any similarities in paternal investment among birds and mammals is the result of convergent evolution (rather than shared ancestry), one can infer that overlapping patterns in fathers’ physiology across taxonomic lines represents natural selection (or other evolutionary processes) repeatedly co-opting similar neurobiological and hormonal pathways, however diverse the ecological settings are in which paternal investment evolves (Gettler, 2014, 2016a; Gray & Anderson, 2010; Storey & Ziegler, 2016). This process is sometimes referred to as “opportunism in evolution” (G. G. Simpson, 1967). Here, it would reflect evolutionary changes emerging through effects on existing physiological substrates that are shared across vertebrates, such as the hypothalamic-pituitary-gonadal axis that produces T and the hypothalamic-pituitary-­ dopaminergic signaling pathways that regulate PRL. Such “opportunistic,” convergent evolutionary processes are generally thought to be a path of less resistance for selection, compared to (in this case) the physiological changes required for de novo fatherhood neuroendocrine adaptations via random mutations in diverse vertebrate taxa (Gettler, 2014). Those evolved foundational neuroendocrine pathways provide the scaffolding through which biosocial or biocultural interfaces emerge in contemporary fathers today. Indeed, our understanding of human fathers’ neuroendocrinology has expanded over the past decade as this evolutionary foundation has been integrated with biocultural approaches and concepts related to parent–child biobehavioral synchrony (Feldman, 2012; Feldman, Monakhov, Pratt, & Ebstein, 2016; Gettler,  2010,  2014; Gray & Anderson,  2010; Trumble et al., 2015; van Anders, 2013; van Anders et al., 2011). As we will discuss in subsequent sections, the results of these studies cumulatively link specific paternal T, PRL, and OT profiles to overall involvement in direct care and specific forms of ­nurturant, sensitive parenting behaviors in those contexts (Feldman, 2012; Gettler,  2014; Gray &

Anderson, 2010; van Anders,  2013; van Anders et al., 2011). Consequently, in light of cross-species comparisons, these patterns are broadly (albeit indirectly) consistent with the notion that nurturant, direct caregiving was sufficiently common and ­beneficial to male reproductive fitness during our evolutionary past that human male neurobiological and endocrine systems have undergone selective pressure to specifically help elicit and respond to sensitive, high-quality father–child social interactions (Gettler, 2014; Gray & Anderson, 2010; Storey & Ziegler, 2016).

Connections Between Men’s Hormones and Fathering testosterone

Drawing influence from the challenge hypothesis, as well as research on behavioral physiology among nonhuman mammalian species in which fathers are involved with care, a large body of literature has explored the variation of human males’ T by life history status (i.e., marriage/partnering and parenthood; Gettler,  2014; Gray, McHale, & Carré, 2016; Storey & Ziegler,  2016; van Anders, 2013; van Anders et al.,  2011). Although the links between elevated T and greater aggressive behavior tend to be either weaker or mixed among humans, compared to what has been observed in many other nonhuman species, a number of observational and experimental studies are consistent with the notion that men with elevated T might be prone to competitive behavior, anger, and reduced empathy (Archer, 2006; Carré & Olmstead, 2015; Gettler & Oka, 2016; Gettler, 2016b; Gray et al., 2016; van Anders, 2013; van Anders et al., 2011). Neuroimaging studies also indicate that elevated T alters the neural communication and connectivity between executive-functioning frontal areas and emotion-facilitating limbic areas, such as the amygdala, when humans are confronted with status challenges and threat. It is thought that such neural profiles may increase the likelihood of reactive ­aggression. In contrast, lower T males generally show greater tendencies toward prosocial, empathetic behavior, and it is thought that reduced T may be beneficial toward romantic and parent–child relationships grounded in warmth and nurturance (Archer, 2006; Carré & Olmstead, 2015; Gettler & Oka, 2016; Gettler, 2016b; Gray et al., 2016; van Anders, 2013; van Anders et al., 2011). Some of the earliest and most influential insights in this domain emerged from a long-term study of U.S.  military personnel. Drawing on longitudinal

data collected over a decade, Mazur and Michalek (1998) showed that U.S.  veterans who became newly married during the study period experienced a significant decline in T.  They also found that married men who had elevated T were more likely to become divorced at future time points in the study. Thus, these results hinted at a bidirectional or transactional relationship between men’s hormones and the social dynamics and outcomes they experienced in their marriages. Building from this work and in a series of foundational cross-cultural studies, others began testing whether T varied by both fatherhood and marital status and how that variation might be influenced by cultural and ecological dynamics in settings around the world (Gray & Campbell, 2009). In those studies and those by other teams, it has since been shown that men who are partnered (e.g., married or cohabitating) men, especially those who are fathers, have lower T than single nonfathers in many global contexts, such as the United States, Senegal, Tanzania, the Philippines, China, and Japan (Alvergne, Faurie, & Raymond, 2009; Gray, Kahlenberg, Barrett, Lipson, & Ellison, 2002; Gray, Yang, & Pope,  2006; Kuzawa, Gettler, Muller, McDade, & Feranil,  2009; Muller, Marlowe, Bugumba, & Ellison, 2009). In contrast, this specific pattern did not hold among men in a Kenyan cultural group in which polygyny was common (Gray, 2003; see also Gray, Ellison, & Campbell, 2007), and in a large, recent study of fathers in Jamaica, where it is culturally common for fathers’ relationship statuses and residence with their children to shift relatively frequently, patterns likewise varied (Gray et al., 2017). In total, these latter findings suggest a moderating effect of societal mating systems and cultural patterning of marriage and partnering on how and whether men’s T varies by life history status. U.S. based studies have also found that men in committed relationships do not have reduced T when they maintain interest in extra-pair sexual (McIntyre et  al.,  2006) or romantic (van Anders & Goldey, 2010) opportunities, which point to the interrelationships between T and dimensions of sociosexuality in some contexts (Puts et al., 2015). Recent, complementary U.S.-based research has also shown cross-partner correlations between T and relationship dynamics, as men and women with higher T reported lower relationship satisfaction and commitment and so did their partners (i.e., those partnered to someone with high T also reported reduced satisfaction and commitment; Edelstein et al., 2014). In a 2017 study, Jamaican fathers were similarly Boyet te and Get tler

187

shown to have lower relationship quality if their T was elevated (Gray et al., 2017). Multiple studies from various cultural settings have also explored the associations between fathers’ T and their involvement with childcare (Gettler, 2014, 2016a; Gray et al., 2016). In the only study to date to explicitly compare these dynamics crossculturally, Muller and colleagues (2009) showed that fathers had lower T than nonfathers among a sample of Hadza hunter-gatherers in Tanzania, whereas among a neighboring group of Datoga pastoralists fathers’ and nonfathers’ T did not differ. These two groups have different models of fathers’ involvement with children, as Datoga fathers have little involvement with young children and Hadza fathers are routinely involved with childcare. These differences might help account for the cross-cultural variability in T’s relationship to fatherhood status (Muller et al.,  2009). Supporting this notion, a number of other subsequent studies have shown that fathers’ T tends to be lower when they are more involved with hands-on childcare (Alvergne et al., 2009; Mascaro, Hackett, & Rilling, 2013; Weisman, Zagoory-Sharon, & Feldman, 2014), a­ lthough that pattern is not ubiquitous (Gray et al., 2017; Gray et al., 2002). Adding complexity to our understanding of the relationship between cultural models of fatherhood, T, and paternal care, recent research suggests men with T functioning that is too low are prone to lower participation in childcare (Gettler et al., 2017). In addition, high T might also facilitate paternal indirect care when competitive or risk-taking behaviors, which are promoted by T (e.g., hunting, predator defense), benefit his offspring (van Anders, 2013). For example, our recent research has shown that, among Bondongo fisher-farmers in the Republic of the Congo, men who were rated has better providers of resources (i.e., more indirect care) had higher T. However, men rated as more likely to provide direct care (e.g., socialization and care of sick children) had middle-range levels of T—neither too high or too low— for their community (Boyette, Lew-Levy, Sarma, and Gettler, 2019). Furthermore, we found that those children whose fathers had middle-range T were in better energetic condition. These findings indicate that father’s provisioning is highly valued in this cultural context and is linked to higher T, although it appears that there are potential costs to having relatively higher levels of T, perhaps because high T fathers direct their effort toward mating. Our findings, along with those of Muller and colleagues (2009) noted earlier, indicate men’s biological 188

Evolution and Human Fatherhood

responses to different care contexts are also sensitive to the cultural context and should be interpreted in regards to factors such as marriage patterns (e.g., polygyny vs. monogamy) and cultural values around men’s roles as fathers. Although most prior research in this area has been cross-sectional, a small number of studies have used longitudinal data to model changes in men’s T across the parenting transition and how shifts in behavior, family dynamics, and T interrelate with one another. Drawing on a large longitudinal study in the Philippines, Gettler and colleagues (Gettler, McDade, Agustin, Feranil, & Kuzawa,  2015; Gettler, McDade, Feranil, & Kuzawa, 2011; Gettler, McKenna, Agustin, McDade, & Kuzawa,  2012) have made many critical contributions in this area. They showed that men transitioning from being single nonfathers to partnered fathers experienced large declines in T (e.g., a median of –34 percent for evening T) across a 4.5-year period. Their declines in T were much larger than men who remained single nonfathers (Gettler et al., 2011). At this same site, fathers’ T and caregiving were shown to change in tandem, with fathers’ T declining if they increased their involvement in care over the 4.5-year follow-up period (Gettler et al.,  2015), and new fathers’ T decreased more steeply if they slept in close proximity to their children rather than separately from them (Gettler, McKenna, et al., 2012). In addition, Gettler, McDade, Agustin, Feranil, and Kuzawa (2013) showed that the greater men’s decline in T as they became newly partnered new fathers, the less frequently they reported having sexual intercourse with their partners. Finally, expanding this research on male psychobiology, Gettler et al. (2017) recently integrated data on an androgen receptor genetic polymorphism that contributes to the physiological effects of T after binding (see review in Ryan & Crespi, 2012). Men who had particularly high levels of androgenicity (i.e., androgen functionality) or low levels of androgenicity were more likely to experience relationship instability. Young men’s androgenicity before they were parents also predicted whether they were later involved as direct caregivers, with moderately androgenic men being more involved, on average, compared to low- or high-androgenic males (Gettler et al., 2017). Elsewhere, a small, intensive study of pregnant women and their partners in the United States has found complementary results, showing that men’s T declines, on average, from the late prepartum to  the early postpartum period. Fathers’ whose T decreased more from the pre- to postpartum period

reported higher relationship satisfaction and investment, as did their partners. Men with more substantial declines in T were also rated as more supportive by their partners after the baby arrived (Edelstein et al., 2015; Edelstein et al., 2017; Saxbe et al., 2016). In a similar study that tracked men at two time points before and after their infants were born, fathers showed declines in T but only if they had scored lower on an index of sensation-seeking during the prepartum period. This suggests that personality or disposition and/or their neurobiological or endocrine underpinnings help shape T responses to parenthood (Perini, Ditzen, Hengartner, & Ehlert, 2012). In this study, fathers who reported more of a decline in T across the peripartum also reported decreased tenderness in their relationship with the mother postpartum, which the authors note could reflect a shift in both parents’ focus to the infant (Perini, Ditzen, Fischbacher, & Ehlert, 2012).

prolactin

Compared to the extensive research that has been done on T’s implications for human social relationships and behaviors, far fewer studies have examined PRL in these areas. PRL is a protein hormone that is produced in the pituitary gland and is regulated by hypothalamic release of dopamine (Freeman, Kanyicska, Lerant, & Nagy,  2000). Among bird species expressing biparental care, PRL has been shown to be elevated among males cooperating with females to raise young, and it is correlated to a diverse array of parenting behaviors, including direct care, such as incubation and feeding, and indirect care, such as provisioning of mothers and offspring (Angelier, Wingfield, Tartu, & Chastel, 2016; Gettler, McDade, Feranil, & Kuzawa, 2012). PRL is similarly commonly elevated among mammalian males when they serve in caregiving roles (Gettler, 2014). For example, longitudinal research on African striped mice, a species in which males can alternate between multiple reproductive tactics, has shown that males’ PRL goes up when they shift from being isolated roamers or group-living nonbreeders to dominant breeders (i.e., becoming fathers; Schradin & Yuen, 2011). Among meerkats, PRL is elevated among males immediately before they opt to serve in “babysitting” roles over foraging (Carlson et al., 2006). Among certain New World primates with intensive paternal care, PRL is elevated in experienced fathers, relative to nonfathers (cotton-top tamarins), or is particularly high during the postpartum period when fathers play critical roles in carrying young

(common marmosets; Storey & Ziegler,  2016). Either suppressing PRL or artificially elevating it well beyond typical levels reduces common marmoset fathers’ responsiveness to infant stimuli and also adjusted PRL’s relationship with T (with the normative pattern being elevated PRL and reduced T), which may be a potential joint-hormone pathway related to care expression (cf. Almond, Brown, & Keverne,  2006; Ziegler, Prudom, Zahed, Parlow, & Wegner,  2009). Marmoset fathers with suppressed PRL also lost substantial weight in this study, which parallels the prior observation that PRL ­increases in expectant fathers contribute to weight gain that helps buffer males from energetic stress ­related to metabolically costly transport of their litters of twins and triplets (Storey & Ziegler, 2016). Among humans, some of the earliest hints that PRL might be related to dimensions of human fathering emerged from research on Canadian fathers. Cross-sectional research targeting expectant and new fathers at different stages across the peripartum period indicated that fathers’ PRL was elevated in the weeks just before their babies were born. In that same study, men were exposed to recorded infant cries, and those who reported feeling the most concerned had higher PRL than less responsive men (Storey, Walsh, Quinton, & Wynne-Edwards, 2000). Similar results were found in a subsequent study of Canadian fathers, as those with higher PRL were more alert and tended to be more positive in response to audio recordings of infant cries. In that study, fathers with more experience caring for infants also had elevated PRL compared to less experienced men (Fleming, Corter, Stallings, & Steiner, 2002). Elsewhere, research conducted in Israel found that fathers’ PRL was generally stable across the first six months postpartum and that fathers with greater PRL tended to play with their infants in a coordinated, exploratory manner, which is beneficial to child development (Gordon, Zagoory-Sharon, Leckman, & Feldman, 2010). In the Philippines, young adult fathers were found to have higher PRL than nonfathers, and PRL was most elevated among fathers with infants (1-year-old or less). However, fathers’ PRL was not correlated to their level of caregiving (Gettler, McDade, et al.,  2012; Gettler et al., 2015). In contrast to these findings, in a comparatively smaller study in Jamaica, men’s PRL did not vary based on whether they were nonfathers, coresidential fathers, or nonresidential visiting fathers (Gray et al., 2007b). There are so few studies on human PRL and social behavior, including fathering, that it would Boyet te and Get tler

189

be premature to make definitive statements about PRL’s influence or responsiveness to paternal care. Some of the existing observations suggest that PRL might mechanistically help elicit positive paternal behaviors, such as beneficial forms of play or responses to infant distress (Fleming et al., 2002; Gordon et al., 2010; Storey et al., 2000). In that sense, they show broad similarities with patterns from other vertebrata taxa, including among nonhuman primates. Findings that fail to demonstrate similar patterns, such as Gettler and colleagues’ findings from the Philippines, are potentially indicative that PRL’s interrelationships with parenting vary across cultures in a similar manner to T (Gettler, McDade, et al., 2012; Gettler et al.,  2015). We would collectively be able to formulate more rigorous hypotheses about the implications of PRL within family systems and across various cultural and ecological contexts if we better understood its effects on other (nonparenting) types of social behaviors and cognitive processes, as we do for T and OT.

oxytocin

OT has been extensively studied for its promotion of certain mammalian social bonds, particularly between mothers and their infants. In that vein, it also has a critical role in lactation physiology, stimulating milk letdown from the mammary gland. In addition to its implications for mammalian mother–infant bonding, OT was shown to contribute to the formation of pair-bonds among (frequently) monogamous vole species, particularly among females, and these monogamous species of voles had more higher densities of OT receptors in dopaminergic pathways related to reward and social motivations and preferences (Carter, 1998; Young, Lim, Gingrich, & Insel,  2001). Building largely from this body of research, early studies on the function of oxytocin in human social behavior and bonding focused extensively on its potential to facilitate and respond to prosocial or nurturing social interactions, such as warm contact between romantic partners or trust and generosity during low-stakes economic games (Bartz, Zaki, Bolger, & Ochsner, 2011). This body of literature continued to grow, with many studies confirming OT’s prosociality and bonding-promoting effects, until researchers began testing whether OT might also facilitate agonistic behaviors, depending on social context, which we return to later in this section (Bartz et al.,  2011; Feldman, 2012). Before exploring the existing research regarding OT and fathering, it is important to point out that 190

Evolution and Human Fatherhood

these studies have methodological limitations, which are often variably addressed (Churchland & Winkielman, 2012; McCullough, Churchland, & Mendez, 2013). Human-based studies of OT, including those we will review later in this section, virtually uniformly measure OT from blood, saliva, or urine, which provide measures of “peripheral” OT, meaning that which is released into circulation outside of the brain. Animal models demonstrate that OT that is released peripherally and that which is released centrally (i.e., in the brain) are sometimes correlated but also disassociated at times (Landgraf & Neumann, 2004). Unlike T and PRL, OT cannot cross from the peripheral circulation across the blood–brain barrier to reach central neuronal pathways. Conse­ quently, the effect of OT on behavior, emotion, and cognition is thought to be primarily caused by its central release into the brain, rather than through peripheral pathways, which constrains the insights that can be gained through the latter, although most prominently for null associations between behavior and OT (Churchland & Winkielman,  2012; Gettler, 2014; McCullough et al., 2013). In addition, many studies of OT’s effects on behavior rely on exogenous doses of OT through intranasal administration. Although triangulation across multiple types of studies and data suggest such exogenous OT likely reaches the brain, questions do remain about the mechanics and pathways involved to affect neural function and (ultimately) behavior (Churchland & Winkielman, 2012; Gettler, 2014). Finally, some scholars have suggested that the assay procedure used to measure OT is often not conducted properly, whereas others have raised questions as to how OT migrates to saliva, through which it is frequently measured, and why blood and salivary OT levels are only modestly correlated (see review in McCullough et al., 2013). Much of the research that has been conducted on fathering and OT has emerged from extensive studies of Israeli families (Feldman, 2012). These studies have shown that fathers and mothers generally have comparable circulating levels of OT, when maternal OT is not measured during breastfeeding episodes, and fathers who served in primary caregiving roles had higher OT than those who were more secondary (Abraham et al., 2014; Feldman, Gordon, & Zagoory-Sharon,  2011). Moreover, maternal and paternal OT tends to be positively correlated within families, so if one ­parent’s OT is elevated, the other’s tends to be as well. These within-family OT correlations also extend to young children (i.e., higher OT mothers

and fathers have children with higher OT; Feldman et  al.,  2011; Feldman, Gordon, Influs, Gutbir, & Ebstein, 2013). This research has yielded many notable insights regarding fathering and OT, including that fathers with higher OT exhibit more positive engagement, communication, and affect synchrony with their children during interactions and they also engage in more affectionate touch. Overall, parents with greater OT report stronger bonding to their infants (Feldman et al., 2011). Feldman and colleagues have provided increasing evidence for the cross-generational transmission of OT profiles, prosocial behavior, and ­parental involvement. For example, fathers with higher OT reported receiving more care from their own parents (Feldman et al., 2012). In a study focusing on social behavior of toddlers, they observed that if mothers had elevated OT, their toddlers also had elevated OT, which was related to how prosocially the t­oddlers behaved with their friends (Feldman et  al., 2013). Finally, in both observational studies of parent–infant OT and experimental work in which fathers were administered intranasal OT, it was found that when parental OT was elevated and they engaged with the infant through affectionate and synchronous interactions, the infants’ OT also spiked significantly (Feldman, Gordon, Schneiderman, Weisman, & Zagoory-Sharon, 2010; Weisman, Zagoory-Sharon, & Feldman, 2012). Although heritable genetic polymorphisms related to OT function likely have implications for parent–child correlations in  OT function (Feldman et al.,  2016), these studies provide evidence for the ways in which OT–social behavior profiles can be transferred across generations. Outside of this extensive work on Israeli fathers, there have been relatively few other studies specifically focusing on OT and fathering. A small study of Jamaican fathers, mentioned earlier, did not find differences in urinary OT between single men, residential fathers, and visiting fathers (Gray, Parkin, & Samms-Vaugh, 2007). Two studies of Dutch fathers in which men received intranasal OT and were observed during in-home interactions with their toddlers showed that elevated OT was related to great stimulation of exploratory and sensitive play and lower paternal hostility (Naber, Poslawsky, van IJzendoorn, Van Engeland, & Bakermans-Kranenburg, 2013; Naber, van Ijzendoorn, Deschamps, van Engeland, & Bakermans-Kranenburg, 2010). Finally, in functional magnetic resonance imaging (fMRI) studies of paternal neural activity in response to infant

stimuli, OT was not correlated to men’s responses to infant cries (compared to control stimuli; Mascaro, Hackett, Gouzoules, Lori, & Rilling, 2013). Similarly, a separate study examining men’s neural responses to child faces along with images of adult men and women, including sexually provocative images of the latter, found limited results, with higher OT being related to greater hippocampus activity during child viewing tasks (Mascaro, Hackett, & Rilling,  2014; Mascaro, Hackett, Gouzoules, & Rilling, 2013). Finally, there is increasingly nuanced research on OT demonstrating that its prosocial-promoting effects are context dependent and that it can promote antisocial, agonistic tendencies in specific situations. A review of this literature is beyond the scope of this chapter, particularly since the studies in question have not focused on fathering (see reviews in Bartz et al., 2011; Ma, Shamay-Tsoory, Han, & Zink, 2016). However, evidence suggests artificially elevating OT can increase gloating and envy, mob mentality, ethnocentrism, and out-group hostility, as well as reaction to provocation (see reviews in Bartz et al.,  2011; Ma et al.,  2016). Exogenous OT  also contributes to reduced tension among nulliparous women in response to infant cries, but only if they had not received harsh parenting during their own childhoods. Exogenous OT had no such tension-reducing effect for women who had negative experiences with their own parents (Bakermans-Kranenburg, van Ijzendoorn, Riem, Tops, & Alink, 2012). Citing this work, Gettler (2014) argued that there are many circumstances during familial interactions, particularly across the span of child development from infancy to adolescence, in which elevated OT would likely not elicit positive interactions from parents or children and might even predispose certain parents to harsh, reactive, or abusive behaviors. The importance of cultural models of family systems, gender norms, and parenting roles in shaping the contexts in which OT’s effects find expression are also largely absent from this literature.

brief summary

The existing literature on human paternal psychobiology is consistent with the idea that males have the neurobiological and neuroendocrine capacity to respond to invested fatherhood with shifts in hormones that likely help elicit nurturant behavior. The expression of these hormonal profiles also­ ­appears to be highly flexible and plastic and not canalized as a result of the transition to parenthood. Boyet te and Get tler

191

Nonadaptationist explanations for these paternal psychobiological profiles are plausible and should be considered. For example, based on behavioral physiological research across the Primate order, Gettler (2014) has suggested that the capacity of human males to respond to fatherhood with neuroendocrine shifts might reflect broader primate psychobiological flexibility, reflecting our evolutionary history as highly social, often tolerant and prosocial, group-living mammals. However, the similarities between the hormonal profiles associated with nurturant, committed partnering and parenting in humans and that of other taxa with evolved biparental care are consistent with the idea that humans have evolved these psychobiological capacities, through direct selection or other evolutionary processes (Gettler,  2014; Gray & Anderson,  2010; Storey & Ziegler, 2016). The common associations between fathers’ lower T and their engagement in childcare, observed across cultures, and the complementary findings from PRL and OT from certain societal settings provide a line of evidence suggesting that such roles may have been evolutionarily important to males’ reproductive fitness and to child outcomes (Gettler, 2014). These patterns are notable because human fathers in contemporary societies engage in relatively little direct care, on average, especially compared to mothers, but they do align with research on child development, largely from Western societies, suggesting that high-quality, sensitive, nurturant parenting from fathers has positive impacts on child well-being (Pleck, 2012). The notion that some fathers might have been involved caregivers during our evolutionary past is compatible with, and does not negate the notion that, they cooperated with mothers in terms of foraging and providing children with resources, which is more commonly emphasized in terms of paternal roles during hominin evolution (Gettler, 2010).

Multilayered Explanations for the Diversity of Male Caregiving

Given what is known about the ultimate and proximate basis for paternal care in humans, what theories have been proposed that explain the wide variation observed in men’s care at both the population and individual level? Life history theory has again been useful in modeling how different features of the socioecological context might inform variation in investment strategy (e.g., emphasis on parenting vs. mating effort) of a group of men such that population-level patterns emerge from the shared contexts of individual men. In the theoretical treatments 192

Evolution and Human Fatherhood

discussed in this section, early-life contexts of stress are emphasized such that psychological, cultural, and neuroendocrine foundations for parenting style are canalized—at least to some degree—during development as a response to features of the environment before a man is a father. Knowledge of paternity certainty, ease of food acquisition and divisibility of shares, relationship with a child’s mother, and the number of other potential alloparents to aid in care may all influence any individual male’s investment decisions, which are in turn moderated by his neuroendocrine systems. How, then, do we explain large-scale trends in male investment across social groups or whole cultures? Draper and Harpending (1982) proposed that developmental plasticity in relation to early-life ­ psychosocial stress may be key to the unfolding of life history strategy. Their theory stems from the observation that father absence during childhood seemed to be associated with “unfavorable” outcomes—early sexual activity, teenage pregnancy, unstable partnerships, and risky behavior in men. From a life history perspective, there is a fundamental trade-off between reproducing now and accumulating resources to invest in offspring later. They reasoned that from a child’s perspective, father a­bsence indicates that success is not dependent on parental investment (i.e., there are risks in the e­nvironment that parental effort cannot reduce or avoid; Quinlan,  2007), such that the most successful strategy would be to mature quickly and invest in reproduction (mating strategy). On the other hand, father presence indicates local resource stability, such that stable pair-bond and ­paternal investment will lead to greater long-term fitness. Draper and Harpending’s (1982) central claim was that humans evolved a psychological ­mechanism that cued the early adoption of one or the other strategy. Such a mechanism would be ­adaptive as long as the environment does not change substantially between generations. Belsky, Steinberg, and Draper (1991) and Chisholm (1993) extend this theory, suggesting that parent–child attachment security is the mechanism that mediates between paternal investment and adult behavior. Securely attached individuals would be more likely to delay reproduction and invest more in building a stable partnership, as they tend to have an internal working model, or stable cognitive representation, of social relationships as stable and worthy of trust (Bowlby, 1969; Hewlett, Lamb, Leyendecker, & Schölmerich, 2000). In contrast, the absence of reliable and

nurturant caregiving is associated with the development of an internal working model of relationships as unstable and based in mistrust and, accordingly, would prompt the adoption of a mating strategy and early reproductive development. There is considerable evidence that father absence and parental conflict without divorce are associated with earlier age at menarche for females (e.g., Bogaert,  2005; Chisholm, Quinlivan, Petersen, & Coall,  2005; Ellis, McFadyen-Ketchum, Dodge, Pettit, & Bates, 1999; Moffitt, Caspi, Belsky, & Silva, 1992; Quinlan, 2003). Only one study has shown a similar association for males (Bogaert  2005). However, for both sexes, there is evidence that mating strategy behaviors are associated with insecure attachment (Belsky et al., 1991). Quinlan (2007) and Quinlan and Quinlan (2007) apply Belsky and colleagues’ (1991) theory to the level of cultural variation. Significantly, their “risk-response model” relates community-level parenting practices to the degree of extrinsic risk faced by a population. Children experience extrinsic risk independent of parenting practices. If local extrinsic risk is high, then parental effort might be wasted, since increases in effort do not increase children’s chances of survival. Parents maximize their longterm fitness through high fertility and low parental investment. If local extrinsic risk is low, then sensitive-nurturant parenting may afford children the best chance of success, and would be associated with relatively low fertility. Thus, “cultures of risk” emerge where populations sharing in high extrinsic risk develop similar psychological tendencies, which support similar cultural models—shared cognitive representations of appropriate behavior—of social relationships, including marriage (i.e., pair-bonds). They found compelling evidence for their integrative model. Using a representative sample of cultures (the Standard Cross-Cultural Sample), Quinlan (2007) showed that cultures exposed to a high pathogen load shared reduced paternal and maternal involvement, and earlier ages of weaning. Additionally, Quinlan and Quinlan (2007) found that father involvement (measured as father sleeping proximity to infants) and parental responsiveness to infant crying were associated with less acceptance of extramarital affairs (unstable pair-bonds) and reduced prevalence of witchcraft (mistrusting social relationships). Furthermore, father involvement and later weaning ages were prevalent in societies with lower levels of aggression and theft. Socialization pressures later in childhood were controlled for, and thus the  established statistical relationships can be

a­ ttributed to early-life parenting behavior (Quinlan & Quinlan, 2007), consistent with the predictions of Draper and Harpending (1982) and Belsky and colleagues (1991). Building from these conceptual models and the demographic, reproductive, and behavioral empirical data that have emerged from them, del Guidice and colleagues (2011) proposed a framework that they termed The Adaptive Calibration Model. This framework rigorously incorporates life history theoretical, developmental biological, and psychobiological perspectives to outline how our bodies’ stress physiology systems filter and internalize relevant ecological and developmental information to shape behavioral trajectories, in feedback loops across the life course, with a particular focus on critical developmental windows and early-life exposures, prior to maturation. Relevant to our focus here, they outline a predictive framework for modeling the ways in which social neuroendocrine systems are plastically shaped across development in domains relevant to fitness, including competition and risk taking, social affiliation and bonding, and parenting. When considered alongside the other lines of evidence and comparative perspectives we have described in this chapter, the adaptive calibration model provides a conceptual basis for new studies aiming to model variation in paternal psychobiology within and across cultures. It likewise informs our understanding of how social and ecological environmental variability may have shaped the expression of fathers’ biology in our evolutionary past. Complementing this approach, two recent reviews and conceptual frameworks have more ­ ­explicitly tackled perspectives on the intergenerational transmission of parenting behavior and the  neurobiological and endocrine pathways that underlie it. Grounded in comparative, evolutionary, and biocultural anthropological approaches, Gettler’s (2016a) DADS model focuses on the role of earlylife social experiences in shaping the function of social neuroendocrine pathways related to fathering. The acronym “DADS” stands for duration, attitude, dedication, and salience, which reflect cognitive and behavioral aspects of fathering that are shaped by the cultural and ecological context in which they find expression. Gettler argues that these dimensions of fatherhood and family life interrelate with broader aspects of cultural systems, such as socialization into gender roles, to provide early-life ­experiences that alter the function and development of evolved social neuroendocrine pathways related to fathering. Boyet te and Get tler

193

Finally, in a recent extensive review of human neuroscience and psychobiology research, Bos (2017) conceptualized parenting-related neuroendocrine pathways along a gradient of “tending and caring” versus “defending and explore.” In terms of intergenerational transfers of neuroendocrine profiles related to parenting, this model begins from heritable genetic polymorphisms that affect the function of these systems (e.g., see Feldman et al.,  2016; Gettler et al., 2017), to maternal–fetal physiological interactions during pregnancy, to the endocrine experiences of birth and the early postpartum period, and to childhood and adolescence, leading to reproductive maturation. Theoretically and conceptually, this framework is strongly grounded in attachment theory and internal working models to focus on the intergenerational transmission of sensitive versus harsh parenting and their underlying biology. Collectively, these models provide related and overlapping but distinctive approaches to understanding the plasticity of human fathering and paternal psychobiology and represent a way forward, toward “evo-devo” and Tinbergen-like evolutionary models of human fatherhood.

Conclusion

To review, the first human fathers who demonstrated some degree of parental investment likely emerged from a general shift from a polygynous mating system to a monogamous mating system during a period of environmental fluctuation, and when resources were patchily distributed. Like the social carnivores, these men, rather than searching for other mates and risking combat with other men, potentially saw greater benefits (or fewer costs) to working with their current mates to bring in food, which reduced women’s duration of lactation and the interbirth interval. Although sometimes their hunting helped display their qualities to other extra-pair mates, men likely tended to direct their provisioning toward their established mates and families. In either case, selection acted upon these qualities. As in New World monkeys, men who took to carrying their offspring further reduced the energetic burden on their mates, and helped to increase women’s reproductive rate. Because humans are quite social and, based on work from contemporary hunter-gatherers, mated pairs are highly mobile (Dyble et al., 2015; Lewis, Vinicius, Strods, Mace, & Migliano, 2014), the other members of a community could at different periods of the life of any family take up responsibilities for care 194

Evolution and Human Fatherhood

of the young, freeing men to pursue other activities. These quite remarkable shifts from a gorilla-like social system to the uniquely human group of multiple mated pairs living together may have been facilitated by the already present neurobiological and neuroendocrine foundation that appears to represent a feature of primate psychobiological flexibility—a flexibility that can claim humanity as its poster child, as attested to by the several theoretical and empirical linkages between environmental risk and degree of paternal investment or proclivities toward direct care.

References

Abraham, E., Hendler, T., Shapira-Lichter, I., Kanat-Maymon, Y., Zagoory-Sharon, O., & Feldman, R. (2014). Father’s brain is sensitive to childcare experiences. Proceedings of the National Academy of Sciences of the United States of America, 111(27), 9792–9797. Achenbach, G. G., & Snowdon, C. T. (2002). Cost of caregiving: Weight loss in captive adult male cotton-top tamarins (Saguinus oedipus) following the birth of infants. International Journal of Primatology, 23(1), 179–189. Almond, R.  E.  A., Brown, G.  R., & Keverne, E.  B. (2006). Suppression of prolactin does not reduce infant care by parentally experienced male common marmosets (Callithrix jacchus). Hormones and Behavior, 49(5), 673–680. Alvergne, A., Faurie, C., & Raymond, M. (2009). Variation in testosterone levels and male reproductive effort: Insight from a polygynous human population. Hormones and Behavior, 56, 491–497. Anderson, K. G., Kaplan, H., Lam, D., & Lancaster, J. (1999). Paternal care by genetic fathers and stepfathers II. Evolution and Human Behavior, 20(6), 433–451. Anderson, K.  G., Kaplan, H., & Lancaster, J. (1999). Paternal care by genetic fathers and stepfathers I.  Evolution and Human Behavior, 20(6), 405–431. Angelier, F., Wingfield, J.  C., Tartu, S., & Chastel, O. (2016). Does prolactin mediate parental and life-history decisions in response to environmental conditions in birds? A review. Hormones and Behavior, 77, 18–29. Archer, J. (2006). Testosterone and human aggression: An evaluation of the challenge hypothesis. Neuroscience and Biobehavioral Review, 30(3), 319–345. Bakermans-Kranenburg, M.  J., van Ijzendoorn, M.  H., Riem, M. M., Tops, M., & Alink, L. R. (2012). Oxytocin decreases handgrip force in reaction to infant crying in females without harsh parenting experiences. Social Cognitive and Affective Neuroscience, 7(8), 951–957. Bartz, J. A., Zaki, J., Bolger, N., & Ochsner, K. N. (2011). Social effects of oxytocin in humans: Context and person matter. Trends in Cognitive Sciences, 15(7), 301–309. Belsky, J., Steinberg, L., & Draper, P. (1991). Childhood experience, interpersonal development, and reproductive strategy: An evolutionary theory of socialization. Child Development, 62(4), 647. https://doi.org/10.2307/1131166 Bliege Bird, R., Smith, E., & Bird, D. (2001). The hunting handicap: Costly signaling in human foraging strategies. Behavioral Ecology and Sociobiology, 50(1), 9–19. https://doi. org/10.1007/s002650100338

Bogaert, A.  F. (2005). Age at puberty and father absence in a national probability sample. Journal of Adolescence, 28, 541–546. Bos, P. A. (2017). The endocrinology of human caregiving and its intergenerational transmission. Development and psychopathology, 29(3), 971–999. Bowlby, J. (1969). Attachment. New York, NY: Basic Books. Boyette, A.  H., Lew-Levy, S., Sarma, M.  S., & Gettler, L.  T. (2018). Testosterone, fathers as providers and caregivers, and child health: Evidence from fisher-farmers in the Republic of the Congo. Hormones and Behavior, 106, 1–11. Bribiescas, R.  G., Ellison, P.  T., & Gray, P.  B. (2012). Male life  history, reproductive effort, and the evolution of the genus Homo: New directions and perspectives. Current Anthropology, 53(S6), S424–S435. https://doi.org/10.1086/ 667538 Burley, N. T., & Johnson, K. (2002). The evolution of avian parental care. Philosophical Transactions of the Royal Society B: Biological Sciences, 357(1419), 241–250. https://doi. org/10.1098/rstb.2001.0923 Carlson, A.  A., Russell, A.  F., Young, A.  J., Jordan, N.  R., McNeilly, A. S., Parlow, A. F., . . . Clutton-Brock, T. (2006). Elevated prolactin levels immediately precede decisions to babysit by male meerkat helpers. Hormones and Behavior, 50, 94–100. Carré, J. M., & Olmstead, N. A. (2015). Social neuroendocrinology of human aggression: Examining the role of competitioninduced testosterone dynamics. Neuroscience, 286, 171–186. Carter, C.  S. (1998). Neuroendocrine perspectives on social attachment and love. Psychoneuroendocrinology, 23(8), 779–818. Chisholm, J.  S. (1993). Death, hope, and sex: Life-history theory and the development of reproductive strategies [and comments and reply]. Current Anthropology, 34(1), 1–24. https://doi.org/10.1086/204131 Chisholm, J. S., Quinlivan, J. A., Petersen, R. W., & Coall, D. A. (2005). Early stress predicts age at menarche and first birth, adult attachment, and expected lifespan. Human Nature, 16(3), 233–265. https://doi.org/10.1007/s12110-005-1009-0 Churchland, P. S., & Winkielman, P. (2012). Modulating social behavior with oxytocin: How does it work? What does it mean? Hormones and Behavior, 61(3), 392–399. Clutton-Brock, T.  H. (1991). The evolution of parental care. Princeton, NJ: Princeton University Press. Clutton-Brock, T., & Isvaran, K. (2006). Paternity loss in contrasting mammalian societies. Biology Letters, 2(4), 513–516. https://doi.org/10.1098/rsbl.2006.0531 Daly, M., & Wilson, M. (1988a). Evolutionary social psychology and family homicide. Science, 242(4878), 519–524. Daly, M., & Wilson, M. (1988b). Homicide. New York, NY: A. de Gruyter. Del Giudice, M., Ellis, B.  J., & Shirtcliff, E.  A. (2011). The Adaptive Calibration Model of Stress responsivity. Neuroscience & Biobehavioral Reviews, 35(7), 1562–1592. Digby, L.  J., Ferrari, S.  F., & Saltzman, W. (2007). The role of competition in cooperatively breeding species. In C. Campbell, A. Fuentes, K. MacKinnon, M. Panger & S. Bearder (Eds.), Primates in perspective (pp. 85–106). New York: Oxford University Press. Dira, S.  J., & Hewlett, B.  S. (2016). Learning to spear hunt among Ethiopian Chabu adolescent hunter-gatherers. In B.  S.  Hewlett & H.  Terashima (Eds.), Social learning and innovation in contemporary hunter-gatherers: Evolutionary

and ethnographic perspectives (pp. 71–82). Tokyo, Japan: Springer Japan. Draper, P., & Harpending, H. (1982). Father absence and reproductive strategy: An evolutionary perspective. Journal of Anthropological Research, 38, 255–279. Dyble, M., Salali, G. D., Chaudhary, N., Page, A., Smith, D., Thompson, J., . . . Migliano, A.  B. (2015). Sex equality can explain the unique social structure of hunter-gatherer bands. Science, 348(6236), 796–798. https://doi.org/10.1126/ science.aaa5139 Edelstein, R. S., Chopik, W. J., Saxbe, D. E., Wardecker, B. M., Moors, A.  C., & LaBelle, O.  P. (2017). Prospective and dyadic associations between expectant parents’ prenatal hormone changes and postpartum parenting outcomes. Developmental psychobiology, 59(1), 77–90. Edelstein, R. S., van Anders, S. M., Chopik, W. J., Goldey, K. L., & Wardecker, B.  M. (2014). Dyadic associations between testosterone and relationship quality in couples. Hormones and Behavior, 65, 401–407. Edelstein, R. S., Wardecker, B. M., Chopik, W. J., Moors, A. C., Shipman, E. L., & Lin, N. J. (2015). Prenatal hormones in first-time expectant parents: Longitudinal changes and within-couple correlations. American Journal of Human Biology, 27, 317–325. Ellis, B. J., McFadyen-Ketchum, S., Dodge, K. A., Pettit, G. S., & Bates, J. E. (1999). Quality of early family relationships and individual differences in the timing of pubertal maturation in girls: A longitudinal test of an evolutionary model. Journal of Personality and Social Psychology, 77(2), 387–401. Feldman, R. (2012). Oxytocin and social affiliation in humans. Hormones and Behavior, 61(3), 380–391. Feldman, R., Gordon, I., Influs, M., Gutbir, T., & Ebstein, R. P. (2013). Parental oxytocin and early caregiving jointly shape children’s oxytocin response and social reciprocity. Neuropsychopharmacology, 38(7), 1154–1162. Feldman, R., Gordon, I., Schneiderman, I., Weisman, O., & Zagoory-Sharon, O. (2010). Natural variations in maternal and paternal care are associated with systematic changes in oxytocin following parent–infant contact. Psychoneuroendocrinology, 35(8), 1133–1141. Feldman, R., Gordon, I., & Zagoory-Sharon, O. (2011). Maternal and paternal plasma, salivary, and urinary oxytocin and parent-infant synchrony: Considering stress and affiliation components of human bonding. Developmental Science, 14(4), 752–761. Feldman, R., Monakhov, M., Pratt, M., & Ebstein, R.  P. (2016). Oxytocin pathway genes: Evolutionary ancient system impacting on human affiliation, sociality, and psychopathology. Biological Psychiatry, 79(3), 174–184. Feldman, R., Zagoory-Sharon, O., Weisman, O., Schneiderman, I., Gordon, I., Maoz, R., . . . Ebstein, R. P. (2012). Sensitive parenting is associated with plasma oxytocin and polymorphisms in the OXTR and CD38 genes. Biological Psychiatry, 72(3), 175–181. Fernandez-Duque, E., Valeggia, C. R., & Mendoza, S. P. (2009). The biology of paternal care in human and nonhuman primates. Annual Review of Anthropology, 38(1), 115–130. https://doi.org/10.1146/annurev-anthro-091908-164334 Fleming, A. S., Corter, C., Stallings, J., & Steiner, M. (2002). Testosterone and prolactin are associated with emotional responses to infant cries in new fathers. Hormones and Behavior, 42(4), 399–413.

Boyet te and Get tler

195

Flinn, M.  V. (1988). Step- and genetic parent/offspring relationships in a Caribbean village. Ethology and Sociobiology, 9(6), 335–369. doi.org/10.1016/0162-3095(88)90026-X Freeman, M. E., Kanyicska, B., Lerant, A., & Nagy, G. (2000). Prolactin: Structure, function, and regulation of secretion. Physiological Reviews, 80(4), 1523–1631. Geary, D.  C., & Flinn, M.  V. (2001). Evolution of human parental behavior and the human family. Parenting: Science & Practice, 1(1/2), 5. Gettler, L. T. (2010). Direct male care and hominin evolution: Why male-child interaction is more than a nice social idea. American Anthropologist, 112(1), 7–21. https://doi.org/10.1111/ j.1548–1433.2009.01193.x Gettler, L. T. (2014). Applying socioendocrinology to evolutionary models: Fatherhood and physiology. Evolutionary Anthropology, 23(4), 146–160. Gettler, L. T. (2016a). Becoming DADS: Considering the role of cultural context and developmental plasticity for paternal socioendocrinology. Current Anthropology, 57(S13), S38–S51. Gettler, L.  T. (2016b). Testosterone, fatherhood, and social networks. In W.  Trevathan & K.  Rosenberg (Eds.), Costly and cute: How helpless newborns made us human (pp. 149–176). Santa Fe, NM: School for Advanced Research. Gettler, L. T., McDade, T. W., Agustin, S. S., Feranil, A. B., & Kuzawa, C. W. (2013). Do testosterone declines during the transition to marriage and fatherhood relate to men’s sexual behavior? Evidence from the Philippines. Hormones and Behavior, 64(5), 755–763. Gettler, L. T., McDade, T. W., Agustin, S. S., Feranil, A. B., & Kuzawa, C. W. (2015). Longitudinal perspectives on fathers’ residence status, time allocation, and testosterone in the Philippines. Adaptive Human Behavior and Physiology, 1(2), 124–149. Gettler, L. T., McDade, T. W., Feranil, A. B., & Kuzawa, C. W. (2011). Longitudinal evidence that fatherhood decreases testosterone in human males. Proceedings of the National Academy of Sciences of the United States of America, 108(29), 16194–16199. Gettler, L. T., McDade, T. W., Feranil, A. B., & Kuzawa, C. W. (2012). Prolactin, fatherhood, and reproductive behavior in human males. American Journal of Physical Anthropology, 148(3), 362–370. Gettler, L. T., McKenna, J. J., Agustin, S. S., McDade, T. W., & Kuzawa, C. W. (2012). Does cosleeping contribute to lower testosterone levels in fathers? Evidence from the Philippines. PLoS One, 7(9), e41559. Gettler, L.  T., & Oka, R.  C. (2016). Aging US males with multiple sources of emotional social support have low testosterone. Hormones and Behavior, 78, 32–42. Gettler, L. T., Ryan, C. P., Eisenberg, D. T., Rzhetskaya, M., Hayes, M. G., Feranil, A. B., . . . Kuzawa, C. W. (2017). The role of testosterone in coordinating male life history strategies: The moderating effects of the androgen receptor CAG repeat polymorphism. Hormones and Behavior, 87, 164–175. Gordon, I., Zagoory-Sharon, O., Leckman, J. F., & Feldman, R. (2010). Prolactin, oxytocin, and the development of paternal behavior across the first six months of fatherhood. Hormones and Behavior, 58(3), 513–518. Gray, P.  B. (2003). Marriage, parenting, and testosterone variation among Kenyan Swahili men. American Journal of Physical Anthropology, 122(3), 279–286.

196

Evolution and Human Fatherhood

Gray, P. B., & Anderson, K. G. (2010). Fatherhood: Evolution and human paternal behavior. Cambridge, MA: Harvard University Press. Gray, P. B., & Campbell, B. C. (2009). Human male testosterone, pair bonding and fatherhood. In P. T. Ellison & P. B. Gray (Eds.), Endocrinology of social relationships (pp. 270–293). Cambridge, MA: Harvard University Press. Gray, P. B., & Crittenden, A. N. (2014). Father Darwin: Effects of children on men, viewed from an evolutionary perspective. Fathering, 12(2), 121–142. Gray, P. B., Ellison, P. T., & Campbell, B. C. (2007). Testosterone and marriage among Ariaal men of northern Kenya. Current Anthropology, 48(5), 750–755. Gray, P. B., Kahlenberg, S. M., Barrett, E. S., Lipson, S. F., & Ellison, P. T. (2002). Marriage and fatherhood are associated with lower testosterone in males. Evolution and Human Behavior, 23(3), 193–201. Gray, P. B., McHale, T. S., & Carré, J. M. (2016). A review of human male field studies of hormones and behavioral reproductive effort. Hormones and Behavior, epub. Gray, P.  B., Parkin, J.  C., & Samms-Vaughan, M.  E. (2007). Hormonal correlates of human paternal interactions: A hospital-based investigation in urban Jamaica. Hormones and Behavior, 52(4), 499–507. Gray, P., Reece, J., Coore Desai, C., Dinall, T., Pellington, S., & Samms Vaughan, M. (2017). Testosterone and Jamaican fathers. Human Nature, 28(2), 201–218. Gray, P.  B., Yang, C.  J., & Pope, H.  G. (2006). Fathers have lower salivary testosterone levels than unmarried men and married non-fathers in Beijing, China. Proceedings of the Royal Society B: Biological Sciences, 273(1584), 333–339. Gurven, M., & Hill, K. (2009). Why do men hunt?: A reevaluation of “man the hunter” and the sexual division of labor. Current Anthropology, 50(1), 51–74. https://doi. org/10.1086/595620 Hawkes, K. (1991). Showing off: Tests of an hypothesis about men’s foraging goals. Ethology and Sociobiology, 12, 29–54. Hawkes, K., & Bliege Bird, R. (2002). Showing off, handicap signaling, and the evolution of men’s work. Evolutionary Anthropology, 11(2), 58–67. https://doi.org/10.1002/ evan.20005 Hawkes, K., O’Connell, J. F., & Blurton Jones, N. G. (2001a). Hadza meat sharing. Evolution and Human Behavior, 22(2), 113–142. https://doi.org/10.1016/S1090-5138(00)00066-0 Hawkes, K., O’Connell, J. F., & Blurton Jones, N. G. (2001b). Hunting and nuclear families: Some lessons from the Hadza about mens work. Current Anthropology, 42(5), 681–709. https://doi.org/10.1086/322559 Hewlett, B. S. (1991). Intimate fathers: The nature and context of Aka pygmy paternal infant care. Ann Arbor, MI: University of Michigan Press. Hewlett, B. S., Lamb, M. E., Leyendecker, B., & Schölmerich, A. (2000). Internal working models, trust and sharing among foragers. Current Anthropology, 41(2), 287–297. Hill, K., Barton, M., & Hurtado, A. M. (2009). The emergence of human uniqueness: Characters underlying behavioral modernity. Evolutionary Anthropology, 18(5), 187–200. https://doi.org/10.1002/evan.20224 Hrdy, S. B. (1977). Infanticide as a primate reproductive strategy. American Scientist, 65(1), 40–49. Hrdy, S. B. (1999). Mother nature: A history of mothers, infants, and natural selection. New York, NY: Pantheon Books.

Hrdy, S. B. (2009). Mothers and others: The evolutionary origins of mutual understanding. Cambridge, MA: Belknap Press of Harvard University Press. Jašarević, E., Bailey, D.  H., Crossland, J.  P., Dawson, W.  D., Szalai, G., Ellersieck, M.  R.,  .  .  .  Geary, D.  C. (2013). Evolution of monogamy, paternal investment, and female life history in Peromyscus. Journal of Comparative Psychology, 127(1), 91–102. https://doi.org/10.1037/a0027936 Kaplan, H., Hill, K., Lancaster, J., & Hurtado, M. (2000). A theory of human life history evolution: Diet, intelligence, and longevity. Evolutionary Anthropology: Issues, News, and Reviews, 9, 156–185. Kleiman, D. G., & Malcolm, J. R. (1981). The evolution of male parental investment in mammals. In D.  J.  Gubernick & P. H. Klopfer (Eds.), Parental care in mammals (pp. 347–387). New York, NY: Plenum Press. Kokko, H., & Jennions, M.  D. (2008). Parental investment, sexual selection and sex ratios. Journal of Evolutionary Biology, 21(4), 919–948. Kuzawa, C. W., Gettler, L. T., Muller, M. N., McDade, T. W., & Feranil, A.  B. (2009). Fatherhood, pairbonding, and testosterone in the Philippines. Hormones and Behavior, 56(4), 429–435. Lancaster, J. B., & Lancaster, C. S. (1983). Parental investment: The hominin adaptation. In D. J. Ortner (Ed.), How humans adapt: A biocultural odyssey (pp. 33–66). Washington, DC: Smithsonian Institution Press. Landgraf, R., & Neumann, I.  D. (2004). Vasopressin and oxytocin release within the brain: A dynamic concept of multiple and variable modes of neuropeptide communication. Frontiers in Neuroendocrinology, 25(3), 150–176. Lee, R.  B., & De Vore, I. (Eds.). (1968). Man the hunter. Chicago, IL: Aldine Publishing. Lewis, H. M., Vinicius, L., Strods, J., Mace, R., & Migliano, A. B. (2014). High mobility explains demand sharing and enforced cooperation in egalitarian hunter-gatherers. Nature Communications, 5, 5789. https://doi.org/10.1038/ ncomms6789 Lordkipanidze, D., Jashashvili, T., Vekua, A., de León, M. S. P., Zollikofer, C. P. E., Rightmire, G. P., . . . Rook, L. (2007). Postcranial evidence from early Homo from Dmanisi, Georgia. Nature, 449(7160), 305–310. https:// doi.org/10.1038/nature06134 Lovejoy, C.  O. (1981). The origin of man. Science, 211(4480), 341–350. Lukas, D., & Clutton-Brock, T. (2012). Cooperative breeding and monogamy in mammalian societies. Proceedings of the Royal Society B: Biological Sciences, 279(1736), 2151–2156. https://doi.org/10.1098/rspb.2011.2468 Lukas, D., & Clutton-Brock, T.  H. (2013). The evolution of social monogamy in mammals. Science, 341(6145), 526–530. https://doi.org/10.1126/science.1238677 Ma, Y., Shamay-Tsoory, S., Han, S., & Zink, C.  F. (2016). Oxytocin and social adaptation: Insights from neuroimaging studies of healthy and clinical populations. Trends in Cognitive Sciences, 20(2), 133–145. Marlowe, F.  W. (1999). Showoffs or providers? The parenting effort of Hadza men. Evolution and Human Behavior, 20, 391–404. Marlowe, F.  W. (2003). A critical period for provisioning by Hadza men: Implications for pair bonding. Evolution and Human Behavior, 24, 217–229.

Mascaro, J.  S., Hackett, P.  D., Gouzoules, H., Lori, A., & Rilling, J. K. (2013). Behavioral and genetic correlates of the neural response to infant crying among human fathers. Social cognitive and affective neuroscience, 9(11), 1704–1712. Mascaro, J. S., Hackett, P. D., & Rilling, J. K. (2013). Testicular volume is inversely correlated with nurturing-related brain  activity in human fathers. Proceedings of the National Academy of Sciences of the United States of America, 110(39), 15746–15751. Mascaro, J. S., Hackett, P. D., & Rilling, J. K. (2014). Differential neural responses to child and sexual stimuli in human fathers and non-fathers and their hormonal correlates. Psychoneuroendocrinology, 46, 153–163. Mazur, A., & Michalek, J. (1998). Marriage, divorce, and male testosterone. Social Forces, 77(1), 315–330. McCullough, M. E., Churchland, P. S., & Mendez, A. J. (2013). Problems with measuring peripheral oxytocin: Can the data on oxytocin and human behavior be trusted? Neuroscience & Biobehavioral Reviews, 37(8), 1485–1492. McHenry, H.  M. (1994). Behavioral ecological implications of early hominin body size. Journal of Human Evolution, 27, 77–87. McIntyre, M., Gangestad, S.  W., Gray, P.  B., Chapman, J.  F., Burnham, T.  C., O’Rourke, M.  T., . . . O’Rourke, M.  T. (2006). Romantic involvement often reduces men’s testosterone levels—but not always: The moderating role of extrapair sexual interest. Journal of Personality and Social Psychology, 91(4), 642–651. Meehan, C.  L. (2005). The effects of residential locality on parental and alloparental investment among the Aka foragers of the Central African Republic. Human Nature, 16(1), 58–80. https://doi.org/10.1007/s12110-005-1007-2 Mendoza, S.  P., & Mason, W.  A. (1986). Parental division of labour and differentiation of attachments in a monogamous primate (Callicebus moloch). Animal Behaviour, 34(5), 1336–1347. https://doi.org/10.1016/S0003-3472(86)80205-6 Moffitt, T.  E., Caspi, A., Belsky, J., & Silva, P.  A. (1992). Childhood experience and the onset of menarche: A test of a sociobiological model. Child Development, 63(1), 47. https:// doi.org/10.2307/1130900 Muller, M. N., Marlowe, F. W., Bugumba, R., & Ellison, P. T. (2009). Testosterone and paternal care in East African foragers and pastoralists. Proceedings of the Royal Society B: Biological Sciences, 276(1655), 347–354. Naber, F.  B., Poslawsky, I.  E., van IJzendoorn, M.  H., Van Engeland, H., & Bakermans-Kranenburg, M. J. (2013). Brief report: Oxytocin enhances paternal sensitivity to a child with autism: A double-blind within-subject experiment with intranasally administered oxytocin. Journal of Autism and Developmental Disorders, 43(1), 224–229. Naber, F., van Ijzendoorn, M. H., Deschamps, P., van Engeland, H., & Bakermans-Kranenburg, M.  J. (2010). Intranasal oxytocin increases fathers’ observed responsiveness during play with their children: A double-blind within-subject experiment. Psychoneuroendocrinology, 35(10), 1583–1586. O’Connell, J.  F., Hawkes, K., & Blurton Jones, N. (1999). Grandmothering and the evolution of Homo erectus. Journal of Human Evolution, 36, 461–485. Palombit, R. A., Seyfarth, R. M., & Cheney, D. L. (1997). The adaptive value of “friendships” to female baboons: Experimental and observational evidence. Animal Behaviour, 54(3), 599–614. https://doi.org/10.1006/anbe.1996.0457

Boyet te and Get tler

197

Perini, T., Ditzen, B., Fischbacher, S., & Ehlert, U. (2012). Testosterone and relationship quality across the transition to fatherhood. Biological Psychology, 90(3), 186–191. Perini, T., Ditzen, B., Hengartner, M., & Ehlert, U. (2012). Sensation seeking in fathers: The impact on testosterone and paternal investment. Hormones and Behavior, 61(2), 191–195. Pleck, J. H. (2012). Integrating father involvement in parenting research. Parenting, 12(2–3), 243–253. Plummer, T. (2004). Flaked stones and old bones: Biological and cultural evolution at the dawn of technology. American Journal of Physical Anthropology, 125(S39), 118–164. https:// doi.org/10.1002/ajpa.20157 Puts, D. A., Pope, L. E., Hill, A. K., Cárdenas, R. A., Welling, L. L., Wheatley, J. R., . . . Breedlove, M. S. (2015). Fulfilling desire: Evidence for negative feedback between men’s testosterone, sociosexual psychology, and sexual partner number. Hormones and Behavior, 70, 14–21. Quinlan, R.  J. (2003). Father absence, parental care, and female reproductive development. Evolution and Human Behavior, 24(6), 376–390. https://doi.org/10.1016/S10905138(03)00039-4 Quinlan, R. J. (2007). Human parental effort and environmental risk. Proceedings of the Royal Society B: Biological Sciences, 274(1606), 121–125. https://doi.org/10.1098/rspb.2006.3690 Quinlan, R. J., & Quinlan, M. B. (2007). Parenting and cultures of risk: A comparative analysis of infidelity, aggression, and witchcraft. American Anthropologist, 109(1), 164–179. https:// doi.org/10.1525/aa.2007.109.1.164 Quinlan, R.  J., & Quinlan, M.  B. (2008). Human lactation, pair-bonds, and alloparents: A cross-cultural analysis. Human Nature, 19(1), 87–102. https://doi.org/10.1007/ s12110-007-9026-9 Ribble, D. O. (2003). The evolution of social and reproductive monogamy in Peromyscus: Evidence from Peromyscus californicus (the California mouse). In U.  H.  Reichard & C.  Boesch (Eds.), Monogamy: Mating strategies and partnerships in birds, humans, and other mammals (pp. 81–92). Cambridge, UK: Cambridge University Press. Roff, D.  A. (2002). Life history evolution. Sunderland, MA: Sinauer Associates. Ruff, C. (2010). Body size and body shape in early hominins— implications of the Gona pelvis. Journal of Human Evolution, 58(2), 166–178. https://doi.org/10.1016/j.jhevol.2009.10.003 Ryan, C.  P., & Crespi, B.  J. (2012). Androgen receptor polyglutamine repeat number: Models of selection and disease susceptibility. Evolutionary Applications, 6(2), 180–196. Saxbe, D. E., Edelstein, R. S., Lyden, H. M., Wardecker, B. M., Chopik, W.  J., & Moors, A.  C. (2016). Fathers’ decline in testosterone and synchrony with partner testosterone during pregnancy predicts greater postpartum relationship investment. Hormones and Behavior, epub, 1–35. Scelza, B. A. (2010). Fathers’ presence speeds the social and repro­ ductive careers of sons. Current Anthropology, 51(2), 295–303. Schradin, C., & Yuen, C. (2011). Hormone levels of male African striped mice change as they switch between alternative reproductive tactics. Hormones and Behavior, 60(5), 676–680. Sear, R., Steele, F., McGregor, I.  A., & Mace, R. (2002). The effects of kin on child mortality in rural Gambia. Demography, 39(1), 43–63. Shenk, M. K., & Scelza, B. A. (2012). Paternal investment and status-related child outcomes: Timing of father’s death affects offspring success. Journal of Biosocial Science, 44(5), 549–569. https://doi.org/10.1017/S0021932012000053

198

Evolution and Human Fatherhood

Simpson, G. G. (1967). The meaning of evolution: A study of the history of life and of its significance for man. New Haven, CT: Yale University Press. Simpson, S.  W., Quade, J., Levin, N.  E., Butler, R., DupontNivet, G., Everett, M., & Semaw, S. (2008). A female Homo erectus pelvis from Gona, Ethiopia. Science, 322(5904), 1089–1092. Smuts, B.  B., & Gubernick, D.  J. (1992). Male-infant relationships in nonhuman primates: Paternal investment of mating effort? In B.  S.  Hewlett (Ed.), Father-child relationships: Cultural and biosocial contexts (pp. 1–29). New York, NY: Aldine de Gruyter. Spence-Aizenberg, A., Di Fiore, A., & Fernandez-Duque, E. (2016). Social monogamy, male–female relationships, and biparental care in wild titi monkeys (Callicebus discolor). Primates, 57(1), 103–112. https://doi.org/10.1007/s10329015-0489-8 Stearns, S. C. (1976). Life-history tactics: A review of the ideas. Quarterly Review of Biology, 51(1), 3–47. Storey, A. E., Walsh, C. J., Quinton, R. L., & Wynne-Edwards, K. E. (2000). Hormonal correlates of paternal responsiveness in new and expectant fathers. Evolution and Human Behavior, 21(2), 79–95. Storey, A.  E., & Ziegler, T.  E. (2016). Primate paternal care: Interactions between biology and social experience. Hormones and Behavior, 77, 260–271. Tinbergen, N. (1963). On aims and methods of ethology. Zeitshcrif Für Tierpscychologie, 20, 410–433. Trivers, R. L. (1972). Parental investment and sexual selection. In B. G. Campbell (Ed.), Sexual selection and the descent of man, 1871–1971 (pp. 136–179). Chicago, IL: Aldine. Trumble, B. C., Jaeggi, A. V., & Gurven, M. (2015). Evolving the neuroendocrine physiology of human and primate cooperation and collective action. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 370(1683), 20150014. doi:10.1098/rstb.2015.0014. van Anders, S.  M. (2013). Beyond masculinity: Testosterone, gender/sex, and human social behavior in a comparative context. Frontiers in Neuroendocrinology, 34(3), 198–210. van Anders, S.  M., & Goldey, K.  L. (2010). Testosterone and partnering are linked via relationship status for women and “relationship orientation” for men. Hormones and Behavior, 58(5), 820–826. van Anders, S. M., Goldey, K. L., & Kuo, P. X. (2011). The steroid/ peptide theory of social bonds: Integrating testosterone and peptide responses for classifying social behavioral contexts. Psychoneuroendocrinology, 36(9), 1265–1275. van Schaik, C. P., & Kappeler, P. M. (2003). The evolution of social monogamy in primates. In U.  H.  Reichard & C.  Boesch (Eds.), Monogamy: Mating strategies and partnerships in birds, humans, and other mammals (pp. 59– 80). Cambridge, UK: Cambridge University Press. Weisman, O., Zagoory-Sharon, O., & Feldman, R. (2012). Oxytocin administration to parent enhances infant physiological and behavioral readiness for social engagement. Biological psychiatry, 72(12), 982–989. Weisman, O., Zagoory-Sharon, O., & Feldman, R. (2014). Oxytocin administration, salivary testosterone, and father– infant social behavior. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 49, 47–52. West, H. E. R., & Capellini, I. (2016). Male care and life history traits in mammals. Nature Communications, 7, 11854. https:// doi.org/10.1038/ncomms11854

Wingfield, J.  C., Hegner, R.  E., Ball, G.  F., & Duffy, A.  M. (1990). The “challenge hypothesis”: Theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. American Naturalist, 136, 829–846. Woodroffe, R., & Vincent, A. (1994). Mother’s little helpers: Patterns of male care in mammals. Trends in Ecology & Evolution, 9(8), 294–297.

Young, L. J., Lim, M. M., Gingrich, B., & Insel, T. R. (2001). Cellular mechanisms of social attachment. Hormones and Behavior, 40(2), 133–138. Ziegler, T.  E., Prudom, S.  L., Zahed, S.  R., Parlow, A.  F., & Wegner, F.  H. (2009). Prolactin’s mediative role in male parenting in parentally experienced marmosets (Callithrix jacchus). Hormones and Behavior, 56, 436–443.

Boyet te and Get tler

199

CH A PT E R

12

Hormones, Sexual Orientation, and Gender Identity

Nicholas C. Neibergall, Alex J. Swanson, and Francisco J. Sánchez

Abstract Gender identity and sexual orientation are two characteristics that play a significant role in human development. This chapter focuses on the potential role that hormones play in their development. First, a review of the direct effects of hormones and endocrine-disrupting chemicals is provided with a focus on animal models and persons born with disorders of sex development. Second, some evidence from association studies linking characteristics that are known or suspected of being influenced by hormones is provided. Although biological research has yielded some intriguing findings—especially from neuroimaging studies—scientists have yet to conclude what specific factors contribute to their development. Keywords:  homosexuality, gender dysphoria, human sexuality, sexual behavior, neuroendocrinology, transgender, intersex, diethylstilbestrol (DES), phthalates, bisphenol A (BPA)

Which restroom should transgender people use, and should they be allowed to serve in the military (Steinmetz, 2017; Stewart & Chiacu, 2017)? Does the ban on sex discrimination in the Civil Rights Act of 1964 protect workers on the basis of their sexual orientation (Shear & Savage, 2017)? At what age should clinicians provide services to transgender youth who request medical treatment to help align their bodies with their gender identity (Fishman, 2017)? Such questions have been the focus of recent articles in the public domain, and this focus has partly been stoked given recent decisions by the current administration in the United States. Although transgender issues are currently dominating the public conversation related to sexual and gender minorities, the questions being asked of this community are ones that previously and primarily focused on nonheterosexual people. When people encounter others who seem to “deviate” from the “norm,” they may assume that there is something “wrong.” Of course, such simple generalizations overlook the fact that when it comes to most complex human traits, there is great variability between people.

Unlike other characteristics for which people have been historically marginalized (e.g., disabilities), many place great emphasis on the biological bases of sexual orientation and gender identity. Some legal decisions have relied on biological research when deliberating cases (see “Opinion of the Court” in Obergefell v. Hodges,  2015, p. 8, for an example). Furthermore, public opinion related to civil rights seems to be influenced by whether there is a biological basis for these traits (e.g., Wood & Bartkowski, 2004). Although science has advanced our understanding of these complex issues, much remains unknown. In this chapter, we provide a brief overview of research related to hormones. Before reviewing this literature, we present some key terms and brief background information.

Key Terms and Background Information

A variety of terms are associated with the lesbian, gay, bisexual, and transgender (LGBT) community. Complicating the issue is that the transgender and gender-nonconforming community (henceforth transgender, for simplicity) is diverse, and the needs 201

of any subset of the transgender community do not necessarily represent the needs of all (e.g., the needs of someone seeking to transition from one gender to the other are more complex than the needs of someone who “cross-dresses” for entertainment or personal enjoyment). An extensive list of terms can be found in Bockting (2008). Here we provide some key definitions as used in this chapter: • Birth sex refers to the sex assigned to a baby at birth, which is typically based on the external genitalia (Vilain, 2000). • Cisgender refers to a person who identifies with his or her birth sex (Schilt & Westbrook, 2009). • Gender identity is the psychological sense of maleness or femaleness (Stoller, 1968). • Homosexual will be used when describing nonheterosexual sexual behavior. Given that LGBT and associated terms are identity terms that not all people adopt (e.g., many men who only have sex with men do not identify as gay), we will selectively use homosexual in reference to human behavior when identity is not a key variable. It should be noted, however, that the term homosexual may be offensive to some people, and one should remain sensitive to the terms used when referring to and working with people (American Psychological Association, 2009). • Intersex refers to conditions characterized by atypical sexual development resulting from genetic, chromosomal, or hormonal anomalies (Vilain, 2000). Although we will use the term, the current classification for these types of conditions is disorders of sex development (Hughes et al., 2006). • Transgender refers to people whose gender identity and/or gender expression differ in important ways from their birth sex (Davidson, 2007). A transgendered person may identify in a variety of ways, including transsexual, cross-dresser, drag king/queen, gender queer, and intersex (see Bockting, 2008, for discussion of various descriptors). • Transsexual refers to a person who lives or desires to live full time as a person opposite to his or her birth sex; often the person takes steps to alter his or her anatomy and physiology so that it is congruent with his or her identity (APA Task Force, 2008). We will use the term transwoman to refer to persons assigned a male birth sex but who identify as female (often referred to as male-tofemale or MTF transsexuals in the literature). Similarly, we will use the term transman to refer to 202

persons assigned a female birth sex but who identify as male (often referred to as female-tomale or FTM transsexuals in the literature). Although there are a variety of psychological components that may factor into sexual orientation (e.g., emotional connection and fantasy vs. actual behavior), most life science research has focused on the behavior and/or self-identity components. The most common way to assess this has been to ask participants to self-identify or with the use of specific items from the Kinsey Scale (Kinsey, Pomeroy, & Martin, 1948). Even though Kinsey’s estimates for the prevalence of nonheterosexual identities is often cited—that 10 percent of men and 5 percent of women are bisexual or gay/lesbian—more recent population estimates suggest that about 5 to 6 ­percent of men and 3 to 4 percent of women are nonheterosexual (Laumann, Gagnon, Michael, & Michaels, 1994; Sell, Wells, & Wypij, 1995). Estimating the prevalence of a transgender identity is challenging, in part because of the various aforementioned identity labels. The majority of research on the transgender community, however, has focused on those diagnosed with gender dysphoria (formerly gender identity disorder). Thus, we will focus on this subset of the transgender community. The American Psychiatric Association (2013) suggested that 1:10,000 to 1:20,000 persons in the United States will identify as a transwoman and that 1:30,000 to 1:50,000 will identify as a transman. Estimates in Asia have suggested that these rates are as high as 1:2,900 for transwomen and 1:8,300 for transmen (Tsoi, 1988). Many hypotheses have been proposed in relation to what contributes to the development of these identities. For both sexual orientation and gender identity, similar hypotheses have been proposed and falsified, including the idea that they develop due to traumatic past experiences (e.g., childhood sexual abuse), problematic parenting, and the absence of  a  specific parent (e.g., Bullough, Bullough, & Smith, 1983; Stevens, Golombok, Beveridge, & the ALSPAC Study Team, 2002). Just as the social sciences have helped dispel some misunderstandings regarding these traits, the life sciences have made some advances in explaining how biological factors contribute to sexual orientation and gender identity. Before proceeding, we offer an important caveat related to the peer-reviewed literature in this area of research: The research on sexual orientation is primarily based on gay men, and the research on gender identity is primarily based on transwomen. Thus, it

Hormones, Sexual Orientation, and Gender Identit y

is less certain to what degree the results generalize to lesbian women and transmen. In addition, those who identify as bisexual and asexual are usually combined into samples of gay men and lesbian women. To some degree this may merely reflect the fact that gay men and transwomen are more prevalent in the general population and because they are chromosomally less complex than those born female (i.e., they only possess one X chromosome). Nevertheless, this skewed representation in research limits how much we can generalize beyond gay men and transwomen.

Hormonal Influence

Sex hormones—specifically androgens and estrogens—play a key role in human development. Both their organizational and activational effects have been extensively studied (see, e.g., Arnold, 2009; see also Hampson, this volume). Even though it is no longer believed that sex hormones are the sole reason for the observed differences between males and females (e.g., Dewing, Shi, Horvath, & Vilain, 2003), they do play a critical role in sexual development. Consequently, the first area of research we focus on addresses the role that hormones play in influencing the development of sexual orientation and gender identity in three key ways: the activational effects of circulating (i.e., endogenous) hormones, the organizational effects of prenatal exposure to hormones, and the influence of exogenous hormones.

Circulating Hormones

Most hormones are circulating hormones. They are produced and released by an endocrine cell in one part of the body; they travel through the bloodstream to another part of the body; and they bind to specific receptors within target cells, which signal or activate the cell to behave in predetermined ways. Given the role that sex hormones play in sexual d ­ evelopment and given sex differences in hormone levels observed among mammalian species, the first area of research we highlight focuses on the role of circulating hormones in sexual orientation and gender identity. Sexual orientation. Early theories regarding sexual orientation centered on the hypothesis that homosexual men and homosexual women had abnormal levels of sex hormones (Meyer-Bahlburg, 1977, 1979). For instance, it was believed that homosexual men had lower levels of testosterone than their heterosexual counterparts, and it was hypothesized that this “deficiency” contributed to an “immature” sexuality (Lurie,  1944). As a result, some early researchers tested whether the administration of exogenous testosterone to homosexual men would lead them

to engage in sexual relations with women instead of men. The opposite, however, was found to occur: Instances of homosexual behavior increased in the majority of the men who were administered testosterone (Glass & Johnson, 1944). Subsequently, limited research has found differences between heterosexual and nonheterosexual adults. The few published reports primarily came from Robert Kolodny and colleagues (1971, 1972), in which they showed that homosexual men had lower levels of testosterone and impaired spermatogenesis compared to controls. Conversely, the majority of studies have shown that homosexual men do not differ in testosterone levels when compared to heterosexual men (Barlow, Abel, Blanchard, & Mevissakalian, 1974; Meyer-Bahlburg, 1984; Pillard, Rose, & Sherwood, 1974; Tourney & Hatfield, 1973), and testosterone levels among homosexual women were found to be comparable to heterosexual women (Dancey,  1990; Downey, Ehrhardt, Schiffman, Dyrenfurth, & Becker,  1987; Gartrell, Loriaux, & Chase, 1977). There is some evidence, however, that among lesbians there may be an association between androgen levels and their self-reported role within a romantic relationship (Singh, Vidaurri, Zambarano, & Dabbs, 1999). That is, there was a paired difference within each dyad, whereby the partner who endorsed more stereotypically masculine traits had higher testosterone levels compared to the partner who endorsed more stereotypically feminine traits (Pearcey, Docherty, & Dabbs, 1996). Gender identity. As with sexual orientation, early hypotheses for explaining transsexual identity related to atypical hormonal ranges. Two early studies reported lower plasma testosterone in transwomen compared to heterosexual men (Starká, Spiová, & Hynie, 1975) and higher testosterone levels in transmen compared to heterosexual women (Spiová & Starká, 1977). Yet, later studies did not find differences in gonadotropin and sex steroid secretions when comparing transwomen to control males (e.g., Spijkstra, Spinder, & Gooren,  1988). Subsequent clinical trials found that prior to hormonal treatment, transwomen are in the normal range for cisgender males (e.g., Dittrich et al., 2005).

Prenatal Exposure to Endogenous Hormones

Most remaining hypotheses linking hormones with sexual orientation and gender identity focus on the prenatal environment. Specifically, there are critical periods of development during which it is essential that sex hormones be present at sufficient levels Neibergall, Swanson, and Sánche z

203

to facilitate the differentiation and organization of the developing embryo—especially the reproductive system and sexually dimorphic brain regions. Otherwise, there can be long-term, irreversible ­effects if the exposure to these hormones occurs too late or at atypical levels. Consequently, several hypotheses suggest that extreme variations in ­prenatal exposure to hormones during this critical period may contribute to the development of nonheterosexual orientations and non-cisgender identities. These hypotheses were primarily derived from animal models. For instance, Phoenix, Goy, Gerall, and Young (1959) found that prenatally manipulating hormone levels in developing guinea pigs resulted in adult animals that displayed sexual behavior common of the opposite sex (e.g., females attempting to mount vs. assuming a receptive position). Similar results were found in rats that had their hormones manipulated either prenatally or perinatally (Davidson, 1969; Larsson & Sodersten, 1971; Paup, Mennin, & Gorski,  1975; Quadagno, Shryne, Anderson, & Gorski, 1972). Although intriguing, using animal models to study these characteristics makes it difficult to draw firm conclusions for two main reasons: First, in most of these animal studies, hormone levels were manipulated significantly beyond the naturally occurring range for a given sex. Second, the observed animal behavior does not compare well to humans, at least in regard to sexual orientation and gender identity. From a behavioral perspective, all that can be concluded is that hormonally manipulated animals exhibited oppositesex sexual behaviors in response to sexual stimuli; nothing can be concluded regarding the animals’ cognitions, affect, or “identity.” A definitive way to test these hypotheses would be to conduct such studies in humans. Of course, stricter ethical and moral parameters that exist for human subjects compared to animal subjects prohibit this type of experimental manipulation. Yet, many people have been exposed to atypical hormone levels in the womb for a variety of reasons. Here we report what this limited research has found related to sexuality. Limited research has been published related to prenatal exposure and gender identity. Classifying this requires participants’ self-report, which cannot be done with animal models. The low co-occurrence of a transgender identity within families (Green, 2000)—a common method for studying the ­inheritance of traits—makes it difficult to conduct common methods (e.g., linkage analysis) at this 204

time. Nevertheless, it is possible that, given the role that hormones play in sex development, they also influence personality traits related to gender identity. Perhaps the closest naturally occurring “experiment” with humans would be studying people who have been diagnosed with a disorder of sex development. Two specific groups have been researched. The first group includes women diagnosed with congenital adrenal hyperplasia (CAH), in which females are prenatally exposed to high levels of androgens produced by their adrenal glands; such women report greater interest in stereotypically masculine interests and behaviors compared to women without CAH (Zucker et al., 1996). Numerous studies examining the sexual orientation of women with CAH have mostly found that fewer such women identify as exclusively or almost exclusively heterosexual compared to control women (e.g., Dittmann, Kappes, & Kappes, 1992; Frisén et al., 2009). This outcome seems to be mediated by severity of the condition, whereby women with the more severe form of CAH (with life-threatening sodium deficiency symptoms) report a higher incidence of samesex attraction compared to women with milder forms of CAH (Meyer-Bahlburg, Dolezal, Baker, & New, 2008). It should be noted, however, that the physical consequences among women with CAH can adversely affect their body image, such that they view themselves as sexually undesirable and are highly self-conscious (Frisén et al.,  2009; Zucker et al., 1996). Consequently, the effects of this condition on sexual orientation are complex. The second group includes persons diagnosed with androgen insensitivity syndrome—a condition resulting from a genetic mutation that impairs the cell receptors for androgen. Persons with this condition inherited an XY sex chromosome complement (which normally leads to the development of a male), yet during the critical period of sexual development in utero, their bodies were unable to utilize androgens, thus impairing the “masculinization” of their bodies. Depending on the severity of the condition, the degree of impairment ranges from mild to complete across all androgen-dependent features (e.g., reproductive organs and specific brain regions). In the extreme, a person with complete androgen insensitivity would develop as a female, and the condition may not be discovered until the adolescent fails to experience menarche. Studies examining sexual orientation in these groups have yielded mixed results. Early reports found that participants identified as heterosexual relative to the gender of rearing versus their chromosomal

Hormones, Sexual Orientation, and Gender Identit y

composition—those raised as girls were attracted to males and those raised as boys were attracted to females (Lev-Ran, 1974; Money, Devore, & Norman, 1986; Money & Dalery, 1976; Money & Ogunro, 1974). Since then, studies found higher levels of homosexual or bisexual behavior among these ­ groups when compared to the general population (Hines, Brook, & Conway, 2004; Johannsen, Ripa, Mortensen, & Main,  2006; Zucker et  al.,  1996). A deeper analysis by Meyer-Bahlburg et al. (2008) suggested that the observed variation in sexual orientation may have been an artifact of the severity of the disorder of sex development. Specifically, women with CAH exposed to extremely high levels of testosterone were more likely to be classified as homosexual or bisexual (31 percent) compared to women with CAH who had been exposed to less testosterone in utero. Although this seems to implicate androgens in human sexual orientation, such an interpretation must be made with caution: Most people with CAH were exposed to levels of androgens far above the average exposure that developing male fetuses, including many of those who identify as homosexual, experience.

Pharmaceuticals and Endocrine-Disrupting Chemicals

A different approach to estimate how hormonal anomalies may affect developing embryos has been to focus on the offspring of mothers given hormonal treatments during pregnancy. Here we offer information on a few chemicals that have been studied, but there are others that may also affect sexual characteristics (e.g., synthetic thyroid medication, certain pesticides, and chlorinated hydrocarbons). Diethylstilbestrol (DES). DES is a synthetic form of estrogen. From the 1940s to 1970s, DES was prescribed to pregnant women because it was believed to curtail complications, including miscarriage and premature labor (Rubin, 2007). In 1971, the FDA discontinued use of DES with pregnant women ­because, rather than preventing complications, it was found to increase the odds of having lifelong problems of the reproductive system for those prenatally exposed (“Selected Item fFrom the FDA Drug Bulletin—November 1971: Diethylstilbestrol Contraindicated in Pregnancy,” 1972). Although DES-exposed males do experience more health problems (e.g., testicular cancer) and structural differences of the genitals (e.g., misplaced urethral opening) compared to unexposed males, DES-exposed females experience proportionately greater problems (e.g., various types of cancer, fertility problems, and early

menopause) compared to unexposed females (Hoover et al., 2011). Psychosocial studies report that, on average, those exposed to DES report interests and behaviors stereotypically associated with their sex (Lish, MeyerBahlburg, Ehrhardt, Travis, & Veridiano, 1992). In terms of sexual orientation, there does not seem to be an association between DES exposure and sexual orientation in men. The results for women exposed to DES, however, are mixed. Based on three combined samples of women exposed to DES (n = 97), Meyer-Bahlburg and colleagues (1995) found that 76 percent reported exclusive heterosexual interests and experiences compared to 94 percent of women not exposed to DES. Yet, based on a larger sample of women exposed to DES (n = 3,946), Titus-Ernstoff and colleagues (2003) found no difference between these women and controls, and reports of exclusively heterosexual behavior were consistent with estimates from the general population. The marked difference between the two studies, however, may have been influenced by the questions used, whereby the former team assessed multiple dimensions of sexuality (e.g., fantasies, sexual attraction, and sexual arousal) and the latter team simply asked about actual sexual contact with others. Lutocyclin. Recently, Reinisch, Mortensen, and Sanders (2017) examined if prenatal exposure to exogenous progesterone—a hormone naturally released to support fertilization and pregnancy—was associated with sexual orientation. Based on medical records at University Hospital in Copenhagen, Denmark, the researchers identified and recruited 17 men and 17 women who had been prenatally exposed to lutocyclin, a bioidentical form of progesterone. Lutocyclin had been administered to their mothers while they were pregnant with the participants to minimize the chance of miscarriage. When compared to a matched group of control men and control women, Reinisch et al. found a higher incidence of identifying as nonheterosexual, increased likelihood of being attracted to the same sex, and increased likelihood of having engaged in same-sex sexual behavior among the exposed group. These initial findings support the possibility that exogenous progesterone exposure during critical periods of in utero development may alter embryonic development in ways that influence later behavior. Plastic compounds. Since the invention of the first synthetic polymer in the mid-1800s, plastic products have become an integral part of everyday life. From children’s toys to components within military Neibergall, Swanson, and Sánche z

205

weapons, plastics can be found just about anywhere. Although there are many advantages to plastics (e.g., they are inexpensive and easy to produce), their disadvantages may go beyond their massive environmental impact due to not being biodegradable. Specifically, compounds used to produce plastics may behave as endocrine disruptors—or chemicals that interfere with the endocrine system—if they enter the human body. Two compounds have been the focus of concern: phthalates and bisphenol A (BPA). Phthalates help make plastic flexible yet strong (e.g., soda bottles, medical tubing, and pacifiers). However, concerns over the potential hazard that phthalates could have on human health began to mount in the 1960s because of reports that transfused blood from plastic storage bags contained phthalates and because of reports of occupational illnesses among workers subjected to prolonged exposure to phthalates (Jaeger & Rubin,  1970; Marx,  1972; Milkov et al., 1973; Perspectives on PAEs, 1973). Subsequently, research demonstrated that phthalates adversely affected numerous mammalian species—especially among male organisms—including disrupting metabolism (Bell, 1982), causing abnormal development of the reproductive system (Howdeshell, Rider, Wilson, & Gray, 2008), inducing testicular atrophy (Oishi & Hiraga,  1980), and impairing spermatogenesis (Parmar, Srivastava, & Seth, 1986). Similarly, BPA is commonly used to make clear plastics hard and shatter resistant (e.g., eyeglasses, food storage containers, and baby bottles). Laboratory scientists unexpectedly discovered that BPA was seeping out of plastic flasks and contaminating the results of experiments (Krishnan, Stathis, Permuth, Tokes, & Feldman, 1993), which led to the discovery that BPA was seeping out of many types of plastics (Feldman & Krishnan, 1995). Soon after, researchers began to study the effects of BPA exposure on animal models; BPA was found to adversely affect sexual development, including disrupting androgen production (Akingbemi, Sottas, Koulova, Klinefelter, & Hardy, 2004), reducing fertility (Herath et al., 2004), causing abnormal ovarian development (Adewale, Jefferson, Newbold, & Patisaul, 2009), and contributing to recurrent miscarriages (Hunt et al., 2003). As evidence began to emerge that traces of phthalates and BPA were being detected in pregnant and nursing mothers (e.g., Sun et al.,  2004; Swan et al., 2005), public concern began to mount regarding how plastics could be affecting infants and children, especially given the coverage by the popular press (Matthews, 2016; O’Callaghan, 2009). Many of these fears were rooted in what research on 206

animals had shown; yet, the translation of animal findings to humans has been complex and contentious, with some arguing that there is no harm so long as doses are kept “low” (Habert, Livera, & Rouiller-Fabre,  2014; Waring & Harris,  2011). Nevertheless, it has been shown that these compounds can cross the placenta into human fetuses (Silva et al., 2004), and several correlational studies have found an association between plastic compounds and numerous sexual traits among humans, including underdevelopment of the external genitalia and incomplete testicular descent (Toppari, Virtanen, Skakkebaek, & Main,  2006; Swan et al.,  2005). Whether or not these chemicals can influence sexual orientation and gender identity remains unknown, but it is certainly possible given the aforementioned disruption of sex development seen in animals models (e.g., Gray et al.,  2000; Maamar et al.,  2015) and given the associations found in sex-related traits in humans (e.g., Percy et al., 2016; Swan et al., 2005). Future research should further investigate these potential associations.

Prenatal Stress

Another area that we will briefly mention is maternal stress during pregnancy (for a more in-depth account, see Deer, Bernard, & Hostinar, this volume). Animal studies have found that stressing pregnant animals, especially during critical periods of development, had significant consequences on the offspring (Allen & Haggett, 1977). For instance, Ward (1972) found that the male offspring of female rats repetitively subjected to an extremely stressful scenario during pregnancy were less likely to engage in mounting behavior and more likely to assume a receptive position to sexual stimuli when compared to the male offspring of rats that had not been stressed during pregnancy. Similarly, Sachser and Kaiser (1996) found that the female offspring of guinea pigs subjected to unstable social environments while pregnant exhibited more male-typical courtship and play behavior compared to the offspring from mothers that had been in stable social environments. Such studies demonstrate long-lasting effects that environmental stressors can have on pregnant animals and their offspring, in part because of the hormones produced when mammals experience distress. Among humans, initial evidence suggested that prenatal stress may also influence sexuality. A team from Germany recruited 60 homosexual men, 40 bisexual men, and 100 heterosexual men, and they asked them to consult with their parents regarding stressful events while their mothers had been pregnant

Hormones, Sexual Orientation, and Gender Identit y

with them. Significantly more of the homosexual and bisexual men reported that their mothers had experienced moderately to severely stressful events— mostly related to World War II—compared to the heterosexual men (68 percent, 40 percent, and 6 percent, respectively; Dörner, Schenk, Schmiedel, & Ahrens, 1983). Yet, subsequent retrospective studies failed to properly replicate this association among gay men and lesbian women (e.g., Bailey, Willerman, & Parks, 1991; Ellis, Ames, Peckham, & Burk, 1988). Furthermore, a longitudinal study of 13,998 pregnancies in England showed no significant relationship between prenatal stress among male offspring and gender atypical behavior, and they found only small correlations among girls (Hines et al., 2002).

Genetics

The role that genes play in our lives is a constantly evolving area of research, especially with the emergence of research on epigenetics. Science has identified several genes that play a critical role in human sex development—especially the role of the SRY gene (i.e., the sex-determining region on the Y chromosome). Yet, there is uncertainty regarding how genes influence sexual orientation and gender identity. Most genetics research in this area has focused on sexual orientation. This is in part because of the higher prevalence of people who identify as gay or lesbian compared to those who identify as transgender, which makes it easier to use established methods that rely on large samples (e.g., for genome-wide association studies; e.g., Sanders et al., 2017) or the co-occurrence of specific traits within families (e.g., for linkage analyses; e.g., Hamer, Hu, Magnuson, Hu, & Pattatucci, 1993) and among twins (e.g., Sánchez, Bocklandt, & Vilain,  2013). Details on general ­genetics research can be found elsewhere (e.g., Ngun et al.,  2011). Here we briefly mention studies that have focused on molecular genetics related to hormones among people who do not have disorders of sex development. Numerous genes that play a role in sex development have been considered as candidate genes for sexual orientation and gender identity. Some of these included CYP19 (converts androgens into estrogens), ERα and ERβ (encodes for estrogen receptors), PGR (encodes for progesterone receptors), and SRD5A2 (converts testosterone into dihydrotestosterone). Yet, molecular studies focused on these have yielded mixed results. For instance, the androgen receptor gene (AR) is located on the X chromosome and it  encodes for the protein that allows the body to ­respond to androgens. This receptor is critical during

key periods of development because androgens differentiate the fetus in male-specific ways. Yet, no significant variations were found in AR when comparing heterosexual and homosexual participants (Macke et al.,  1993). In regard to gender identity, Hare and her colleagues (2009) found that transwomen had a significant variation of the AR gene that would cause the receptors to be less effective; however, two subsequent groups failed to replicate that finding (Fernández et al., 2014b; Ujike et al., 2009). Overall, molecular studies that have focused on genes that affect the endocrine system among persons without disorders of sex development have yet to provide evidence that they influence sexual orientation or gender identity. Yet, emerging technologies and our evolving understanding of what influences gene expression may lead the way to solving the question of how our DNA influences these traits (Rice, Friberg, & Gavrilets, 2016). Furthermore, it is likely that different mechanisms are involved for different subsets of the LGBT community, whereby genes that contribute to a transwoman identity, for instance, will be markedly different from genes that contribute to a transman identity (Fernández et al., 2014a).

Association Studies

Studies focused directly on hormones have yielded more questions than answers, in part because of the  limitations of studying people versus animals. Although it may be some time before technology advances enough to make definitive conclusions, here we highlight association studies that have received substantial attention and for which hormones may have played a role. These studies have relied on cross-­ sectional, correlational approaches, which prohibit causative conclusions. Research in this area encompasses a broad range of topics, each with varying degrees of differences that are potentially related to hormones.

Birth Order Studies

In 1996, Blanchard and Bogaert first reported a phenomenon known as the fraternal birth order effect. They found evidence that, on average, homosexual men had a greater number of older brothers when compared to heterosexual men (Blanchard & Bogaert, 1996, 1997). In several follow-up studies, they concluded that each older brother increases the odds that a male will be homosexual by 33 percent (Blanchard & Bogaert, 2004). Although this sounds exceptionally large, this increase is relative to the small baseline rate of approximately 2.5 percent among the general population for men. Nevertheless, Neibergall, Swanson, and Sánche z

207

a statistical analysis suggested that an estimated 15 percent of homosexual men could attribute their sexual orientation to this effect (Cantor, Blanchard, Paterson, & Bogaert,  2002), though several caveats exist (e.g., the trend is limited to right-handed men). Interestingly, the same effect is not found for homosexual women, whereby older sisters do not increase the likelihood of homosexuality (Blanchard & Lippa, 2007). Since the publication of these findings, researchers have opined why this relationship exists among men. The leading theory suggests that the mother’s immune system may be affecting the developing brain of male fetuses via an immune response to malespecific biomolecules produced by the male fetus (Blanchard, 2001). Specifically, the developing testes produce H-Y antigens that play a hormone-like role in differentiating the reproductive system and are foreign to a woman’s body (Graves, 2010; Wolf, 1998). Thus, a mother’s immune system may become more efficient at producing antibodies to “attack” a male fetus with each successive male conception, which can adversely affect the “masculinization” of the brain (Blanchard & Bogaert,  1996; Blanchard & Klassen, 1997). Yet, there is no firm data to support this hypothesis. Some research has investigated the possibility of a fraternal birth order effect among transgender people. However, the limited findings seem to only show a trend among transwomen who are attracted to men (Blanchard & Sheridan,  1992; Poasa, Blanchard, & Zucker, 2004). Further evidence for this effect was found among a non-Western sample of transwomen in Spain and Turkey (Gómez-Gil et al., 2011; Bozkurt, Bozkurt, & Somnez, 2015). However, there were mixed results in a Korean sample (Zucker, Blanchard, Kim, Pae, & Lee, 2006). Consequently, the limited evidence within the trangender community so far seems to suggest that this effect is related to sexual orientation and not gender identity given that the findings have been limited to those who were sexually attracted to people who were of the same birth sex as the participant.

Finger-Length Ratio

One of the most frequently cited lines of studies relates to the finger-length ratio between the index finger (2D) and ring finger (4D). This ratio has been shown to be sexually dimorphic, whereby men on average have a shorter 2D than 4D, and women have about equal lengths (Manning, Scutt, Wilson, & Lewis-Jones,  1998; Phelps,  1952). The dominant 208

hypothesis for this sex difference—which lacks direct evidence—is that the ratio is sensitive to prenatal androgen exposure (Manning et al.,  1998). Consequently, studies have been conducted to determine if this trend is associated with sexual orientation and gender identity. In terms of sexual orientation, several studies initially found a difference: Both homosexual women and homosexual men were found to have a digit ratio more consistent with the opposite sex when compared to heterosexual controls (Williams et al., 2000). Yet, subsequent studies either found no difference between groups (e.g., Lippa, 2003; Kraemer et al., 2006) or the finding was found to be in the opposite direction of what was expected (e.g., homosexual men had more “masculine” ratios compared to heterosexual men; Robinson & Manning,  2000). In terms of gender identity, two studies found the 2D:4D ratio to be “feminized” in transwomen compared to control males, but did not find a “masculinized” pattern among transmen (Kraemer et al, 2006; Schneider, Pickel, & Stalla, 2006). A separate study found transmen had a more “masculinized” pattern but did not  find a “feminized” pattern among transwomen (Leinung & Wu, 2017). Thus, as with sexual orientation, uncertainty remains regarding the degree to which 2D:4D is associated with gender identity. Interest in the finger-length ratio continues with a variety of studies associating this ratio with other human traits. Yet, the debate about whether this ratio, which plays no critical role in sexual reproduction, is sensitive to androgens continues (Berenbaum, Bryk, Nowak, Quigley, & Moffat,  2009; Wallien, Zucker, Steensma, & Cohen-Kettenis, 2008).

Brain Research

Perhaps the strongest biological findings come from brain studies. Intrigue in this area began with findings among animal models. In addition to the aforementioned studies in which hormones were manipulated in rodents, a curious finding was reported among sheep, where approximately 8 percent of male rams exhibited a sexual preference for other rams (Roselli, Reddy, & Kaufman, 2011). Studies based on autopsied brains found that the sexually dimorphic ­nucleus in the preoptic area of the hypothalamus (SDN-POA) in these male rams was comparable in  size to female rams (Katz, Price, Wallach, & Zenchak, 1988; Roselli, Larkin, Resko, Stellflug, & Stormshak,  2004). This area is associated with sexual behavior and its size is sensitive to hormones

Hormones, Sexual Orientation, and Gender Identit y

during critical periods of development (Rhees, Shryne, & Gorski, 1990a, 1990b). Consequently, as science continues to explore how brain development affects animal behavior (e.g., Ngun et al.,  2014), ­attention has also been given to the role of brain structures in humans.

(p. 2041). Although promising, these results should be taken with caution because in both studies the transwomen had varying degrees of exposure to hormonal treatments, which were shown to alter the brain anatomy of transgender patients (Hulshoff Pol et al., 2006).

Autopsy studies. Initially, three sexually dimorphic brain structures were found to correlate with increased rates of homosexuality: the third interstitial nucleus of the anterior hypothalamus (INAH-3; LeVay,  1991), the anterior commissure (Allen & Gorski, 1992), and the arginine vasopressin n ­ euronal population of the suprachiasmatic nucleus (Swaab, Zhou, Fodor, & Hofman, 1997). For each of these brain areas, homosexual men were found to have structures that were more similar to control females than heterosexual males. LeVay’s (1991) finding received the most attention and criticism due to possible confounds. He obtained his results via postmortem autopsy analyses in which all of his nonheterosexual participants had died from AIDS complications. A subsequent study by Byne and colleagues (2001) found that AIDS was not associated with alterations in INAH-3 and that there was a statistical trend in the direction that LeVay found when comparing the brains of heterosexual and nonheterosexual men.

Brain imaging. Whereas earlier autopsy studies were unclear due to confounds and the difficulty in conducting them—chiefly because they relied on people donating their bodies to science—advances in brain imaging have led to multiple studies based on significantly larger sample sizes. Numerous regions have been studied in relation to sexual orientation and gender identity, including cortical thickness, the corpus callosum, cerebral asymmetry, and the putamen. Here we briefly highlight functional studies that have focused on responses to two specific types of stimuli. Readers interested in information on other areas are referred elsewhere (LeVay, 2016).

In regard to gender identity, an early key study (Zhou, Hofman, Gooren, & Swaab, 1995) compared the autopsied brains of heterosexual men, heterosexual women, homosexual men, and six transwomen. Results showed that the sexually dimorphic bed nucleus of the stria terminalis of the hypothalamus (BSTc) in homosexual men was similar in size to heterosexual men in the sample, and the BSTc of both these groups of men was significantly larger than heterosexual women. This showed that sexual orientation was not associated with this area. However, the size of the BSTc in the transwomen was significantly smaller than both groups of men and more similar in size to the heterosexual women (Zhou et al., 1995). A subsequent study on the brains of the transwomen found that they had neuronal density in the BSTc that was consistent with that of heterosexual women (Kruijver, Fernández-Guasti, Fodor, Kraan, & Swaab, 2002). Considering the results of these studies together, Kruijver and his colleagues suggested that “transsexualism may reflect a form of  brain hermaphroditism such that this limbi ­nucleus itself is structurally sexually differentiated opposite to the transsexual’s genetic and genital sex”

First, researchers have compared how brains r­espond to scent. Numerous species secrete pheromones or signaling chemicals to influence the behavior of organisms within range of detecting the signal (e.g., to signal an alarm or availability for breeding). Although the existence of human pheromones remains a disputed topic (Wyatt,  2015), researchers have examined how men and women respond to cross-sex scents given that humans have a good sense of smell. For instance, a Swedish research team found that the cerebral activation patterns for homosexual men smelling male perspiration was both similar to patterns shown by heterosexual women and significantly different from the patterns of heterosexual men (Savic, Berglund, & Lindström, 2005). Likewise, the researchers found that the cerebral activation patterns for homosexual women smelling male perspiration and female urine was both similar to patterns shown by heterosexual men and significantly different from the patterns of heterosexual women (Berglund, Lindström, & Savic, 2006). Finally, the Swedish group found that among a sample of transwomen who were attracted to other women, their response to odors was “in between” both male and female controls (Berglund, Lindström, DhejneHelmy, & Savic, 2008). Second, researchers have compared how brains respond to visual sexual stimuli. Numerous studies have found that brain activation while viewing erotic videos demonstrated arousal consistent with their sexual orientation (Hu et al.,  2008; Ponseti et al., 2006; Rieger, Chivers, & Bailey, 2005; Safron Neibergall, Swanson, and Sánche z

209

et  al.,  2007) and aversion when inconsistent with  their sexual orientation (Paul et al.,  2008). Although this finding was reported among a group of transwomen, the effect was not as marked as the findings among cisgender participants (Gizewski et al., 2009).

Conclusion

There continues to be fascination over what factors contribute to the development of our gender identity and sexual orientation. In this chapter, we highlighted past and current biological research related to each with an emphasis on the role of hormones. Although most research in this area has focused on people who develop in atypical ways relative to the general population, understanding these characteristics is relevant for all human beings. Moreover, advances in science—including epigenetics and neuroplasticity—continue to unlock some of the mysteries of human development and spark further fascination regarding the influence of hormones during critical periods in our lives (e.g., Aure, Fuss, Höhne, Stall, & Siever, 2014; Chavarria, Sánchez, Chou, Thompson, & Luders, 2014). It may be some time before scientists can provide any definitive causal conclusions, but there is continued excitement in this area given the pace at which technology is advancing.

References

Adewale, H. B., Jefferson, W. N., Newbold, R. R., & Patisaul, H.  B. (2009). Neonatal bisphenol-A exposure alters rat reproductive development and ovarian morphology without impairing activation of gonadotropin-releasing hormone neurons. Biology of Reproduction, 81, 690–699. doi:10.1095/ biolreprod.109.078261 Akingbemi, B.  T., Sottas, C.  M., Koulova, A.  I., Klinefelter, G.  R., & Hardy, M.  P. (2004). Inhibition of testicular steroidogenesis by the xenoestrogen bisphenol A is associated with reduced pituitary luteinizing hormone secretion and decreased steroidogenic enzyme gene expression in rat Leydig cells. Endocrinology, 145, 592–603. doi:10.1210/en.2003-1174 Allen, L. S., & Gorski, R. A. (1992). Sexual orientation and the size of the anterior commissure in the human brain. Proceedings of the National Academy of Sciences of the United States of America, 89, 7199–7202. Allen, T.  O., & Haggett, B.  N. (1977). Group housing of pregnant mice reduces copulatory receptivity of female progeny. Physiology & Behavior, 19, 61–68. doi:10.1016/00319384(77)90160-3 American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author. American Psychological Association. (2009). Publication manual of the American Psychological Association (6th ed.). Washington, DC: Author.

210

American Psychological Association (2009). Report of the Task Force on Gender Identity and Gender Variance. Washington, DC: Author. Arnold, A. P. (2009). The organizational-activational hypothesis as a foundation for a unified theory of sexual differentiation of all mammalian tissue. Hormones & Behavior, 55, 570–578. doi:10.1016/j.yhbeh.2009.03.011 Bailey, J. M., Willerman, L., & Parks, C. (1991). A test of the maternal stress theory of human male homosexuality. Archives of Sexual Behavior, 20, 277–293. doi:10.1007/ BF0154184 Barlow, L. H., Abel, G. G., Blanchard, E. B., & Mevissakalian, M. (1974). Plasma testosterone levels and male homosexuality: A failure to replicate. Archives of Sexual Behavior, 3, 571–575. Bell, F. P. (1982). Effects of phthalate esters on lipid metabolism in various tissues, cells and organelles in mammals. Environmental Health Perspectives, 45, 41–50. Berenbaum, S. A., Bryk, K. K., Nowak, N., Quigley, C. A., & Moffat, S. (2009). Fingers as a marker of prenatal androgen exposure. Endocrinology, 150, 5119–5124. doi:10.1210/en.20090774 Berglund, H., Lindström, P., Dhejne-Helmy, C., & Savic, I. (2008). Male-to-female transsexuals show sex-atypical hypothalamus activation when smelling odorous steroids. Cerebral Cortex, 18, 1900–1908. doi:10.1093/cercor/bhm216 Berglund, H., Lindström, P., & Savic, I. (2006). Brain response to putative pheromones in lesbian women. Proceedings of the National Academy of Sciences of the United States of America, 103, 8269–8274. doi:10.1073/pnas.0600331103 Blanchard, R. (2001). Fraternal birth order and the maternal immune hypothesis of male homosexuality. Hormones & Behavior, 40, 105–114. doi:10.1006/hbeh.2001.1681 Blanchard, R., & Bogaert, A. F. (1996). Homosexuality in men and number of older brothers, American Journal of Psychiatry, 153, 27–31. doi:10.1176/ajp.153.1.27 Blanchard, R., & Bogaert, A. F. (1997). Additive effects of older brothers and homosexual brothers in the prediction of marriage and cohabitation. Behavioral Genetics, 27, 45–54. Blanchard, R., & Bogaert, A. F. (2004). Proportion of homosexual men who owe their sexual orientation to fraternal birth order: An estimate based on two national probability samples. American Journal of Human Biology, 16, 151–157. doi:10.1002/ ajhb.20006 Blanchard, R., & Klassen, P. (1997). H-Y antigen and homosexuality in men. Journal of Theoretical Biology, 185, 373–378. doi:10.1006/jtbi.1996.0315 Blanchard, R., & Lippa, R. A. (2007). Birth order, sibling sex ratio, handedness, and sexual orientation of male and female participants in a BBC internet research project. Archives of  Sexual Behavior, 36, 163–176. doi:10.1007/s10508-0069159-7 Blanchard, R., & Sheridan, P. (1992). Sibship size, sibling sex ratio, birth order, and parental age in homosexual and nonhomosexual gender dysphorics. Journal of Nervous & Mental Disease, 180, 40–47. Bockting, W.  O. (2008). Psychotherapy and the real-life experience: From gender dichotomy to gender diversity. Sexologies, 17, 211–224. doi:10.1016/j.sexol.2008.08.001 Bozkurt, A., Bozkurt, O. H., & Sonmez, I. (2015). Birth order and sibling sex ratio in a population with high fertility: Are Turkish male to female transsexuals different? Archives of Sexual Behavior, 44, 1331–1337. doi:10.1007/s10508-014-0425-9

Hormones, Sexual Orientation, and Gender Identit y

Bullough, V. L., Bullough, B., & Smith, R. (1983). A comparative study of male transvestites, male to female transsexuals, and male homosexuals. Journal of Sex Research, 19, 238–257. Byne, W., Tobet, S., Mattiace, L., Lasco, S., Kemether, E., Edgar, M.  A., . . . Jones, L.  B. (2001). The interstitial nuclei of the human anterior hypothalamus: An investigation of variation within sex, sexual orientation and HIV status. Hormones & Behavior, 40, 86–92. doi:10.1006/hbeh.2001.1680 Cantor, J. M., Blanchard, R., Paterson, A. D., & Bogaert, A. F. (2002). How many gay men owe their sexual orientation to fraternal birth order? Archives of Sexual Behavior, 31, 63–71. Dancey, C. P. (1990). Sexual orientation in women: An investigation of hormonal and personality variables. Biological Psychiatry, 30, 251–264. Davidson, J. M. (1969). Effects of estrogen on the sexual behavior or male rats. Endocrinology, 84, 1365–1372. doi:10.1210/endo84-6-1365 Dewing, P., Shi, T., Horvath, S., & Vilain, E. (2003). Sexually dimorphic gene expression in mouse brain precedes gonadal differentiation. Molecular Brain Research, 118, 82–90. doi:10.1016/S0169-328X(03)00339-5 Dittmann, R.  W., Kappes, M.  E., & Kappes, M.  H. (1992). Sexual behavior in adolescent and adult females with congenital adrenal hyperplasia. Psychoneuroendocrinology, 17, 153–170. doi:10.1016/0306-4530(92)90054-B Dittrich, R., Binder, H., Cupisti, S., Hoffmann, I., Beckmann, M. W., & Mueller, A. (2005). Endocrine treatment of maleto-female transsexuals using gonadotropin-releasing hormone agonist. Experimental and Clinical Endocrinology & Diabetes, 113, 586–592. doi:10.1055/s-2005-865900 Dörner, G., Schenk, B., Schmiedel, B., & Ahrens, L. (1983). Stressful events in prenatal life of bi- and homosexual men. Experimental & Clinical Endocrinology, 81(1), 83–87. Downey, J., Ehrhardt, A. A., Schiffman, M., Dyrenfurth, I., & Becker, J. (1987). Sex hormones in lesbian and heterosexual women. Hormones & Behavior, 21, 347–357. doi:10.1016/0018506X(87)90019-5 Ellis, L., Ames, M. A., Peckham, W., & Burke, D. (1988). Sexual orientation of human offspring may be altered by severe maternal stress during pregnancy. Journal of Sex Research, 25, 152–157. Feldman, D., & Krishnan, A. (1995). Estrogens in unexpected places: Possible implications for researchers and consumers. Environmental Health Perspectives, 103(Suppl. 7), 129–133. Fernández, R., Esteva, I., Gómez-Gil, E., Rumbo, T., Almaraz, M.  C., Roda, E.,  .  .  .  Pásaro, E. (2014a). The (CA)n polymorphism of ERβ gene is associated with FtM transsexualism. Journal of Sexual Medicine, 11, 720–728. doi:10.1111/jsm.12398 Fernández, R., Esteva, I., Gómez-Gil, E., Rumbo, T., Almaraz, M.  C., Roda, E., . . . Pásaro, E. (2014b). Association study of ERβ, AR, and CYP19A1 genes and MtF transsexualism. Journal of Sexual Medicine, 11, 2986–2994. doi:10.1111/jsm.12673 Fishman, M. (2017, June 6). “I am a girl”: One child’s journey to “truegender.” USA Today. Retrieved from http://www. usatoday.com Frisén, J., Nordenstrom, A., Falhammar, H., Filipsson, H., Holmdahl G, Janson, P.  O., . . . Nordenskjold, A. (2009). Gender role behavior, sexuality, and psychosocial adaptation in women with congenital adrenal hyperplasia due to CYP21A2 deficiency. Journal of Clinical Endocrinology & Metabolism, 94, 3432–3439. doi:10.1210/jc.2009-0636

Gartrell, N.  K., Loriaux, D.  L., & Chase, T.  N. (1977). Plasma testosterone in homosexual and heterosexual women. American Journal of Psychiatry, 134, 1117–1119. doi:10.1176/ajp.134.10.1117 Gizewski, E. R., Krause, E., Schlamann, M., Happich, F., Ladd, M.  E., Forsting, M., & Senf, W. (2009). Specific cerebral activation due to visual erotic stimuli in male-to-female transsexuals compared with male and female controls: An fMRI study. Journal of Sexual Medicine, 6, 440–448. doi:10.1111/j.1743-6109.2008.00981.x Glass, S.  J., & Johnson, R.  W. (1944). Limitations and complications of organotherapy in male homosexuality. Journal of Clinical Neuroendocrinology, 4, 550–554. Gómez-Gil, E., Esteva, I., Carrasco, R., Almaraz, M. C., Pasaro, E., Salamero, M., & Guillamon  A. (2011). Birth order and ratio of brothers to sisters in Spanish transsexuals. Archives of Sexual Behavior, 40, 505–510. doi:10.1007/s10508-010-9614-3 Graves, J. A. M. (2010). Review: Sex chromosome evolution and the expression of sex-specific genes in the placenta. Placenta, 31, S27–S32. doi:10.1016/j.placenta.2009.12.029 Gray, L.  E., Jr., Ostby, J., Furr, J., Price, M., Veeramachaneni, D.  N.  R., & Parks, L. (2000). Perinatal exposure to the phthalates DEHP, BBP, and DINP, but not DEP, DMP, or DOTP, alters sexual differentiation of the male rat. Toxicological Sciences, 58, 350–365. doi:10.1093/toxsci/58.2.350 Green, R. (2000). Family cooccurrence of “gender dysphoria”: Ten sibling or parent–child pairs. Archives of Sexual Behavior, 29, 499–507. Habert, R., Livera, G., & Rouiller-Fabre, V. (2014). Man is not a big rat: Concerns with traditional human risk assessment of phthalates based on their anti-androgenic effects observed in the rat foetus. Basic & Clinical Andrology, 24, 14. doi:10.1186/2051-4190-24–14 Hamer, D. H., Hu, S., Magnuson, V. L., Hu, N., & Pattatucci, A.  M. (1993). A linkage between DNA markers on the X chromosome and male sexual orientation. Science, 261, 321–327. doi:10.1126/science.8332896 Hare, L., Bernard, P., Sánchez, F. J., Vilain, E., Kennedy, T., & Harley, V.  R. (2009). Androgen receptor (AR) repeat length polymorphism associated with male-to-female transsexualism. Biological Psychiatry, 65, 93–96. doi:10.1016/j.biopsych.2008.08.033 Herath, C. B., Jin, W., Watanabe, G., Arai, K., Suzuki, A. K., & Taya, K. (2004). Adverse effects of environmental toxicants, octylphenol and bisphenol A, on male reproductive functions in pubertal rats. Endocrine, 25, 163–172. doi:10.1385/ ENDO:25:2:163 Hines, M., Brook, C., & Conway, G.  S. (2004). Androgen and psychosexual development: Core gender identity, sexual orientation, and recalled childhood gender role behavior in women and men with congenital adrenal hyperplasia (CAH). Journal of Sex Research, 41, 75–81. doi:10.1080/00224490409552215 Hines, M., Johnston, K., Golombok, S., Rust, J, Stevens, M., Golding, J., & the ALSPAC Study Team. (2002). Prenatal stress and gender role behavior in girls and boys: A longitudinal, population study. Hormones & Behavior, 42, 126–134. doi:10.1006/hbeh.2002.1814 Hoover, R. N., Hyer, M., Pfeiffer, R. M., Adam, E., Bond, B., Cheville, A. L., . . . Troisi, R. (2011). Adverse health outcomes in women exposed in utero to diethylstilbestrol. New England Journal of Medicine, 365, 1304–1314. doi:10.1056/ NEJMoa1013961

Neibergall, Swanson, and Sánche z

211

Howdeshell, K. L., Rider, C. V., Wilson, V. S., & Grey, L. E., Jr. (2008). Mechanisms of action of phthalate esters, individually and in combination, to induce abnormal reproductive development in male laboratory rats. Environmental Research, 108, 168–176. doi:10.1016/j.envres.2008.08.009 Hu, S. H., Wei, N., Wang, Q. D., Yan, L. Q., Wei, E. Q., Zhang, M. M., . . . Xu Y. (2008). Patterns of brain activation during visually evoked sexual arousal differ between homosexual and heterosexual men. American Journal of Neuroradiology, 29, 1890–1896. doi:10.3174/ajnr.A1260 Hughes, I.  A., Houk, C., Ahmed, S.  F., Lee, P.  A., LWPES Consensus Group, & ESPE Consensus Group. (2006). Consensus statement on management of intersex disorders. Archives of Disease in Childhood, 91, 554–563. doi:10.1136/ adc.2006.098319 Hulshoff Pol, H. E., Cohen-Kettenis, P. T., van Haren, N. E. M., Peper, J.  S., Brans, R.  G.  H., Cahn, W., . . . Kahn, R.  S. (2006). Changing your sex changes your brain: Influences of testosterone and estrogen on adult human brain structure. European Journal of Endocrinology, 155, S107– S114. doi:10.1530/eje.1.02248 Hunt, P. A., Koehler, K. E., Susiarjo, M., Hodges, C. A., Ilagan, A., Voigt, R.  C., . . . Hassold, T.  J. (2003). Bisphenol A exposure causes meiotic aneuploidy in the female mouse. Current Biology, 13, 546–553. doi:10.1016/S0960-9822 (03)00189-1 Jaeger, R.  J., & Rubin, R.  J. (1970). Contamination of blood stored in plastic packs [Letter to the Editor]. Lancet, 296, 151. doi:10.1016/S0140-6736(70)92734-0 Johannsen, T.  H., Ripa, C.  P.  L., Mortensen, E.  L., & Main, K. M. (2006). Quality of life in 70 women with disorders of sex development. European Journal of Endocrinology, 155, 877–885. doi:10.1530/eje.1.02294 Katz, L. S., Price, E. O., Wallach, S. J. R., & Zenchak, J. J. (1988). Sexual performance of rams reared with or without females after weaning. Journal of Animal Science, 66, 1166–1173. Kinsey, A. C., Pomeroy, W. B., & Martin, C. E. (1948). Sexual behavior in the human male. Bloomington, IN: Indiana University Press. Kraemer, B., Noll, T., Delsignore, A., Milos, G., Schnyder, U., & Hepp, U. (2006). Finger length ratio (2D:4D) in adults with gender identity disorder. Archives of Sexual Behavior, 38, 359–363. doi:10.1007/s10508-007-9262-4 Krishnan, A. V., Stathis, P., Permuth, S. F., Tokes, L., & Feldman, D. (1993). Bisphenol-A: An estrogenic substance is released from polycarbonate flasks during autoclaving. Endocrinology, 132, 2279–2286. doi:10.1210/endo.132.6.8504731 Kruijver, F.  P.  M., Fernández-Guasti, A., Fodor, M., Kraan, E. M., & Swaab, D. F. (2002). Male-to-female transsexuals have female neuron numbers in a limbic nucleus. Journal of Clinical Endocrinology & Metabolism, 85, 2034–2041. doi:10.1210/jcem.85.5.6564 Larsson, K., & Södersten, P. (1971). Lordosis behavior in male rats treated with estrogen in combination with tetrabenazine and nialamide. Psychopharmacologia, 21, 13–16. Laumann, E. O., Gagnon, J. H., Michael, R. T., & Michaels, S. (1994). The social organization of sexuality. Chicago, IL: University of Chicago Press. Leinung, M., & Wu, C. (2017). The biological basis of transgender identity: 2D:4D finger length ratios implicate a role for prenatal androgen activity. Endocrine Practice, 23, 669–671. doi:10.4158/EP161528.OR

212

LeVay, S. (1991). A difference in hypothalamic structure between heterosexual and homosexual men. Science, 253, 1034–1037. doi:10.1126/science.1887219 LeVay, S. (2016). Gay, straight, and the reason why: The science of sexual orientation (2nd ed.). New York, NY: Oxford University Press. Lev-Ran, A. (1974). Sexuality and educational levels of women with the late treated adrenogenital syndrome. Archives of Sexual Behavior, 3, 27–32. Lippa, R.  A. (2003). Are 2D:4D finger-length ratios related to sexual orientation? Yes for men, no for women. Journal of Personality & Social Psychology, 85, 179–188. doi:10.1037/00223514.85.1.179 Lish, J. D., Meyer-Bahlburg, H. F. L., Ehrhardt, A. A., Travis, B.  G., & Veridiano, N.  P. (1992). Prenatal exposure to diethylstilbestrol (DES): Childhood play behavior and adult gender-role behavior in women. Archives of Sexual Behavior, 21, 423–441. doi:10.1007/BF01542268 Lurie, L.  A. (1944). The endocrine factor in homosexuality: Report of treatment of 4 cases with androgen hormone. American Journal of Medical Sciences, 208, 176–186. Maamar, M. B., Lesné, L., Desdoits-Lethimonier, C., Coiffec, I., Lassurguère, J., Lavoué, V.,  .  .  .  Jégou, B. (2015). An investigation of the endocrine-disruptive effects of bisphenol A in human and rat fetal testes. PLoS One, 10, e0117226. doi:10.1371/journal.pone.0117226 Macke, J. P., Hu, N., Hu, S., Bailey, M., King, V. L., Brown, T.,  .  .  .  Nathans, J. (1993). Sequence variation in the androgen receptor gene is not a common determinant of male sexual orientation. American Journal of Human Genetics, 53, 844–852. Manning, J. T., Scutt, D., Wilson, J., & Lewis-Jones, D. I. (1998). The ratio of 2nd to 4th digit length: A predictor of sperm numbers and concentrations of testosterone, luteinizing hormone and oestrogen. Human Reproduction, 13, 3000–3004. Marx, J.  L. (1972). Phthalic acid esters: Biological impact uncertain. Science, 178, 46–47. Matthews, S. (2016, December 7). Why men should stay away from plastic. Daily Mail. Retrieved from http://www.dailymail. co.uk Meyer-Bahlburg, H.  F.  L. (1977). Sex hormones and male homosexuality: A comparative perspective. Archives of Sexual Behavior, 6, 297–325. Meyer-Bahlburg, H.  F.  L. (1979). Sex hormones and female homosexuality: A critical examination. Archives of Sexual Behavior, 8, 101–119. Meyer-Bahlburg, H. F. L. (1984). Psychoendocrine research on sexual orientation. Current status and future options. Progress in Brain Research, 61, 375–398. doi:10.1016/S00796123(08)64448-9 Meyer-Bahlburg, H.  F.  L., Dolezal, C., Baker, S.  W., & New, M. I. (2008). Sexual orientation in women with classical or non-classical congenital adrenal hyperplasia as a function of degree of prenatal androgen excess. Archives of Sexual Behavior, 37, 85–99. doi:10.1007/s10508-007-9265-1 Meyer-Bahlburg, H.  F.  L., Ehrhardt, A.  A., Rosen, L.  R., Gruen, R. S., Veridiano, N. P., Vann, F. H., & Neuwalder, H.  F. (1995). Prenatal estrogens and the development of homosexual orientation. Developmental Psychology, 31, 12–21. doi:10.1037/0012-1649.31.1.12 Milkov, L. E., Aldyreva, M. V., Popova, T. B., Kopukhova, K. A., Makarenko, Y. L., Malyar, L. M., & Shakhova, T. K. (1973). Health status of workers exposed to phthalate plasticizers in

Hormones, Sexual Orientation, and Gender Identit y

the manufacture of artificial leather and films based on PVC resins. Environmental Health Perspectives, 3, 175–178. Money, J., & Dalery, J. (1976). Iatrogenic homosexuality: Gender identity in seven 46, XX chromosomal females with hyperadrenocortical hermaphroditism born with a penis. Three reared as boys, four reared as girls. Journal of Homosexuality, 1, 357–371. Money, J., Devore, H., & Norman, B. F. (1986). Gender identity and gender transposition: Longitudinal outcome study of 32 male hermaphrodites assigned as girls. Journal of Sex & MaritalTherapy,12,165–181.doi:10.1080/00926238608415404 Money, J., & Ogunro, C. (1974). Behavioral sexology: Ten cases of genetic male intersexuality with impaired prenatal and pubertal androgenization. Archives of Sexual Behavior, 3, 181–205. Ngun, T.  C., Ghahramani, N.  M., Creek, M.  M., WilliamsBurris, S. M., Bareseghyan, H., Itoh, Y., . . . Vilain, E. (2014). Feminized behavior and brain gene expression in a novel mouse model of Klinefelter syndrome. Archives of Sexual Behavior, 43, 1043–1057. doi:10.1007/s10508-014-0316-0 Ngun, T. C., Ghahramani, N., Sánchez, F. J., Bocklandt, S., & Vilain, E. (2011). The genetics of sex differences in brain and behavior. Frontiers in Neuroendocrinology, 32, 227–246. doi:10.1016/j.yfrne.2010.10.001 O’Callaghan, T. (2009, November 19). Can plastic chemicals cause effeminate behavior in boys? Time. Retrieved from http://healthland.time.com Obergefell v. Hodges, 576 U.S. ___ (2015). Retrieved from http:// www.supremecourt.gov Oishi, S., & Hiraga, K. (1980). Testicular atrophy induced by phthalic acid esters: Effect on testosterone and zinc concentrations. Toxicology & Applied Pharmacology, 53, 35–41. doi:10.1016/0041-008X(80)90378-6 Parmar, D., Srivastava, S.  P., & Seth, P.  K. (1986). Effect of di(2-ethylhexyl)phthalate (DEHP) on spermatogenesis in adult rats. Toxicology, 42, 47–55. Paup, D. C., Mennin, S. P., & Gorski, R. A. (1975). Androgenand estrogen-induced copulatory behavior and inhibition of luteinizing hormone (LH) secretion in the male rat. Hormones & Behavior, 6, 35–46. Paul, T., Schiffer, B., Zwarg, T., Krüger, T. H. C., Karama, S., Schedlowski, M., . . . Gizewski, E. R. (2008). Brain response to visual sexual stimuli in heterosexual and homosexual males. Human Brain Mapping, 29, 726–735. doi:10.1002/ hbm.20435 Pearcey, S. M., Docherty, K. J., & Dabbs, J. M. (1996). Testosterone and sex role identification in lesbian couples. Physiology & Behavior, 60, 1033–1035. doi:10.1016/0031-9384(96)00132-1 Percy, Z., Xu, Y., Sucharew, H., Khoury, J. C., Calafat, A. M., Braun, J. M., . . . Youtlon, K. (2016). Gestational exposure to phthalates and gender-related play behaviors in 8-year-old children: An observational study. Environmental Health, 15, 87. doi:10.1186/s12940-016-0171-7 Perspectives on PAEs [Editorial]. (1973). Environmental Health Perspectives, 3, 1. Phelps, V. R. (1952). Relative index finger length as a sex-influenced trait in man. American Journal of Human Genetics, 4, 72–89. Phoenix, C. H., Goy, R. W., Gerall, A. A., & Young, W. C. (1959). Organizing action of prenatally administered testosterone propionate on the tissue mediating mating behavior in the female guinea pig. Endocrinology, 65, 369–382. Pillard, R.  C., Rose, R.  M., & Sherwood, M. (1974). Plasma testosterone levels in homosexual men. Archives of Sexual Behavior, 3, 453–458.

Poasa, K. H., Blanchard, R., & Zucker, K. J. (2004). Birth order in transgendered males from Polynesia: A quantitative study of Samoan Fa’afafine. Journal of Sex & Marital Therapy, 30, 13–23. Ponseti, J., Bosinski, H.  A., Wolff, S., Peller, M., Jansen, O., Mehdorn, H.  M., . . . Siebner, H.  R. (2006). A functional endophenotype for sexual orientation in humans. NeuroImage, 33, 825–833. doi:10.1016/j.neuroimage.2006.08.002 Quadagno, D.  M., Shryne, J., Anderson, C., & Gorski, R.  A. (1972). Influence of gonadal hormones on social, sexual, emergence, and open field behaviour in the rat (Rattus norvegicus). Animal Behaviour, 20, 732–740. Reinisch, J.  M., Mortensen, E.  L., & Sanders, S.  A. (2017). Prenatal exposure to progesterone affects sexual orientation in humans. Archives of Sexual Behavior, 46, 1239–1249. doi:10.1007/s10508-016-0923-z Rhees, R. W., Shryne, J. E., & Gorski, R. A. (1990a). Onset of the hormone-sensitive perinatal period for sexual differentiation of the sexually dimorphic nucleus of the preoptic area in female rats. Developmental Neurobiology, 21, 781–786. doi:10.1002/neu.480210511 Rhees, R. W., Shryne, J. E., & Gorski, R. A. (1990b). Termination of the hormone-sensitive period for differentiation of the sexually dimorphic nucleus of the preoptic area in male and female rats. Developmental Brain Research, 52, 17–23. doi:10.1016/0165-3806(90)90217-M Rice, W.  R., Friberg, U., & Gavrilet, S. (2016). Sexually antagonistic epigenetic marks that canalize sexually dimorphic development. Molecular Ecology, 25, 1812–1822. doi:10.1111/ mec.13490 Rieger, G., Chivers, M. L., & Bailey, J. M. (2005). Sexual arousal patters of bisexual men. Psychological Science, 16, 579–584. doi:10.1111/j.1467–9280.2005.01578.x Robinson, S. J., & Manning, J. T. (2000). The ratio of 2nd to 4th digit length and male homosexuality. Evolution and Human Behavior, 21, 333–345. Roselli, C. E., Larkin, K., Resko, J. A., Stellflug, J. N., & Stormshak, F. (2004). The volume of a sexually dimorphic nucleus in the ovine medial preoptic area/anterior hypothalamus varies with sexual partner preference. Endocrinology, 145, 473–483. doi:10.1210/en.2003–1098 Roselli, C.  E., Reddy, R., & Kaufman, K. (2011). The development of male-oriented behavior in rams. Frontiers in Neuroendocrinology, 32, 164–169. doi:10.1016/j.yfrne. 2010.12.007 Rubin, M.  M. (2007). Antenatal exposure to DES: Lessons learned . . . future concerns. Obstetrical & Gynecological Survey, 62, 548–555. doi:10.1097/01.ogx.0000271138.31234.d7 Sachser, N., & Kaiser, S. (1996). Prenatal social stress masculinizes the females’ behaviour in guinea pigs. Physiology & Behavior, 60, 589–594. doi:10.1016/S0031-9384(96)80036-9 Safron, A., Barch, B., Bailey, J.  M., Gitelman, D.  R., Parrish, T. B., & Reber, P. J. (2007). Neural correlates of sexual arousal in homosexual and heterosexual men. Behavioral Neuroscience, 121, 237–248. doi:10.1037/0735-7044.121.2.237 Sánchez, F. J., Bocklandt, S., & Vilain, E. (2013). The relationship between help-seeking attitudes and masculine norms among monozygotic male-twins discordant for sexual orientation. Health Psychology, 32, 52–56. doi:10.1037/a0029529 Sanders, A. R., Beecham, G. W., Guo, S., Dawood, K., Rieger, G., Badner, J.  A., . . . Martin, E.  R. (2017). Genome-wide association study of male sexual orientation. Scientific Reports, 7, 16950. doi:10.1038/s41598-017-15736-4

Neibergall, Swanson, and Sánche z

213

Savic, I., Berglund, H., & Lindström, P. (2005). Brain response to putative pheromones in homosexual men. Proceedings of the National Academy of Sciences of the United States of America, 102, 7356–7361. Schilt, K., & Westbrook, L. (2009). Doing gender, doing heteronormativity: “Gender normals,” transgender people, and the social maintenance of heterosexuality. Gender & Society, 23, 440–464. doi:10.1177/0891243209340034 Schneider, H. J., Pickel, J., & Stalla, G. K. (2006). Typical female 2nd-4th finger length (2D:4D) ratios in male-to-female transsexuals-Possible implications for prenatal androgen exposure. Psychoneuroendocrinology, 31, 265–269. doi:10.1016/j. psyneuen.2005.07.005 Selected Item From the FDA Drug Bulletin—November 1971: Diethylstilbestrol Contraindicated in Pregnancy (1972). California Medicine, 116(2), 85–86. Sell, R. L., Wells, J. A., & Wypij, D. (1995). The prevalence of homosexual behavior and attraction in the United States, the United Kingdom and France: Results of national populationbased samples. Archives of Sexual Behavior, 24, 235–248. doi:10.1007/BF01541598 Shear, M. D., & Savage, C. (2017, July 27). In one day, Trump administration lands 3 punches against gay rights. New York Times. Retrieved from http://www.nytimes.com Silva, M.  J., Barr, D.  B., Reidy, J.  A., Malek, N.  A., Hodge, C. C., Caudill, S. P., . . . Calafat, A. (2004). Urinary levels of seven phthalate metabolites in the U.S. population from the national health and nutrition examination survey (NHANES) 1999–2000. Environmental Health Perspectives, 112, 331–338. Singh, D., Viaurri, M., Zambarano, R. J., & Dabbs, J. M., Jr. (1999). Lesbian erotic role identification: Behavioral, morphological, and hormonal correlates. Journal of Personality & Social Psychology, 76, 1035–1049. doi:10.1037/00223514.76.6.1035 Spijkstra, J.  J., Spinder, T., & Gooren, L.  J.  G. (1988). Shortterm patterns of pulsatile luteinizing hormone secretion do not differ between male-to-female transsexuals and heterosexual men. Psychoneuroendocrinology, 13, 279–273. Spiová, L., & Starká, L. (1977). Plasma testosterone values in transsexual women. Archives of Sexual Behavior, 6, 477–481. Starká, L., Spiová, I., & Hynie, J. (1975). Plasma testosterone in male transsexuals and homosexuals. Journal of Sex Research, 11, 134–138. Steinmetz, K. (2017, February 22). President Trump just rolled back guidelines that protected transgender students. Time Magazine. Retrieved from http://www.time.com Stevens, M., Golombok, S., Beveridge, M., & the ALSPAC Study Team. (2002). Does father absence influence children’s gender development? Findings from a general population study of preschool children. Parenting: Science & Practice, 2, 47–60. doi:10.1207/S15327922PAR0201_3 Stewart, P., & Chiacu, D. (2017, July 26). Trump to ban transgender U.S. military personnel, reversing Obama. Reuters. Retrieved from http://www.reuters.com Sun, Y., Irie, M., Kishikawa, N., Wada, M., Kuroda, N., & Nakashima, K. (2004). Determination of bisphenol A in human breast milk by HPLC with column-switching and fluorescence detection. Biomedical Chromatography, 18, 501–507. doi:10.1002/bmc.345 Swaab, D. F., Zhou, J. N., Fodor, M., & Hofman, M. A. (1997). Sexual differentiation of the human hypothalamus: Differences according to sex, sexual orientation, and transsexuality. In L.  Ellis & L.  Ebertz (Eds.), Sexual orientation: Toward

214

biological understanding (pp. 129–150). Westport, CT: Praeger Publishers/Greenwood Publishing Group. Swan, S. H., Main, K. M., Liu, F., Stewart, S. L., Kruse, R. L., Calafat, A.  M., . . . the Study for Future Families Research Team. (2005). Decrease in anogenital distance among male infants with prenatal phthalate exposure. Environmental Health Perspectives, 113, 1056–1061. doi:10.1289/ehp.8100 Titus-Ernstoff, L., Perez, K., Hatch, E.  E., Troisi, R., Palmer, J.  R., Hartge, P.,  .  .  .  Hoover, R. (2003). Psychosexual characteristics of men and women exposed prenatally to diethylstilbestrol. Epidemiology, 14, 155–160. doi:10.1097/01. EDE.0000039059.38824.B2 Toppari, J., Virtanen, H., Skakkebaek, N.  E., & Main, K.  M. (2006). Environmental effects on hormonal regulation of testicular descent. Journal of Steroid Biochemistry & Molecular Biology, 102, 184–186. doi:10.1016/j.jsbmb.2006.09.020 Tourney, G., & Hatfield, L. M. (1973). Androgen metabolism in schizophrenics, homosexuals, and normal controls. Biological Psychiatry, 6, 23–26. Tsoi, W. F. (1988). The prevalence of transsexualism in Singapore. Acta Psychiatrica Scandinavica, 78, 501–504. Ujike, H., Otani, K., Nakatsuka, M., Ishii, K., Sasaki, A., Oishi, T., . . . Kuroda, S. (2009). Association study of gender identity disorder and sex hormone-related genes. Progress in NeuroPsychopharmacology & Biological Psychiatry, 33, 1241–1244. doi:10.1016/j.pnpbp.2009.07.008 Vilain, E. (2000). Genetics of sexual development. Annual Review of Sex Research, 11, 1–25. doi:10.1080/10532528.2000.10559783 Wallien, M.  S.  C., Zucker, K.  J., Steensma, T.  D., & CohenKettenis, P. T. (2008). 2D:4D finger-length ratios in children and adults with gender identity disorder. Hormones & Behavior, 54, 450–454. doi:10.1016/j.yhbeh.2008.05.002 Ward, I. L. (1972). Prenatal stress feminizes and demasculinizes the behavior of males. Science, 175(4017), 82–84. doi:10.1126/ science.175.4017.82 Waring, R. H., & Harris, R. M. (2011). Endocrine disrupters—a threat to women’s health? Maturitas, 68, 111–115. doi:10.1016/j. maturitas.2010.10.008 Williams, T.  J., Pepitone, M.  E., Christensen, S.  E., Cooke, B.  M., Huberman, A.  D., Breedlove, N.  J., . . . Breedlove, S.  M. (2000). Finger-length ratios and sexual orientation. Nature, 404, 455–456. doi:10.1038/35006555 Wolf, U. (1998). The serologically detected H-Y antigen revisited. Cytogenetics & Cell Genetics, 80, 232–235. Wood, P.  B., & Bartkowski, J.  P. (2004). Attribution style and public policy attitudes toward gay rights. Social Science Quarterly, 85, 58–74. doi:10.1111/j.0038-4941.2004.08501005.x Wyatt, T. D. (2015). The search for human pheromones: The lost decades and the necessity of returning to first principles. Proceedings of the Royal Society B: Biological Sciences, 282, 20142994. doi:10.1098/rspb.2014.2994 Zhou, J.  N., Hofman, M.  A., Gooren, L.  J., & Swaab, D.  F. (1995). A sex difference in the human brain and its relation to transsexuality. Nature, 378, 68–70. doi:10.1038/378068a0 Zucker, K. J., Blanchard, R., Kim, T-K., Pae, C-U., & Lee, C. (2006). Birth order and sibling ration in homosexual transsexual South Korean men: Effects of the male-preference stopping rule. Psychiatric & Clinical Neurosciences, 61, 529–533. doi:10.1111/j.1440-1819.2007.01703.x Zucker, K. J., Bradley, S. J., Oliver, G., Blake, J., Fleming, S., & Hood, J. (1996). Psychosexual development of women with congenital adrenal hyperplasia. Hormones & Behavior, 30, 300–318. doi:10.1006/hbeh.1996.0038

Hormones, Sexual Orientation, and Gender Identit y

CH A PT E R

13

Intersexual and Intrasexual Competition and Their Relation to Jealousy

Abraham P. Buunk, Karlijn Massar, Pieternel Dijkstra, and Ana Maria Fernandez

Abstract This chapter discusses sex differences in intersexual competition and describes particularly the ­consequences of such competition for conflict between the sexes, as well as for sex differences in mate guarding and, relatedly, in the types of infidelity that evoke jealousy, including online infidelity. It also discusses individual differences in jealousy as related to attachment styles and describes the effects of height, hormones, and the menstrual cycle on jealousy. Next, the chapter moves on to intrasexual competition and discusses, among other topics, intrasexual competition among men and among women, the role of sex differences in rival characteristics in evoking jealousy, the role of attachment styles and hormones, and individual differences in intrasexual competitiveness. Keywords:  intrasexual competition, intersexual competition, women, jealousy, infidelity, hormones

Intrasexual competition refers to rivalry with same-sex others that is ultimately driven by the motive to obtain and maintain access to mates. In his theorizing on sexual selection, Darwin (1871) suggested that intrasexual competition led to important ­adaptations to compete effectively with same-sex conspecifics to attract mates. It has often been noted that in most species males can sire offspring with a single sexual act, whereas females need to invest in the gestation and birth of her offspring and, after birth, in lactation and childcare. As a consequence, according to Trivers’s (1972) well-known Parental Investment Theory, females are a scarce resource over which males compete (Andersson, 1994). Therefore, in most species males usually engage in quite fierce competition with other males over the access to ­females, whereas females show fewer signs of intrasexual competition. However, even in mammals where males invest little in their offspring, female competition for resources and mates is widespread (Stockley & Bro-Jørgensen, 2011). Because overall, human males also need—at least compared to other species—to invest heavily in their offspring, human

females are also competing with same-sex conspecifics to attract and keep mates who have the willingness and potential to invest in offspring (e.g., Campbell, 2013). Nevertheless, despite the fact that human males do invest heavily in their offspring after birth (Geary, 2000), according to Trivers’s (1972) Parental Investment Theory, in part the reproductive interests of males and females do not overlap, resulting in a considerable degree of intersexual competition. Due to their higher physiological ­investments and the costs associated with these investments, both a woman and her offspring would benefit most from a long-term, highly investing male, whereas for men, short-term mating might yield more reproductive benefits than for women, even when they are already investing in offspring in a long-term relationship. Indeed, numerous studies have shown that, compared to women, men report a greater desire for short-term relationships (e.g., Schmitt, Shackelford, & Buss, 2001), have more permissive attitudes toward casual sex (Petersen & Hyde, 2010), are more interested in short-term 215

extra-dyadic sexual relationships (Buunk,  1980), and report desiring up to three times as many ­lifetime sexual partners (Buss & Schmitt,  1993; Schmitt, 2003). In a now-famous study, R. D. Clark and Hatfield (1989) showed that 75 percent of the men, but 0 percent of the women, approached by an attractive stranger of the opposite sex consented to a request to have sex that evening— clearly illustrating the possible degree of intersexual conflict. In the present chapter, we first discuss intersexual competition and describe particularly the consequences of such competition for sex differences in mate guarding and, relatedly, in the types of infidelity that evoke jealousy, including online infidelity. We also pay attention to individual differences in jealousy as related to attachment styles, and to the effects of height, hormones, and the menstrual cycle on jealousy. Next, we move on to intrasexual competition and discuss, among others, intrasexual competition among men and among women, the role of sex differences in rival characteristics in evoking jealousy, the role of ­attachment styles and hormones, and individual differences in intrasexual competition.

Intersexual Competition

Conflict between the sexes. The pursuit of qualitatively different mating strategies and the behaviors that follow from this may lead to considerable conflict between the sexes—after all, women might not be willing to consent to men’s desires for casual sexual relationships, and men may not always give in to women’s desire for commitment. Women will  more  often withhold sex—since this is the ­resource men value most—until they can be sure of a man’s commitment, whereas men will more often deceive women about their commitment to obtain sexual access (e.g., Haselton, Buss, Oubaid, & Angleitner, 2005). Buss (1989) focused on the specific emotions evoked by these kinds of conflict and showed that women were most likely to be upset and ­angered by men’s greater sexual assertiveness and persistence. Men, on the other hand, were more likely to be angered and upset by women’s tendency to require an emotional commitment before engaging in sexual behaviors. More recent studies have shown that men, particularly those who are focused on short-term mating and who rate themselves as attractive, tend to overperceive women’s sexual i­nterest (i.e., the overperception bias; Haselton & Buss, 2000). Conversely, women tend to underestimate men’s desire for a committed 216

relationship (i.e., the commitment skepticism bias). Conflicts over sexual access—men’s greater desire and women’s greater hesitation—can cause anger and frustration in men (Shackelford & Goetz, 2004), which in turn increases the likelihood of sexual ­coercion. Jealousy-evoking events. The notion of intersexual competition has important implications for sex differences in the events that evoke jealousy in more or less committed relationships. Overall, men will be more interested than women in engaging in uncommitted extra-dyadic sex as this may enhance their reproductive success. However, such behavior when engaged in by their female partner would be particularly threatening for men. Because from an evolutionary perspective men have over the course of evolution faced the problem of paternity confidence, and women of securing the partner’s investment of resources, one would predict that male jealousy would be specifically focused on the sexual aspects of the partner’s extra-dyadic activities, and female jealousy on the emotional involvement of the partner with the rival (e.g., Buss, Shackelford, Choe, Buunk, & Dijkstra, 2000; Bjorklund & Shackelford, 1999). For men, any act of intercourse of his partner with a third person is a potential threat to his reproductive success. In contrast, for women, an act of intercourse by her partner may especially, or only, be a threat when the investment of the partner in the relationship is in jeopardy. Indeed, when the partner has been unfaithful a number of times while maintaining his commitment, a woman may, under some conditions, adapt to her partner’s infidelity (see, e.g., Buunk, 1995). To test the gender difference predicted by e­ volutionary theorizing, Buss, Larsen, Westen, and Semmelroth (1992) developed a research paradigm in which participants were presented with dilemmas, in which they had to choose a partner’s sexual unfaithfulness or a partner’s emotional unfaithfulness as the most upsetting event. Buss et  al. did indeed find that more men than women selected a partner’s sexual infidelity as the most upsetting event, whereas more women than men reported a partner’s emotional infidelity as the most upsetting event. This pattern of results has since then been replicated many times across cultures, and for both conventional and online infidelity (e.g., Buss et al., 1999; Buunk, Angleitner, Oubaid, & Buss,  1996; Dunn & McLean,  2015; Groothof, Dijkstra, & Barelds, 2009). Schützwohl and his colleagues conducted a series of studies that shed a more specific

Intersexual and Intrasexual Competition

light on the cognitive processes that take place before, during, and following responses to a partner’s infidelity, and found that these cognitive processes were in line with the sex difference in distress reported by Buss et al. (1992). First, Schützwohl (2006) found that, even before any incident of infidelity has occurred, women more than men are preoccupied with thoughts about a partner’s emotional infidelity, whereas men more than women are more occupied with thoughts about a partner becoming sexually unfaithful. Once suspecting a partner’s ­ ­infidelity, men more than women tend to seek information that may indicate a partner’s sexual ­infidelity, whereas women more than men tend to seek information that is indicative of emotional infidelity. Once individuals notice they start to feel  jealous, for  feelings of jealousy to become ­intolerable, men need significantly fewer cues to sexual infidelity than women, whereas women need significantly fewer cues to emotional infidelity than men (Schützwohl, 2005). Finally, following an ­incident of infidelity, men and women differ in what they remember about the incident. When unexpectedly asked about an imaginary story about their partners’ infidelity participants listened to a  week ago, Schützwohl and Koch (2004) found that women recalled significantly more cues to emotional than to sexual infidelity, whereas men recalled significantly more cues to sexual than to emotional infidelity. Remarkably, in Sweden and Norway, both among the most egalitarian societies in the world, strong sex differences in response to emotional and sexual infidelity have been found (Bendixen, Kennair, & Buss, 2015; Walum, Larsson, Westberg, Lichtenstein, & Magnusson, 2013). Contradictory to what a learning perspective would predict— that is, an attenuation of the differences between men and women in the domain of jealousy—an evolutionary perspective holds that in these societies sex differences might be expressed to a larger degree due to expected larger paternal investments. Bendixen, Kennair, and Buss (2015) argued that as men’s investment in childrearing increases, for women the costs of losing these investments to another woman and for men the costs of paternal uncertainty also increase. It must be noted that the findings with the paradigm developed by Buss et al. (1999) have been questioned (e.g., DeSteno & Salovey,  1996; Harris, 2003), assuming that most gender differences would disappear when actual infidelity experiences and ­responses on continuous measures (rather than forced-choice dilemmas)

would be considered. However, a meta-analysis by Sagarin et al. (2012) shows that across 45 independent samples, the gender differences emerge on continuous measures and in response to experienced infidelities. Indeed, sex differences have been found in studies using different methods, such as reaction times (Schützwohl,  2005) or audio records of partner interrogations in the face  of an actual infidelity threat (Kuhle,  2011). Noteworthy is the finding that the sex difference in emotional and sexual ­infidelity generalizes to online infidelity. More specifically, Guadagno and Sagarin (2010) found sex differences both when participants were asked to choose between the classic scenarios of sexual and emotional infidelity, and when they were asked to choose between a scenario in which the partner engaged in cybersex and a scenario in which the partner formed a close emotional attachment with someone else online. The role of conception risk. Nevertheless, when there is no risk of conception—such as when an infidelity between an opposite-sex partner occurs with a same-sex rival—the sex differences in jealousy tend to be attenuated (e.g., Harris, 2002; Sagarin, Becker, Guadagno, Nicastle, & Millevoi,  2003). For instance, the sex difference disappears when men are compared to women who are using hormone-based birth control. The use of hormone-based birth control may change women’s psychological and social behavior in ways that differ from the influence of natural estrogens or progesterone. This is, among other things, reflected in the intensity to which women respond to information about a partner’s infidelity, especially in the case of emotional infidelity. More specifically, Geary, DeSoto, Hoard, Sheldon, and Cooper (2001) found that the majority (70 percent) of women not using hormone-based birth control indicated that the emotional infidelity of their partner was more distressing than their partner’s sexual infidelity, whereas only about one-half (56 percent) of women using a hormone-based birth control responded in the same way. In this latter case, women no longer significantly differed from men in their choice of the type of infidelity they experience as most distressing. When there is no risk of conception, the reproductive stakes of ­infidelity are much lower than when sex is highly likely to result in pregnancy and to have long-term consequences. Especially under these last conditions—high risk of pregnancy in the context of infidelity—the universal sex difference seems to emerge. Illustrative

Buunk, Massar, Dijkstra, and Fernande z

217

in this context is a study by Scelza (2014), who studied the responses to emotional and sexual infidelity among the Himba, a small-scale, natural fertility population in the northwest of Namibia where rates of infidelity are high and, at the time of study, only 30 percent of inhabitants had ever heard of a modern method of contraception, and no one actually used it. Scelza found that although Himba men and women both reported being significantly more distressed about the possibility of a partner’s sexual infidelity than a partner’s emotional infidelity, men were more likely to choose sexual infidelity as more distressing than women, whereas women were more likely to choose emotional infidelity as more distressing than men. Also, other factors may moderate the gender difference in emotional versus sexual jealousy. For example, the gender difference has been found to be attenuated when controlling for variables such as personal ­experiences as a victim or perpetrator of infidelity (e.g., Bendixen, Kennair, Ringheim, et al., 2015; Tagler, 2010), attachment style (Burchell & Ward, 2011; Levy & Kelly,  2009), and participant age (IJzerman et al., 2014). More specifically, IJzerman et al. (2014) found the gender difference in samples consisting of individuals in their early 20s, but not in samples of individuals over 50. Likewise, Tagler (2010) found that, in contrast to individuals who had not experienced adultery in their current or past relationships, among men and women who did experience adultery, no gender difference emerged in the choice of most upsetting infidelity scenario. In a similar vein, whereas the gender difference among individuals with a secure attachment style is relatively small, among individuals with a dismissing attachment style it is relatively large (e.g., Levy & Kelly, 2009). Mate guarding. The male focus on preventing sexual contact between their partners and other males is quite manifest in the extreme forms of mate guarding, in which males in many cultures may engage. In his now-classic review, Murdock (1967) noted that only 4 out of 849 societies did not show any sign of male mate guarding (i.e., keeping close tabs on their mates, sometimes even when she is urinating or defecating). A recurrent finding is the relatively high incidence of physical violence that is used by men as a means of mate guarding or as an expression of jealousy, both in response to their partner and in response to same-sex rivals, in the latter case turning into intrasexual competition (Graham & Wells, 2001; M. Wilson & Daly, 1993). 218

Likewise, it has often been suggested that jealousy as a motive in homicide also seems more common among male than among female perpetrators (e.g., Stöckl & DeVries, 2013). However, after reviewing the literature, Harris (2003) concluded that the existing data on homicide offers no support for the belief that male murderers are more often motivated by jealousy than female murderers. Moreover, several conditions may moderate the occurrence of homicide by men due to jealousy. For instance, Mize, Shackelford, and Shackelford (2009) found that men are more likely to kill their partners by beating when the relationship is dating or nonmarital cohabiting versus legal marriage. According to these authors, a lack of formal commitment in the form of a marriage certificate may fuel feelings of jealousy that may give rise to higher levels of aggression that, consequently, are more likely to result in murder. The incidence of male partner violence due to mate guarding and jealousy may also differ as a function of culture. There are considerable cultural differences in the frequency and intensity of mate guarding. For example, in cultures where parents control the mate choice of their offspring, mate guarding is much more common (Buunk & Castro Solano,  2012), and honor cultures often establish strict norms for female chastity and virginity, with a transgression of these norms being perceived as a threat to a family or husband’s honor. Illustrative is the Arab expression that a man’s honor “lies ­between the legs of a woman.” As a consequence, especially in cultures where honor is a salient organizing theme, such as most Arab countries (Vandello & Cohen, 2003), physical violence against women is seen as an acceptable and appropriate means of restoring male reputation and identity when damaged by a female’s (potential) infidelity. In addition to physical aggression, veiling is used in many Arab honor cultures as a method of mate guarding by which men try to reduce the risk of a partner’s infidelity. Recently, Pazhoohi and Hosseinchari (2014) studied the effectiveness of different types of veiling on ratings of female attractiveness. This study examined three types of veils that cover up the female body to different degrees: (1) a black hijab headscarf that covers up the hair, neck, and shoulders with tight clothing covering bodily curves; (2) a head scarf plus official clothing covering bodily curves; and (3) the chador, a traditional wide body-length fabric that covers the whole body except the face. Results

Intersexual and Intrasexual Competition

showed that Muslim men were more motivated to  view women exhibiting less veiling and rated them more attractive than those women whose bodily curves were less apparent, and that, thus, veiling seems to constitute an effective mate-guarding tactic for men. Although often neglected in the literature, women may also engage in mate guarding, and not only to retain a mate. In our ancestral past, and even in the not-so-distant past, sexually transmitted infections (STIs) constituted a major threat, more for women than for men, because such diseases might lead to infertility (e.g., Mackey & Immerman, 2001). Currently, because extra-dyadic sex often happens without a condom (e.g., Dew, Brubaker, & Hays, 2006), individuals engaging in extra-dyadic sex run the risk of getting infected with STIs and of infecting their partner (e.g., Buunk & Bakker, 1997; Pulerwitz, Izazola-Licea, & Gortmaker, 2001), which is particularly risky for women of adulterous spouses because STIs are much more easily transmitted from males to females than vice versa. In addition, when infected with an STI and others learn about this, women run the risk of being perceived as immoral and promiscuous (Neal, Lichtenstein, & Brodsky, 2010), decreasing their attractiveness as a partner. Although men also run the risk of being stigmatized when infected with an STI, women are more severely stigmatized than men, even when their partner is responsible for infecting them (Lichtenstein, Hook, & Sharma, 2005). Mate retention. Men and women differ also in the types of tactics they employ to retain their mate when jealous. Among others, Buss and Shackelford (1997) showed that mate retention parallels mate preferences, such that men will attempt to retain their mate by engaging in resource displays like buying expensive gifts for their partner, whereas women are more likely to enhance their appearance. Moreover, men more than women try to keep their partner away from other men and will verbally and physically threaten their rivals (thus showing direct intrasexual competition), whereas women are more likely to flirt with other men to evoke jealousy in their partners. More recently, research (e.g., Miner, Starratt, & Shackelford, & Starratt,  2009) has shown that a man’s mate value is associated with specific types of mate retention tactics: Men with a high mate value are more likely to use mate retention behaviors that focus on enticing a woman to stay invested in the current relationship (e.g., by

buying gifts), whereas men with a low mate value are more likely to engage in behaviors that function by inflicting or threatening to inflict costs on a woman for not remaining invested in the current relationship (e.g., by isolating her). Mate retention intensifies when the costs of losing a partner increase. For instance, Buss and Shackelford (1997) found that both men married to attractive women and women married to highincome men reported more frequent and intense mate guarding. Also, men’s mate retention efforts increase the more time they spend apart from their partners (Shackelford, Goetz, McKibbin, & Starratt, 2007) and when their partner is near ovulation—a time when a female infidelity would be most costly (Gangestad, Thornhill, & Garver,  2002). Across animal species, the use of aggression can be seen as a form of punishment that deters the targeted individual from repeating a behavior that conflicts with the interests of the aggressor (Clutton-Brock & Parker, 1995). Similarly, among humans, sexual coercion can be seen as an extreme mate retention tactic, aimed at preventing the partner from committing a sexual infidelity, and thereby reducing the risk of cuckoldry. Sexual coercion as the result of sexual jealousy can take on subtle forms (e.g., threatening to leave the relationship or withholding gifts if a partner does not consent to sexual intercourse) but may escalate to partner-directed violence and forcibly raping one’s partner (Goetz, Shackelford, Romero, Kaighobadi, & Miner, 2008). Influence of life history and attachment. In life history theory, a distinction is made between two strategies: an r- versus K-strategy (Roff, 1992; Stearns, 1992). The r-strategy is characterized by fast development: short growth, early sexual maturation, having many offspring of low quality, and an overall short lifespan. The K-strategy, in contrast, is characterized by slow development: a long growth span, late sexual maturation, few offspring of high quality, and an overall long lifespan. A disrupted attachment history would likely foster an r-strategy, accompanied by relatively high levels of jealousy. However, few studies have examined the relationship between adult jealousy and attachment history in terms of objective indicators, such as separation from a parent during childhood. Using the scales developed by Buunk (1997), Van Brummen-Girigori, Buunk, and Dijkstra (2016) found that on Curaçao, a Caribbean island where relatively many children grow up without a father in the home, women who were abandoned by their father during childhood

Buunk, Massar, Dijkstra, and Fernande z

219

reported significantly more anxious and possessive jealousy (an index of mate guarding) than females who grew up in the presence of their father. Many other studies have shown that individuals with an insecure attachment style are more jealous than individuals with a secure attachment style (e.g., ­ M.  J.  Miller, Denes, Diaz, & Buck,  2014), independent of the influence of personality characteristics such as self-esteem and neuroticism (Buunk, 1997). In particular, individuals with an anxious-ambivalent attachment style have been found to experience greater jealousy in response to a Facebook wall post sent by a rival for their partner’s attention (e.g., Fleuriet, Cole, & Guerrero, 2014; Marazziti et al., 2010), possibly because communication on Facebook (e.g., a winking emoticon) is often quite ambiguous in terms of a partner finding someone else appealing. This ambiguity may trigger jealousy especially in those already anxious about their relationship. As a consequence, such individuals tend to closely monitor their partner’s activities on Facebook (e.g., Muise, Christofides, & Desmarais, 2014; Marshall, Bejanyan, Di Castro, & Lee, 2013). As an anxiousambivalent attachment style implies a “clinging” to the relationship out of fear of losing the partner, the link between this style and jealousy seems self-­ evident. Less self-evident is the finding that avoidantly attached individuals are often relatively jealous. A possible explanation is that such individuals are actually quite dependent on their partner but feel that they are not meeting the needs of their partner by their distant attitude, and are therefore concerned with losing their partner. Influence of physical characteristics. Not only psychological but also physical characteristics may be ­related to jealous mate guarding. The reproductive advantages of height for males are apparent in the female preference for taller males (Kurzban & Weeden,  2005; Pawlowski,  2003; Shepperd & Strathman, 1989). Indeed, taller men receive more replies to dating announcements (Pawlowski & Koziel, 2002), have more physically attractive girlfriends (Feingold, 1982), and have more reproductive success (Mueller & Mazur, 2001; Nettle, 2002; Pawlowski, Dunbar, & Lipowicz, 2000). Given that height is highly heritable (one recent estimate from a study using the Danish Twin Registry found heritability coefficients of 0.69 for men and 0.81 for women; Schousboe et al., 2004), females choosing tall males are more likely to have tall male offspring, who in turn would be preferred by females. Male height has been found to be correlated with physical 220

health, as well as with morphological symmetry (Manning, 1995; Silventoinen, Lahelma, & Rahkonen, 1999). Tall men and women of medium height have the highest mate value and are considered the most attractive. For example, in speed dating experiments tall men and women of medium height receive the most positive responses from the opposite sex (Stulp, Buunk, Kurzban, & Verhulst, 2013). It seems self-evident that individuals with a high mate value will feel they have less to fear from rivals. Indeed, Buunk, Park, Zurriaga, Klavina, and Massar (2008) found that as men were taller, they reported on a single-item, global measure less jealousy, whereas average-height women reported lower levels of jealousy than short and tall women (see also Buunk, Pollet, Klavina, Figueredo, & Dijkstra, 2009). Not only are objective differences in height related to jealousy, but also subjective ­beliefs about one’s body: When individuals believe  their body is less attractive, they tend to be more jealous (e.g., Ambwani & Strauss, 2007). Although the effects of height on jealousy may be mediated by psychological processes, there is also clear evidence for a more direct biological basis of jealousy. First, twin research suggests that emotional and sexual jealousy have a strong genetic component (Walum et al., 2013). Second, several studies have shown that among women, jealousy is associated with hormone levels. Geary et al. (2001) showed that among women, jealousy correlated with estrogen concentration assessed in the second week of the cycle. Cobey et al. (2012) found with a withinsubjects design that in both single and partnered women, jealousy varied as a function of menstrual phase, with higher levels of jealousy reported when women were fertile than when they were nonfertile. An explanation for this is that, given the importance of male investment, protection, and provisioning during and after pregnancy (for a review, see Geary, 2005a), the risk of losing, or not receiving, the necessary investment from one’s mate would be especially threatening for women when they are fertile. Consequently, women will be especially keen at preventing the involvement of their partner with another woman. The findings by Cobey et al. (2012) were replicated in the Afro-Caribbean population of Curaçao, where it was found that especially preventive or possessive jealousy (an index of mate guarding) was higher among women who were fertile. In addition, this type of jealousy was higher the later the age of the first menarche, also suggesting hormonal effects (Buunk & Van Brummen-Girigori, 2016).

Intersexual and Intrasexual Competition

Jealousy and Intrasexual Competition

Jealousy in response to a rival. Because jealousy often implies the explicit presence of a rival when two ­individuals are vying for the attention of the same other, jealousy has a strong component of intrasexual competition. Arnocky, Sunderani, Miller, and Vaillancourt (2012) found indeed that among women, intrasexual competition is enhanced among those with relatively high jealousy levels, who made more appearance comparisons and engaged more often in competitor derogation than their less jealous peers. When looking at the themes that individuals from different cultures mention when asked to provide a description of a jealousy-evoking event, four common dominant themes emerge: (1) fear of or actual infidelity, (2) violated expectations concerning a partner’s time and commitment, (3) a partner paying attention to a rival through social media, and (4) loss of self-esteem due to a partner paying attention to a rival (Zandbergen & Brown, 2015). In line with this, a study in the Netherlands (Buunk, 1981) found that, although this was by far not the most important concern, 20 percent of the individuals who had experienced actual infidelity via their partner having an extra-dyadic affair said they felt threatened because they regarded the rival in certain respects as better than themselves. Thus, across different cultures, the elicitors of jealousy all seem to reflect in part feelings of competition with a rival. Sex differences in the importance of rival characteristics. Overall, a rival who possesses qualities that are believed to be important to the opposite sex or to one’s partner tends to evoke more feelings of jealousy than a rival who does not possess those qualities (e.g., DeSteno & Salovey, 1996; Dijkstra & Buunk, 1998). Men and women report comparable amounts of jealousy as their rivals possess more self-relevant attributes, such as intelligence, popularity, athleticism, and certain professional skills (e.g., DeSteno & Salovey,  1996; Rustemeyer & Wilbert,  2001). However, given the sex differences in mate preferences, with physical attractiveness more valued by men, and status- and dominance-related characteristics more valued by women, from an evolutionary psychological perspective one would expect women to feel more jealous than men when their rival surpasses them on bodily and facial attractiveness, and men to feel more jealous than women when their rival possesses status- and dominance-related characteristics. Based on a series of open interviews, Dijkstra and Buunk (2002) constructed a 56-item

inventory of threatening rival characteristics, composed of five factors: social dominance, physical attractiveness, seductive behavior, physical dominance, and social status. The main finding was that jealousy in men was, more than in women, evoked by the rival’s social and physical dominance, whereas jealousy in women was, more than in men, evoked more by the rival’s physical attractiveness. These findings were obtained in Dutch college samples and were cross-validated in Dutch community samples. In addition, Dijkstra and Buunk (1998) presented participants with a scenario in which their partner was flirting with the opposite-sex individual. Next, participants received a profile of the rival who was either high or low in physical attractiveness and either high or low in dominance. Jealousy in men was particularly influenced by the rival’s social dominance, whereas jealousy in women was particularly influenced by the rival’s physical attractiveness. Subsequent research has shown that such differences occur especially in the case of emotional rather than sexual jealousy (Buunk & Dijkstra, 2003). Moreover, the assessment of the threat of a rival seems such a basic process that it may be perceived outside cognitive awareness. For example, in a study focusing on rivals’ bodily features, Massar and Buunk (2009) used a subliminal priming procedure to expose participants to silhouettes of bodies that were either attractive (a waist-to-hip ratio [WHR] of 0.7 for women and a shoulder-tohip ratio [SHR] of 1.4 for men) or unattractive (a WHR of 0.9 for women and an SHR of 1.2 for men). The results showed that even though participants indicated not being aware of the content of the primes, their jealousy was influenced by the primes: Both men and women responded with the most jealousy after being exposed to the attractive body shapes. Similar findings were obtained in a study in which words relating to attractiveness or social status were used as subliminal primes (Massar, Buunk, & Dechesne, 2009), such as pretty, slender, success, and money. The results from this study revealed the predicted sex differences in the characteristics that evoke jealousy. That is, whereas women’s jealousy increased after exposure to the attractiveness words but not after priming with social dominance words, males showed the reverse pattern and reported increased jealousy after being primed with words relating to social dominance but not after priming with attractiveness words. Finally, a study using photographs of an attractive or an unattractive female (Massar & Buunk,  2010) showed that exposure to a female face outside their awareness

Buunk, Massar, Dijkstra, and Fernande z

221

affected women’s jealousy such that they reported significantly higher levels of jealousy after exposure to an attractive face than after priming with an unattractive face. Cross-cultural similarities and differences. Similar sex differences in the threatening nature of rival characteristics as in the Netherlands have been observed in  studies in other cultures, including the United States, South Korea (Buss et al., 2000), Spain, and Argentina (Buunk, Solano, Zurriaga, & González, 2011). However, although a factor analysis in a sample of young people in Iraqi Kurdistan (Buunk & Dijkstra, 2015) showed exactly the same five dimensions or rival characteristics as found by Dijkstra and Buunk (2002), in contrast to the Dutch, the Kurdish men and women did not differ in which characteristics evoked the most jealousy. A possible explanation for this lack of gender differences is that the overall high level of jealousy was much higher, and that this may have overruled the effects of specific rival characteristics. Nevertheless, at least in Western cultures, sex differences in the rival characteristics occur not only in intimate relationships but also in relationships at work with one’s supervisor in which a rival interfered. A study in Argentina showed that among men, physical dominance of a same-sex rival evoked more jealousy than among females, and among females physical attractiveness evoked more jealousy than among males (Buunk, Aan ‘t Goor, & Castro Solano, 2010). This occurred only when the supervisor was of one’s own sex. These findings suggest that intrasexual competition has a dynamic of its own and is induced more by the presence of same-sex others than by the presence of opposite-sex others (cf. Campbell, 2002; Geary, 1998). Hormonal and physical characteristics. Although there is thus far no evidence that testosterone is associated with jealousy, jealousy in response to particular types of rivals seems to be clearly associated with prenatal exposure to male hormones. Such exposure affects the second-to-fourth-digit ratio (2D:4D), with masculinity associated with a lower and femininity with a higher ratio. Park, Wieling, Buunk, and Massar (2008) found that men with more feminine 2D:4D ratios were most jealous in response to physically and socially dominant rivals, whereas women with more masculine 2D:4D ratios were most jealous in response to physically attractive rivals. In a related vein, a study by Buunk et al. (2008) in Spain showed that height was differently related among men and women to the threatening nature of rival characteristics. As males were shorter, they 222

responded with more jealousy to physically dominant, socially dominant, and physically attractive rivals. Although females responded with more jealousy in response to physically attractive, physically dominant, and high-social-status rivals, averageheight women (i.e., those most preferred by men) tended to be less jealous of physically attractive rivals but more jealous of rivals with “masculine” characteristics of physical dominance and social status. An explanation for this finding is that women may be least jealous of rivals who have features they have themselves and most jealous of rivals who have features they do not have themselves. As women of average height are, as noted above, preferred by men, in part because they tend to be more fertile and healthy (Nettle,  2002; Silventoinen et al., 1999), they would be less jealous of women with features signaling fertility and health such as physical attractiveness but more jealous of women possessing “masculine” features.

Male Intrasexual Competition

Intrasexual competition occurs among males and females to establish their position in a hierarchy. Dominance hierarchies revolve around relative rather than absolute positions, and individuals are more sensitive about getting ahead of another rather than about achieving an absolute position (Buunk & Ybema, 1997). S. E. Hill and Buss (2006) report studies that show that men and women possess a positional bias, making them attend to the positional rather than to the absolute value of resources that are known to affect survival or reproduction, and to personal attributes that affect others’ abilities to acquire such resources. When choosing between having an absolutely larger income or an income that was absolutely less but larger than one’s rivals’ incomes, both men and women chose the greater positional income. Moreover, the positional bias seems to be sex differentiated, evident in the finding that women preferred to be less attractive in an absolute sense but more attractive than their rivals (e.g., scoring a 5 when rivals score a 3) over being more attractive in an absolute sense but less attractive than their rivals (e.g., scoring a 7 when rivals score a 9). Among humans, male–male intrasexual competition is a quite complex and multifaceted phenomenon. Overall, men compete with other ­ men for access to reproductive resources, including resources such as political influence and social status that can be converted into reproductive opportunities, either because these are directly ­

Intersexual and Intrasexual Competition

a­ ttractive to females or because these help conquer rival males (Barkow, 1999; Sidanius & Pratto, 1999). In preindustrial societies in which male–male competition has been studied, it has consistently been found that a man’s status is directly related to his reproductive success (Betzig,  1982,  1986). Even in contemporary Western society, high-income men have more biological children than low-income men, whereas among women the opposite is true (Hopcroft, 2005; Nettle & Pollet, 2008). Costly signaling. One type of intrasexual competition is quite indirect and consists of showing off to females those characteristics that may signal good genetic quality. This type of intrasexual competition is driven by intersexual selection (i.e., the selection by females for male traits that are indicators of good genes). For instance, in birds of paradise, females’ preference for male ornamentation has resulted in intrasexual competition between males for the most attractive plumage (Andersson, 1994). In a similar vein, one way in which intrasexual competition among males can take place is via ritualized displays. As Veblen noted in 1899, conspicuous consumption and conspicuous leisure might be ways of engaging in status competition. Saad and Vongas (2009) found that men’s testosterone levels are responsive to fluctuations in their status as triggered by acts of conspicuous consumption. That is, male testosterone levels ­increased after driving an expensive sports car, whereas they decreased after driving an old family sedan, and male testosterone levels increased when men’s social status was threatened by the wealth displays of a male rival in the presence of a female. This study suggests that showing off by means of conspicuous consumption is an evolved mechanism for responding to intrasexual challenges. Building on Veblen (1934 [1899]), G.  F.  Miller (2000) suggested that conspicuous consumption could be seen as a handicap signal. Handicap signaling refers to the evolution of an honest signal, which cannot be copied because it is costly to produce (Zahavi & Zahavi,  1997). In this way, for example, human males could show off and signal to other men and potential mates: “I can afford all this.” In relation to this, Saad and Gill (2001) conducted a two-person ultimatum game in which one was the allocator and the other the recipient, and the allocator had to split a given sum of money with the recipient. The recipient could either accept or reject the offer. If accepted, both players received their respective splits, but if rejected, neither of them got anything. The results showed

that men made more generous offers when pitted against a woman as opposed to a man. Women, on the other hand, made equal offers independently of the sex of the recipient. In the same vein, Buunk and Massar (2014) had participants engage in a series of decomposed social games in which they had to divide resources between themselves and either a same-sex or an opposite-sex other. Males behaved more competitively toward another man than toward a woman, whereas women did not distinguish between men and women in their degree of competitiveness. At the same time, men behaved more prosocially toward women than women did toward men. In addition, after dividing resources between themselves and another man in the decomposed game task, men showed higher levels of intrasexual competitiveness than after dividing resources between themselves and a woman. There is abundant evidence that male costly signaling via conspicuous consumption is common in traditional societies (Bird, Smith, & Bird, 2000; Hawkes & Bliege Bird,  2002). Some have even argued that men’s hunting has evolved, in part, as a signaling strategy (Hawkes & Bliege Bird,  2002). Hunting is not necessarily an effective activity. Although it may result in obtaining food and warding off starvation, often the time spent hunting could in many cases be better allocated to gathering food. Hunting does, however, appear to  provide some cues about an individual’s quality to relevant audiences: good hunters are subsequently preferred as mates or allies. Also, in modern society men use conspicuous consumption as a strategy of intrasexual competition (G.  F.  Miller,  2009). Lycett and Dunbar (2000) demonstrated this by showing that mobile phones could be construed as lekking devices (i.e., as in grouse; e.g., Gibson & Bradbury, 1985), whereby human males aggregate and conspicuously display their features to attract females. In this study, males were more ­inclined to conspicuously display their mobile phones as the composition of their group became more male biased. Thus, men may aggregate in groups and compete via ritualized displays, such as by showing off their mobile phones. Nevertheless, intrasexual competition among males may become more salient and prevalent in the presence of women, which tends to make men more aware of their status, and more eager to ­demonstrate that they can beat other men. For ­example, an experiment showed that men increased their cooperation in an economic game when observed by women (Iredale, Van Vugt, & Dunbar,

Buunk, Massar, Dijkstra, and Fernande z

223

2008). Thus, men may exhibit competitive altruism: they compete by being generous and forego individual benefits (Van Vugt, Roberts, & Hardy, 2007). Behaving altruistically may improve one’s reputation and status; others often attribute charisma to those who sacrifice their own needs to those of others or the group (De Cremer & Van Knippenberg, 2004). Physical dominance and aggression. Especially among young males with few resources, intrasexual competition is to a large extent driven by direct physical competition (Barber,  1995; K.  Hill & Hurtado, 1996; Kemper, 1990). Males may engage in fierce threats and fights to attain a high status in the group, and to prevent other males from access to females, and as such maintain exclusive sexual access to females (Andersson, 1994). Among primates such as baboons, such conflicts are common (Walters & Seyfarth,  1987; Wrangham & Peterson, 1996). It requires individuals to be healthy, physically strong, and aggressive (i.e., exactly those characteristics that are signaled by an athletic body; Frederick & Haselton, 2007). The strategy of physical dominance is effective in p ­ articular for younger men: Whereas they do not possess a high status yet in a larger social context, they are at their peak with regard to health and fitness (Kemper, 1990). The degree to which men are able to engage in the strategy of physical dominance is related to their physical appearance. Athletic men have been found to be relatively competitive (Quinn & Wilson, 1989) and to like physical adventure, exercise, risk, and car speed relatively more than nonathletes (Child,  1950; Quinn & Wilson, 1989; Sheldon & Stevens, 1942). Compared to men with a less athletic body build, athletic men show lower anxiety and may therefore be high in sensation seeking, and have been found to engage often in impulsive, aggressive, antisocial, and disorderly behavior (see Domey, Duckworth, & Morandi, 1964; Verdonck & Walker, 1976). The male face also signals the degree of physical dominance (Zuckerman, 1986). For instance, men who are high in testosterone—an important hormone regulating aggressive behavior—have larger jaws and a more prominent brow ridge than other men, and as a result are perceived as more masculine and dominant (Penton-Voak & Chen, 2004). According to Frederick and Haselton (2007), a man’s body morphology signals his fitness and the presence of genes that could potentially increase his reproductive success (see also Geary, 2005a). An athletic build characterized by a high degree of ­muscularity and broad shoulders, for i­nstance, could 224

demonstrate that a male is in a good condition. Bar, Neta, and Linz (2006) showed that first impressions of males, especially those that are perceived as a threat, are usually made within the first 39 milliseconds, solely on the basis of visual information. Research using the zero-­ acquaintance paradigm, in which participants are asked to judge personality attributes of people based on short, silent video clips of often no more than 30 seconds, shows that people are often quite accurate when making judgments about, for instance, someone’s self-esteem, status, and level of altruism (e.g., Yeagley, Morling, & Nelson, 2007). In addition, body-related stereotypes that men with an athletic (or mesomorph) body build are stronger, more sportive, more competitive, more dominant, healthier, and more energetic (Butler, Ryckman, Thornton, & Bouchard, 1993; Lerner & Korn, 1972; Ryckman, Robbins, Kaczor, & Gold, 1989) tend to be relatively accurate. As a result, during intrasexual competition, body-related stereotypes may facilitate men to form fast and relatively accurate impressions of their rivals and, in so doing, may enhance competitive success and  prevent a possible loss of status (Fiske, 1992). More specifically, as in many other species, by ­assessing the physical features and deriving conclusions from these features in other males, men may challenge those who can be beaten so as not to miss out on opportunities to raise their status (or their reproductive success in other ways) that could be available, while it may prevent them from competing with superior males, as that would be a waste of energy and would bring substantial costs. In support of this line of reasoning, Tiedens and Fragale (2003) found that men changed their behavior when being in the same room as a male who, due to his bodily posture, was perceived as dominant. When confronted with such a male, men behaved relatively submissively. Although one might assume that physical dominance would not matter in modern organizations with only white-collar work, there is substantial evidence that it still does. Height, especially, has a greater effect on attaining status in organizations than is often thought, whereby taller men tend to  attain higher positions in organizations. In humans, height is one of the first features that others notice and is associated with status. For instance, one study found that full professors were 0.47 inches taller than associate professors, who were 0.26 inches taller than assistant professors, who were 1.24 inches taller than the average nonacademic (Hensley, 1993). The relationship between height and status also leads individuals to distort

Intersexual and Intrasexual Competition

their perceptions of men’s height and, as a result, to hold—relatively accurate—stereotypes about height (P.  R.  Wilson,  1968; Jackson & Ervin,  1992). Research, for instance, shows that the same male is perceived to be taller as his status increases: when, for instance, a man is described as a student, he is estimated to be about 2.5 inches shorter than when he is described as a professor (P. R. Wilson, 1968). Eminence. However, males compete not only through conspicuous consumption and physical dominance but also through achieving eminence, which refers to the elevated rank that is achieved, more gradually, through socially approved accomplishments, such as education and political career making. It requires individuals to be intelligent and  to invest in intellectual activities, rather than in  their physical prowess. There is some evidence that a lean and relatively weak body in terms of muscularity—a so-called ectomorph body—is positively correlated with cognitive abilities (Case & Paxon, 2006), which may translate into higher wages (Judge & Cable, 2004; Loh, 1993). In addition, ectomorphism has been found to correlate positively with interest in high status and intellectually challenging vocations, such as school superintendent, physician, minister, lawyer, and researcher (Cupcea, 1939; Deabler, Hart, & Willis,  1975; Garn & Gertler, 1950; Tanner, 1954), and negatively with lower status and less intellectually challenging occupations, such as bus driver (Deabler et al., 1975). Stereotypes about ectomorph men seem, at least partially, to parallel actual relationships between the ectomorph body build and personality traits. For instance, ectomorph men are usually perceived— with relative accuracy—as more intelligent and scholarly than men with different body builds (Butler et al., 1993; Ryckman et al., 1989). With age, as men’s physical dominance declines, the strategy of eminence will have greater success (Kemper, 1990; K. Hill & Hurtado, 1996). That is, men relying on the strategy of eminence, although they may not be successful early in life, often reach their peak later in life (Buss, 1994). Therefore, a trade-off with age seems to take place between the strategies of physical dominance and eminence. In contrast to the strategy of physical dominance, the strategy of eminence requires individuals to delay the gratification of needs. This can be placed within the earlier mentioned life history framework (Roff,  1992; Stearns,  1992), with physical dominance reflecting an r-strategy, as it is characterized

by fast development and a short-term strategy, and eminence reflecting a K-strategy, characterized by slow development and a long-term strategy.

Female Intrasexual Competition

Although female sociality and same-sex bonding is a recurring characteristic of group social living that substantially increases fitness (e.g., Cheney, Silk, & Seyfarth, 2012), it also results in a higher rate of samesex reproductive rivals, which affects female mating motivation and intrasexual competition for mates and their resources. This competition may in turn cause group instability and intrasexual aggression. Indeed, rivalry among females in socially monogamous species is a well-documented phenomenon. In many socially living species, such as baboons (Huchard & Cowlishaw, 2011) and bonobos (Hohmann & Fruth, 2003), rates of female–female aggression increase as the proportion of especially fertile females in the group increases, with pregnant females showing the highest rates of aggression. In insects, for example, intrasexual competition extends to precopulatory struggles for accessing a mate, postcopulatory competition for access to and protection of the eggs, and the destruction of rival fertilized cells (Buss, 1988); in mammals a similar pattern of female competition for resources and mates is the norm (Stockley & Bro-Jørgensen, 2011). In the human species, female intrasexual competition over males is also quite prevalent (Campbell, 2004, 2013; Fisher, 2004) and often centers on access to mates, but also concerns status enhancement or avoidance of victimization. Burbank (1987), who surveyed 137 societies in the Human Relations Area File, concluded that competition over mates was the single most frequent reason for female–female fights (121 out of the 297 fights for which reasons were recorded). However, according to Campbell (2013), sexual selection in women shaped intrasexual competition to favor indirect aggression over direct female–female combat, largely because in all known cultural groups, the mother is the principal caregiver of the offspring. Indirect aggression and rival derogation. To elaborate on the previous reasoning, in social species, females provide most of the obligatory parental care, and the death of the mother results in a dramatic decrease in the survival rate of the offspring and the potential diverting or loss of resources from male investment (Campbell, 2004). The costs of direct competition are thus higher for females than for males, and as a result, women are more risk averse than men and have a lower fear threshold, which prevents women’s

Buunk, Massar, Dijkstra, and Fernande z

225

involvement in direct physical aggression (Campbell, 1999; Hrdy,  2009). Therefore, women’s competition for the provision of resources and the access to suitable available mates often takes on the form of self-enhancement and indirect aggression in the form of rival derogation, as well as other ways of manipulating the mate and the rival (Campbell, 2004; Fisher,  2004; Fisher & Cox, 2011). To be more specific, women compete with each other by  employing two strategies at the same time: As well as advertising their own strong points through the enhancement of their appearance (e.g., by wearing make-up or tight clothes), they engage in indirect aggression toward other women to damage their reputation, mate value, and social standing (Campbell, 2004). Women’s indirect aggression is mainly directed toward intrasexual rivals, particularly young, attractive, and sexually available females, who may be able to capture the mating effort and resources of the most valuable men available at a given time (Vaillancourt, 2013). Although self-promotion is an effective tactic to retain the attention and investments of a mate, rival derogation allows women to prevent direct confrontation with their rivals, thereby lowering the risk of physical harm while simultaneously attempting to lower their rival’s mate value. For example, Li, Smith, Yong, and Brown (2014) reviewed evidence that young women’s desire for thinness and the sometimes resulting ­development of eating disorders may be the result of young girls’ self-promotion tactics to appear physically fit and attractive to potential mates, but may at the same time be a result of competition with other physically attractive women (see also Fisher & Cox,  2011). Rival derogation often directly targets other females’ mate value, and since female mate value is to a large extent determined by her physical attractiveness, attractive women are the largest threat, and much of women’s intrasexual competition revolves around derogating other women’s appearance. Fink, Klappauf, Brewer, and Shackelford (2014) report that in a context of intrasexual competition, the physical features men value the most in women are perceived by other women as most threatening: a feminine face, larger breasts, and a low WHR. As a way to counter the threat of such attractive rivals, women report often making derogatory comments about the unattractiveness of highly appealing other women and verbal derogation tends to focus mainly on their appearance and sexual health. These acts of rival 226

derogation can take on various forms and range from quite subtle—taking over conversations involving the rival—to blatant, such as pointing out a rival’s flaws to a male in whom one is interested (Fisher & Cox,  2011). Indeed, attractiveness increases the odds of being the victim of verbal derogation up to 35 percent for adolescent girls, and girls’ recent sexual activity increases the risk for indirect aggression (e.g., social exclusion or spreading rumors; Leenaars, Dane, & Marini, 2008). In addition to derogating women’s attractiveness, females tend to derogate other women’s intelligence or professional competence. Indeed, the “tall poppy syndrome” refers to the tendency for successful women to be  “clipped” down to the level of less successful female peers through verbal attacks and gossip (e.g., Dellasega,  2009). Recently, Grabe, Bas, Pagano, and Samson (2012) found that when women were confronted with an attractive and sexualized female news presenter, they tended to derogate her intelligence and her efforts to enhance her appearance, whereas men’s reactions showed an opposite pattern. In social contexts with a skewed sex ratio, and where suitable male partners are thus scarce, rival derogation has been shown to specifically target rivals’ sexuality: gossiping and accusing other women of being promiscuous, “easy,” or sexually unfaithful (Campbell, 2004). Since men looking for a long-term mate place a premium on finding a reproductively exclusive mate, accusing competitors of promiscuity and sexual availability may be a useful weapon in such contexts (Campbell, 2004). Experimental research (Arnocky, Ribout, Mirza, & Knack,  2014) indicates that mere perceptions of mate scarcity—induced by a bogus newspaper article—increased women’s intrasexual competitiveness, jealousy, and willingness to aggress indirectly against a same-sex rival. Moreover, in mate-scarce contexts, there is a risk that female indirect aggressive behaviors—including rival ­ derogation—escalate into female–female direct ­ (physical) aggression (Campbell, 2004). Regarding the effectiveness of these intrasexual competitive tactics, research has shown that attractive women have a credibility advantage over their less attractive peers; after hearing derogatory remarks about another woman’s appearance, men tended to decrease their ratings of the facial attractiveness of this person when the remarks came from an attractive woman (Fisher, 2004). Fisher, Shaw, Worth, Smith, and Reeve (2010) report that such derogatory comments cause men to devalue

Intersexual and Intrasexual Competition

not only a woman’s attractiveness but also her kindness, trustworthiness, and overall desirability as a mate. In sum, attractive women seem to have a strategic advantage over less attractive women when it comes to influencing male perceptions of other women’s attractiveness. This also provides an additional explanation for the finding that attractive females are often the target of derogation, whereas women who are unattractive or have a lower self-perceived mate value are more often the perpetrators of same-sex indirect aggression (Arnocky et al., 2012; Vaillancourt, 2013). However, although rival derogation may be an effective strategy to ­influence perceptions of a rival’s attractiveness or sexual permissiveness, it simultaneously decreases the perpetrator’s reputation or mate value for men, as well as her appeal as a same-sex ally for women (Fisher et al., 2010; Vaillancourt, 2013). Moreover, women are aware of the fact that direct, blatant, or unreasonable attacks on another woman’s appearance are inappropriate and likely to be counterproductive (Fisher, Cox, & Gordon, 2009). Indirect—or social—aggression refers to the purposeful, but often covert, manipulation of interpersonal relationships and includes behaviors such as social exclusion, breaking confidences, spreading rumors, and gossip. Relative to direct, and especially physical, aggression, these acts are less risky since there is a lower likelihood of retaliation or social consequences, and are equally or more often employed by women than by men (Hess & Hagen,  2006). For example, Benenson et al. (2013) observed that “the formation of temporary exclusionary coalitions provides an elegant means by which females, either directly or indirectly, can minimize competition without incurring large costs” (p. 5). The overall goal of the various forms of indirect aggression usually is to exclude rivals from one’s social group while simultaneously damaging their ability to form or maintain a social network of their own (Geary & Flinn,  2002). In  addition to social exclusion, other intrasexual competition strategies involve manipulation of the mate and the potential rivals (Fisher & Cox, 2011). Manipulation of a potential mate functions to secure his exclusive attention relative to rivals and varies from lying to a potential mate about the availability or interest of another woman to socializing with the potential mate’s friends. Manipulation of the rival, on the other hand, involves behaviors that are aimed at other women with the intent to make them appear less attractive or interesting, and varies from telling other

women suitable mates are not interested in them to making the rival women believe these men are bad lovers. Gossip, or spreading rumors, is a common intrasexual competitive tactic used by women ­ and  tends to focus on topics like other women’s appearance or sexual conduct (for a review, see McAndrew, 2014). Gossip between females is more negative than gossip between men or mixed-sex pairs, and women in particular respond encouragingly and positively to gossip they hear from their female friends (Leaper & Holliday, 1995). Younger women tend to engage more in gossip than older women: In a study among women aged 20 to 50, Massar, Buunk, and Rempt (2012) found that the younger women were, the more likely they were to engage in derogatory gossip about a woman who was interested in the same man as they were. The finding that younger women engage more in ­intrasexual competition is consistently reported in  the literature. For example, in a study conducted in South America, younger women reported significantly more intrasexual competition than women who were past their reproductive peak (Fernandez, Muñoz-Reyes, & Dufey,  2014), and premenopausal women are more likely to derogate other women’s attractiveness than are postmenopausal women (Vukovic et al., 2009). Hormonal influences. The effects of hormones on intrasexual competition, as well as the influence of intrasexual rivals on hormone levels, have increasingly been studied over the past years. Ovulatory shifts in intrasexual competition have by now been firmly established. For example, Fisher (2004) showed that women tested during the fertile phase of their cycle rated other women’s facial attractiveness as lower compared to menstruating women. Moreover, women are less willing to bargain with other women when they are fertile (Lucas, Koff, & Skeath, 2007), and tend to report increased dehumanization of other women—but not men or elderly women— during the ovulatory phase of their cycle (Piccoli, Foroni, & Carnaghi, 2013). Moreover, during fertile days in their cycle, women “dress to impress” (i.e., they dress more revealing and sexy than during nonfertile days of their cycle; Durante, Li, & Haselton, 2008). These results suggest that to maximize the likelihood of obtaining a mate with suitable qualities when this is most relevant (i.e., when conception risk is highest), women are driven by a  motivation to outcompete attractive rivals and show  increased intrasexual attitudes

Buunk, Massar, Dijkstra, and Fernande z

227

and behaviors. Research further suggests that these effects may be driven by estrogen levels in particular: Piccoli, Cobey, and Carnaghi (2014) showed that as the level of synthetic estrogen in women’s hormonal contraceptives was higher, their intrasexual competition also increased, in particular their tendency to objectify other women. Similarly, it has been established that among women using hormonal contraceptives, higher levels of synthetic estrogen— but not synthetic progesterone—predict higher levels of self-reported jealousy (Cobey, Pollet, Roberts, &  Buunk,  2011) and increased mate guarding tendencies (Welling, Puts, Roberts, Little, & ­ Burriss, 2012). In addition to their own menstrual cycle influencing intrasexual competition, a same-sex rival’s fertility status might increase women’s intrasexual competition and mate retention efforts. In yellow baboons, for example, females try to monopolize the attention of males when same-sex rivals are fertile (Wasser & Barash, 1983). In a similar vein, Krems, Neel, Neuberg, Puts, and Kenrick (2016) showed that women in a relationship who were exposed to photographs of other women taken during their ovulatory cycle phase reported they would try to avoid or even socially exclude these women, but only when they rated their own partners as highly desirable. Moreover, the presence of  especially fertile intrasexual competitors may affect women’s hormonal levels. It already has been established that anticipation of competition (e.g., watching or playing soccer matches) elevates testosterone levels not only among men (e.g., Van der Meij et al.,  2012) but also among women (Oliveira, Gouveia, & Oliveira, 2009). Recent research shows that intrasexual or social competition also affects female testosterone levels. For example, Maner and McNulty (2013) report that women who were exposed to the scent of a fertile woman subsequently displayed higher levels of testosterone than women exposed to the scent of a nonfertile woman. A study tracking women’s testosterone and intrasexual competition levels during five weeks (Hahn, Fisher, Cobey, DeBruine, & Jones, 2016) revealed that women reported greater intrasexual competitiveness in the test sessions when their testosterone levels were high (as it is near ovulation). There was no association between other hormones (estradiol, progesterone, cortisol) and intrasexual competition, suggesting a unique role for testosterone in regulating women’s intrasexual competitiveness (see also Cobey, Klipping, & Buunk, 2013). 228

Individual Differences in Intrasexual Competitiveness

Although it seems clear that both sexes tend to compete largely with same-sex others, there are important individual differences in the extent to which males and females engage in intrasexual competition. Some individuals seem to have as their major goal to “beat” others, whereas others seem to have as their major goal to develop collaborative relationships with others. As noted by Buunk and Fisher (2009), some evolutionary psychologists have argued that individual differences in such adaptations are merely noise. According to Tooby and Cosmides (1990), “Heritable variation in a trait generally signals a lack of adaptive significance” (p. 38 [italics in original]). However, other authors have suggested that such heritable variation may continue to exist because ­individual differences may reflect equally adaptive strategies (e.g., Buss, 1991; Gangestad & Simpson, 1990; MacDonald,  1995). From an evolutionarypsychological point of view, individual differences such as these may exist for several reasons. First, combinations of specific individual differences may result in equally viable behavioral strategies (Penke, Denissen, & Miller,  2007). Although each behavioral strategy has its specific costs and benefits, the net effect may be the same (Nettle, 2006). For instance, both competition and altruism may help individuals gain higher group status and can, as such, both be considered adaptive strategies. Second, Figueredo et al. (2005) argued that personality ­differences may be adaptive in social competition because of the operation of frequency-dependent selection. Frequency-dependent selection implies that a single optimal strategy does not exist, and that various distinct strategies may all be heritable. Different strategies may have developed because, under different conditions, different strategies may be adaptive. For example, in a population with predominantly cooperative individuals, there would be a niche for competitive individuals, and vice versa. Indeed, it seems probable that being strongly intrasexually competitive and selfish may be adaptive under certain conditions, when, for instance, life expectancy is low, others are low in altruism as well, and the level of social organization is low. Yet the same competitive behavior may be maladaptive under other conditions, for instance, when others are high in altruism or in complex social groups (Rushton, 1985). Findings from studies on species as diverse as great tits (Parus major) to big-horn sheep (Oviscanadensis) demonstrate that traits such as aggressiveness toward conspecifics, boldness, and risk

Intersexual and Intrasexual Competition

taking have different fitness payoffs in different environments (Dingemanse & Réale, 2005). Finally, it has been argued that individuals are genetically predisposed to have personality characteristics that, to a certain extent, are malleable. As a result, situational demands may push individuals to develop certain strategies and traits over other ones (Penke et al., 2007; Saad, 2007), a phenomenon Gangestad and Simpson (2000) refer to as strategic pluralism. To assess individual differences in intrasexual competition, Buunk and Fisher (2009) developed the Intrasexual Competition Scale (ICS). The scale assesses intrasexual competition as an attitude that may be referred to as intrasexual competitiveness and concerns the degree to which individuals view the confrontation with same-sex individuals in competitive terms. It implicates a number of phenomena that have been well described in the psychological literature, albeit not in a mating context, including the desire to outperform others rather than to perform well (Van Yperen, 2003); the desire to view oneself as better than others (cf. self-enhancement; Zuckerman & O’Loughlin, 2006); envy and frustration when others are better off and negative feelings toward such others (Smith & Kim, 2007); and malicious pleasure or schadenfreude when high achievers (“tall poppies”) lose face (Feather, 1994). The ICS operationalized these phenomena, particularly on dimensions relevant to mating, and only formulated with respect to same-sex others. However, in addition, following up on a study by Luxen and Van de Vijver (2005), who showed that women often reject attractive women as candidates for a position in their department, questions were included on the resistance to having others with higher mate value as close colleagues. The 12item scale was constructed simultaneously in the Netherlands and Canada and proved to be sex neutral, to possess high reliability, to have a high degree of cross-national equivalence, and to be ­related to self-report of sibling rivalry in one’s childhood (Buunk & Fisher, 2009). Intrasexual competitiveness appears among men in a different way rooted in personality than among women. In a study among Canadian students, Buunk and Fisher (2009) assessed the relationship between intrasexual competitiveness and the Big Five personality traits, using Costa and McCrae’s (1992) scales for the Big Five personality traits, including neuroticism, extraversion, openness, agreeableness, and conscientiousness. These traits have been recognized in many species (see also Andersson, 1994; Buss, 1991). Buunk and Fisher (2009) found

that intrasexual competitiveness was associated with a lack of agreeableness and neuroticism (although in a regression neuroticism was only marginally significant) among women, and with a high level of neuroticism and extraversion among men, with neuroticism making the strongest contribution. In a study among 140 adults in Uruguay with a mean age of nearly 37, these findings were replicated and were even stronger, with clear independent significant effects of lack of agreeableness and neuroticism ­ among women, and extraversion and neuroticism among men (Buunk, FornariBucksath, & Cordero, 2017). Thus, an intrasexual competitive attitude is typical for neurotic extraverted men and for neurotic nonagreeable women. A possible theoretical explanation for this sex difference is that among males, these are the traits that were the first individual differences to evolve in freely moving species (Figueredo et al., 2005), and that intrasexual competition among males has a longer evolutionary history than among females, due to which the adaptive value of different levels of the same trait may have had more time to evolve (cf. Nettle, 2006). As noted by Costa and McCrae (1992), the disagreeable or antagonistic person is egocentric, skeptical of others’ intentions, and competitive rather than cooperative. Given their longer evolutionary history of intrasexual competition, men may find agreeing with statements reflecting negative attitudes to other men relatively normal, whereas women need to have a disagreeable personality in order to agree with such statements. Indeed, more than women, men seem to consider competition with same-sex others a more normal fact of life, which is for them not incompatible with maintaining collaborative relationships with same-sex others (cf. Buunk & Massar, 2014; Campbell, 2013). Although intrasexual competitiveness is clearly an individual difference characteristic, the level of  it is dependent on environmental conditions. First, among both sexes, intrasexual competitiveness will be enhanced when there is a relative shortage of  potential acceptable partners. Correspondingly, Buunk, Stulp, and Ormel (2014) found in a large representative sample of adolescent women that such women were c­ onsistently more intrasexually competitive the higher the socioeconomic status of their parents, assumedly because young women are nowadays on average more highly educated than young men, intensifying female competition over highly educated males. In contrast, males with parents with the lowest socioeconomic status tended to be more intrasexually competitive than those with parents with a medium socioeconomic

Buunk, Massar, Dijkstra, and Fernande z

229

status, assumedly because females can “marry up,” and lowly educated males have to compete not only with other males with a low education but also with males with a higher education. A second type of condition that will enhance intrasexual competitiveness is ecological conditions (e.g., Nettle, 2006) such as a low life expectancy, a low perceived chance of attaining a high status in the long run, and a low mate value. In line with this, Van Brummen-Girigori and Buunk (2016) found that girls who had grown up without their biological father before the age of 14 reported overall more intrasexual competitiveness than girls who had grown up in intact families (while there were no differences in socioeconomic level between both groups). In  addition, this intrasexual competitiveness was strongly correlated with a variety of nonverbal strategies to attract males, probably mostly for short-term mating, including direct flirtation, the use of hairstyles with waves, the use of facial make-up, the use of conspicuous nail care, and active and restless behavior in the presence of males. Intrasexual competitiveness was a significant mediator between father absence and the expression of most nonverbal seduction strategies. Although human females high in intrasexual competitiveness may try to enhance their attractiveness to attract males, in many species males in intrasexually competitive situations tend to e­ nhance their body size. Various studies have provided evidence of the association between dominance and an expanded posture both in primates and in humans (Weisfeld & Linkey,  1985). Consistently, Duguid and Goncalo (2012) reported that the e­ xperience of power influenced individuals’ perceptions of their own height, such that individuals experiencing more power overestimated their height. In a similar vein, Mailhos, Buunk, and Del Arca (2017) reasoned that males high in intrasexual competitiveness might overestimate their own height. Indeed, in a sample of junior soccer players from a First Division Uruguayan soccer team, they found that intrasexual competitiveness was positively correlated with a bias in height report; that is, the higher the level of intrasexual competitiveness, the more the self-report of one’s own height exceeded one’s actual height.

Conclusion

Competition is a basic phenomenon in nature, and in this chapter we have tried to outline the evolutionary and biological basis, as well as the manifestations

230

and nature of intersexual and intrasexual competition among humans. Among many species, including humans, the basic reproductive interests of males and females do in part diverge, resulting in  intersexual conflict. Specifically relevant in the present context, both men and women in steady ­relationships may for various reasons be interested in involvement with opposite-sex partners other than their steady partner, which may be a threat to the reproductive interests of the steady partner. This conflict manifests in not only the occurrence of jealousy as such, but also the different foci of men and women in jealousy, with men focusing more on sexual infidelity and women more on emotional infidelity. In the case of the presence of actual rivals, jealousy has a strong component of intrasexual competition, with men and women paying attention to different characteristics of the rivals. These sex differences in attention to specific rival characteristics even occur outside conscious awareness, and have a physical and hormonal basis. Intrasexual competition is, however, more than jealousy; it is manifest in many ways and differs considerably between men and women. Male intrasexual competition is evolutionarily probably more ancient and is more characterized by physical dominance and direct aggression, conspicuous consumption, and aims of attaining eminence, whereas female intrasexual competition is more characterized by indirect aggression, self-promotion, and rival derogation. Both male and female intrasexual competitions have hormonal bases, with the fertile stage of the cycle playing an important role among women. Finally, we have paid attention to a phenomenon that has thus far received little attention—the considerable individual differences in the tendency to be intrasexually competitive, which are related to personality factors and environmental conditions. In sum, the present chapter outlined the multifaceted nature of intrasexual and intrasexual competition, including hormonal, physical, cognitive, sex-specific, and personality aspects.

References

Ambwani, S., & Strauss, J. (2007). Love thyself before loving others? A qualitative and quantitative analysis of gender differences in body image and romantic love. Sex Roles, 56, 13–21. Andersson, M.  B. (1994). Sexual selection. Princeton, NJ: Princeton University Press. Arnocky, S., Ribout, A., Mirza, R. S., & Knack, J. M. (2014). Perceived mate availability influences intrasexual competition, jealousy and mate-guarding behavior. Journal of Evolutionary Psychology, 12, 45–64.

Intersexual and Intrasexual Competition

Arnocky, S., Sunderani, S., Miller, J., & Vaillancourt, T. (2012). Jealousy mediates the relationship between attractiveness comparison and females indirect aggression. Personal Relationships, 19, 290–303. doi:10.1111/j.1475-6811.2011. 01362.x Bar, M., Neta, M., & Linz, H. (2006). Very first impressions. Emotion, 6(2), 269–278. Barber, N. (1995). The evolutionary psychology of physical attractiveness: Sexual selection and human morphology. Ethology and Sociobiology, 16, 395–424. Barkow, J.  H. (1989). Darwin, sex, and status. Biological approaches to mind and culture. Toronto, Canada: University of Toronto Press. Bendixen, M., Kennair, L. O., & Buss, D. M. (2015). Jealousy: Evidence of strong sex differences using both forced choice and continuous measure paradigms. Personality and Individual Differences, 86, 212–216. Bendixen, M., Kennair, L. E. O., Ringheim, H. K., Isaksen, L., Pedersen, L., Svangtun, S., & Hagen, K. (2015). In search of moderators of sex differences in forced-choice jealousy responses: Effects of 2D: 4D digit ratio and relationship infidelity experiences. Nordic Psychology, 67, 272–284. Benenson, J.  F. (2013). The development of human female competition: Allies and adversaries. Philosophical Transactions of the Royal Society B, 368, 20130079. Benenson, J.  F., Markovits, H., Hultgren, B., Nguyen, T., Bullock, G., & Wrangham, R. (2013). Social exclusion: More important to human females than males. PLoS One, 8, e55851. doi:10.1371/journal.pone.0055851. Betzig  L. (1982). Despotism and differential reproduction. Ethology and Sociobiology, 3, 209–221. Betzig, L. (1986). Despotism and differential reproduction: A Darwinian view of history. New York, NY: Aldine. Bird, R. B., Smith, E. A., & Bird, D. W. (2001). The hunting handicap: Costly signaling in human foraging strategies. Behavioral Ecology and Sociobiology, 50, 9–19. Bjorklund, D.  F., & Shackelford, T.  K. (1999). Differences in parental investment contribute to important differences between men and women. Current Directions in Psychological Science, 8, 86–89. Burbank, V.  K. (1987). Female aggression in cross-cultural perspective. Behavior Science Research, 21, 70–100. Burchell, J. L., & Ward, J. (2011). Sex drive, attachment style, relationship status and previous infidelity as predictors of sex differences in romantic jealousy. Personality and Individual Differences, 51, 657–661. Buss, D. M. (1988). The evolution of human intrasexual compe­ tition: Tactics of mate attraction. Journal of Personality and Social Psychology, 54, 616–628. doi:10.1037/00223514.54.4.616 Buss, D. M. (1989). Sex differences in human mate preferences: Evolutionary hypotheses tested in 37 cultures. Behavioral and Brain Sciences, 12, 1–49. Buss, D.  M. (1991). Evolutionary personality psychology. In M.  R.  Rosenzweig, L.  W.  Porter, M.  R.  Rosenzweig, & L.  W.  Porter (Eds.), Annual review of psychology (Vol. 42, pp. 459–491). Palo Alto, CA: Annual Reviews. Buss, D. M. (1994). The evolution of desire: Strategies of human mating. New York, NY: Basic Books. Buss, D. M., Larsen, R. J., Westen, D., & Semmelroth, J. (1992). Sex differences in jealousy: Evolution, physiology, and psychology. Psychological Science, 3, 251–255.

Buss, D. M., & Schmitt, D. P. (1993). Sexual strategies theory: An evolutionary perspective on human mating. Psychological Review, 100, 204–232. Buss, D.  M., & Shackelford, T.  K. (1997). Susceptibility to infidelity in the first year of marriage. Journal of Research in Personality, 31, 193–221. Buss, D.  M., Shackelford, T.  K., Choe, J., Buunk, A.  P., & Dijkstra, P. (2000). Distress about mating rivals. Personal Relationships, 7, 235–243. Buss, D. M., Shackelford, T. K., Kirkpatrick, L. A., Choe, J. C., Lim, H. K., Hasegawa, M., . . . Bennett, K. (1999). Jealousy and the nature of beliefs about infidelity: Tests of competing hypotheses about sex differences in the United States, Korea, and Japan. Personal Relationships, 6, 125–150. Butler, V. J., Ryckman, R. M., Thornton, B., & Bouchard, R. L. (1993). Assessment of the full content of physique stereotypes with a free-response format. Journal of Social Psychology, 133, 147–162. Buunk, A. P. (1995). Sex, self-esteem, dependency and extradyadic sexual experiences as related to jealousy responses. Journal of Social and Personal Relationships, 12, 147–153. Buunk, B.  P. (1997). Personality, birth order, and attachment styles as related to various types of jealousy. Personality and Individual Differences, 23, 997–1006. Buunk, B.  P. (1980). Extramarital sex in the Netherlands: Motivations in social and marital context. Alternative Lifestyles, 3, 11–39. Buunk, B.  P. (1981). Jealousy in sexually open marriages. Alternative Lifestyles, 4, 357–372. Buunk, A.  P., Aan ‘t Goor, J., & Castro Solano, A. (2010). Intrasexual competition at work: Sex differences in the jealousy-evoking effect of rival characteristics in work settings. Journal of Social and Personal Relationships, 27, 671–684. Buunk, B. P., Angleitner, A., Oubaid, V., & Buss, D. M. (1996). Sex differences in jealousy in evolutionary and cultural perspective: Tests from the Netherlands, Germany, and the United States. Psychological Science, 7, 359–363. Buunk, A.  P., & Bakker, A.  B. (1997). Commitment to the relationship, extradyadic sex, and AIDS-preventive behavior. Journal of Applied Social Psychology, 27, 1241–1257. Buunk, A.  P., Fornari Bucksath, A., & Cordero, S. (2017). Intrasexual competitiveness and personality traits: A study in Uruguay. Personality and Individual Differences, 108, 178–181. Buunk, A. P., & Castro Solano, A. (2012). Mate guarding and parental influence on mate choice. Personal Relationships, 19, 103–112. Buunk, B. P., & Dijkstra, P. (2003). Men, women and infidelity: Sex differences in extradyadic sex and jealousy. In J. Duncombe, K. Harrison, G. Allan, & D. Marsden (Eds.), The state of affairs. Explorations in infidelity and commitment (pp. 103–120). Mahwah, NJ: Erlbaum. Buunk, A.  P., & Dijkstra, P. (2015). Rival characteristics that provoke jealousy: A study in Iraqi Kurdistan. Evolutionary Behavioral Sciences, 9, 116–127. Buunk, A.  P., & Fisher, M. (2009). Individual differences in intrasexual competition. Journal of Evolutionary Psychology, 1, 37–48. Buunk, A. P., & Massar, K. (2014). A night on the town: When the importance of mate acquisition overrides intrasexual competition. Anthropological Review, 77, 273–285.

Buunk, Massar, Dijkstra, and Fernande z

231

Buunk, A. P., Park, J. H., Zurriaga, R., Klavina, L., & Massar, K. (2008). Height predicts jealousy differently for men and women. Evolution and Human Behavior, 29, 133–139. Buunk, A.  P., Pollet, T.  V., Klavina, L., Figueredo, A.  J., & Dijkstra, P. (2009). Height among women in curvilinearly related to life history strategy. Evolutionary Psychology, 7, 545–559. Buunk, A.  P., Solano, A., Zurriaga, R., & González, P. (2011). Gender differences in the jealousy-evoking effect of rival characteristics: A study in Spain and Argentina. Journal of Cross-Cultural Psychology, 42, 323–339. Buunk, A. P., Stulp, G., & Ormel, J. (2014). Parental social status and intrasexual competitiveness among adolescents. Evolutionary Psychology, 12, 1022–1037. Buunk, A. P., & Van Brummen-Girigori, O. (2016). Menstrual cycle effects on jealousy: A study in Curaçao. Evolution, Mind and Behavior, 14(1), 43–54. Buunk, A.  P., & Ybema, J.  F. (1997). Social comparisons and occupational stress: The identification-contrast model. In B. P. Buunk & F. X. Gibbons (Eds.), Health, coping, and wellbeing: Perspectives from social comparison theory (pp. 359–388). Mahwah, NJ: Lawrence Erlbaum Associates Publishers. Campbell, A. (1999). Staying alive: Evolution, culture, and women's intrasexual aggression. Behavioral and Brain Sciences, 22, 203–214. Campbell, A. (2002). A mind of her own: The evolutionary psychology of women. Oxford, UK: Oxford University Press. Campbell, A. (2004). Female competition: Causes, constraints, content, and contexts. Journal of Sex Research, 41, 16–26. Campbell, A. (2013). A mind of her own: The evolutionary psychology of women (2nd ed.). New York, NY: Oxford University Press. Case, A., & Paxon, C. (2006). Height, health, and cognitive function at older ages. American Economic Review: Papers & Proceedings, 98, 463–467. Cheney, D. L., Silk, J. B., & Seyfarth, R. M. (2012). Evidence for  intrasexual selection in wild female baboons. Animal Behaviour, 84, 21–27. Child, I. L. (1950). The relation of somatotype to self-ratings on Sheldon’s temperamental traits. Journal of Personality, 18, 440–453. Clark, R.  D., & Hatfield, E. (1989). Gender differences in receptivity to sexual offers. Journal of Psychology and Human Sexuality, 2, 39–55. Clutton-Brock, T.  H., & Parker, G.  A. (1995). Punishment in animal societies. Nature, 373, 209–216. Cobey, K.  D., Buunk, A.  P., Roberts, C.  S., Klipping, C., Appels, N., Zimmerman, Y., . . . Pollet, T.V. (2012). Reported jealousy differs as a function of menstrual cycle stage and contraceptive pill use: A within-subjects investigation. Evolution and Human Behavior, 33, 395–401. Cobey, K. D., Klipping, C., & Buunk, A. P. (2013). Hormonal contraceptive use lowers female intrasexual competition in pair-bonded women. Evolution and Human Behavior, 34, 294–298. Cobey, K. D., Pollet, T. V., Roberts, S. C., & Buunk, A. P. (2011). Hormonal birth control use and relationship jealousy: Evidence for estrogen dosage effects. Personality and Individual Differences, 50, 315–317. Costa, P., & McCrae, R. (1992). The neo PI-R professional manual. Odessa, FL: Psychological Assessment Resources. Cupcea, S. (1939). Constitutie morfologica si intelligenta. Revista de Psihologie, 2, 169–176.

232

Darwin, C. (1871). The descent of man and selection in relation to sex. London, UK: John Murray. De Cremer, D., & Van Knippenberg, D. (2004). Leader selfsacrifice and leadership effectiveness: The moderating role of  leader self-confidence. Organizational Behavior and Human Decision Processes, 95(2), 140–155. Deabler, H. L., Hartl, E. M., & Willis, C. A. (1975). Physique and personality: Somatotype and vocational interest. Perceptual and Motor Skills, 41, 382. Dellasega, C.  A. (2009). Bullying among nurses. American Journal of Nursing, 109, 52–58. DeSteno, D. A., & Salovey, P. (1996). Evolutionary origins of sex differences in jealousy? Questioning the fitness of the model. Psychological Science, 7, 367–372. Dew, B., Brubaker, M., & Hays, D. (2006). From the altar to the Internet: Married men and their online sexual behavior. Sexual Addiction & Compulsivity, 13, 195–207. Dijkstra, P., & Buunk, B. P. (1998). Jealousy as a function of rival characteristics: An evolutionary perspective. Personality and Social Psychology Bulletin, 24, 1158–1166. Dijkstra, P., & Buunk, B.  P. (2002). Sex differences in the jealousy-evoking effect of rival characteristics. European Journal of Social Psychology, 32, 829–852. Dingemanse, N. J., & Réale, D. (2005). Natural selection and animal personality. Behaviour, 142, 1165–1190. Domey, R.  G., Duckworth, J.  E., & Morandi, A.  J. (1964). Taxonomies and correlates of physique. Psychological Bulletin, 62, 411–426. Duguid, M.  M., & Goncalo, J.  A. (2012). Living large: The powerful overestimate their own height. Psychological Science, 23, 36–40. Dunn, M.  J., & McLean, H. (2015). Jealousy-induced sex differences in eye gaze directed at either emotional- or sexual infidelity–related mobile phone messages. Cyberpsychology, Behavior, and Social Networking, 18, 37–40. Durante, K. M., Li, N. P., & Haselton, M. G. (2008). Changes in women’s choice of dress across the ovulatory cycle: Naturalistic and laboratory task-based evidence. Personality and Social Psychology Bulletin, 34, 1451–1460. Feather, N.  T. (1994). Attitudes towards high achievers and reactions to their fall: Theory and research concerning tall poppies. Advances in Experimental Social Psychology, 26, 1–73. Feingold, A. (1982). Do taller men have prettier girlfriends? Psychological Reports, 50, 810. Fernandez, A.  M., Muñoz-Reyes, J.  A., & Dufey, M. (2014). BMI, age, mate value, and intrasexual competition in Chilean women. Current Psychology, 33, 435–450. doi:10.1007/s12144-014-9221-x Figueredo, A.  J., Sefcek, J.  A., Vasquez, G., Brumbach, B.  H., King, J. E., & Jacobs, W. J. (2005). Evolutionary personality psychology. In D. M. Buss (Ed.), The handbook of evolutionary psychology (pp. 851–877). Hoboken, NJ: Wiley. Fink, B., Klappauf, D., Brewer, G., & Shackelford, T. K. (2014). Female physical characteristics and intra-sexual competition in women. Personality and Individual Differences, 58, 138–141. Fisher, M. L. (2004). Female intrasexual competition decreases female facial attractiveness. Proceedings of the Royal Society B: Biological Sciences, 271, 283–285. doi:10.1098/rsbl.2004.0160 Fisher, M., & Cox, A. (2011). Four strategies used during intrasexual competition for mates. Personal Relationships, 18, 20–38. doi:10.1111/j.1475-6811.2010.01307.x Fisher, M., Cox, A., & Gordon, F. (2009). Self-promotion versus competitor derogation: The influence of sex and romantic

Intersexual and Intrasexual Competition

relationship status on intrasexual competition strategy selection. Journal of Evolutionary Psychology, 7, 287–308. Fisher, M., Shaw, S., Worth, K., Smith, L., & Reeve, C. (2010). How we view those who derogate: Perceptions of female competitor derogators. Journal of Social, Evolutionary, and Cultural Psychology, 4, 265–276. Fiske, S.  T. (1992). Stereotypes work . . . but only sometimes: Comment on how to motivate the “unfinished mind.” Psychological Inquiry, 3, 161–162. Fleuriet, C., Cole, M., & Guerrero, L.  K. (2014). Exploring Facebook: Attachment style and nonverbal message characteristics as predictors of anticipated emotional reactions to Facebook postings. Journal of Nonverbal Behavior, 38, 429–450. Frederick, D., & Haselton, M. G. (2007). Why is muscularity sexy? Tests of the fitness indicator hypothesis. Personality and Social Psychology Bulletin, 33, 1167–1183. Gangestad, S.  W., & Simpson, J.  A. (1990). Towards an evolutionary history of female sociosexual variation. Journal of Personality, 58, 70–96. Gangestad, S.  W., & Simpson, J.  A. (2000). The evolution of human mating: Trade-offs and strategic pluralism. Behavioral and Brain Sciences, 23, 573–587. Gangestad, S.  W., Thornhill, R., & Garver, C.  E. (2002). Changes in women’s sexual interests and their partners’ mate retention tactics across the menstrual cycle: Evidence for shifting conflicts of interest. Proceedings of the Royal Society of London, Series B, 269, 975–982. Garn, S. M., & Gertler, M. M. (1950). An association between type of work and physique in an industrial group. American Journal of Physical Anthropology, 8, 387–397. Geary, D.  C. (1998). Male, female: The evolution of human sex differences. Washington, DC: American Psychological Association. Geary, D.  C. (2000). Evolution and proximate expression of human paternal investment. Psychological Bulletin, 126, 55–77. Geary, D.  C. (2005a). Evolution of paternal investment. In D.  M.  Buss (Ed.), The evolutionary psychology handbook (pp. 483–505). New York, NY: Wiley. Geary, D.  C. (2005b). The origin of mind: Evolution of brain, cognition, and general intelligence. Washington, DC: American Psychological Association Press. Geary, D. C., DeSoto, C. M., Hoard, M. K., Sheldon, M. S., & Cooper, L.  M. (2001). Estrogens and relationship jealousy. Human Nature, 12, 299–320. Geary, D.  C., & Flinn, M.  V. (2002). Sex differences in behavioral and hormonal response to social threat: Commentary on Taylor et al. (2000). Psychological Review, 109, 745–750. Gibson, R.  M., & Bradbury, J.  W. (1985). Sexual selection in lekking sage grouse: Phenotypic correlates of male mating success. Behavioral Ecology and Sociobiology, 18, 117–123. Goetz, A. T., Shackelford, T. K., Romero, G. A., Kaighobadi, F., & Miner, E.  J. (2008). Punishment, proprietariness, and paternity: Men’s violence against women from an evolutionary perspective. Aggression and Violent Behavior, 13, 481–489. Grabe, M. E., Bas, O., Pagano, L. A., & Samson, L. (2012). The architecture of female competition: Derogation of a sexualized female news anchor. Journal of Evolutionary Psychology, 10, 107–133. doi:10.1556/jep.10.2012.3.2 Graham, K., & Wells, S. (2001). The two worlds of aggression for men and women. Sex Roles, 45, 595–622.

Groothof, H.  A., Dijkstra, P., & Barelds, D.  P. (2009). Sex differences in jealousy: The case of Internet infidelity. Journal of Social and Personal Relationships, 26, 1119–1129. Guadagnom, R. E., & Sagarin, B. J. (2010). Sex differences in jealousy: An evolutionary perspective on online infidelity. Journal of Applied Social Psychology, 40, 2636–2655. Hahn, A. C., Fisher, C. I., Cobey, K. D., DeBruine, L. M., & Jones, B.  C. (2016). A longitudinal analysis of women’s salivary testosterone and intrasexual competitiveness. Psycho­ neuroendocrinology, 64, 117–122. Harris, C. R. (2002). Sexual and romantic jealousy in heterosexual and homosexual adults. Psychological Science, 13, 7–12. Harris, C.  R. (2003). A review of sex differences in sexual jealousy, including self-report data, psychophysiological responses, interpersonal violence, and morbid jealousy. Personality and Social Psychology Review, 7, 102–128. Haselton, M. G., & Buss, D. M. (2000). Error management theory: A new perspective on biases in cross-sex mind reading. Journal of Personality and Social Psychology, 78, 81–91. Haselton, M.  G., Buss, D.  M., Oubaid, V., & Angleitner, A. (2005). Sex, lies, and strategic interference: The psychology of deception between the sexes. Personality and Social Psychology Bulletin, 31, 3–23. Hawkes, K., & Bliege Bird, R. (2002). Showing off, handicap signaling, and the evolution of men’s work. Evolutionary Anthropology, 11, 58–67. Hensley, W. E. (1993). Height as a measure of success in academe. A Journal of Human Behavior, 30, 40-46. Hess, N. H., & Hagen, E. H. (2006). Sex differences in indirect aggression: Psychological evidence from young adults. Evolution and Human Behavior, 27, 231–245. Hill, K., & Hurtado, A. M. (1996). Ache life history. New York, NY: Walter de Gruyter. Hill, S. E., & Buss, D. M. (2006). Envy and positional bias in the evolutionary psychology of management. Managerial and Decision Economics, 27, 131–143. Hohmann, G., & Fruth, B. (2003). Intra- and inter-sexual aggression by bonobos in the context of mating. Behaviour, 140, 1389–1413. Hopcroft, R. L. (2005). Sex, status, and reproductive success in the contemporary United States. Evolution and Human Behavior, 27, 104–120. Hrdy, S. B. (2009). The woman that never evolved. Cambridge, MA: Harvard University Press. Huchard, H., & Cowlishaw, G. (2011). Female–female aggression around mating: An extra cost of sociality in a multimale primate society. Behavioral Ecology, 22, 1003–1011. IJzerman, H., Blanken, I., Brandt, M.  J., Oerlemans, J.  M., Van den Hoogenhof, M. M., Franken, S. J., & Oerlemans, M.  W. (2014). Sex differences in distress from infidelity in  early adulthood and in later life. Social Psychology, 45, 202–208. Iredale, W., Van Vugt, M., & Dunbar, R. I. M. (2008). Showing off in humans: Male generosity as mate signal. Evolutionary Psychology, 6, 386–392. Jackson, L.  A., & Ervin, K.  S. (1992). Height stereotypes of women and men: The liabilities of shortness for both sexes. Journal of Social Psychology, 132, 433–445. Judge, T.  A., & Cable, D.  M. (2004). The effect of physical height on workplace success and income: Preliminary test of a theoretical model. Journal of Applied Psychology, 89, 428–441.

Buunk, Massar, Dijkstra, and Fernande z

233

Kemper, T. D. (1990). Social structure and testosterone: Explorations of the socio-bio-social chain. New Brunswick, NJ: Rutgers University Press. Krems, J. A., Neel, R., Neuberg, S. L., Puts, D. A., & Kenrick, D.  T. (2016). Women selectively guard their (desirable) mates from ovulating women. Journal of Personality and Social Psychology, 110, 551–573. Kuhle, B. X. (2011). Did you have sex with him? Do you love her? An in vivo test of sex differences in jealous interrogations. Personality and Individual Differences, 51, 1044–1047. Kurzban, R., & Weeden, J. (2005). Hurrydate: Mate preference in action. Evolution and Human Behavior, 26, 227–244. Leaper, C., & Holliday, H. (1995). Gossip in same-gender and cross-gender friends’ conversations. Personal Relationships, 2, 237–246. Leenaars, L.  S., Dane, A.  V., & Marini, Z.  A. (2008). Evolutionary perspectives on indirect victimization in adolescence: The role of attractiveness, dating and sexual behavior. Aggressive Behavior, 34, 404–415. Lerner, R. M., & Korn, S. J. (1972). The development of bodybuild stereotypes in males. Child Development, 3, 908–920. Levy, K. N., & Kelly, K. M. (2009). Sex differences in jealousy: A contribution from attachment theory. Psychological Science, 21, 168–173. Li, N.  P., Smith, A.  R., Yong, J.  C., & Brown, T.  A. (2014). Intrasexual competition and other theories of eating restriction. In V. A. Weekes-Shackelford & T. K. Shackelford (Eds.), Evolutionary perspectives on human sexual psychology and behavior (pp. 323–346). New York, NY: Springer. Loh, E. S. (1993). The economic effects of physical appearance. Social Science Quarterly, 74, 420–438. Lucas, M.  M., Koff, E., & Skeath, S. (2007). Pilot study of relationship between fertility risk and bargaining. Psychological Reports, 101(1), 302–310. Luxen, M. F., & van de Vijver, F. J. R. (2005). Facial attractiveness, sex and personnel selection: When evolved preferences matter. Journal of Organizational Behavior, 27, 241–255. Lycett, J., & Dunbar, R. I. M. (2000). Mobile phones as lekking devices among human males. Human Nature, 11, 93–104. MacDonald, K. (1995). Evolution, the 5-factor model, and levels of personality. Journal of Personality, 63, 525–567. Mackey, W. C., & Immerman, R. S. (2001). Restriction of sexual activity as a partial function of disease avoidance: A cultural response to sexually transmitted diseases. Cross-Cultural Research, 35, 400–423. Mailhos, A., Buunk, A. P., & del Arca, D. (2017). High intrasexual competition is related to inflated height reports in junior soccer players. Personality and Individual Differences, 113, 229–234. Maner, J.  K., & McNulty, J.  K. (2013). Attunement to the fertility status of same-sex rivals: Women’s testosterone responses to olfactory ovulation cues. Evolution and Human Behavior, 34, 412–418. Manning, J. T. (1995). Fluctuating asymmetry and body weight in men and women: Implications for sexual selection. Ethology and Sociobiology, 16, 145–153. Marazziti, D., Consoli, G., Albanese, F., Laquidara, E., Baroni, S., & Dell’Osso, M.  C. (2010). Romantic attachment and subtypes/dimensions of jealousy. Clinical Practice and Epidemiology in Mental Health, 6, 53–58. Marshall, T.  C., Bejanyan, K., Di Castro, G., & Lee, R.  A. (2013). Attachment styles as predictors of Facebook-related jealousy and surveillance in romantic relationships. Personal Relationships, 20, 1–22.

234

Massar, K., & Buunk, A.  P. (2009). Rivals in the mind’s eye: Jealous responses after subliminal exposure to body shapes. Personality and Individual Differences, 46, 129–134. Massar, K., & Buunk, A. P. (2010). Judging a book by its cover: Jealousy after subliminal priming with attractive and unattractive faces. Personality and Individual Differences, 49, 634–638. Massar, K., Buunk, A. P., & Dechesne, M. (2009). Jealousy in the blink of an eye: Jealous reactions following subliminal exposure to rival characteristics. European Journal of Social Psychology, 39, 768–779. Massar, K., Buunk, A. P., & Rempt, S. (2012). Age differences in women’s tendency to gossip are mediated by their mate value. Personality and Individual Differences, 52, 106–109. McAndrew, F. T. (2014). The “sword of a woman”: Gossip and female aggression. Aggression and Violent Behavior, 19, 196–199. Miller, G. F. (2000). The mating mind. London, UK: Heinemann. Miller, G. F. (2009). Spent: Sex, evolution, and consumer behavior. New York, NY: Viking Adult. Miller, M. J., Denes, A., Diaz, B., & Buck, R. (2014). Attachment style predicts jealous reactions to viewing touch between a romantic partner and close friend: Implications for Internet social communication. Journal of Nonverbal Behavior, 38, 451–476. Miner, E. J., Starratt, V. G., & Shackelford, T. K. (2009). Mate value of romantic partners predicts men’s partner-directed verbal insults. Personality and Individual Differences, 46, 135–139. Mize, K.  D., Shackelford, T.  K., & Shackelford, V.  A. (2009). Hands-on killing of intimate partners as a function of sex and relationship status/state. Journal of Family Violence, 24, 463–470. Mueller, U., & Mazur, A. (2001). Evidence of unconstrained directional selection for male tallness. Behavioral Ecology and Sociobiology, 50, 302–311. Muise, A., Christofides, E., & Desmarais, S. (2014). “Creeping” or just information seeking? Gender differences in partner monitoring in response to jealousy on Facebook. Personal Relationships, 21, 35–50. Murdock, G.  P. (1967). Ethnographic atlas. Pittsburgh, PA: University of Pittsburgh Press. Nettle, D. (2002). Height and reproductive success in a cohort of British men. Human Nature, 13, 473–491. Nettle, D. (2006). The evolution of personality variation in humans and other animals. American Psychologist, 61, 622– 631. Nettle, D., & Pollet, T.  V. (2008). Natural selection on male wealth in humans. American Naturalist, 172, 658–666. Oliveira, T., Gouveia, M.  J., & Oliveira, R.  F. (2009). Testosterone responsiveness to winning and losing experiences in female soccer players. Psychoneuroendocrinology, 34, 1056–1064. Park, J. H., Wieling, M. B., Buunk, A. P., & Massar, K. (2008). Sex-specific relationship between digit ratio (2D: 4D) and romantic jealousy. Personality and Individual Differences, 44, 1039–1045. Pawlowski, B. (2003). Variable preferences for sexual dimorphism in height as a strategy for increasing the pool of potential partners in humans. Proceedings of the Royal Society of London B, 270, 709–712. Pawlowski, B., Dunbar, R.  I.  M., & Lipowicz, A. (2000). Tall men have more reproductive success. Nature, 403, 156.

Intersexual and Intrasexual Competition

Pawlowski, B., & Koziel, S. (2002). The impact of traits offered in personal advertisements. Evolution and Human Behavior, 23, 139–149. Pazhoohi, F., & Hosseinchari, M. (2014). Effects of religious veiling on Muslim men’s attractiveness ratings of Muslim women. Archives of Sexual Behavior, 43, 1083–1086. Penke, L., Denissen, J.  J.  A., & Miller, G.  F. (2007). The evolutionary genetics of personality. European Journal of Personality, 21(5), 549–587. Penton-Voak, I.  S., & Chen, J.  Y. (2004). High salivary testosterone is linked to masculine male facial appearance in humans. Evolution and Human Behavior, 25, 229–241. Petersen, J. L., & Hyde, J. S. (2010). A meta-analytic review of research on gender differences in sexuality, 1993–2007. Psychological Bulletin, 136, 21–38. Piccoli, V., Cobey, K.  D., & Carnaghi, A. (2014). Hormonal contraceptive use and the objectification of women and men. Personality and Individual Differences, 66, 44–47. Piccoli, V., Foroni, F., & Carnaghi, A. (2013). Comparing group dehumanization and intra-sexual competition among nor­ mally ovulating women and hormonal contraceptive users. Personality and Social Psychology Bulletin, 39, 1600–1609. Pulerwitz, J., Izazola-Licea, J.  A., & Gortmaker, S.  L. (2001). Extrarelational sex among Mexican men and their partners’ risk of HIV and other sexually transmitted diseases. American Journal of Public Health, 91, 1650–1652. Quinn, T. J., & Wilson, B. R. (1989). Somatotype and type A behavior in college-age adults. Psychological Reports, 65, 15–18. Roff, D. A. (1992). The evolution of life histories. New York, NY: Springer. Rushton, P. J. (1985). Differential K theory: The sociobiology of individual and group differences. Personality and Individual Differences, 6, 441–452. Rustemeyer, R., & Wilbert, C. (2001). Jealousy within the perspective of a self-evaluation maintenance theory. Psychological Reports, 88, 799–804. Ryckman, R. M., Robbins, M. A., Kaczor, L. M., & Gold, J. A. (1989). Male and female raters’ stereotyping of male and female physiques. Personality and Social Psychology Bulletin, 15, 244–251. Saad, G. (2007). A multitude of environments for a consilient Darwinian meta-theory of personality: The environment of evolutionary adaptedness, local niches, the ontogenetic environment and situational contexts. European Journal of Personality, 21, 624–626. Saad, G., & Gill, T. (2001). Sex differences in the ultimatum game: An evolutionary psychology perspective. Journal of Bioeconomics, 3, 171–193. Saad, G., & Vongas, J.  G. (2009). The effect of conspicuous consumption on men’s testosterone levels. Organizational Behavior and Human Decision Processes, 110, 80–92. Sagarin, B. J., Becker, V. D., Guadagno, R. E., Nicastle, L. D., & Millevoi, A. (2003). Sex differences (and similarities) in jealousy: The moderating influence of infidelity experience and sexual orientation of the infidelity. Evolution and Human Behavior, 24, 17–23. Sagarin, B.  J., Martin, A.  L., Coutinho, S.  A., Edlund, J.  E., Patel, L., Skowronski, J.  J., & Zengel, B. (2012). Sex differences in jealousy: A meta-analytic examination. Evolution and Human Behavior, 33, 595–614. Scelza, B.A. (2014). Jealousy in a small-scale, natural fertility population: The roles of paternity, investment and love

in  jealous response. Evolution and Human Behavior, 35, 103–108. Schmitt, D. P. (2003). Universal sex differences in the desire for sexual variety: Tests from 52 nations, 6 continents, and 13  islands. Journal of Personality and Social Psychology, 85, 85–104. Schmitt, D. P., Shackelford, T., & Buss, D. M. (2001). Are men really more “oriented” toward short-term mating than women? A critical review of theory and research. Psychology, Evolution, and Gender, 3, 211–239. Schousboe, K., Visscher, P. M., Erbas, B., Kyvik, K. O., Hopper, J. L., Henriksen, J. E., et al. (2004). Twin study of genetic and environmental influences on adult body size, shape, and composition. International Journal of Obesity, 28, 39–48. Schützwohl, A. (2005). Sex differences in jealousy: The processing of cues to infidelity. Evolution and Human Behavior, 26, 288–299. Schützwohl, A. (2006). Sex differences in jealousy: Information search and cognitive preoccupation. Personality and Individual Differences, 40, 285–292. Schützwohl, A., & Koch, S. (2004). Sex differences in jealousy: The recall of cues to sexual and emotional infidelity in personally more and less threatening context conditions. Evolution and Human Behavior, 25, 249–257. Shackelford, T. K., & Goetz, A. T. (2004). Men’s sexual coercion in intimate relationships: Development and initial validation of the sexual coercion in intimate relationships scale. Violence and Victims, 19, 541–556. Shackelford, T. K., Goetz, A. T., McKibbin, W. F., & Starratt, V. G. (2007). Absence makes the adaptations grow fonder: Proportion of time apart from partner, male sexual psychology, and sperm competition in humans (Homo sapiens). Journal of Comparative Psychology, 121, 214–220. Sheldon, W.  H., & Stevens, S.  S. (1942). The varieties of temperament: A psychology of constitutional differences. Oxford, UK: Harper. Shepperd, J. A., & Strathman, A. J. (1989). Attractiveness and height: The role of stature in dating preference, frequency of dating, and perceptions of attractiveness. Personality and Social Psychology Bulletin, 15, 617–627. Sidanius, J., & Pratto F. (1999). Social dominance. Cambridge, UK: Cambridge University Press. Silventoinen, K., Lahelma, E., & Rahkonen, O. (1999). Social background, adult body-height and health. International Journal of Epidemiology, 28, 911–918. Smith, R.  H., & Kim, S.  H. (2007). Comprehending envy. Psychological Bulletin, 133, 46–64. Stearns, S. C. (1992). The evolution of life histories. Oxford, UK: Oxford University Press. Stöckl, H., & DeVries, K. (2013). Global estimates of homicide risk related to intimate partner violence: Authors’ reply. Lancet, 382, 1625–1626. Stockley, P., & Bro-Jørgensen, J. (2011). Female competition and its evolutionary consequences in mammals. Biological Reviews, 86, 341–366. doi:10.1111/j.1469-185x.2010.00149.x Stulp, G., Buunk, A. P., Kurzban, R., & Verhulst, S. (2013). The height of choosiness: Mutual mate choice for stature results in sub-optimal pair formation for both sexes. Animal Behaviour, 86, 37–46. Tagler, M.  J. (2010). Sex differences in jealousy: Comparing the influence of previous infidelity among college students and adults. Social Psychological and Personality Science, 1, 353–360.

Buunk, Massar, Dijkstra, and Fernande z

235

Tanner, J. M. (1954). Physique and choice of a career. Eugenics Review, 46, 149–157. Tiedens, L.  Z., & Fragale, A.  R. (2003). Power moves: Complementarity in dominant and submissive nonverbal behavior. Journal of Personality and Social Psychology, 84(3), 558–568. Tooby, J., & Cosmides, L. (1992). The psychological foundations of culture. In J. Tooby (Ed.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 19–136). New York, NY: Oxford University Press. Trivers, R. L. (1972). Parental investment and sexual selection. In B.  Campbell (Ed.), Sexual selection and the descent of man: 1871–1971 (pp. 136–179). Chicago, IL: Aldine. Vaillancourt, T. (2013). Do human females use indirect aggression as an intrasexual competition strategy? Philosophical Transactions of the Royal Society B, 368, 20130080. Van Brummen-Girigori, O., & Buunk, A. P. (2016). Intrasexual competitiveness and non-verbal seduction strategies to attract males: A study among teenage girls from Curaçao. Evolution and Human Behavior, 37, 134–141. Van Brummen-Girigori, O., Buunk, A.  P., & Dijkstra, P. (2016). Father abandonment and jealousy: A study among women on Curaçao. Personality and Individual Differences, 96, 181–184. Van der Meij, L., Almela, M., Hidalgo, V., Villada, C., IJzerman, H., van Lange, P. A., & Salvador, A. (2012). Testosterone and cortisol release among Spanish soccer fans watching the 2010 World Cup final. PLoS One, 7, e34814. Van Vugt, M., Roberts, G., & Hardy, C. (2007). Competitive altruism: Development of reputation-based cooperation in groups. In R. I. M. Dunbar & L. Barrett (Eds.), Handbook of evolutionary psychology (pp. 531–540). Oxford, UK: Oxford University Press. Van Yperen, N. W. (2003). Task interest and actual performance: The moderating effects of assigned and adopted purpose goals. Journal of Personality and Social Psychology, 85, 1006–1015. Vandello, J.  A., & Cohen, D. (2003). Male honor and female fidelity: Implicit cultural scripts that perpetuate domestic violence. Journal of Personality and Social Psychology, 84, 997–1010. Veblen, T.B., (1899 [1934]). The theory of the leisure class. New York, NY: Modern Library. Verdonck, P. F., & Walker, R. N. (1976). Body build and behavior in emotionally disturbed Dutch children. Genetic Psychology Monographs, 94(1), 149–173. Vukovic, J., Jones, B. C., DeBruine, L. M., Little, A. C., Feinberg, D. R., & Welling, L. M. (2009). Circum-menopausal effects

236

on women’s judgments of facial attractiveness. Biology Letters, 5, 62–64. Walters, J.  R., & Seyfarth, R.  M. (1987). Conflict and cooperation. In B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham, & T. T. Struhsaker (Eds.), Primate societies (pp. 306–318). Chicago, IL: University of Chicago Press. Walum, H., Larsson, H., Westberg, L., Lichtenstein, P., & Magnusson, P.  E. (2013). Sex differences in jealousy: A population-based twin study in Sweden. Twin Research and Human Genetics, 16, 941–947. Wasser, S. K., & Barash, D. P. (1983). Reproductive suppression among female mammals: Implications for biomedicine and sexual selection theory. Quarterly Review of Biology, 58, 513–538. Weisfeld, G. E., & Linkey, H. E. (1985). Dominance displays as indicators of a social success motive. In S.  L.  Ellyson &  J.  F.  Dovidio (Eds.), Power, dominance, and nonverbal behavior (pp. 109–128). New York, NY: Springer-Verlag. Welling, L.  M., Puts, D.  A., Roberts, S.  C., Little, A.  C., & Burriss, R. P. (2012). Hormonal contraceptive use and mate retention behavior in women and their male partners. Hormones and Behavior, 61, 114–120. Wilson, P.  R. (1968). Perceptual distortions of height as a function of ascribed academic status. Journal of Social Psychology, 74, 97–102. Wilson, M., & Daly, M. (1993). An evolutionary psychological perspective on male sexual proprietariness and violence against wives. Violence and Victims, 8, 271–294. Wrangham, R. W., & Peterson, G. (1996). Demonic males: Apes and the origins of human violence. Boston, MA: Houghton, Mifflin and Company. Yeagley, E., Morling, B., & Nelson, M. (2007). Nonverbal zeroacquaintance accuracy of self-esteem, social dominance orientation, and satisfaction with life. Journal of Research in Personality, 41(5), 1099–1106. Zahavi, A., & Zahavi, A. (1997). The handicap principle: A missing piece of Darwin’s puzzle. Oxford, UK: Oxford University Press. Zandbergen, D. L., & Brown, S. G. (2015). Culture and gender differences in romantic jealousy. Personality and Individual Differences, 72, 122–127. Zuckerman, M. (1986). On the meaning and implications of facial prominence. Journal of Nonverbal Behavior, 10, 215–229. Zuckerman, M., & O’Loughlin, R. E. (2006). Self-enhancement by social comparison: A prospective analysis. Personality and Social Psychology Bulletin, 32, 751–760.

Intersexual and Intrasexual Competition

CH A PT E R

14

Synthetic Hormones The Influence of Hormonal Contraceptives and Hormone Replacement Therapy on Aspects of Women’s Mating Psychology

Amanda C. Hahn and Kelly D. Cobey

Abstract Following the invention of the hormonal contraceptive pill in the mid-20th century, there has been a rise in exogenous hormone use worldwide. Across the lifespan, many women will utilize synthetic hormones in the form of hormonal contraceptives and/or hormone replacement therapy. It is estimated that 100 million women worldwide use combined oral contraceptives, whereas 20 million women worldwide use hormone replacement therapy. Although extensive research has been done investigating the health side effects of these synthetic hormones, relatively little is known about their potential cognitive and behavioral consequences. This chapter reviews knowledge regarding the potential impact of these synthetic hormones on women’s psychology. Much of this work derives from the field of evolutionary psychology, which considers potential adaptive functions of behavior and their underlying mechanisms. This chapter emphasizes the need for randomized within-subject clinical trials to better understand the true causal effects they may have on women’s behavior. Keywords:  hormonal contraceptives, pill, estrogen, progesterone, testosterone, hormone replacement therapy

The notion that women’s mating psychology is ­influenced by their hormonal status has received a great deal of attention from both scientists and the media over the last several decades. Early work in this area relied primarily on inferred or estimated hormone levels by focusing on changes across the menstrual cycle. This work produced landmark papers in Nature (Penton-Voak et al., 1999), Psychological Science (Gangestad, Simpson, Cousins, Garver-Apgar, & Christensen,  2016; Macrae, Alnwick, Milne, & Schloerscheidt, 2016), Proceedings of the Royal Society Biology (Gangestad, Thornhill, & Garver,  2002), and Proceedings of the National Academy of Sciences (Dreher et al.,  2007), and has been the source of much recent debate (see Welling & Burriss, this volume, for an in-depth review of this topic). More recent work has begun to focus on measured hormone levels in an attempt to clarify the potential underlying hormonal mechanisms of observed cyclical shifts in women’s mating psychology (Ditzen,

Palm-Fischbacher, Gossweiler, Stucky, & Ehlert, 2017; Grebe, Emery Thompson, & Gangestad, 2016; Hahn, Fisher, Cobey, DeBruine, & Jones, 2016; Holzleitner et al., 2017; Roney & Simmons, 2013, 2016), including several “big N” studies using very large samples (e.g., Jones, Hahn, Fisher, Wang, Kandrik, Han, et al.,  2018; Jones, Hahn, Fisher, Wang, Kandrik, & DeBruine, 2018; Marcinkowska, Galbarczyk, & Jasienska, 2018). Although new research continues to elucidate the role of endogenous (i.e., naturally occurring) hormones in regulating behavior, modern medicine has allowed for the artificial influence of hormones through the consumption of synthetic, exogenous hormones—the most widespread examples of this are the use of the combined oral contraceptive pill (“the pill”) and hormone replacement therapy (HRT). The current chapter explores our knowledge of how the use of these two interventions ­impacts women’s psychology, with a focus on aspects 237

of women’s relationship or mating psychology. After providing an overview of the current evidence for how these interventions are thought to influence women’s psychology, we provide a critique of the study designs used to generate this evidence base. We conclude with a call to action for researchers to remedy these design limitations to improve the quality of evidence in this area. Given their widespread use, it is our view that women would benefit from continued discovery on how intake of exogenous hormones impacts their psychology.

pill report discontinuing use due to side effects (Mosher & Jones, 2010). In light of the ubiquitous nature of hormonal contraceptives, and the pill in particular, it is critical that scientific research works to better understand how synthetic hormones may or may not impact women’s psychology. Here, we outline existing evidence on this topic. We emphasize findings describing effects of the COC pill, simply because such studies are most represented in the literature.

Combined Oral Contraceptives—“The Pill”

Hormonal contraceptives alter endogenous hormone levels through the introduction of exogenous steroid hormones, namely, estrogens and progestins. All forms of hormonal contraception contain ­synthetic progestins, whereas not all forms of hormonal contraception contain synthetic estrogens. Forms of hormonal contraception that administer both estrogen and progestins are termed combined (oral) contracep­ tives. Hormonal contraceptive use leads to a reduction of endogenous estradiol and progesterone levels via negative feedback mechanisms (Sahlberg, Landgren, & Axelson,  1987) and have been shown to suppress endogenous testosterone levels (Zimmerman, Eijkemans, Coelingh Bennink, Blankenstein, & Fauser, 2014; see Box 14.1). Estrogen, progesterone, and testosterone have all been implicated in the regulation of social-emotional behavior and women’s psychology (Bos, Panksepp, Bluthé, & Honk, 2012; Bos, Terburg, & van Honk, 2010; Ellison & Gray, 2009; Garver-Apgar, Gangestad, & Thornhill, 2008; Hahn et al., 2016; Pisanski et al., 2014; Roney & Simmons, 2013; Roney, Simmons, & Gray, 2011; Van Wingen, Ossewaarde, Bäckström, Hermans, & Fernández, 2011; Wang, Hahn, Fisher, DeBruine, & Jones, 2014; Welling et al., 2007). The suppression of these endogenous hormones and introduction of synthetic exogenous hormones could, therefore, result in potential b­ ehavioral and/or psychological side effects of hormonal contraceptive use.2 Indeed, recent reviews have suggested that synthetic exogenous hormones may affect the regulation of affective responses and social-emotional b­ ehavior (Hamstra, De Rover, De Rijk, & Van der Does, 2014; Montoya & Bos, 2017; Oinonen & Mazmanian,  2002; Radke & Derntl, 2016).

Described as the invention that defined the 20th century,1 “the pill” was first introduced in the early 1960s as a method of preventing unintended pregnancy (Burrows, Basha, & Goldstein,  2012). This medical advancement revolutionized the lives of women, allowing them increased access to education and a stronger foothold in the workforce (Bailey, 2006; Traulsen, Haugbølle, & Bissell, 2003). Hormonal contraception is even credited with narrowing the gender wage gap (Bailey & Hershbein, 2012; Chiappori & Oreffice, 2015). Since its induction, the pill has been the leading form of contraception in the United States (Mosher, Martinez, Chandra, & Abma,  2004), and is used by more than 100 million women worldwide (Petitti, 2003). The pill is now available in many forms and dosages, and combined contraceptives can be delivered via alternative routes of administration (e.g., the transdermal contraceptive patch, the vaginal contraceptive ring). Although we will touch on hormonal contraceptives other than the combined oral contraceptive (COC) pill, COCs are the focus of this review, given that the vast majority of ­research regarding potential psychological effects of contraceptive use has focused on women using COCs. In the United States, a staggering 82 percent of women have used the pill at some point in their lifetime (Mosher & Jones, 2010), and research suggests that modern adolescent girls have begun using hormonal contraceptives earlier in the lifespan, typically shortly after the onset of puberty (Parkes, Wight, Henderson, Stephenson, & Strange, 2009). Although physical side effects of hormonal contraceptive use have been the focus of a tremendous amount of research (reviewed in Welling,  2013), relatively little is known about the potential psychological consequences of hormonal contraceptive use. Yet, 30 percent of women who have used the 1

 http://www.economist.com/node/347484.

238

Synthetic Hormones

How Do Hormonal Contraceptives Work?

2  Assuming synthetic, exogenous hormones affect neural path­ways in a similar way to endogenous hormones, allowing for some effect on psychological outcomes. That these synthetic hormones are able to act on brain systems to suppress ovulation centrally (Frye,  2006) suggests they are, indeed, capable of affecting neural pathways.

Box 14.1.  Ovarian Hormone Levels During the Menstrual Cycle and Combined Oral Contraceptive Use The typical menstrual cycle is approximately 28 days in length and is characterized by fluctuations in the ovarian steroid hormones estradiol (hereafter simply referred to as estrogen) and progesterone. The cyclical fluctuation in  these hormones is regulated by the hypothalamic-pituitary-ovarian (HPO) axis. The cycle begins at the onset of menstrual bleeding (i.e., the menstrual phase), at which time both estrogen and progesterone levels are low. After menstrual bleeding ceases, the follicular phase begins and follicle-stimulating hormone (FSH), released from the pituitary, promotes the growth of immature egg follicles in the ovary. The follicular phase is characterized by a steady increase in estrogen levels that peak just before ovulation. A surge in luteinizing hormone (LH) instigates the release of a mature follicle at ovulation, typically around day 14 of the cycle. The corpus luteum, a temporary structure, forms at the follicle release site and produces high amounts of progesterone (and some estrogen) in the second half of the cycle, referred to as the luteal phase. Note that although estrogen levels may dip after ovulation, they remain relatively high during the luteal phase. If implantation of a fertilized egg does not occur, the corpus luteum regresses, causing both progesterone and estrogen levels to steeply decline in the final days of the cycle. This hormone decline triggers the onset of menstruation and the beginning of a new menstrual cycle. The primary mechanism of action of the combined oral contraceptive (COC) pill is to suppress ovulation (American College of Gynecologists and Obstetricians, 1998) via negative feedback mechanisms that block the release of the gonadotropins (i.e., FSH and LH) that normally promote growth and release of the ovarian follicle each month. COCs contain exogenous estrogen and progestin that suppress the HPO axis. The suppression of these hormones decreases ovarian activity and results in lower levels of endogenous estrogen and progesterone (in the context of high levels of exogenous estrogen and progestin). Notably, however, reactivation of the HPO axis can occur during the pill-free interval, called the pill break, that is characteristic of many COC regimens (Van Heusden & Fauser, 1999).

Hormone concentration

(A)

5

10

14

22

25

(B)

5

10 14 22 Menstrual cycle (days)

Figure 14.1  Representative levels of endogenous estrogen (gray lines), progesterone (black lines), luteinizing hormone (LH; light gray lines), and follicle-stimulating hormone (FSH; dotted lines) levels across the menstrual cycle in (A) normally cycling women and (B) women using combination oral contraceptive (COCs). By providing a steady daily level of both progestin (a substitute for progesterone) and ethyl estradiol (a substitute for endogenous estradiol), oral contraceptives prevent gonadotropin-releasing hormone (GnRH) secretion from the hypothalamus, blocking a signal to the pituitary gland to produce FSH and LH. Because FSH stimulates the ovaries to grow egg follicles and LH triggers ovulation, their absence causes the ovary to be relatively dormant, and no egg is produced to a point where it could be released. Hormonal contraception thus maintains the menstrual cycle at the same late phase of the natural cycle on a continuous basis. Image & caption adapted with permission from Alvergne and Lummaa (2010).

Hahn and Cobey

239

Hormonal Contraception and Mate Preferences

Perhaps the most well-researched aspect of potential COC effects on women’s psychology is of the study of mate preferences (reviewed in Alvergne & Lummaa, 2010; Welling,  2013). Popular media commonly makes bold claims about the potential behavioral side effects of hormonal contraceptives based on ­research on this topic—claims including “Women choose different partners when on the Pill,”3 “Birth control pills affect women’s taste in men,”4 and “How the Pill could be seriously ruining your love life,”5 to highlight a few. Although these headlines imply a causal link between the use of synthetic hormones and changes in women’s psychology, relatively little causal evidence actually exists. Existing studies on this topic suggest that women using COCs may have a greater interest in  shortterm mating opportunities (Guillermo, Manlove, Gray, & Zava, 2010) and may not show the same cyclical fluctuation in putative cues of men’s mate quality including masculinity preferences (Feinberg, DeBruine, & Jones,  2008; Puts,  2006; Smith et al., 2009), symmetry preferences (Gangestad & Thornhill, 1998; Thornhill, 1999), and preferences for cues of health (Jones et al.,  2005), including preferences for major histocompatibility complex (MHC) dissimilarity (Roberts, Gosling, Carter, & Petrie, 2008; Wedekind & Seebeck, 1995). Because these mate preferences are thought to be adaptive (i.e., potentially enhance reproductive success and/ or offspring quality), some researchers have argued that COC use may disrupt women’s adaptive mate preferences, which could have downstream reproductive consequences (reviewed in Welling, 2013), including ­effects on relationship stability (Birnbaum, Birnbaum, & Ein-Dor, 2017; Roberts et al., 2012; Taggart, Hammett, & Ulloa, 2016), sexual behavior (Cobey, Havlicek, Klapilová, & Roberts,  2016; Roberts et al., 2012), and potentially even offspring health (Birnbaum et al., 2017). Indeed, if COCs are affecting women’s mate preferences, it is possible that attraction toward an existing partner could change over time if a woman alters her contraceptive status (i.e., initiates or discontinues COC use). This possibility raises the potential for psychological

3  http://www.telegraph.co.uk/women/sex/9173543/Womenchoose-different-partners-when-on-the-Pill.html 4  https://www.scientificamerican.com/article/birth-controlpills-affect-womens-taste/ 5  http://www.cosmopolitan.com/uk/love-sex/relationships/ news/a31308/gone-right-off-him-blame-the-pill/

240

Synthetic Hormones

side effects of the pill not currently addressed by ­traditional medical research looking at side effects during use. Retrospective studies investigating this issue have found that COC use does appear to be related to relationship stability, or lack thereof (Birnbaum et al., 2017; Roberts et al.,  2012). Roberts et al. (2012) found that women who used COCs at the time of partner choice reported lower sexual satisfaction (across multiple measures) than did women who did not use COCs at the time of partner choice (note, however, that these same women reported higher general satisfaction with their partner choice than did women not using COCs at the time of partnership formation). However, another study by this same group found that women who met their partner while using COCs reported significantly higher partner-specific attraction and sexual desire during pregnancy than did women who met their partner while naturally cycling (Cobey et al., 2016). This seemingly opposing pattern of results has been attributed to changes in contraceptive congruency, rather than simply COC use, whereby women are predicted to be more satisfied with their partner choice if their current contraceptive or hormonal status is congruent with their status at the time of relationship formation (see Roberts et al., 2014 and Jern et al., 2018, for an expanded discussion of the congruency hypothesis). With respect to relationship stability, results are again mixed. Roberts et al. (2012) found that separation rates were lower in couples who met while the woman was using COCs—23.6 percent of couples who met when the woman was using COCs separated compared to a 33.3 percent separation rate among couples who met when the woman was not using COCs. A more recent study, however, found the opposite pattern of results, reporting separation among 11.5 percent of the couples who met while the woman was using COCs compared to a 3.5 percent separation rate among couples who met when the woman was not using COCs (Birnbaum et al.,  2017). Curiously, ­although they did not find that relationship dissolution was more common among COC users, Roberts et al. (2012) did find that if a separation occurred for those couples who met when the woman was using COCs, the woman was disproportionately more likely to have initiated the separation (84.8 percent of separations initiated by the woman, whereas only 73.6 percent of separations among the non-COC group were initiated by the woman). These results suggest that COC use could have some effect on partner choice; however, the question

remains as to whether or not COC use actually causes any change in women’s mate preferences. Few studies have explored this issue within women, and those that have yield equivocal results. Little, Burriss, Petrie, Jones, and Roberts (2013) examined preferences for cues of facial masculinity in 18 women before COC use (during the follicular phase of their menstrual cycle, when masculinity preferences are believed to be at their peak) and after initiating COC use. Here, participants were asked to alter men’s faces along a masculinity continuum to make the face most attractive in a short- and longterm mating context (i.e., all participants completed both the short- and long-term task). When comparing the women’s masculinity preferences before and during COC use, they found that women’s masculinity preferences decreased following COC initiation, regardless of the relationship context sought. Similarly, Roberts et al. (2008) found that preferences for MHC dissimilarity in odor samples ­decreased in a sample of 37 women who initiated COC use, although they did not detect any significant MHC-associated preferences at either the preor post-COC initiation test sessions. Together, these two studies do provide evidence for the notion that initiation of COC use may cause a disruption in women’s adaptive mate preferences. However, a recent study by Jones, Hahn, Fisher, Wang, Kandrik, Han, et al. (2018) tested for potential COC use effects on women’s facial masculinity preferences in both a short-term and long-term mating context, using multiple measures during both COC use and a period without COC use in a sample of 45 women. They did not detect any changes in women’s facial masculinity preferences as a function of COC use. This study incorporated both women who initiated COC use, testing their preferences before and after using COCs, and women who ceased COC use, testing their preferences during COC use and after they had discontinued using COCs. Regardless of the direction of the change (i.e., initiating or discontinuing COCs), no change in women’s masculinity preferences was observed. This study also reports analyses of women’s preferences for additional facial cues, including facial symmetry, facial prototypicality, and perceived health (see study supplemental materials) and again found no effect of COC use on women’s preferences for any of these traits. It is difficult to draw any definitive conclusions about the impact of synthetic hormones on women’s mate preferences from the limited longitudinal research conducted thus far. The evidence to date regarding the causal effects of COC use on women’s

mate preferences is ambiguous, often limited by use of between-participant study designs, low power, and a failure to control for COC brand and other potential confounding variables. Comparing women who use the pill to women who do not is problematic as these groups of women may differ from one another on a whole host of factors, including age, relationship status, sexual activity, sociosexual orientation, and so forth. These potentially confounding variables are not always controlled for in the ­existing literature, especially within the same model to account for potential interaction effects. Although there are numerous clinical trials addressing the impact of COCs on women’s physiology, comparably powered randomized trails addressing the potential for changes in women’s psychology and behavior during COC use are only just beginning to be conducted.

Hormonal Contraception and Sexual Behavior

Sexual behavior6 is another aspect of women’s ­psychology that may potentially be subject to the influence of synthetic hormones given the reported impact of circulating endogenous hormone levels on various aspects of female sexual behavior in both humans (Grebe, Emery Thompson, & Gangestad, 2016; Jones, Hahn, Fisher, Wang, Kandrik, & DeBruine, 2018; Matteo, 1984; Roney & Simmons, 2013; see Cappelletti & Wallen,  2016, or MottaMena & Puts, 2017, for recent reviews of this literature) and nonhuman primates (Wallen, 1984, 1990, 2001). Indeed, concern about the impact of COC use on women’s sexual behavior, particularly concern regarding decreased libido, has been expressed for almost as long as hormonal contraceptives have been available. An early report from the Royal College of General Practitioners Survey (1974) found that women using the COC pill were nearly four times more likely to complain of sexual difficulties than women using other, nonhormonal methods of contraception. Since this early report, a number of between-subject studies have reported similar decreases in aspects of sexual behavior in women using the COC pill as compared to women not using hormonal contraception (Adams, Gold, & Burt, 1978; Alexander, 1990; Bancroft & Sartorius, 1990; McCoy & Matyas, 1996; Wallwiener et al., 2010). One potential explanation for the observed negative effects of COCs on women’s sexual behavior 6   The term sexual behavior is used in this review to broadly encompass sex drive or libido, sexual activity, and sexual functioning.

Hahn and Cobey

241

is that COCs could be altering women’s sexual ­cognition. Studies comparing attention to sexual stimuli (Rupp & Wallen,  2007) and neural responses to sexual stimuli (Abler et al.,  2013) or images of one’s romantic partner (Scheele, Plota, & Stoffel-Wagner, 2015) among COC users and nonusers have suggested that COC use may be linked to decreased processing of sexually relevant stimuli. Although many studies report a decline in women’s sexual behavior that is assumed to be the result of COC use, systematic reviews of the existing literature indicate that the evidence for an effect of COCs on female sexual behavior is highly mixed, including reported negative and positive effects, ­reported mixed effects, and/or reports that there are no effects of COCs on sexual behavior (Bancroft & Sartorius, 1990; Bancroft, Sanders, Warner, & Loudon, 1987; Burrows et al.,  2012; Davis & Castaño, 2012; Dei, Verni, Bigozzi, & Bruni, 1997; Pastor, Holla, & Chmel, 2013; Robinson, Dowell, Pedulla, & McCauley, 2004; Zethraeus et al., 2016). Indeed, one review noted that COC users reported an increase in sexual desire in 15 studies, no impact on sexual desire in 12 studies, and a decrease in sexual desire in 9 studies (reviewed in Pastor et al., 2013). In addition to these mixed findings across studies, there is also often considerable variation within the population of a single study, whereby large proportions of women within a study sample experience an increase or decrease in desire, whereas others are apparently unaffected by COC use (Burrows et al.,  2012; Davis & Castaño,  2012; Pastor et al.,  2013; Schaffir,  2006). For example, in a study of 61 women, Graham, Bancroft, Doll, Greco, and Tanner (2007) found that one-third of  the women tested reported negative sexual side effects, one-third reported positive sexual side effects, and one-third reported no sexual side effects associated with COC use. One possibility for these mixed findings is that there are multiple forms of COC, meaning that there is no single “pill” to study—the available COCs on the market today differ vastly in terms of  both their chemical composition and delivery schedule (Mark, Leistner, & Garcia, 2016). A recent study found that women in the United States alone reported using over 80 different COC brands (Hall & Trussell, 2012), highlighting the potential variability that may exist within a study sample. Another possibility is that differences in methodology between studies, or in instrumentation used to assay sexual desire and related variables, may contribute to the lack of clarity regarding an effect of COCs on 242

Synthetic Hormones

women’s sexual behavior. Different patterns of results have been observed when considering uncontrolled, prospective studies versus placebo-controlled studies (Davis & Castaño, 2012). Additionally, differences in the length of time women have been using COCs when post-COC initiation data is collected may further influence reported findings. For example, Caruso et al. (2011) reported significant improvements in aspects of women’s sexual functioning when comparing women’s responses before COC use and six cycles after COC initiation. The majority of these effects, however, were nonsignificant when the same women were assessed only three cycles after COC initiation. There may even be placebo effects to consider; one study administered a contraceptive placebo to 147 young women. These women were surveyed monthly over the next year and 29.5 percent reported decreased libido due to their “pill use” (Aznar-Ramos, Giner-Velázquez, Lara-Ricalde, & Martínez-Manautou, 1969). These mixed findings have led researchers to often conclude that the effects of COC use on women’s sexual behavior are complex and likely due to a combination of psychological, biological, and social factors. Although undoubtedly complex, it is critical that researchers continue to investigate the potential impact of COC use on women’s sexual behavior given that perceived changes in sexual behavior, ­especially libido, are often reported as important contributors to discontinuation of COCs and/or switching contraceptive methods (Sanders, Graham, Bass, & Bancroft,  2001). Because the quality of evidence obtained from a well-designed within­ subject study is higher than that of a between-subject study, we will emphasize these studies for the purposes of this review. Several studies have explored potential changes in women’s sexual behavior utilizing a within-subject design, although methods for assessing sexual behavior vary across these studies, including interviewer ratings of sexual functioning and self-administered questionnaires to assess various aspects of sexual ­behavior (Bancroft et al., 1987; Caruso et al., 2004, 2005; Graham, Ramos, Bancroft, & Maglaya, 1995; Graham & Sherwin, 1993; Greco, Graham, Bancroft, Tanner, & Doll, 2007; Sabatini & Cagiano, 2006; Sanders et al., 2001; Strufaldi et al., 2010). Although these studies report mixed findings regarding various aspects of sexual behavior (see Table  14.1), there is some evidence for a negative impact of COC initiation on sexual desire or interest (Caruso et al., 2004; Graham et al., 1995; Graham & Sherwin,  1993; Sabatini & Cagiano, 2006;

Table 14.1.  Summary of Findings Across Within-Subject Studies Exploring the Impact of Combined Oral Contraceptive Initiation on Aspects of Women’s Sexual Behavior Study

N

Assessment of sexual behavior

COC formulation

Sexual desire or interest

Sexual activity

Sexual arousal

Sexual enjoyment or satisfaction

Orgasm ­frequency

Graham et al., 1993

45

Visual analog scale ratings

35 µg EE 0.1 mg norethisterone

Decreased

—

—

—

—

Graham et al., 1995

50

IRSF SEQ

30 µg EE 0.15 mg levonorgestrel

No change

—

76

IRSF

30 µg EE 0.25 mg norgestimate OR 0.180 mg, 0.215 mg, 0.250 mg norgestimate

Decreased (Scottish sample) Decreased

—

Sanders et al., 2001

Decreased (marginal effect in Filipino sample) Decreased

—

—

—

SEQ Caruso et al., 2004

48

PEQ

15 µg EE 60 µg gestodene

Decreased

Decreased

Decreased

Decreased

No change

Caruso et al., 2005

80

PEQ

30 µg EE 3 mg drospirenone

No change

Increased

Increased

Increased

Increased

Guida et al., 2005

28

IRSF

20 μg of EE 150 μg desogestrel

Increased

Increased

—

Increased

Increased

Oranratanaphan et al., 2006)

42

FSFI

30 µg EE 3 mg drospirenone

Increased

—

Increased

Increased

No change

Oranratanaphan et al., 2006

44

FSFI

Increased

—

Increased

Increased

 

Sabatini et al, 2006

94

Self-report

20 μg of EE 75 μg gestodene

20 µg EE 100 µg levonorgestrel

Decreased

—

—

Decreased

—

Sabatini et al., 2006

92

Self-report

15 µg EE 60 µg gestodene

Decreased

—

—

Decreased

—

Graham et al., 2007

61

IRSF SDI

25 µg or 35 µg EE 0.18, 0.215, and 0.25 mg norgestimate

No change

No change

No change or increased (groups lumped)

No change or increased (groups lumped)

No change

(continued )

Table 14.1.  Continued Study

N

Assessment of sexual behavior

COC formulation

Sexual desire or interest

Sexual activity

Sexual arousal

Sexual enjoyment or satisfaction

Orgasm ­frequency

Greco et al., 2007

24

SDI SEQ

25 µg EE 0.18, 0.215, and 0.25 mg norgestimate

No change

—

—

—

—

Greco et al., 2007

24

SDI SEQ

35 µg EE 0.18, 0.215, and 0.25 mg norgestimate

No change

—

—

—

—

Westhoff et al., 2007

1716

Interview

Varied

—

—

—

No change

—

Caruso et al., 2009

56

SPEQ

30 µg EE 2 mg chlormadinone a­ cetate

No change

—

No change

Increased

Increased

Brucker et al., 2010

1063 Interview

20 µg EE 2 mg chlormadinone a­ cetate

No change

—

—

—

—

Strufaldi et al., 2010

49

FSFI

30 µg EE 150 µg levonorgestrel

Increased

—

No change

Increased§

No change

Strufaldi et al., 2010

48

FSFI

20 µg EE 100 µg levonorgestrel

Increased§

—

No change

No change

No change

Caruso et al., 2011

54

SPEQ 3 cycles post-COC initiation

20 µg EE 3 mg drospirenone 21/7 regime

No change

No change

No change

No change

No change

Caruso et al., 2011

54

SPEQ

20 µg EE

No change

Increased

Increased

Increased

No change

6 cycles post-COC initiation

3 mg drospirenone 21/7 regime

Caruso et al., 2011

61

SPEQ 3 cycles post-COC initiation

20 µg EE 3 mg drospirenone 24/4 regime

Increased

No change

Increased

No change

Increased

Caruso et al., 2011

61

SPEQ 6 cycles post-COC initiation

20 µg EE 3 mg drospirenone 24/4 regime

Increased

Increased

Increased

Increased

Increased

Battaglia et al., 2012

22

MFSQ

30 µg EE

–-

Decreased

—

—

Decreased

3 mg drospirenone Studies reporting on combined contraceptives with alternate administration routes Guida et al., 2005

20

IRSF

NuvaRing 15 μg/day EE 120 μg/day etonogestrel

Increased

Increased

—

Increased

Increased

Sabatini et al., 2006

94

Self-report

NuvaRing 15 μg/day EE 120 μg/day etonogestrel

Increased

—

—

Increased

—

Note: A study may appear multiple times if it reports analyses of multiple COC brands. § A reported effect that failed to reach statistical significance. FSFI, Female Sexual Function Index; IRSF, Interviewer Ratings of Sexual Functioning; MFSQ, McCoy Female Sexuality Questionnaire; SDI, Sexual Desire Inventory; SEQ, Side-Effects Questionnaire; (S)PEQ, (Short) Personal Experience Questionnaire.

Sanders et al., 2001; Zethraeus et al., 2016), sexual activity (Caruso et al., 2004; Graham et al., 1995; Sanders et al., 2001), sexual arousal (Caruso et al., 2004; Zethraeus et al., 2016), and enjoyment or satisfaction (Caruso et al.,  2004; Sabatini & Cagiano, 2006). The evidence for a potential negative impact of  COC initiation on women’s sexual behavior is strongest for assessments of sexual desire or interest, although methodological differences in assessing sexual behavior may contribute to conflicting findings here. Studies using interviewer ratings of sexual functioning (IRSF) and those using self-­ administered questionnaires regarding sexual desire have consistently reported negative effects of COC initiation on sexual desire/interest (Caruso et al., 2004; Graham et al.,  1995; Graham & Sherwin, 1993; Sabatini & Cagiano,  2006; Sanders et al., 2001); however, one study utilizing the Female Sexual Function Index (FSFI) reported a positive impact of COC initiation on sexual desire/interest in two separate samples (Strufaldi et al.,  2010). Importantly, however, many of these same studies report no effect of COC initiation on some of the individual aspects of sexual behavior assessed (Caruso et al., 2004, 2005; Graham et al., 1995; Greco et al., 2007; Strufaldi et al., 2010). The findings from Graham et al. (1995) suggest that cultural and/or individual differences may partially contribute to the discrepant findings reported across studies. They performed a placebo-controlled, double-blind study to explore differences among COC and progestin (P)-only pill users (note: only the results from the COC group are discussed here). Within the COC group, 25 of the women were ­recruited in Scotland and 25 women were recruited in the Philippines. A negative effect of COC initiation on sexual activity and sexual interest was consistently observed across the four months following COC initiation, but only in the Scottish women. No change in sexual interest or activity was reported among the Filipino women. The authors offer two possible explanations for this cultural difference— the first, that the Scottish sample reported more positive sexual experiences at baseline than did the Filipino sample, which would allow for a greater scope of negative outcomes for the Scottish women; the second, that any lack of consistency in assessment across the samples could have impacted the accuracy of the results, especially in the Filipino sample, where translation is an issue. Another possibility, however, is that COCs may affect individuals differently and such differences may be impacted by 246

Synthetic Hormones

regional and ethnic variation. However, relatively few studies have investigated potential individual or cultural differences. One study has suggested that prenatal testosterone exposure, measured through the second-to-fourth-digit ratio (2D:4D), may be linked to the experience of negative side effects and, of particular relevance here, the experience of changes in sex drive (Oinonen, 2009). Another study has reported evidence of individual differences in COC-related changes in affect, with women younger than age 20 reporting greater negative affect following COC initiation (Bancroft et al., 1987). Although these results add to an already-complex picture, they highlight the importance of considering individual factors when assessing potential psychobehavioral consequences of synthetic hormones.

Hormonal Contraception and Other Aspects of Relationship Behavior

Although sexual behavior and mate preferences are among the best studied components of the psychobehavioral effects of synthetic hormones, a number of additional aspects of relationship-related behavior have also been studied in the last few ­decades. Intrasexual competition (i.e., competition between members of one sex for access to mating opportunities with members of the opposite sex; Andersson, 1994) occurs in females of many species and has been shown to be influenced by endogenous hormone levels (Cobey & Hahn, 2017), suggesting that this aspect of relationship behavior may also be susceptible to the influence of synthetic hormones. Indeed, nonhuman primates have shown COClinked decreases in intrasexual competition for access to a male conspecific (Shively, Manuck, Kaplan, & Koritnik, 1990) and elimination of cyclical changes in aggressive behaviors (Sarfaty, Margulis, & Atsalis, 2012). In humans, COC initiation has similarly been shown to reduce intrasexual competition, at least in partnered women (Cobey, Klipping, & Buunk, 2013). In a sample of 14 partnered women tested before and after COC initiation, Cobey et al. (2013) observed a significant reduction in reported intrasexual competition following COC initiation. Intriguingly, although intrasexual competition appears to be decreased by COC use, self-reported mate ­retention behaviors have been observed to be higher in COC users than nonusers (Welling, Puts, Roberts, Little, & Burriss, 2012), although this has yet to be investigated within subject as a function of COC initiation. Another aspect of relationship behavior that may be influenced by both endogenous and exogenous

hormones is relationship jealousy (i.e., thoughts, emotions, or behaviors that occur as a result of the perceived threat of losing a potential mate to an actual or imagined rival). One early study reported that COC users report greater levels of relationship jealousy than do nonusers (Geary, DeSoto, Hoard, & Sheldon,  2001). Subsequent work utilizing a within-subject design found that reported levels of jealousy were not affected by COC initiation when compared to those same women’s reported jealousy during the fertile phase of their natural menstrual cycle, but were significantly increased following COC initiation when compared to those same women’s reported jealousy during the nonfertile phase of their natural menstrual cycle (Cobey et al., 2012). The study sample was modest and the finding was also qualified by an interaction with relationship status, whereby the impact of COC initiation was only seen among partnered women.

Effects of Dosage, Formulation, and Route of Administration

Since its introduction in the early 1960s, the pill has undergone significant changes in formulation. The initial pill of the early 1960s was G. D. Searle & Company’s Enovid, a combination pill containing 150 μg of mestranol (an estrogen) and 10 mg of norethynodrel (a progestin), approved by the U.S. Food and Drug Administration (FDA) in 1959 for distribution in the United States. Shortly thereafter, Bayer Schering introduced Anovlar, a combined pill containing 50 μg ethinylestradiol (an estrogen) and 4 mg norethisterone acetate (a progestin), which became available in Australia and Europe in 1961 (Rabe et al., 2011). The combined oral contraceptive pill has continued to be developed since its initial release; today, the combined oral contraceptive pill contains lower concentrations of the synthetic hormones than were present in its initial form. Current formulations typically contain between 20 and 35 μg ethinylestradiol (EE) and varying levels of synthetic progestins, classified by their generation (Pletzer & Kerschbaum,  2014). The older generation progestins (e.g., desogestrel, levonorgestrel) are androgenic (i.e., they are capable of exerting androgenic effects), whereas the newer progestins (e.g., drospirenone) are antiandrogenic and bind specifically to the progesterone receptor. Additionally, there are monophasic pills, which ­deliver a constant level of synthetic hormones across the 21-day regimen, and multiphasic pills, which deliver various phases of progesterone doses across the 21-day regimen. In light of these differences in

both the specific synthetic progestin used and ­variation in the dosage (of both estrogen and progestin) across different brands of the combined oral contraceptive pill, collapsing data from all “pill users” into a single, homogenous group may present a significant design concern. A large-scale study of COC use in the United States recently demonstrated that women aged 15 to 44 report using over 80 different brands of COCs (Hall & Trussell, 2012). In this sample of over 12,000 women, high-dose estrogen pills were more common than low-dose, 58 percent of women used a COC with an older generation progestin, and two-thirds used monophasic pills. This study highlights the importance of considering the specific COC used in future research, and of the need for within-participant ­research examining the potential differential impact of different COCs. Recent studies have begun to investigate potential differences among contraceptive users as a function of the estrogen dosage of the combined oral contraceptive used. Circulating levels of EE have been positively linked to aspects of sexual desire in young women (Jones, Hahn, Fisher, Wang, Kandrik, & DeBruine,  2018; Roney & Simmons,  2013), suggesting that COCs with higher doses of EE may be associated with more positive outcomes regarding women’s sexual desire or interest. The EE dosage varies from 15–5 μg in the studies summarized in Table  14.1. Although both studies reporting the lowest EE dose (15 μg EE) were associated with negative effects on reported sexual desire/interest, it should also be noted that women who initiated use of the vaginal ring, which also contains 15 μg/day EE, report positive effects on various aspects of their sexual behavior (Guida et al.,  2005; Sabatini & Cagiano, 2006). Studies investigating the effects of COCs with higher EE doses were not consistently associated with positive effects on women’s sexual desire/interest or other aspects of sexual behavior. Increases in sexual desire/interest were observed in women using COCs with 20 μg and 30 μg EE; ­decreases in sexual desire/interest were observed in women using COCs with 15 μg, 20 μg, 30 μg, and 35 μg EE; and no change in sexual desire/interest was observed in women taking COCs with 20 μg, 25 μg, 30 μg, and 35 μg EE. Regarding other aspects of relationship behavior, a more consistent pattern of EE-dose effects has emerged based on current evidence. That is, estrogen may act to increase mating behaviors associated with partner ­fidelity. The EE dose in COCs has been positively linked to reported jealousy (Cobey, Pollet, Roberts, Hahn and Cobey

247

& Buunk,  2011) and mate retention behavior (Welling et al., 2012), as well as aspects of women’s personality, such as neuroticism and extroversion (Welling, 2016). However, there were no effects of the progestin dose in the COC used observed on any of these behaviors, although it should be noted that all these studies used a between-participants design. When considering potential effects of the progestin component of COCs, the story becomes increasingly complex because the specific progestin used and the dose varies widely across COC brands (Hall & Trussell, 2012). Circulating levels of endogenous progesterone have been negatively linked to aspects of sexual desire in young women (Jones, Hahn, Fisher, Wang, Kandrik, & DeBruine, 2018; Roney & Simmons,  2013), suggesting that synthetic progestin may be associated with some negative outcomes regarding women’s sexual desire or interest. One study has reported significantly more sexual thoughts and fantasies in women using a triphasic COC compared with a monophasic COC—the difference in pill composition being a lower dose of progestin in the triphasic regimen (McCoy & Matyas, 1996). This finding supports the notion that higher progestin doses could be linked to negative outcomes regarding women’s sexual behavior. These results should be interpreted with caution, however, given that the study was done retrospectively. In addition to the potential for various progestin doses to impact behavior, COCs contain a variety of synthetic progestins. Given the differences in the androgenic properties of these synthetic progestins, there is the potential for these to differentially affect aspects of women’s sexual behavior. Indeed, various synthetic progestins have been shown to have significantly different effects on circulating concentrations of sex hormone– binding globulin (Jung-Hoffmann, Heidt, & Kuhl, 1988). The studies shown in Table 14.1 suggest that women who use COCs that utilize the antiandrogen chlormadinone acetate (CMA) report no change in sexual desire/interest (Brucker et al., 2010; Caruso, Rugolo, Agnello, Romano, & Cianci, 2009), whereas those using the antiandrogen drospirenone (DRSP) report no change or positive effects. The other progestins listed (desogestrel, gestodene, levonorgestrel, norethisterone, and norgestimate) are primarily androgenic and were not associated with any clear pattern of changes in sexual desire/interest (Schneider, 2003). One study directly compared women who initiated use of Yasmin (containing an antiandrogenic progestin, DRSP) versus women who initiated use of Meliane 248

Synthetic Hormones

(containing an androgenic progestin, g­ estodene), but did not observe any significant d ­ ifferences in post-COC initiation reports of sexual desire, sexual arousal, orgasm frequency, or general sexual satisfaction (Oranratanaphan & Taneepanichskul, 2006). Given that estrogens and progestins are predicted to have opposing effects on general sexual desire (Roney & Simmons, 2013), it is difficult to assess the effects of COCs, which contain both estrogen and progestins. A recent study investigating the effects of synthetic hormones on sexual behavior in a large sample of Norwegian women developed a novel procedure for estimating the effective dose of both the synthetic estrogen and progestin on continuous scales while simultaneously controlling for the androgenic properties of the progestin (Grøntvedt, Grebe, Kennair, & Gangestad, 2017). Although they did not observe any direct effects of EE or P-dose on  reported frequency of sexual intercourse using this method, Grøntvedt et al. did demonstrate an interaction between these synthetic hormones and loyalty/faithfulness on women’s reported frequency of sexual intercourse. As EE levels decreased and progestin levels increased, women’s loyalty/faithfulness became more positively associated with frequency of sexual intercourse. There were mixed results for the effect of the androgenic activity on frequency of sexual intercourse, with no association observed in Study 1, but a significant negative association observed in Study 2 and reported in many of the models included in the supplemental materials (Grøntvedt et al., 2017). Future research is needed to attempt to clarify the impact of the androgenic effects of contraceptives on sexual interests. Although the pill was the first hormonal contraceptive (HC) on the market, today there exists a wide variety of HC options, including long-lasting methods (e.g., the implant, Depo-Provera, the intrauterine device [IUD]). These long-lasting methods have gained popularity due to their superior performance in preventing unwanted pregnancy (Winner et al., 2012). A recent survey in the United States found that 22 percent of women report having used an injectable, long-lasting contraceptive (Mosher & Jones, 2010). While the pill is a COC containing both an estrogen and a progestin, many of these additional contraceptive methods rely solely on progestin (Rivera, Yacobson, & Grimes, 1999). As such, studying the effects of these additional HCs offers novel insight into the potential psychobehavioral effects of synthetic progesterone. Decreased libido and other sexual side effects are commonly cited as reasons for discontinuing use of progesterone-only

HCs (Sergent, Clamageran, Bastard, Verspyck, & Marpeau, 2004). Indeed, the Depo-Provera injection has even been used to decrease sexual drive in men suffering from pedophilia (Murray, 2010). Empirical studies directly assessing the effects of these alternate HCs on women’s behavior are much less common than studies assessing COCs, and those that do exist primarily rely on between-subject designs. A recent comparative study assessing multiple forms of HCs in a sample of nearly 10,000 women found that women who use Depo-Provera and the implant more commonly report lack of interest in sex when compared to women using the nonhormonal copper IUD (Boozalis, Tutlam, Robbins, & Peipert, 2016). However, previous studies of both Chinese (Li et al., 2004) and Chilean (Barnhart, Furman, Pommer, Coutinho, & Devoto, 1997) women did not report any negative effects of the contraceptive implant on women’s sexual behavior, and analyses of clinical trial data for women initiating use of the implant suggests that only 3 to 5 percent of women report experiencing decreased libido (Brache, Faundes, Alvarez, & Cochon, 2002). Studies of additional progesterone-only methods of hormonal contraception have also failed to find negative effects on women’s sexual behavior. Women using P-only oral contraceptives report no change in  libido following HC initiation (Graham et al., 1995), and women using the hormonal IUD7 do not report changes in their sexual interest (Boozalis et al., 2016) or sexual functioning (Li et al., 2004). In their study of the estrogenic and progesterogenic effects of contraceptives, Grøntvedt et al. (2017) tested for an effect of pill versus other delivery (see supplemental materials), but no significant difference in frequency of sexual intercourse was observed for women using the pill versus other HCs.

Hormone Replacement Therapy in Menopause

In addition to the common use of synthetic hormones in reproductively aged women as a form of contraception, synthetic hormones are used by older women as hormone replacement therapy (HRT) to help ease the menopause transition. Menopause is the permanent cessation of menstruation that results from the loss of ovarian follicular activity (World Health Organization,  1996). Menopause typically occurs around age 50 to 51 in women (McKinlay, Brambilla, & Posner, 1992) and is characterized by 7   The hormonal IUD acts locally rather than systemically and delivers a much smaller dose of synthetic progesterone.

the decline of circulating levels of estrogen and ­progesterone (Hunter,  1990), with a later decline in testosterone (Burger, Dudley, & Cui, 2000). Women undergoing menopause or those who are postmenopausal often report a variety of cognitive and behavioral symptoms, including changes in mood, memory, and libido (González, Viáfara, Caba, & Molina, 2004; reviewed in Pearce, Hawton, & Blake, 1995). Many medical professionals prescribe HRT during the menopause transition to ­alleviate some of these physical and psychological discomforts (Stadberg, Mattsson, & Milsom, 2009; Taavoni, Unesie Kafshgiry, Shahpoorian, & Mahmoudie, 2005). HRT first became available in the 1940s with the introduction of Premarin, an estrogen-only supplement derived from horse urine. Premarin was often combined with progestin-containing Provera until the release of Prempro, which included both estrogen and progestin. HRT usage rates grew to upward of 30 million in the mid-1970s (Kennedy, Baum, & Forbes, 1985) through the early 1990s (Wysowski, 1995), with an estimated 40 percent of perimenopausal women reporting having used HRT at some point during menopause at that time (Keating, Cleary, Rossi, Zaslavsky, & Ayanian, 1999), primarily during the age 50 to 54 window (Rosenberg, 1998). HRT usage suffered a sharp decline after the publication of preliminary findings of the Women’s Health Initiative (Writing Group for the Women’s Health Initiative Investigators, 2002) suggested that the risks of combined HRT exceeded their benefits because HRT may be linked to increased rates of breast cancer, heart disease, and stroke. Some of these findings were further supported by the Million Woman Study (Million Women Study Collaborators, 2003; for an expanded discussion of the history of these studies see Hersh, Stefanick, & Stafford, 2004; Jyotsna,  2013). Sales of HRT dropped by nearly 50 percent following the publication of these studies (Jyotsna, 2013). However, this work has been criticized for overestimating the risks associated with HRT use (Jyotsna, 2013). More recent evidence from the Women’s Health Initiative (Manson et al., 2013) and additional longitudinal studies (Schierbeck et al., 2012) has demonstrated that HRT may actually benefit women’s health, leading clinicians and researchers to re-evaluate recommendations for HRT use (North American Menopause Society, 2010). Today there are over 50 different types of HRT available to women, with varying synthetic hormone doses and administration methods (Women’s Health Concern, 2015). Hahn and Cobey

249

How Does Hormone Replacement Therapy Work?

The purpose of HRT is to compensate for the natural decrease in estradiol production by the ovaries. Similar to hormonal contraceptives, HRT regimens may consist of some combination of synthetic estrogen alone or a combination of synthetic estrogen and progestin (Greendale et al., 1999), with many women also incorporating the use of a synthetic testosterone supplement (Hersh et al.,  2004). HRT administration is typically achieved through an oral or transdermal route, although additional administration routes exist (Voican et al., 2012). The addition of a synthetic progestin is primarily used to balance estrogen’s proliferative effects on the endometrium (Voican et al., 2012), whereas the addition of testosterone may decrease the risk of breast cancer (Dimitrakakis, Jones, Liu, & Bondy, 2004) and improve aspects of well-being and sexuality (Flöter, Nathorst-Böös, Carlström, & von Schoultz, 2009). Indeed, the addition of testosterone is thought to specifically address issues of reduced sexual desire during menopause (but see Dennerstein, Dudley, Hopper, & Burger, 1997; Flöter et al., 2009; Simon et al., 2005).

Hormone Replacement Therapy and Sexual Behavior

Use of HRT supplementation during menopause has been suggested to improve women’s sexual function and genital health (Borissova, Kovatcheva, Shinkov, & Vukov,  2001; González et al.,  2004; Kingsberg, 1998; Taavoni et al., 2005). Indeed, two studies have reported that postmenopausal HRT users fared better than nonusers in all aspects of their sex life, including libido, sexual activity and orgasm frequency, sexual satisfaction, and sexual pleasure (Borissova et al., 2001; Taavoni et al., 2005), while an additional study found that including testosterone supplementation further enhanced sexual satisfaction when compared to an estrogen-only HRT regimen (Simon et al.,  2005). In a longitudinal study of 154 women who elected to use HRT during menopause and 130 women who did not, Taavoni et al. (2005) assessed women’s libido, sexual activity, pleasure and frequency of orgasm, and attitudes toward sex on multiple occasions, providing a withinwoman analysis of these changes and the potential effects of HRT use. Women using HRT fared better in all aspects of their sex life—with the majority of these women reporting that their libido, sexual activity, sexual pleasure, sexual satisfaction, and frequency 250

Synthetic Hormones

of orgasm were unchanged or improved in comparison with their own experience during the premenopausal period. Conversely, the majority of women in the non-HRT group reported decreases in all measured aspects of their sex lives. For all variables measured, these differences were statistically significant, suggesting that HRT may protect or shield women from the negative impact of menopause on their sexual behavior. It should be noted, however, that the potential issue of women’s motivation for using HRT cannot be controlled for here. Indeed, these groups of women differed in their views on the importance of sex—nearly 86 percent of women in the HRT group believed sex is relatively important in a relationship, whereas only 42 percent of women in the non-HRT group reported the same. Additional work has supported the notion that HRT use improves aspects of women’s sexual behavior postmenopause; HRT use is associated ­ with improved sexual function, although not always consistently across variables used to assess sexual ­behavior (González et al., 2004). However, women using HRT report higher sexual desire (Borissova et al., 2001; Donaldson, Welling, & Reeve, 2017; but see González et al., 2004), sexual arousal (Borissova et al., 2001; but see González et al., 2004), sexual satisfaction (Donaldson et al., 2017; González et al., 2004), orgasm frequency/experience (Borissova et al., 2001; Donaldson et al., 2017; González et al., 2004), and sociosexual orientation (Donaldson et al., 2017).

Hormone Replacement Therapy and Mate Preferences

Evolutionary psychologists have argued that ­menopause is accompanied by a shift from matingoriented psychology to family-oriented psychology (Hawkes, O’Connell, Jones, Alvarez, & Charnov, 1998). Indeed, mating-related behaviors, such as intrasexual competition, appear to be diminished in postmenopausal women (Vukovic et al., 2009). Given this purported shift in women’s psychology during menopause, we might expect other aspects of women’s reproductive psychology to undergo changes during the menopause transition and to be subject to effects of HRT, although the potential effects of HRT on many of these additional relationship behaviors (e.g., intrasexual competition) have yet to be investigated. HRT users report significantly more in-pair and extra-pair sexual interest than nonusers (Donaldson et al., 2017). They also report slightly increased feelings of jealousy, although this difference was not significant (Donaldson et al.,

2017). Still, relatively little work has been conducted in this area and future research is needed to further our understanding of the potential effects of synthetic hormones on aspects of menopausal women’s psychology.

Conclusion

This chapter described the impact of women’s ­exogenous hormone use on their psychology. The chapter focused on two of the most commonly used hormonal interventions: the pill and HRT. Based on the currently available research, it appears there is no consensus on how the pill or HRT influences women’s psychology and behavior. Indeed, we do not currently have a coherent understanding of how women’s endogenous hormones impact psychology, much less exogenous hormones. Some particular topics, such as mate preferences and sexual desire, have been studied more extensively than others. It is possible that in such topic areas where more extensive literature exists, a future ­systematic review and meta-analysis may provide insight into the pattern of findings across studies. Unfortunately, we suspect that one key finding of this review would be that there is limited evidence based on weak study designs, modest sample sizes, and inconsistent instrumentation to measure variables of interest. Here we emphasized the explicit advantages of within- versus between-participant designs on these topics, and further note the need to measure rather than estimate women’s hormone levels. Interpretations of studies of the pill or HRT also need to carefully consider how additional variables, such as women’s motivation for using these interventions, impact findings. Many fruitful areas remain to be examined in future research. Studies are only beginning to emerge that consider how various dosages and combinations of COCs impact women’s psychology; rigorously designed work on this topic should be pursued. In addition, studies examining hormonal contraceptives administered via other routes ­addressing women’s psychology remain to be conducted. Through the explicit integration of evolutionary thinking into clinical research (i.e., evolutionary medicine), we hope to see evolution-based hypotheses on topics related to women’s psychology that have otherwise been neglected studied using more robust clinical methods. The millions of women globally who use the pill and HRT would benefit from discovery on this topic; this may lead to better understanding of the changes these interventions cause

and potentially offer insight into how these interventions may be modified to reduce side effects.

References

Abler, B., Kumpfmüller, D., Grön, G., Walter, M., Stingl, J., & Seeringer, A. (2013). Neural correlates of erotic stimulation under different levels of female sexual hormones. PloS One, 8(2), e54447. http://doi.org/10.1371/journal.pone.0054447 Adams, D. B., Gold, A. R., & Burt, A. D. (1978). Rise in femaleinitiated sexual activity at ovulation and its suppression by oral contraceptives. New England Journal of Medicine, 299(21), 1145–1150. http://doi.org/10.1056/NEJM197811232992101 Alexander, G. (1990). Testosterone and sexual behavior in oral contraceptive users and nonusers: A prospective study. Hormones and Behavior, 24(3), 388–402. Alvergne, A., & Lummaa, V. (2010). Does the contraceptive pill alter mate choice in humans? Trends in Ecology & Evolution, 25(3), 171–179. http://doi.org/10.1016/j.tree.2009.08.003 American College of Obstetricians and Gynecologists. (1998). Statement on contraceptive methods. Washington, DC: Author. Andersson, M. B. (1994). Sexual selection. Princeton, NJ: Princeton University Press. Aznar-Ramos, R., Giner-Velázquez, J., Lara-Ricalde, R., & Martínez-Manautou, J. (1969). Incidence of side effects with contraceptive placebo. American Journal of Obstetrics and Gynecology, 105(7), 1144–1149. http://doi.org/10.1016/00029378(69)90142-2 Bailey, M.  J. (2006). More power to the pill: The impact of contraceptive freedom on women’s life cycle labor supply. Quarterly Journal of Economics, 121(1), 289–320. https://doi. org/10.1093/qje/121.1.289 Bailey, M.  J., & Hershbein, B. (2012). The opt-in revolution? Contraception and the gender gap in wages. American Economic Journal: Applied Economics, 4, 225–254. http://doi. org/10.1257/app.4.3.225 Bancroft, J., Sanders, D., Warner, P., & Loudon, N. (1987). The effects of oral contraceptives on mood and sexuality: A comparison of triphasic and combined preparations. Journal of Psychosomatic Obstetrics & Gynecology, 7(1), 1–8. http:// doi.org/10.3109/01674828709019585 Bancroft, J., & Sartorius, N. (1990). The effects of oral contraceptives on well-being and sexuality. Oxford Reviews of Reproductive Biology, 12, 57–92. Barnhart, K.  T., Furman, I., Pommer, R., Coutinho, E., & Devoto, L. (1997). Changes in the menstrual bleeding of users of a subdermal contraceptive implant of nomegestrol acetate (Uniplant) do not influence sexual frequency, sexual desire, or sexual enjoyment. Fertility and Sterility, 67(2), 244–249. http://doi.org/10.1016/S0015-0282(97)81905-X Battaglia, C., Battaglia, B., Mancini, F., Busacchi, P., Paganotto, M. C., Morotti, E., & Venturoli, S. (2012). Sexual behavior and oral contraception: A pilot study. Journal of Sexual Medicine, 9(2), 550–557. http://doi.org/10.1111/j.1743-6109.2011. 02597.x Birnbaum, S., Birnbaum, G.  E., & Ein-Dor, T. (2017). Can contraceptive pill affect future offspring’s health? The implications of using hormonal birth control for human evolution. Evolutionary Psychological Science, 3(2), 89–96. http://doi.org/10.1007/s40806-016-0074-4 Boozalis, A., Tutlam, N.  T., Robbins, C.  C., & Peipert, J.  F. (2016). Sexual desire and hormonal contraception. Obstetrics

Hahn and Cobey

251

& Gynecology, 127(3), 563–572. http://doi.org/10.1097/ AOG.0000000000001286 Borissova, A.  M., Kovatcheva, R., Shinkov, A., & Vukov, M. (2001). A study of the psychological status and sexuality in middle-aged Bulgarian women: Significance of the hormone replacement therapy (HRT). Maturitas, 39(2), 177–183. Bos, P. A., Panksepp, J., Bluthé, R.-M., & Honk, J. V. (2012). Acute effects of steroid hormones and neuropeptides on human social–emotional behavior: A review of single administration studies. Frontiers in Neuroendocrinology, 33(1), 17–35. http:// doi.org/10.1016/j.yfrne.2011.01.002 Bos, P.  A., Terburg, D., & van Honk, J. (2010). Testosterone decreases trust in socially naïve humans. Proceedings of the  National Academy of Sciences of the United States of America, 107(22), 9991–9995. http://doi.org/10.1073/pnas. 0911700107 Brache, V., Faundes, A., Alvarez, F., & Cochon, L. (2002). Nonmenstrual adverse events during use of implantable contra­ ceptives for women: Data from clinical trials. Contraception, 65(1), 63–74. http://doi.org/10.1016/S0010-7824(01)00289-X Brucker, C., Hedon, B., The, H. S., Höschen, K., Binder, N., & Christoph, A. (2010). Long-term efficacy and safety of a monophasic combined oral contraceptive containing 0.02 mg ethinylestradiol and 2 mg chlormadinone acetate administered in a 24/4-day regimen. Contraception, 81(6), 501–509. http:// doi.org/10.1016/j.contraception.2010.01.011 Burger, H. G., Dudley, E. C., & Cui, J. (2000). A prospective longitudinal study of serum testosterone, dehydroepiandro­ sterone sulfate, and sex hormone-binding globulin levels through the menopause transition. Journal of Clinical Endocrinology and Metabolism, 85, 2832–2838. Burrows, L. J., Basha, M., & Goldstein, A. T. (2012). The effects of hormonal contraceptives on female sexuality: A review. Journal of Sexual Medicine, 9(9), 2213–2223. http://doi. org/10.1111/j.1743–6109.2012.02848.x Cappelletti, M., & Wallen, K., (2016). Increasing women’s sexual desire: The comparative effectiveness of estrogens and androgens. Hormones & Behavior, 78, 178–193. Caruso, S., Agnello, C., Intelisano, G., Farina, M., Di Mari, L., & Cianci, A. (2004). Sexual behavior of women taking lowdose oral contraceptive containing 15 μg ethinylestradiol/ 60 μg gestodene. Contraception, 69(3), 237–240. http://doi. org/10.1016/j.contraception.2003.11.001 Caruso, S., Agnello, C., Intelisano, G., Farina, M., Di Mari, L., Sparacino, L., & Cianci, A. (2005). Prospective study on sexual behavior of women using 30 μg ethinylestradiol and 3 mg drospirenone oral contraceptive. Contraception, 72(1), 19–23. http://doi.org/10.1016/j.contraception.2005.02.002 Caruso, S., Iraci Sareri, M., Agnello, C., Romano, M., Presti, Lo, L., Malandrino, C., & Cianci, A. (2011). Conventional vs. extended-cycle oral contraceptives on the quality of sexual life: Comparison between two regimens containing 3 mg drospirenone and 20 µg ethinyl estradiol. Journal of Sexual Medicine, 8(5), 1478–1485. http://doi.org/10.1111/j.1743-6109. 2011.02208.x Caruso, S., Rugolo, S., Agnello, C., Romano, M., & Cianci, A. (2009). Quality of sexual life in hyperandrogenic women treated with an oral contraceptive containing chlormadinone acetate. Journal of Sexual Medicine, 6(12), 3376–3384. http:// doi.org/10.1111/j.1743-6109.2009.01529.x Chiappori, P. A., & Oreffice, S. (2015). Birth control and female empowerment: An equilibrium analysis. Journal of Political Economy, 116(1), 113–140. http://doi.org/10.1086/529409

252

Synthetic Hormones

Cobey, K. D., Buunk, A. P., Roberts, S. C., Klipping, C., Appels, N., Zimmerman, Y., . . . & Pollet, T.  V. (2012). Reported jealousy differs as a function of menstrual cycle stage and contraceptive pill use: A within-subjects investigation. Evolution and Human Behavior, 33(4), 395–401. http://doi. org/10.1016/j.evolhumbehav.2011.12.001 Cobey, K.  D., & Hahn, A.  C. (2017). The endocrinology of female competition. In M.  L.  Fisher (Ed.), The Oxford handbook of women and competition (pp. 371–396). Oxford University Press. Cobey, K.  D., Havlicek, J., Klapilová, K., & Roberts, S.  C. (2016). Hormonal contraceptive use during relationship formation and sexual desire during pregnancy. Archives of Sexual Behavior, 45(8), 2117–2122. http://doi.org/10.1007/ s10508-015-0662-6 Cobey, K. D., Klipping, C., & Buunk, A. P. (2013). Hormonal contraceptive use lowers female intrasexual competition in pair-bonded women. Evolution and Human Behavior, 34(4), 294–298. http://doi.org/10.1016/j.evolhumbehav.2013.04.003 Cobey, K.  D., Pollet, T.  V., Roberts, S.  C., & Buunk, A.  P. (2011). Hormonal birth control use and relationship jealousy: Evidence for estrogen dosage effects. Personality and Individual Differences, 50(2), 315–317. http://doi. org/10.1016/j.paid.2010.09.012 Davis, A.  R., & Castaño, P.  M. (2012). Oral contraceptives and libido in women. Annual Review of Sex Research, 15(1), 297–320. Dei, M., Verni, A., Bigozzi, L., & Bruni, V. (1997). Sex steroids and libido. European Journal of Contraception & Reproductive Health Care, 2(4), 253–258. http://doi.org/10.3109/ 13625189709165303 Dennerstein, L., Dudley, E.  C., Hopper, J.  L., & Burger, H. (1997). Sexuality, hormones and the menopausal transition. Maturitas, 26(2), 83–93. http://doi.org/10.1016/S03785122(96)01093-6 Dimitrakakis, C., Jones, R. A., Liu, A., & Bondy, C. A. (2004). Breast cancer incidence in postmenopausal women using testosterone in addition to usual hormone therapy. Menopause, 11(5), 531. Ditzen, B., Palm-Fischbacher, S., Gossweiler, L., Stucky, L., & Ehlert, U. (2017). Effects of stress on women’s preference for male facial masculinity and their endocrine correlates. Psychoneuroendocrinology, 82, 67–74. Donaldson, S. H., Welling, L. L. M., & Reeve, S. D. (2017). The influence of hormone replacement therapy on mating psychology among post-menopausal women. Personality and Individual Differences, 115, 13–18. http://doi.org/10.1016/j. paid.2016.10.038 Dreher, J.-C., Schmidt, P. J., Kohn, P., Furman, D., Rubinow, D., & Berman, K.  F. (2007). Menstrual cycle phase modulates reward-related neural function in women. Proceedings of the National Academy of Sciences of the United States of America, 104(7), 2465–2470. http://doi.org/10.1073/ pnas.0605569104 Ellison, P. T., & Gray, P. B. (Eds.) (2009). Endocrinology of social relationships. Harvard University Press. Feinberg, D.  R., DeBruine, L.  M., Jones, B.  C., & Little, A. C. (2008). Correlated preferences for men’s facial and vocal  masculinity. Evolution and Human Behavior, 29(4), 233–241. Flöter, A., Nathorst-Böös, J., Carlström, K., & von Schoultz, B. (2009). Addition of testosterone to estrogen replacement therapy in oophorectomized women: Effects on sexuality and

well-being. Climacteric, 5(4), 357–365. http://doi.org/ 10.1080/cmt.5.4.357.365 Frye, C. A. (2006). An overview of oral contraceptives: Mechanism of action and clinical use. Neurology, 66(3), S29–S36. http:// doi.org/10.1212/WNL.66.66_suppl_3.S29 Gangestad, S. W., Simpson, J. A., Cousins, A. J., Garver-Apgar, C. E., & Christensen, P. N. (2016). Women’s preferences for male behavioral displays change across the menstrual cycle. Psychological Science, 15(3), 203–207. http://doi.org/10.1111/ j.0956–7976.2004.01503010.x Gangestad, S.  W., & Thornhill, R. (1998). Menstrual cycle variation in women’s preferences for the scent of symmetrical men. Proceedings of the Royal Society of London B: Biological Sciences, 265(1399), 927–933. http://doi.org/10.1098/rspb. 1998.0380 Gangestad, S.  W., Thornhill, R., & Garver, C.  E. (2002). Changes in women’s sexual interests and their partner’s mate–retention tactics across the menstrual cycle: Evidence for shifting conflicts of interest. Proceedings of the Royal Society of London B: Biological Sciences, 269(1494), 975–982. http://doi.org/10.1098/rspb.2001.1952 Garver-Apgar, C. E., Gangestad, S. W., & Thornhill, R. (2008). Hormonal correlates of women’s mid-cycle preference for the scent of symmetry. Evolution and Human Behavior, 29(4), 223–232. http://doi.org/10.1016/j.evolhumbehav.2007. 12.007 Geary, D. C., DeSoto, M. C., Hoard, M. K., Sheldon, M. S., & Cooper, M.  L. (2001). Estrogens and relationship jealousy. Human Nature, 12(4), 299–320. González, M., Viáfara, G., Caba, F., & Molina, E. (2004). Sexual function, menopause and hormone replacement therapy (HRT). Maturitas, 48(4), 411–420. http://doi.org/10.1016/j. maturitas.2003.10.005 Graham, C. A., Bancroft, J., Doll, H. A., Greco, T., & Tanner, A. (2007). Does oral contraceptive-induced reduction in free testosterone adversely affect the sexuality or mood of women? Psychoneuroendocrinology, 32(3), 246–255. http://doi.org/ 10.1016/j.psyneuen.2006.12.011 Graham, C. A., Ramos, R., Bancroft, J., Maglaya, C., & Farley, T. M. (1995). The effects of steroidal contraceptives on the well-being and sexuality of women: A double-blind, placebocontrolled, two-centre study of combined and progestogenonly methods. Contraception, 52(6), 363–369. Graham, C., & Sherwin, B. (1993). The relationship between mood and sexuality in women using an oral contraceptive as  a treatment for premenstrual symptoms. Psychoneuro­ endocrinology, 18(4), 273–281. http://doi.org/10.1016/03064530(93)90024-F Grebe, N.  M., Emery Thompson, M., & Gangestad, S.  W. (2016). Hormonal predictors of women’s extra-pair vs. in-pair sexual attraction in natural cycles: Implications for extended sexuality. Hormones and Behavior, 78, 211–219. http://doi. org/10.1016/j.yhbeh.2015.11.008 Greco, T., Graham, C.  A., Bancroft, J., Tanner, A., & Doll, H. A. (2007). The effects of oral contraceptives on androgen levels and their relevance to premenstrual mood and sexual interest: A comparison of two triphasic formulations containing norgestimate and either 35 or 25 μg of ethinyl estradiol. Contraception, 76(1), 8–17. http://doi.org/10.1016/ j.contraception.2007.04.002 Greendale, G. A., Reboussin, B. A., Sie, A., Singh, H. R., Olson, L. K., Gatewood, O., et al. (1999). Effects of estrogen and estrogen–progestin on mammographic parenchymal density.

Annals of Internal Medicine, 130(4, Part 1), 262–268. http:// doi.org/10.7326/0003-4819-130-4_Part_1-19990216000003 Grøntvedt, T.  V., Grebe, N.  M., Kennair, L.  E.  O., & Gangestad, S.  W. (2017). Estrogenic and progestogenic effects of hormonal contraceptives in relation to sexual behavior: Insights into extended sexuality. Evolution and Human Behavior, 38(3), 283–292. http://doi.org/10.1016/j. evolhumbehav.2016.10.006 Guida, M., Di Spiezio Sardo, A., Bramante, S., Sparice, S., Acunzo, G., Tommaselli, G. A., et al. (2005). Effects of two types of hormonal contraception—oral versus intravaginal— on the sexual life of women and their partners. Human Reproduction, 20(4), 1100–1106. http://doi.org/10.1093/ humrep/deh686 Guillermo, C. J., Manlove, H. A., Gray, P. B., Zava, D. T., & Marrs, C. R. (2010). Female social and sexual interest across the menstrual cycle: The roles of pain, sleep and hormones. BMC Women’s Health, 10(1), 19. Hahn, A. C., Fisher, C. I., Cobey, K. D., DeBruine, L. M., & Jones, B. C. (2016). A longitudinal analysis of women’s salivary testosterone and intrasexual competitiveness. Psychoneuro­ endocrinology, 64, 117–122. http://doi.org/10.1016/j.psyneuen. 2015.11.014 Hall, K.  S., & Trussell, J. (2012). Types of combined oral contraceptives used by US women. Contraception, 86(6), 659–665. http://doi.org/10.1016/j.contraception.2012.05.017 Hamstra, D. A., De Rover, M., De Rijk, R. H., & Van der Does, W. (2014). Oral contraceptives may alter the detection of emotions in facial expressions. European Neuropsycho­ pharmacology, 24(11), 1855–1859. http://doi.org/10.1016/ j.euroneuro.2014.08.015 Hawkes, K., O’Connell, J. F., Jones, N. G. B., Alvarez, H., & Charnov, E.  L. (1998). Grandmothering, menopause, and the evolution of human life histories. Proceedings of the National Academy of Sciences of the United States of America, 95(3), 1336–1339. Hersh, A. L., Stefanick, M. L., & Stafford, R. S. (2004). National use of postmenopausal hormone therapy: Annual trends and response to recent evidence. JAMA, 291(1), 47–53. http:// doi.org/10.1001/jama.291.1.47 Holzleitner, I. J., Hahn, A. C., Fisher, C., Lee, A., O’Shea, K. J., Jones, B. C. & DeBruine, L. M. (2017, preprint). Hormonal regulation of women’s prosocial, but not sexual, responses to kinship cues. BioRXiv. https://doi.org/10.1101/192054 Hunter, M. S. (1990). Emotional well-being, sexual behaviour and hormone replacement therapy. Maturitas, 12(3), 299–314. Jones, B. C., Hahn, A., Fisher, C., Wang, H., Kandrik, M., & DeBruine, L. (2018). General sexual desire, but not desire for uncommitted sexual relationships, tracks changes in women’s hormonal status. Psychoneuroendocrinology, 88, 153–157. doi: 10.1016/j.psyneuen.2017.12.015 Jones, B. C., Hahn, A. C., Fisher, C. I., Wang, H., Kandrik, M., Han, C., et al. (2018). No compelling evidence that preferences for facial masculinity track changes in women’s hormonal status. Psychological Science, 29, 996–1005. https://doi.org/ 10.1177/0956797618760197 Jones, B. C., Perrett, D. I., Little, A. C., Boothroyd, L., Cornwell, R.  E., Feinberg, D.  R., et al. (2005). Menstrual cycle, pregnancy and oral contraceptive use alter attraction to apparent health in faces. Proceedings of the Royal Society of London B: Biological Sciences, 272(1561), 347–354. http:// doi.org/10.1098/rspb.2004.2962

Hahn and Cobey

253

Jung-Hoffmann, C., Heidt, F., & Kuhl, H. (1988). Effect of two oral contraceptives containing 30 μg ethinylestradiol and 75 μg gestodene or 150 μg desogestrel upon various hormonal parameters. Contraception, 38(6), 593–603. http://doi.org/ 10.1016/0010-7824(88)90044-3 Jyotsna, V. P. (2013). Postmenopausal hormonal therapy: Current status. Indian Journal of Endocrinology and Metabolism, 17(Suppl. 1), S45. http://doi.org/10.4103/2230-8210.119504 Keating, N. L., Cleary, P. D., Rossi, A. S., Zaslavsky, A. M., & Ayanian, J. Z. (1999). Use of hormone replacement therapy by postmenopausal women in the United States. Annals of Internal Medicine, 130(7), 545–553. http://doi.org/10.7326/ 0003-4819-130-7-199904060-00002 Kennedy, D. L., Baum, C., & Forbes, M. B. (1985). Noncontra­ ceptive estrogens and progestins: Use patterns over time. Obstetrics & Gynecology, 65(3), 441. Kingsberg, S.  A. (1998). Postmenopausal sexual functioning: A  case study. International Journal of Fertility and Women’s Medicine, 43(2), 122–128. Li, R. H. W., Lo, S. S. T., Teh, D. K. G., Tong, N.-C., Tsui, M.  H.  Y., Cheung, K.-B., & Chung, T.  K.  H. (2004). Impact of common contraceptive methods on quality of life  and sexual function in Hong Kong Chinese women. Contraception, 70(6), 474–482. http://doi.org/10.1016/ j.contraception.2004.06.010 Little, A. C., Burriss, R. P., Petrie, M., Jones, B. C., & Roberts, S.  C. (2013). Oral contraceptive use in women changes preferences for male facial masculinity and is associated with partner facial masculinity. Psychoneuroendocrinology, 38(9), 1777–1785. http://doi.org/10.1016/j.psyneuen.2013. 02.014 Macrae, C. N., Alnwick, K. A., Milne, A. B., & Schloerscheidt, A. M. (2016). Person perception across the menstrual cycle: Hormonal influences on social-cognitive functioning. Psychological Science, 13(6), 532–536. http://doi.org/10.1111/ 1467-9280.00493 Manson, J.  E., Chlebowski, R.  T., Stefanick, M.  L., Aragaki, A.  K., Rossouw, J.  E., Prentice, R.  L., et al. (2013). Menopausal hormone therapy and health outcomes during the intervention and extended poststopping phases of the Women’s Health Initiative Randomized Trials. JAMA, 310(13), 1353–1368. http://doi.org/10.1001/jama.2013.278040 Marcinkowska, U. M., Galbarczyk, A., & Jasienska, G. (2018). La donna è mobile? Lack of cyclical shifts in facial symmetry, and facial and body masculinity preferences—A hormone based study. Psychoneuroendocrinology, 88, 47–53. Mark, K.  P., Leistner, C.  E., & Garcia, J.  R. (2016). Impact of contraceptive type on sexual desire of women and of men partnered to contraceptive users. Journal of Sexual Medicine, 13(9), 1359–1368. http://doi.org/10.1016/j.jsxm.2016.06.011 Matteo, S. (1984). Increased sexual activity during the midcycle portion of the human menstrual cycle. Hormones and Behavior, 18(3), 249–255. http://doi.org/10.1016/0018506X(84)90014-X McCoy, N. L., & Matyas, J. R. (1996). Oral contraceptives and sexuality in university women. Archives of Sexual Behavior, 25(1), 73–90. http://doi.org/10.1007/BF02437907 McKinlay, S. M., Brambilla, D. J., & Posner, J. G. (1992). The normal menopause transition. Maturitas, 14(2), 103–115. http://doi.org/10.1016/0378-5122(92)90003-M Million Women Study Collaborators. (2003). Breast cancer and hormone-replacement therapy in the Million Women Study.

254

Synthetic Hormones

Lancet, 362(9382), 419–427. http://doi.org/10.1016/S01406736(03)14065-2 Montoya, E. R., & Bos, P. A. (2017). How oral contraceptives impact social-emotional behavior and brain function. Trends in Cognitive Sciences, 21(2), 125–136. http://doi.org/10.1016/j. tics.2016.11.005 Mosher, W., & Jones, J. (2010). Use of contraception in the United States: 1982–2008. Vital and Health Statistics. Series 23, Data From the National Survey of Family Growth, 29, 1–44. Mosher, W. D., Martinez, G. M., Chandra, A., Abma, J. C., & Willson, S. J. (2004). Use of contraception and use of family planning services in the United States: 1982–2002. Advance Data, 350(1), 1–36. Motta-Mena, N.  V., & Puts, D.  A. (2017). Endocrinology of human female sexuality, mating, and reproductive behavior. Hormones & Behavior, 91, 19–35. Murray, J. B. (2010). Psychopharmacological therapy of deviant sexual behavior. Journal of General Psychology, 115(1), 101–110. http://doi.org/10.1080/00221309.1988.9711093 North American Menopause Society. (2010). Estrogen and progestogen use in postmenopausal women: 2010 position statement of the North American Menopause Society. Menopause, 17(2), 242–255. Oinonen, K. A. (2009). Putting a finger on potential predictors of oral contraceptive side effects: 2D:4D and middlephalangeal hair. Psychoneuroendocrinology, 34(5), 713–726. http://doi.org/10.1016/j.psyneuen.2008.11.009 Oinonen, K. A., & Mazmanian, D. (2002). To what extent do oral contraceptives influence mood and affect? Journal of Affective Disorders, 70(3), 229–240. http://doi.org/10.1016/ S0165-0327(01)00356-1 Oranratanaphan, S., & Taneepanichskul, S. (2006). A double blind randomized control trial, comparing effect of drospirenone and gestodene to sexual desire and libido. Journal of the Medical Association of Thailand, 89, S17–S22. Parkes, A., Wight, D., Henderson, M., Stephenson, J., & Strange, V. (2009). Contraceptive method at first sexual intercourse and subsequent pregnancy risk: Findings from a secondary analysis of 16-year-old girls from the RIPPLE and SHARE studies. Journal of Adolescent Health, 44(1), 55–63. http://doi.org/10.1016/j.jadohealth.2008.06.006 Pastor, Z., Holla, K., & Chmel, R. (2013). The influence of combined oral contraceptives on female sexual desire: A systematic review. European Journal of Contraception & Reproductive Health Care, 18(1), 27–43. http://doi.org/10.310 9/13625187.2012.728643 Pearce, J., Hawton, K., & Blake, F. (1995). Psychological and sexual symptoms associated with the menopause and the effects of hormone replacement therapy. British Journal of Psychiatry, 167(2), 163–173. http://doi.org/10.1192/bjp.167.2.163 Penton-Voak, I. S., Perrett, D. I., Castles, D. L., Kobayashi, T., Burt, D.  M., Murray, L.  K., & Minamisawa, R. (1999). Menstrual cycle alters face preference. Nature, 399(6738), 741–742. http://doi.org/10.1038/21557 Petitti, D.  B. (2003). Combination estrogen–progestin oral contracep­tives. New England Journal of Medicine, 349(15), 1443–1450. Pisanski, K., Hahn, A.  C., Fisher, C.  I., DeBruine, L.  M., Feinberg, D. R., & Jones, B. C. (2014). Changes in salivary estradiol predict changes in women’s preferences for vocal

masculinity. Hormones and Behavior, 66(3), 493–497. http:// doi.org/10.1016/j.yhbeh.2014.07.006 Pletzer, B.  A., & Kerschbaum, H.  H. (2014). 50 years of hormonal contraception—time to find out, what it does to our brain. Frontiers in Neuroscience, 8(256), 1–6. Puts, D. A. (2006). Cyclic variation in women’s preferences for masculine traits. Human Nature: An Interdisciplinary Biosocial Perspective, 17(1), 114–127. Rabe, T., Goeckenjan, M., Ahrendt, H., Crosignani, P., Dinger, J., Mueck, A., et al. (2011). Oral contraceptive pills: Combinations, dosages and the rationale behind 50 years or  oral hormonal contraceptive development. Journal of Reproductive Medicine and Endocrinology, 8(1), 58–129. Radke, S., & Derntl, B. (2016). Affective responsiveness is influenced by intake of oral contraceptives. European Neuropsychopharmacology, 26(6), 1014–1019. http://doi.org/ 10.1016/j.euroneuro.2016.03.004 Rivera, R., Yacobson, I., & Grimes, D. (1999). The mechanism of action of hormonal contraceptives and intrauterine contraceptive devices. American Journal of Obstetrics and Gynecology, 181(5), 1263–1269. http://doi.org/10.1016/S00029378(99)70120-1 Roberts, S. C., Gosling, L. M., Carter, V., & Petrie, M. (2008). MHC-correlated odour preferences in humans and the use of oral contraceptives. Proceedings of the Royal Society of London B: Biological Sciences, 275(1652), 2715–2722. http://doi. org/10.1098/rspb.2008.0825 Roberts, S. C., Klapilová, K., Little, A. C., Burriss, R. P., Jones, B.  C., DeBruine, L.  M., et al. (2012). Relationship satisfaction and outcome in women who meet their partner while using oral contraception. Proceedings of the Royal Society of London B: Biological Sciences, 279(1732), 1430–1436. http://doi.org/10.1098/rspb.2011.1647 Roberts, S. C., Little, A. C., Burriss, R. P., Cobey, K. D., Klapilová, K., Havlicek, J., et al. (2014). Partner choice, relation­ship satisfaction, and oral contraception. Psychological Science, 25(7),1497–1503.http://doi.org/10.1177/0956797614532295 Robinson, S.  A., Dowell, M., Pedulla, D., & McCauley, L. (2004). Do the emotional side-effects of hormonal contraceptives come from pharmacologic or psychological mechanisms? Medical Hypotheses, 63(2), 268–273. http:// doi.org/10.1016/j.mehy.2004.02.013 Roney, J. R., & Simmons, Z. L. (2013). Hormonal predictors of sexual motivation in natural menstrual cycles. Hormones and Behavior, 63(4), 636–645. http://doi.org/10.1016/j.yhbeh.2013. 02.013 Roney, J. R., & Simmons, Z. L. (2016). Within-cycle fluctuations in progesterone negatively predict changes in both in-pair and extra-pair desire among partnered women. Hormones & Behavior, 81, 45–52. Roney, J. R., Simmons, Z. L., & Gray, P. B. (2011). Changes in estradiol predict within-women shifts in attraction to facial cues of men’s testosterone. Psychoneuroendocrinology, 36(5), 742–749. http://doi.org/10.1016/j.psyneuen.2010.10.010 Rosenberg, L. (1998). Correlates of postmenopausal female hormone use among black women in the United States. Obstetrics & Gynecology, 91(3), 454–458. Royal College of General Practitioners. (1974). Oral contraceptives and health. An interim report. London, UK: Sir Isaac Pitman and Sons. Rupp, H. A., & Wallen, K. (2007). Sex differences in viewing sexual stimuli: An eye-tracking study in men and women.

Hormones and Behavior, 51(4), 524–533. http://doi.org/10.1016/ j.yhbeh.2007.01.008 Sabatini, R., & Cagiano, R. (2006). Comparison profiles of cycle control, side effects and sexual satisfaction of three hormonal contraceptives. Contraception, 74(3), 220–223. http://doi. org/10.1016/j.contraception.2006.03.022 Sahlberg, B.-L., Landgren, B.-M., & Axelson, M. (1987). Metabolic profiles of endogenous and ethynyl steroids in plasma and urine from women during administration of oral contraceptives. Journal of Steroid Biochemistry, 26(5), 609–617. http://doi.org/10.1016/0022-4731(87)90014-8 Sanders, S. A., Graham, C. A., Bass, J. L., & Bancroft, J. (2001). A prospective study of the effects of oral contraceptives on  sexuality and well-being and their relationship to discontinuation. Contraception, 64(1), 51–58. http://doi. org/10.1016/S0010-7824(01)00218-9 Sarfaty, A., Margulis, S.  W., & Atsalis, S. (2012). Effects of combination birth control on estrous behavior in captive western lowland gorillas, Gorilla gorilla gorilla. Zoo Biology, 31(3), 350–361. http://doi.org/10.1002/zoo.20401 Schaffir, J. (2006). Hormonal contraception and sexual desire: A  critical review. Journal of Sex & Marital Therapy, 32(4), 305–314. http://doi.org/10.1080/00926230600666311 Scheele, D., Plota, J., Stoffel-Wagner, B., Maier, W., & Hurlemann, R. (2015). Hormonal contraceptives suppress oxytocin-induced brain reward responses to the partner’s face. Social Cognitive and Affective Neuroscience, 11(5), 767–774. Schierbeck, L.  L., Rejnmark, L., Tofteng, C.  L., Stilgren, L., Eiken, P., Mosekilde, L., et al. (2012). Effect of hormone replacement therapy on cardiovascular events in recently postmenopausal women: Randomised trial. BMJ, 345(2), e6409–e6409. http://doi.org/10.1136/bmj.e6409 Schneider, H.  P.  G. (2003). Androgens and antiandrogens. Annals of the New York Academy of Sciences, 997(1), 292–306. http://doi.org/10.1196/annals.1290.033 Sergent, F., Clamageran, C., Bastard, A., Verspyck, E., & Marpeau, L. (2004). Acceptability of the etonogestrelcontaining contraceptive implant (Implanon). Journal De Gynecologie, Obstetrique Et Biologie De La Reproduction, 33, 407–415. http://doi.org/10.1016/S0368-2315(04)96548-9 Shively, C. A., Manuck, S. B., Kaplan, J. R., & Koritnik, D. R. (1990). Oral contraceptive administration, interfemale relation­ ships, and sexual behavior in Macaca fascicularis. Archives of  Sexual Behavior, 19(2), 101–117. http://doi.org/10.1007/ BF01542226 Simon, J., Braunstein, G., Nachtigall, L., Utian, W., Katz, M., Miller, S., et al. (2005). Testosterone patch increases sexual activity and desire in surgically menopausal women with hypoactive sexual desire disorder. Journal of Clinical Endocrinology & Metabolism, 90(9), 5226–5233. http://doi. org/10.1210/jc.2004–1747 Smith, F.  G., Jones, B.  C., Little, A.  C., DeBruine, L.  M., Welling, L.  L.  M., Vukovic, J., & Conway, C.  A. (2009). Hormonal contraceptive use and perceptions of trust modulate the effect of relationship context on women’s preferences for sexual dimorphism in male face shape. Journal of Evolutionary Psychology, 7(3), 195–210. http://doi.org/ 10.1556/JEP.7.2009.3.1 Stadberg, E., Mattsson, L.-Å., & Milsom, I. (2009). The prevalence and severity of climacteric symptoms and the use of different treatment regimens in a Swedish population. Acta

Hahn and Cobey

255

Obstetricia Et Gynecologica Scandinavica, 76(5), 442–448. http://doi.org/10.3109/00016349709047826 Strufaldi, R., Pompei, L.  M., Steiner, M.  L., Cunha, E.  P., Ferreira, J.  A.  S., Peixoto, S., & Fernandes, C.  E. (2010). Effects of two combined hormonal contraceptives with the same composition and different doses on female sexual function and plasma androgen levels. Contraception, 82(2), 147–154. http://doi.org/10.1016/j.contraception.2010.02.016 Taavoni, S., Unesie Kafshgiry, M., Shahpoorian, F., & Mahmoudie, M. (2005). Hormone replacement therapy: Post-menopausal sex life and attitudes towards sex. Psychogeriatrics, 5(1), 9–14. http://doi.org/10.1111/j.1479–8301. 2005.00080.x Taggart, T.  C., Hammett, J.  F., & Ulloa, E.  C. (2016). Oral contra­ceptive use associated with increased romantic relation­ ship satisfaction. Psi Chi Journal of Psychological Research, 21(3), 193–199. Thornhill, R. (1999). The scent of symmetry a human sex phero­ mone that signals fitness? Evolution and Human Behavior, 20(3), 175–201. http://doi.org/10.1016/S1090-5138(99)00005-7 Traulsen, J.  M., Haugbølle, L.  S., & Bissell, P. (2003). (5) Feminist theory and pharmacy practice. International Journal of Pharmacy Practice, 11(1), 55–68. http://doi.org/10.1211/ 002235702865 Van Heusden, A. M., & Fauser, B. C. J. M. (1999). Activity of the pituitary-ovarian axis in the pill-free interval during use of low-dose combined oral contraceptives. Contraception, 59(4), 237–243. http://doi.org/10.1016/S0010-7824(99)00025-6 Van Wingen, G. A., Ossewaarde, L., Bäckström, T., Hermans, E. J., & Fernández, G. (2011). Gonadal hormone regulation of the emotion circuitry in humans. Neuroscience, 191, 38–45. http://doi.org/10.1016/j.neuroscience.2011.04.042 Voican, A., Francou, B., Novac, L., Chabbert-Buffet, N., Canonico, M., Meduri, G., et al. (2012). Pharmacology of hormone replacement therapy in menopause. In L. Gallelli (Ed.), Pharmacology (pp. 313–338). Rijeka, Croatia: InTech. Vukovic, J., Jones, B. C., DeBruine, L. M., Little, A. C., Feinberg, D. R., & Welling, L. L. M. (2009). Circum-menopausal effects on women’s judgements of facial attractiveness. Biology Letters, 5(1), 62–64. http://doi.org/10.1098/rsbl.2008.0478 Wallen, K. (1984). Periovulatory changes in female sexual behavior and patterns of ovarian steroid secretion in group-living rhesus monkeys. Hormones and Behavior, 18(4), 431–450. http://doi. org/10.1016/0018-506X(84)90028-X Wallen, K. (1990). Desire and ability: Hormones and the regulation of female sexual behavior. Neuroscience & Biobeh­ avioral Reviews, 14(2), 233–241. http://doi.org/10.1016/ S0149-7634(05)80223-4 Wallen, K. (2001). Sex and context: Hormones and primate sexual motivation. Hormones and Behavior, 40(2), 339–357. http://doi.org/10.1006/hbeh.2001.1696 Wallwiener, C. W., Wallwiener, L. M., Seeger, H., Mück, A. O., Bitzer, J., & Wallwiener, M. (2010). Prevalence of sexual dysfunction and impact of contraception in female German medical students. Journal of Sexual Medicine, 7(6), 2139– 2148. http://doi.org/10.1111/j.1743-6109.2010.01742.x Wang, H., Hahn, A. C., Fisher, C. I., DeBruine, L. M., & Jones, B.  C. (2014). Women’s hormone levels modulate the motivational salience of facial attractiveness and sexual

256

Synthetic Hormones

dimorphism. Psychoneuroendocrinology, 50, 246–251. http:// doi.org/10.1016/j.psyneuen.2014.08.022 Wedekind, C., Seebeck, T., Bettens, F., & Paepke, A. J. (1995). MHC-dependent mate preferences in humans. Proceedings of the Royal Society of London B: Biological Sciences, 260(1359), 245–249. Welling, L. L. M. (2013). Psychobehavioral effects of hormonal contraceptive use. Evolutionary Psychology, 11(3), 718–742. http://doi.org/10.1177/147470491301100315 Welling, L. L. M. (2016). Synthetic hormone dose in hormonal contraceptives predicts individual differences in personality. Social Behavior Research and Practice, 1(1), 13–16. Welling, L.  L.  M., Jones, B.  C., DeBruine, L.  M., Conway, C. A., Law Smith, M. J., Little, A. C., et al. (2007). Raised salivary testosterone in women is associated with increased attraction to masculine faces. Hormones and Behavior, 52(2), 156–161. http://doi.org/10.1016/j.yhbeh.2007.01.010 Welling, L. L. M., Puts, D. A., Roberts, S. C., Little, A. C., & Burriss, R. P. (2012). Hormonal contraceptive use and mate retention behavior in women and their male partners. Hormones and Behavior, 61(1), 114–120. Westhoff, C.  L., Heartwell, S., Edwards, S., Zieman, M., Stuart,  G., Cwiak, C., et al. (2007). Oral contraceptive discontinuation: Do side effects matter? American Journal of Obstetrics and Gynecology, 196(4), 412.e1–412.e7. http://doi. org/10.1016/j.ajog.2006.12.015 Winner, B., Peipert, J.  F., Zhao, Q., Buckel, C., Madden, T., Allsworth, J.  E., & Secura, G.  M. (2012). Effectiveness of long-acting reversible contraception. New England Journal of Medicine, 366(21), 1998–2007. http://doi.org/10.1056/ NEJMoa1110855 Women’s Health Concern. (2015). HRT: Benefits and risks. Women’s health concern fact sheet. Retrieved from https:// www.womens-health-concern.org/help-and-advice/factsheets/ hrt-know-benefits-risks/ World Health Organization. (1996). Research on the menopause in  the 1990s: Report of a WHO Scientific Group. World Health Organization Technology Report Series, 866, 1–107. Retrieved from http://apps.who.int/iris/handle/10665/41841 Writing Group for the Women’s Health Initiative Investigators. (2002). Risks and benefits of estrogen plus progestin in healthy postmenopausal women: Principal results from the Women’s Health Initiative Randomized Controlled Trial. JAMA, 288(3), 321–333. http://doi.org/10.1001/jama.288.3.321 Wysowski, D. (1995). Use of menopausal estrogens and medroxy­ progesterone in the United States, 1982–1992. Obstetrics &  Gynecology, 85(1), 6–10. http://doi.org/10.1016/00297844(94)00339-F Zethraeus, N., Dreber, A., Ranehill, E., Blomberg, L., Labrie, F., von Schoultz, B., . . . & Hirschberg, A. L. (2016). Combined oral contraceptives and sexual function in women—a doubleblind, randomized, placebo-controlled trial. The Journal of Clinical Endocrinology & Metabolism, 101(11), 4046–4053. Zimmerman, Y., Eijkemans, M.  J.  C., Coelingh Bennink, H. J. T., Blankenstein, M. A., & Fauser, B. C. J. M. (2014). The effect of combined oral contraception on testosterone levels in healthy women: A systematic review and meta-analysis. Human Reproduction Update, 20(1), 76–105. http://doi.org/ 10.1093/humupd/dmt038

P A R T 

Social and Affective Behavior

3

CH A PT E R

15

The Endocrinology of Social Relationships and Affiliation

Christine Anderl, Shimon Saphire-Bernstein, and Frances S. Chen

Abstract Scientific understanding of the neuroendocrine foundations of human social behavior has grown substantially in recent decades. Methodological advances now allow for empirical research on neuroendocrine contributors to both stable differences between individuals (e.g., personality traits) and fluctuations within individuals (e.g., situational influences) in how social relationships are formed and maintained. This chapter will provide an overview of recent research documenting the role of (1) oxytocin and vasopressin; (2) estrogen, progesterone, and testosterone; and (3) cortisol on romantic, parental, and friendship relationships in humans. Keywords:  social behavior, relationships, affiliation, hormones, peptides, steroids

Introduction

The desire to establish and maintain warm and ­intimate social bonds with other individuals is considered a basic human need (Baumeister & Leary, 1995) and is typically referred to as the affiliation motive in the psychological literature (e.g., McClelland, 1987). According to Feldman (2012), affiliation hereby consists of two key aspects: On the one hand, it refers to a close, selective interpersonal relationship, such as the bond between a parent and child (i.e., a parent–child relationship) or between two romantic partners (i.e., an intimate relationship); on the other hand, it refers to the processes or actions required to establish and maintain these relationships, such as caregiving/nurturing, but also more general processes such as behavioral approach, emotion recognition, and trustworthy behavior toward others. Inspired by decades of research establishing that the endocrine system is crucially involved in regulating social affiliation in nonhuman animals, research has recently begun to address the relationship between hormones and social affiliation in humans (cf. McCall & Singer, 2012).

Neuropeptides and Human Relationships

The neuropeptides oxytocin (OT) and vasopressin (AVP) have long been associated with bonding and close relationships in animal research (reviewed in Young, Gobrogge, Liu, & Wang,  2011), although greater attention has typically been paid to OT (Carter, 1998, 2014). Synthesized primarily in the preoptic and paraventricular nuclei of the hypothalamus, OT and AVP are released peripherally into the bloodstream via the pituitary and centrally via direct projections to a number of brain areas ­involved in social cognition. The actions of OT are exclusively mediated by a single oxytocin receptor, which is ­expressed throughout the brain and body. By contrast, there are three known cognate receptors for AVP, including two (the vasopressin receptors 1a  and 1b) that are expressed in the brain and a third (the vasopressin receptor 2) that is primarily ­expressed in the kidney and regulates renal function (for reviews see Ebstein, Knafo, Mankuta, Chew, & Lai, 2012; Feldman, Monakhov, Pratt, & Ebstein, 2016). This section discusses research linking OT and AVP to parental behavior and to romantic

259

relationships in humans, but first we begin with a brief description of the methods most commonly used to assess the functioning of the OT and AVP systems.

Oxytocin and Vasopressin Assessment Methods

Peripheral assessment of OT and AVP. The interplay of OT and AVP with close relationships in humans has most often been studied by assessing peripheral levels of these hormones (i.e., in blood, saliva, or urine). Of these peripheral pathways, human OT has been most commonly studied in plasma, which refers to the portion of blood that remains after anticoagulant agents have been removed. However, the literature on plasma OT has recently been marked by controversy, with the most heated debate centering on extraction, a procedure that aims to separate OT from the other proteins and molecules contained in the plasma (McCullough, Churchland, & Mendez, 2013). A thorough review of the methodological details is beyond the scope of the present chapter (see McCullough et al.,  2013; Robinson, Hazon, Lonergan, & Pomeroy,  2014; Szeto et al.,  2011), but the issue must be raised ­because the vast majority of research on close relationships and plasma OT has not used extraction (reviewed later in the sections on “Oxytocin and Human Parental Behavior” and “Oxytocin and Human Romantic Intimate Relationships”). Debate on the topic is ongoing, and some have argued that it is preferable to test unextracted OT because extraction removes a large portion of the oxytocin that is present in plasma and other body fluids (Carter,  2014; Carter et al.,  2007). The critics are correct in noting that OT values obtained from unextracted plasma are typically 10- to 100-fold larger than those obtained from extracted plasma (McCullough et al., 2013), but this does not constitute evidence that assays on unextracted plasma are invalid per se. Moreover, it is noteworthy that ­intranasal OT administration increases levels of OT as measured using unextracted plasma (Weisman, Zagoory-Sharon, & Feldman, 2012) and extracted plasma (Domes et al., 2010; Gossen et al., 2012; Striepens et al., 2013). Thus, OT measured from unextracted plasma appears to bear some relation to actual levels of OT as manipulated by experimenters. Nevertheless, it bears consideration that much of the foundational research on the OT–relationships link is currently in doubt as a result of this ongoing 260

controversy over extraction.1 Therefore, when ­discussing studies of peripheral OT, we will briefly summarize the more abundant findings from studies measuring OT in unextracted samples before going on to highlight findings using extracted samples wherever these are available. Pharmacological manipulation of OT (and AVP). In the past decade, research on oxytocin in humans has been greatly advanced by the widespread adoption of intranasal oxytocin administration as a means of experimentally manipulating central OT levels under controlled laboratory settings. Various studies have now documented, respectively, that increased levels of OT are detected in rhesus macaque cerebrospinal fluid (CSF) after intranasal administration (Chang, Barter, Ebitz, Watson, & Platt, 2012), that the structurally similar neuropeptide AVP is detected in human CSF after intranasal administration (Born et  al.,  2002), and, most recently and most relevantly, that OT levels in human CSF rise after intranasal administration of OT (Striepens et al., 2013). This method has now been combined with a wide array of tasks and measures targeting a variety of specific cognitions and behaviors (for broad reviews, see Bos, Panksepp, Bluthé, & van Honk,  2012; Striepens, Kendrick, Maier, & Hurlemann, 2011). A similar approach has been used to study the ­effects of AVP, albeit to a much lesser extent (cf. Bos et al., 2012). Although beyond the scope of the present review, it should be noted that the question of whether intranasally administered neuropeptides are able to cross the blood–brain barrier continues to be debated (see Leng & Ludwig, 2016). Other critics note that larger sample sizes and a greater emphasis on replication are needed (Walum, Waldman, & Young, 2016), and it is hard to disagree with these recommendations. These criticisms notwithstanding, however, intranasal administration remains as the best established method for ­experimentally manipulating levels of OT and AVP in humans, so its use is likely to continue for the foreseeable future.

 Additional methodological concerns have been raised regarding between-study variation in the type of assay used (radioimmunoassay vs. enzyme-linked immunoassay; Szeto et al.,  2011) and in the chemicals used as preservatives in the tubes used to collect the plasma (lithium heparin vs. EDTA; see Robinson et al.,  2014). However, these factors do not appear to affect the results as strongly as extraction, which has been the focus of the controversy and is the only factor considered here.

1

Endocrinology of Social Rel ationships and Affiliation

The genetics of the OT-AVP system. Several candidate genes in the OT-AVP system have been studied in relation to close relationships in humans, and the focus in this review is on two genes that have been most extensively examined: the vasopressin receptor 1a gene (AVPR1A) and the oxytocin receptor gene (OXTR; for reviews see Ebstein et al., 2012; Feldman et al., 2016). In view of the need for brevity, we do not discuss the influence of other genes here, but we note that research has begun to accumulate on the effects of variants in the gene encoding the CD38 protein involved in OT secretion (Bartz & McInnes, 2007) on parental (Feldman et al., 2012) and romantic relationships (Algoe & Way,  2014). Additional work has investigated variants in the genes encoding OT and AVP2 with respect to ­maternal behavior (e.g., Mileva-Seitz et al., 2013). The research on parenting in relation to these genes and to other genetic markers unrelated to the OT-AVP system has recently been reviewed elsewhere (see Mileva-Seitz, Bakermans-Kranenburg, & van IJzendoorn, 2016). Candidate gene studies and studies relying on individual single nucleotide polymorphisms (SNPs) have been criticized for being underpowered to detect effects that, when present, are generally quite small. For this reason, we focus on questions of sample size and replication in our coverage of this line of research.

Oxytocin and Human Parental Behavior

Peripheral OT and parental behavior. Studies of OT in unextracted plasma have found increased levels in men and women who have recently become parents (Ulmer-Yaniv et al.,  2016). Moreover, a series of studies using behavioral coding of parent–infant ­interactions have found correlations of basal plasma OT levels with affectionate parenting behavior in  mothers (Feldman, Weller, Zagoory-Sharon, & Levine, 2007; Gordon, Zagoory-Sharon, Leckman, & Feldman, 2010a) and with stimulatory parenting behavior in fathers (Gordon et al., 2010a). Subsequent studies found that the latter two trends emerged for measures of OT in unextracted saliva as well, but not for urinary OT (Feldman, Gordon, & ZagoorySharon, 2011), and that plasma OT in fathers was also associated with higher levels of father–infant coordinated exploratory play and affect synchrony   The two genes, known by the symbols OXT and AVP, are close to one another in the genome and are therefore best treated as a single locus.

2

during interactions with their 6-month-old infants (Gordon, Zagoory-Sharon, Leckman, & Feldman, 2010b). Further work found that plasma OT levels were positively correlated with the extent of affectionate contact provided by mothers and fathers during a 15-minute play period with their­ 4- to 6-month-old infants (Apter-Levi, ZagoorySharon, & Feldman, 2014). Parental interactions with infants have also been associated with increases in OT relative to baseline, but this tendency appears to vary depending on individual differences in the parent. Mothers providing high levels of ­affectionate contact during an interaction with their 4- to 6-month-old infant child showed increases in plasma and salivary OT following the ­interaction, and fathers displaying higher levels of stimulatory contact during the ­interaction showed similar elevations in plasma and salivary OT following the interaction (Feldman, Gordon, Schneiderman, Weisman, & Zagoory-Sharon, 2010). Fewer studies have been published on OT in e­xtracted plasma and parental behavior, but one such study found that mothers with higher levels of separation anxiety and depression during pregnancy had significantly lower plasma OT postpartum (Eapen et al., 2014). Another study using extracted plasma did not report on associations with basal levels of OT, but did find that a mother’s OT response following exposure to her infant was significantly associated with increased gaze toward the infant and decreased gaze away from the infant during a subsequent interaction (Kim, Fonagy, Koos, Dorsett, & Strathearn, 2014). Additional research using extracted plasma to replicate the myriad associations found with unextracted plasma is greatly needed. Intranasal OT and parental behavior. In consonance with the findings from unextracted plasma, intranasal OT infusion has been found to increase o­ bserver-rated parental sensitivity and warmth in fathers’ i­ nteractions with their children (Naber, Poslawsky, van IJzendoorn, van Engeland, & Bakermans-Kranenburg, 2013; Naber, van IJzendoorn, Deschamps, van Engeland, & Bakermans-Kranenburg, 2010; Weisman et al., 2012). However, OT administration to mothers with postnatal depression increased their self-reported intention to use a “harsh caregiving strategy” in ­response to the sound of an infant crying (Mah, van IJzendoorn, Out, Smith, & Bakermans-Kranenburg, 2017). A previous report from this group had indicated that postnatally depressed mothers reported greater depressive symptoms following intranasal

Anderl, Saphire-Bernstein, and Chen

261

OT, but also that OT administration was associated with increased self-reported levels of bonding with the infant (Mah, van IJzendoorn, Smith, & BakermansKranenburg, 2013). OXTR and parental behavior. One of the first and most influential studies associating variants in OXTR with social behavior investigated the influence of the SNP rs53576 on sensitive mothering in a relatively small sample (N = 159; BakermansKranenburg & van IJzendoorn, 2008). These authors reported a significant negative effect of the A allele at rs53576 on sensitive mothering when entered into a simultaneous regression with the serotonin transporter gene–linked promoter repeat variant (entered on its own, there was a similar but marginal trend). Subsequent efforts to replicate this finding have been mixed. One study has replicated the ­association of the G allele of rs53576 with warm and sensitive parenting in a large sample of mothers (N = 500), finding a significant additive effect of the A allele such that A-allele homozygotes reported the least maternal sensitivity, heterozygotes reported intermediate levels of sensitivity, and G-allele homozygotes reported the highest levels of sensitivity (Klahr, Klump, & Burt, 2015). This effect appeared to be sex specific in the latter study (fathers were tested too, but there were no significant effects). Encouraging as this replication may be, it should be noted that the precise contrast used (additive effect of the A allele) was different from the one used in the original study (A-allele carriers vs. G-allele homozygotes), and the largest difference appeared to be ­between A-allele homozygotes and G-allele carriers. However, other studies have not replicated the effect of the G allele at rs53576 on sensitive parenting. In one study with a small sample (N = 35), the A-allele homozygotes demonstrated the highest levels of sensitive parenting (Michalska et al., 2014), whereas a second study with a larger sample (N > 700) found nonsignificantly reduced levels of maternal sensitivity in A-allele homozygotes but nonsignificantly elevated levels of maternal sensitivity in heterozygotes, relative to G-allele homozygotes (Cents et al., 2014). Lastly, a study with a sample of 201 mothers suggests that the effect of rs53576 may depend on the social context; mothers who were G-allele homozygotes demonstrated the highest levels of maternal sensitivity in the presence of low levels of interparental conflict, but they showed the lowest levels of maternal sensitivity in the presence of high levels of interparental conflict (Sturge-Apple, Cicchetti, Davies, & Suor,  2012). Additional research and 262

well-conducted meta-analyses will be necessary before firm conclusions can be drawn regarding the influence of rs53576 on human parenting behavior. A number of additional SNPs in OXTR have been linked with parenting behavior. Feldman and colleagues (2012) found that two OXTR SNPs were significantly associated with sensitive parenting (N = 272 mothers and fathers): a SNP in intron 3  named rs2254298 and a SNP in the 3-prime untranslated region of the gene named rs1042778. Most recently, rs968389, a SNP in the first exon of OXTR, was also associated with observer-coded maternal warmth, as well as with extracted plasma OT (Mehta et al., 2016). This variant is particularly intriguing because of its location in a probable regulatory region for the gene, and because the authors provided functional validation by associating the SNP with known regulatory elements and demonstrating a genotype-based difference in transcription levels of the gene. However, it must be noted that the sample in this study was relatively small (N = 127) for a genetic association study.

Oxytocin and Human Intimate Relationships

Peripheral OT and intimate relationships. In monogamous prairie voles, OT has long been implicated in maternal behavior and partner preference formation (for reviews see Carter, 2014; Young et al., 2011). In  humans, however, research on OT in unextracted plasma has yielded seemingly contradictory ­results. On one hand, some researchers have r­ eported that plasma OT levels are elevated in members of newly formed long-term romantic partnerships (Schneiderman, Zagoory-Sharon, Leckman, & Feldman,  2012; Schneiderman, Kanat-Maymon, Zagoory-Sharon, & Feldman, 2014), and one study found that a relationship partner’s plasma OT levels were correlated with the other partner’s degree of empathic behavior during a laboratory-based discussion of problems in the relationship (Schneiderman, Kanat-Maymon, Zagoory-Sharon, & Feldman, 2014). However, other research has found a seemingly contradictory pattern of results, such that levels of plasma OT were elevated in women whose relationships were distressed (Taylor et al.,  2006; Taylor, Saphire-Bernstein, & Seeman, 2010), and there was no association between measures of relationship quality and plasma OT in men (Taylor et al., 2010). In the latter study, the authors proposed that the increased OT might serve to increase the motivation to either repair the relationship or find a new one.

Endocrinology of Social Rel ationships and Affiliation

Thus, OT might be an indication of the need for greater bonding, whether in the context of a newly formed relationship (Schneiderman et al., 2012) or in the context an established relationship that is no longer as close and reliable as it might once have been (as in Taylor et al., 2010). Recently, Grebe and colleagues (2017) sought to harmonize these findings by proposing the “identify and invest” model of OT, which posits that OT levels are elevated when important relationships are threatened or are in some way in need of attention. The authors presented initial support for this theoretical framework in two studies investigating levels of unextracted salivary OT. Changes in salivary OT following the completion of a task in which participants were asked to write down thoughts about their romantic partner were positively correlated with the extent of an individual’s commitment to the relationship, but they were negatively correlated with the individual’s perception of his or her partner’s commitment to the relationship. For more on this research, see Grebe and Gangestad (this volume). Studies measuring OT in extracted plasma and saliva have also found significant associations of plasma OT with measures of intimate relationship quality (Holt-Lunstad, Birmingham, & Light, 2015; Smith et al., 2013; Turner, Altemus, Enos, Cooper, & McGuiness, 1999). One of these studies reported significant positive correlations between relationship quality and OT in plasma and saliva (Holt-Lunstad et al., 2015), whereas a second study found that OT levels in plasma were elevated in people with distressed primary relationships (Turner et al., 1999), and a third study found a negative but nonsignificant association between relationship quality and plasma OT (Smith et al., 2013). Intranasal OT and intimate relationships. Two studies have assessed the ability of OT to reduce the negative consequences of intracouple conflict (Ditzen et al., 2009, 2013). In the first of these, OT was given to both members of a couple, followed by a “standard instructed couple conflict discussion” (Ditzen et al., 2009). Relative to couples that received placebo, couples that received OT showed more positive communication behavior, relative to negative behavior, during the conflict discussion. In addition, the couples that received OT showed lower levels of cortisol following the conflict discussion. These patterns were evident in both men and women. By contrast, a second study using the same procedure found that intranasal OT was related to decreased levels of salivary α-amylase following couple conflict

in female partners but increased levels of salivary α-amylase following couple conflict in male partners (Ditzen et al., 2013). This increase in salivary α-amylase in males was associated with increased emotional arousal and positive behavior during the conflict discussion, whereas no such effects were found in females. These two studies were the first to explore the possibility of using intranasal OT as an adjunct to couples therapy, a possibility that is worthy of additional, more extensive investigation (cf. Wudarczyk, Earp, Guastella, & Savulescu, 2013). Another connection between OT and relationships that has recently been explored concerns the operation of relationship maintenance ­mechanisms, which are cognitive processes that serve to maintain and enhance intimate relationships. In one study of men in long-term relationships, those given ­intranasal OT preferred to maintain greater interpersonal distance from a female experimenter (but not from a male experimenter), relative to partnered men given a placebo (Scheele et al., 2012). A second study used an approach/avoidance paradigm (pushing or pulling on a joystick in response to positive or negative stimuli, respectively) to show that partnered men given OT were slower to “approach” an image of an attractive alternative than were partnered men given placebo. The authors interpreted these ­behavioral indicators of avoidance following OT administration as arising from a relationship defense mechanism that guides committed individuals away from the threat to the relationship posed by attractive alternatives. Subsequent work from this group has shown increased activation in the nucleus accumbens (a region of the brain primarily associated with reward processing) when viewing images of one’s partner’s face following intranasal OT relative to placebo both in partnered men (Scheele et al., 2013) and in naturally cycling partnered women (but not in partnered women using hormonal contraceptives; Scheele, Plota, Stoffel-Wagner, Maier, & Hurlemann,  2016). These latter findings suggest that intranasal OT may also increase the tendency for partnered individuals to enhance perceptions of their partners’ attractiveness and positive characteristics. Other, similar work designed to investigate this question using a relationships memory paradigm did not find increased recall of positive (nor decreased recall of negative) memories involving one’s partner, but did find that participants given OT ­recalled fewer positive memories of their past partners (Cardoso, Kalogeropoulos, Brown, Orlando, & Ellenbogen, 2016). This might be construed as

Anderl, Saphire-Bernstein, and Chen

263

contributing to relationship maintenance by devaluing previous romantic partners, but the results would be stronger if significant effects had also been detected for memories of the current partner. Even so, the totality of the evidence to date suggests that OT may support or enable the increased expression of  both positive (partner-enhancing) and negative (alternative-derogating) relationship maintenance mechanisms. OXTR and intimate relationships. Walum and ­colleagues (2012) found an association between the 3-prime-UTR-proximal SNP rs7632287 and partner bonding in an initial “discovery” sample (N = 1,240), and the same SNP was also significantly associated with a global measure of relationship quality in a subsequent “replication” sample (N = 1,771). Carriers of the minor A allele reported lower levels of partner bonding or relationship quality, relative to G-allele homozygotes. Other SNPs, including rs53576, rs2254298, and rs1042778, were not significantly associated with partner bonding in this study. A second study on the role of OXTR in intimate relationships investigated the effects of five SNPs in  the gene on communicative empathy ­between ­relationship partners (N = 60 couples) during the ­completion of a standardized laboratory paradigm designed to study intimate relationships (Schneiderman, Kanat-Maymon, Ebstein, & Feldman, 2014). Communicative empathy was rated by independent judges based on video recordings of participants’ behavior in a support-giving interaction with their relationship partners. Possessing more “risk” alleles across the five SNPs was significantly correlated with less empathic communication while supporting a relationship partner, both in males and in females. However, rs1042778 was the only SNP that was significantly associated with empathic communication in tests of single-marker associations. Other research supports a role for rs1042778 in social relationships generally (Creswell et al., 2015) and, as noted earlier, in sensitive parenting in particular (Feldman et al., 2012).

Arginine Vasopressin and Parental Relationships

Peripheral and intranasal AVP and parental behavior. Very little research has examined the relationship between AVP and human parental behavior. Plasma AVP was correlated with the extent of stimulatory contact by both mothers and fathers during three-way family interactions including both parents and their 264

child (Apter-Levi et al., 2014). Similarly, intranasal AVP administered to expectant fathers increased their interest (as indexed by viewing time) in baby-related avatars, relative to controls (Cohen-Bendahan, Beijers, van Doornen, & de Weerth, 2015), although there was no significant effect of intranasal OT on expectant fathers’ interest in baby-related avatars. AVPR1A and parental behavior. A microsatellite marker3 known as RS3 has been associated with measures of maternal behavior in a handful of studies (Avinun, Ebstein, & Knafo,  2012; Bisceglia et  al.,  2012; Leerkes, Su, Calkins, Henrich, & Smolen, 2017). However, there is an impediment to integrating the findings across these studies owing to inconsistencies in the approach to code RS3, which can have 15 to 20 alleles in any given sample. Some researchers have zeroed in on specific alleles, particularly one known as 334, as prospective risk alleles (e.g., Avinun et al., 2012; Walum et al., 2008), whereas others have adopted a splitting approach to derive “short” and “long” alleles (e.g., Bisceglia et al., 2012; Leerkes et al., 2016). Using the former approach, Avinun and colleagues (2012; N = 135 mothers and their twins) found that mothers who were carriers of the 334 allele demonstrated reduced maternal sensitivity during a laboratory-based “free play” session with their 3.5-year-old children. Other studies have used the splitting approach, however, finding in one case that mothers with two copies of the long allele (which includes 334 in these studies) is associated with reduced maternal sensitivity (Bisceglia et al.,  2012; N = 151 mothers) and in ­another case that long allele carriers engaged in less positive cognitive responses to infant crying, which in turn predicted lower levels of sensitive maternal behavior (Leerkes et al., 2016; N = 207 mothers). Unfortunately, there is no legitimate basis for splitting the continuously distributed RS3 marker, as there is no functional evidence to support such a physiological threshold. Moreover, the application of a general rule, such as “split allele groups at the median,” may hamper the ability of scholars to compare findings across samples with different medians and therefore with different split points. Future studies would be well advised to utilize the allele-wise approach instead of the questionable splitting ­ ­approach (cf. Walum et al., 2008).

  A microsatellite is a tract of repetitive DNA in which certain DNA sequence motifs are repeated.

3

Endocrinology of Social Rel ationships and Affiliation

Arginine Vasopressin and Human Intimate Relationships

Peripheral and intranasal AVP and intimate relationships. Peripheral AVP has been far less frequently studied, relative to peripheral OT, but a few investigations have yielded suggestive findings. Taylor and colleagues (2010) found a positive correlation between plasma AVP levels and distress to men’s primary pair-bond relationships. Interestingly, these effects were not found in women, suggesting that AVP may play a sex-specific role in influencing human reproductive strategies. Another study found that, in both men and women, plasma AVP levels were positively correlated with a self-report measure of social support received from one’s spouse and negatively correlated with the frequency of negative social interactions with their spouse over the preceding month (Gouin et al., 2012). Both of these studies measured AVP in unextracted plasma, and so the criticisms reviewed earlier regarding the use of unextracted plasma in OT assessment may well be applicable to the measurement of AVP (see McCullough et al.,  2013). In general, the state of knowledge regarding the measurement of peripheral AVP is considerably less advanced by comparison with the measurement of OT. The literature on intranasal AVP remains similarly underdeveloped in comparison to the literature on intranasal OT. However, a pioneering study on military families struggling with posttraumatic stress disorder (PTSD) found that AVP administration improved the ability of the male partner to detect the female partner’s expressions of anger (i.e., they showed more rapid attentional engagement with their partner’s expressions of anger; Marshall, 2013). Whether the effect would extend to relationship functioning more broadly or to populations not characterized by PTSD remains to be determined. AVPR1A and intimate relationships. The first documented association between a genetic variant and an aspect of human intimate relationships ­investigated the length of the microsatellite marker RS3 in a sample of 552 Swedish twins and tested whether carriers of any of the most frequently ­observed alleles had significantly different scores on measures of relationship quality, relative to noncarriers of that particular allele (Walum et al., 2008). This study found that the 334 allele was associated with significantly lower scores on an ad hoc measure of partner bonding and on a well-established measure

of marital satisfaction. Further exploration of 334 allele carriers showed that they were significantly more likely to be in unmarried cohabiting relationships rather than in marriages and that they were significantly more likely to be at risk for divorce. Importantly, the effects of the 334 allele were only found to be significant for the male participants in the study, whereas a similar pattern was tested but not found for the twins’ female partners, who were also typed in the investigation. This suggests that the 334 allele may only influence the relationships of male carriers, but not those of female carriers. The only attempt to directly replicate the findings of Walum et al. (2008) remains unpublished, although preliminary results appear negative (Jern,  Westberg, & Walum,  2016; N=647 men). However, some degree of corroboration has been provided by an investigation into SNPs associated with the prevalence of drug use disorders in a large case-control study (Maher et al., 2011). The authors typed a few hundred SNPs in and around several dozen preselected candidate genes in a discovery sample of 359 males with substance use disorder and 138 male controls and found that several tightly linked SNPs in the second exon of AVPR1A were significantly associated with both drug use disorder and a measure of relationship quality. The authors then selected one of the tightly linked SNPs, rs11174811, and functionally validated it using bioinformatics analyses, which showed that the SNP affects the ability of certain micro-RNA molecules (miRNAs) to bind with the DNA at that particular spot, thus reducing the ability of these miRNAs to interfere with the transcription of the gene. Finally, Maher and colleagues (2011) replicated the association of rs11174811 with drug use disorder in two ­additional samples, and in one of these samples (N = 2,231 European American male twins) found that the SNP was also associated with a measure of marital warmth (the other follow-up sample did not measure relationship quality). This study is somewhat limited by the fact that it is difficult to infer the ­direction of the effects of rs11174811 on relationship quality in the two samples based on the information presented. Even so, this study strengthens the case for AVPR1A as a candidate gene influencing the quality of human intimate relationships. On a somewhat different note, Zietsch and colleagues (Zietsch, Westberg, Santtila, & Jern, 2015) found nominally significant associations between several SNPs in AVPR1A, including rs11174811, and

Anderl, Saphire-Bernstein, and Chen

265

extra-pair mating, defined as whether or not women in committed relationships reported having more than one sex partner during the course of the preceding year (the same association was not significant in men in this study). Moreover, this study found that overall variation in AVPR1A was also significantly associated with extra-pair mating according to a gene-based test. However, it should be noted that the overall incidence of extra-pair mating in this sample was relatively rare. Thus, while female A-allele homozygotes at rs11174811 were relatively more likely to report extra-pair mating, the total number of A-allele homozygotes in the sample was only 13 (out of an overall sample of 1,563 females), with 5 of those having engaged in extra-pair mating (accounting for 5.4 percent of the 92 women reporting extra-pair mating) and the remaining ­ 8 having not engaged in extra-pair mating (accounting for only 0.5 percent of the 1,471 women who did not engage in extra-pair mating). Whether this finding is related to the results of the study by Maher and colleagues (2011) remains to be determined. No significant associations were found ­between the RS3 variant in AVPR1A and extra-pair mating in this study.

Sex Hormones and Human Relationships

Besides the neuropeptides oxytocin and vasopressin, sex steroid hormones, that is, androgens (e.g., testosterone), estrogens (e.g., estradiol), and progestogens (e.g., progesterone), show associations with human social behavior. Originally mainly discussed in the context of sexual maturation and the (physiological) regulation of sexual behavior and pregnancy, a potential bidirectional relationship between these hormones and the formation and maintenance of close social bonds (especially intimate relationships and parenting) and more general social-affiliative processes has been increasingly recognized over the last decade. Sex steroid hormones are primarily released by the gonads (i.e., testes in men and ovaries in women) and to a lesser degree by adrenal glands; furthermore, they can be produced both centrally and ­peripherally through conversion from precursor hormones. Whereas androgens, estrogens, and progestogens are all present in both sexes, pronounced sex differences exist in the concentrations of these hormones; androgen levels are higher in men and estrogens and progestogens are higher in women (see, e.g., van Anders & Gray, 2007). Sex steroid hormone action occurs via binding to classspecific receptors (i.e., androgen, estrogen, and progesterone receptors) that are widespread in the 266

brain, including limbic areas known to be crucially involved in the regulation of social-affiliative behavior such as the amygdala and the hippocampus (see, e.g., Cushing, Perry, Musatov, Ogawa, & Papademetriou,  2008; Hebbard, King, Malsbury, & Harley, 2003).

Sex Hormone Assessment Methods

Most of the knowledge on the relationship between sex steroid hormones and social-affiliative behavior to date has been established in animal models, but methodological advances now allow for systematic empirical research on this link directly in humans. The three most commonly used methods to study relationships between sex steroid hormones and social behavior in humans are briefly introduced and discussed in the following. Peripheral assessment of sex hormones. Sex steroid hormones can be measured peripherally via blood serum/ spots, saliva, or urine samples, thereby a­llowing the examination of associations between endogenous levels of these hormones and behavior. Peripheral assessment is widely used in the study of sex ­ ­hormone–behavior relationships and has led to great advances in understanding the role of sex steroid hormones in social-affiliative behavior. Most commonly, this method is used to relate differences in baseline levels between individuals to stable interindividual differences in social behavior at longer time scales (i.e., over weeks or months). However, peripheral assessment also renders it possible to i­nvestigate associations between the reactivity (i.e., deflection from baseline concentration after a hormone-eliciting intervention; e.g., Carré, Campbell, Lozoya, Goetz, & Welker, 2013; Welling, Moreau, Bird, Hansen, & Carré, 2016) or natural fluctuations (e.g., during the menstrual cycle; see, e.g., Derntl, Kryspin-Exner, Fernbach, Moser, & Habel,  2008) of a specific sex steroid hormone with social behavior. Despite its multiple valuable applications for the study of sex hormone–behavior associations, several limitations and challenges should be kept in mind when planning or interpreting empirical studies using peripheral hormone assessment (for more detailed discussions of these challenges, see, e.g., Granger, Shirtcliff, Booth, Kivlighan, & Schwartz, 2004; Shirtcliff, Granger, & Likos, 2002; Shirtcliff et al.,  2000). For instance, although sex hormone concentrations obtained with the different peripheral measurement methods (i.e., blood serum/spots, saliva, or urine samples) are correlated, they are not identical and the validity of these methods seems to

Endocrinology of Social Rel ationships and Affiliation

differ depending on sex (cf. Shirtcliff et al., 2002); as a result, findings are sometimes hard to compare between studies and sexes. Additionally, peripheral levels do not reflect the (regionally specific) availability of the respective hormone in the brain, and individual differences in receptor sensitivity and (regionally specific) density are expected to moderate hormone–behavior associations (cf. van Wingen et al., 2008). Moreover, sex hormone levels are not static, but are subject to substantial fluctuations due to biological rhythms (e.g., menstrual cycle, diurnal patterns) and context dependence. Therefore, when investigating relationships between sex hormone concentrations and stable interindividual differences in behavior (e.g., personality traits), it is of crucial importance to thoroughly control for these factors. Furthermore and most important, as natural hormone levels are measured without direct experimental manipulation, this method inevitably precludes causal conclusions about hormone–behavior effects. Menstrual cycle variation in sex hormones. Another method that has become increasingly common in the study of sex hormones and affiliation is examining social processes across the female menstrual cycle and relating it to the naturally occurring, systematic fluctuations in sex hormone levels. As this method provides a systematic and ecologically valid way to investigate hormone–behavior relationships within the same individual, it is often referred to as a “natural experiment” (e.g., Eisenlohr-Moul, DeWall, Girdler, & Segerstrom, 2015). The natural hormonal variations observed across the cycle are essential for reproduction, since sexual intercourse can only result in conception during a relatively short fertile time window around ovulation. Consequently, it has been suggested that natural selection may have established cycle-related psychological adaptations that help to enhance reproductive fitness by promoting behaviors that increase the likelihood of mating with males of high genetic quality during the fertile phase and/or support the fetus’s odds of  survival after conception/in the luteal phase (e.g., Conway et al., 2007; Gangestad, Thornhill, & Garver-Apgar 2005). Therefore, from an evolutionary psychology perspective, investigating behavioral fluctuations across the menstrual cycle is particularly interesting. Studies examining hormone–behavior associations across the menstrual cycle have applied a ­variety of approaches. Common approaches include between- and within-subjects designs comparing behavior between phases of the menstrual cycle

that are characterized by different hormonal profiles (with low estradiol and progesterone levels in the early follicular phase, high estradiol and low progesterone levels in the late follicular/ovulatory phase, and high estradiol and progesterone levels in the midluteal phase), often combined with peripheral sex hormone assessment (e.g., Derntl et al., 2008) or counting methods to estimate individual hormone levels on a single test day based on typical hormone fluctuation patterns across the menstrual cycle (e.g., Anderl et al.,  2015). All of these approaches have their unique benefits and limitations. For instance, within-subjects designs are typically preferable to between-subjects approaches as they take into account interindividual differences in baseline ­ hormone levels and variability across the cycle. However, they are susceptible to memory and learning effects (and other effects related to repeated testing), a problem that might be particularly pronounced in the social domain. Similarly, whereas counting methods are more cost efficient than methods involving actual hormone measurement and often less susceptible to belief or memory effects than methods that require pretest determination of cycle phases or repeated testing, they are less accurate due to high inter- and intrapersonal variability in cycle lengths and therefore only recommended in large samples and/or in combination with an ovulation test (cf. Gangestad et al.,  2016). Importantly and independent of the particular approach used, whereas hormone–behavior associations across the menstrual cycle may help to identify likely hormonal candidates driving these relationships, causal interpretations are generally not possible for findings from menstrual cycle studies because sex steroid hormonal levels are not experimentally manipulated with this method (for more detail on menstrual cycle effects, see Welling & Burriss, this volume). Pharmacological manipulation of sex hormone action. It is possible to investigate the effects of androgens, estrogens, and progestogens on social-affiliative processes by manipulating the levels of these hormones in pharmacological experiments (typically via exogenous sublingual or subcutaneous administration of the hormone itself, although several more advanced methods/combinations of methods have been introduced recently; see, e.g., Goetz et al., 2014). When appropriately applied in a (double-blind) placebo-controlled design, systematic pharmacological manipulation is the only ­approach available for research in humans that allows for causal inference about how a specific hormone

Anderl, Saphire-Bernstein, and Chen

267

a­ ffects social behavior. Therefore, this approach is generally promising for the study of sex hormone effects on social behavior and has been used over the last decade at least for testosterone (reviewed in Bos et al., 2012). When interpreting findings from administration studies, it should, however, be kept in mind that it presently remains unknown what proportion of administered hormones in fact enters the brain/ specific brain areas of interest (cf. McCall & Singer, 2012). Furthermore, administration studies induce relatively unnatural stimulations with regard to dosage and dynamics of absorption, so it is currently unclear how the distribution of exogenously delivered hormones relates to their endogenous release and whether some of the described effects for one hormone may actually be generated by their metabolites rather than by the hormone itself (see, e.g., Bos et al., 2012).

Sex Hormones in Intimate Relationships

Converging lines of empirical evidence suggest that the role of sex steroid hormones for psychological processes may go beyond their modulating role on sexual motivation and behavior. In line with the theoretical assumption that there is an evolutionary trade-off between the amplifying effects of testosterone for “mating success” (i.e., the successful establishment of sexual relationships) and its deleterious effects on caregiving/nurturance, often referred to as the challenge hypothesis (Wingfield, Hegner, Dufty, & Ball, 1990), research suggests that there are bidirectional associations between testosterone and committed romantic relationships that may be sex dependent (for reviews see, e.g., Gray, McHale, & Carré,  2017; Roney & Gettler,  2015; van Anders & Gray, 2007). Testosterone. Several cross-sectional studies show that testosterone levels are lower in men who are married or in a committed romantic relationship compared to men who are single; partnered men with high testosterone have furthermore been found to report lower relationship quality and to be more likely to divorce compared to partnered men with low testosterone (e.g., Booth & Dabbs, 1993; van Anders & Goldey,  2010). Importantly, a recent study comparing men and women who were single, in a long-distance relationship, or in a same-city ­relationship found that single men had higher ­testosterone levels compared to both long-distance and same-city partnered men; this has been interpreted as evidence that differences in men’s testos268

terone levels may r­eflect the strength of romantic commitment rather than regular direct contact with a romantic partner. Interestingly, in the same study, women and men differed in this respect as same-city partnered women had lower testosterone levels compared to both single women and women in long-distance r­elationships (van Anders & Watson, 2007). Generally, findings regarding the relationship between testosterone and intimate partnerships are more mixed among women than in men: Whereas some studies suggest that relationship satisfaction and commitment are negatively related to testosterone in women (e.g., van Anders & Goldey, 2010), others indicate that the association between testosterone levels and relationship status might be reversed in women, at least during the early stages of the relationship (e.g., Marazziti & Canale, 2004). Notably, recent studies provide first evidence for dyadic associations between both partners’ testosterone levels; in a study in romantic couples, relationship satisfaction and commitment were negatively related to their partners’ testosterone concentrations in both women and men, consistent with the notion that high testosterone may be detrimental for maintenance of nurturing/close social relationships (Edelstein, van Anders, Chopik, Goldey, & Wardecker, 2014). In sum, the available evidence supports the ­assumption that there is a (negative) bidirectional relationship between testosterone and commitment in intimate relationships, at least among men, in accordance with the challenge hypothesis. Interestingly, in contrast to the substantial empirical evidence on the association between the “male” sex hormone testosterone and romantic relationships, studies on the relationship between the “female” sex hormones estradiol and progesterone and ­romantic commitment are almost entirely lacking at present. There is a clear need to address this topic in future research.

Sex Steroid Hormones in ParentalBehavior

Testosterone. The challenge hypothesis posits that there are also bidirectional links with testosterone levels and parental caregiving/nurturance. Evidence from a variety of species that exhibit high paternal investment in caregiving supports this notion ­(reviewed in Saltzman & Ziegler, 2014). In humans, both mothers and fathers have been found to have  lower testosterone levels than nonparents in cross-sectional comparisons, again in accordance with findings on romantic relationships (e.g., Barrett

Endocrinology of Social Rel ationships and Affiliation

et al., 2013; Gray, Kahlenberg, Barrett, Lipson, & Ellison,  2002; see also Boyette & Gettler, this volume). Importantly, two recent longitudinal investigations provided first evidence on the directionality of the observed relationship (i.e., the question whether men with lower testosterone levels are more likely to become fathers and/or whether fatherhood suppresses testosterone levels). Specifically, in a large, representative study in Philippine men, single men with higher testosterone levels were more likely to become fathers within the following 4.5 years; after transition to fatherhood, however, their testosterone levels decreased. These decreases in testosterone were furthermore substantially stronger in fathers who interacted more regularly with their children (Gettler, McDade, Feranil, & Kuzawa, 2011). Correspondingly, expectant U.S. fathers’ testosterone (and estradiol) levels were found to decrease throughout their partner’s pregnancy, and larger declines in these hormones were associated with greater involvement in caregiving toward their child postpartum (Edelstein et al.,  2017). Together, these findings support the notion that higher testosterone levels may be beneficial for mating success but detrimental for caregiving. Thus, the decrease in testosterone levels typically observed in (expectant) fathers may serve as a proximate mechanism preparing men to provide nurturant care for their infants, thereby ultimately increasing their chances of survival. However, in apparent contrast to these findings, studies on testosterone reactions to baby cries in men (Fleming, Corter, Stallings, & Steiner,  2002; Storey, Walsh, Quinton, & Wynne-Edwards, 2000) and on the effects of testosterone administration on the neural responses to baby cries in women (Bos, Hermans, Montoya, Ramsey, & van Honk, 2010) suggest that baby cries in fact increase parental testosterone levels, which in turn seem to enhance neural responsivity to the cries, putatively to prepare for parental caregiving behavior. A recent study may, however, help to resolve the apparent paradox: van Anders, Tolman, and Volling (2012) found that men’s testosterone levels only increased when they were listening to infant cries when no caregiving response was possible. In contrast, when put in a situation in which it was possible to show a nurturing response, men’s testosterone levels in fact decreased in the presence of baby cries. This pattern of results has been interpreted in light of the steroid/peptide theory of social bonds (van Anders, Goldey, & Kuo, 2011), suggesting that the typically observed negative association between testosterone levels and parenting in fact only holds for nurturant behavior,

whereas it is reversed in situations that elicit challenge or threat. Estradiol and progesterone. In addition to the empirical evidence relating parenting behavior to testosterone levels, a number of studies suggest that both estradiol and progesterone might also play a role in modulating parental behavior, even though the specific contributions of the particular hormones and effects of timing are not well understood at present. For instance, whereas the levels of these hormones in expecting mothers were not related to feelings of attachment toward their child throughout pregnancy, the pattern of change in the ratio of estradiol to progesterone from early to late pregnancy was associated with feelings of attachment postpartum in a longitudinal investigation. In particular, mothers showing a more pronounced decline in the ratio of estradiol to progesterone (or of estradiol levels) from pregnancy to the early postpartum period reported lower feelings of attachment toward their newborn child (Fleming, Ruble, Krieger, & Wong, 1997). Furthermore, higher quality maternal behavior at one year postpartum was found to be predicted by unique gestational profiles of estradiol, progesterone, and the estrogen-to-progesterone ratio throughout pregnancy in another longitudinal study (Glynn, Davis, Sandman, & Goldberg, 2016). Thus, similar to testosterone, estradiol and progesterone may play a  role in modulating paternal ­behavior. However, future research will be needed to better understand the contributions of each particular hormone and their interactions, as well as the degree to which the observed effects are moderated by timing (e.g., during different stages of pregnancy or postpartum).

Sex Hormones and General SocialAffiliative Processes

In addition to their putative role in regulating specific kinds of social relationships such as intimate and parental partnerships, sex hormones may be involved in modulating more general social processes or motivational states that can facilitate or impede the successful initiation and maintenance of close social relationships (without necessarily being considered to be “affiliation” or “bonding” hormones per se; see Gangestad & Grebe, 2016). Indeed, there is increasing evidence from well-controlled laboratory studies that several general social processes that are known to be important for the formation and maintenance of close social bonds, such as social approach– avoidance motivation, social cognitive and evaluative

Anderl, Saphire-Bernstein, and Chen

269

processes, and cooperative behavior, may be modulated by sex steroid hormones in humans. Testosterone. Testosterone administration alters social evaluations by both enhancing ratings of facial dominance and reducing ratings of other individuals’ facial trustworthiness in those who are high in dispositional trust (Bos, Terburg, & van Honk, 2010; Olsson Kopsida, Sorjonen, & Savic,  2016). Converging lines of evidence suggest that, at least in women, testosterone is involved in regulating social approach and avoidance tendencies (e.g., Enter, Spinhoven, & Roelofs,  2014), reduces e­mpathic mimicry of emotional facial expressions (Hermans, Putman, & van Honk, 2006), and decreases the ability to recognize emotions and infer mental states in others (e.g., Olsson et al., 2016; van Honk et al., 2011; for reviews, see Bos et al., 2012; Eisenegger, Haushofer, & Fehr, 2011). In ­accordance with these findings, a recent study provided initial evidence that lower basal testosterone levels and decreases in testosterone throughout a get-to-know-you task with a stranger were associated with increased closeness toward this stranger after the task (Ketay, Welker, & Slatcher, 2016). Findings from a study conducted in a large, U.S.  sample of middle-aged to older men further suggest that men who feel emotionally supported by a greater number of individuals have lower testosterone levels than men who report less emotional support, independent of their marital and parenting status (Gettler & Oka,  2016); this supports the view that by lowering social cognitive abilities, testosterone might impede the maintenance of supportive social relationships. In contrast, however, a recent study conducted among members of a male rugby team found that team members with higher endogenous testosterone levels were actually more popular than team members with lower testosterone levels, at least when their cortisol levels were low (Ponzi, Zilioli, Mehta, Maslov, & Watson,  2016). These studies may indicate that testosterone levels and the successful establishment and maintenance of social relationships are associated, but that the direction of this link may be moderated by the specific social setting (e.g., a more or less competitive environment and/or a mixed-sex vs. purely male group). Estradiol. Whereas effects of testosterone on general social-affiliative processes such as social cognition, evaluation, and motivation have been studied quite intensively in placebo-controlled administration 270

studies over the last decade, less is known about the degree to which estradiol and progesterone are involved in these processes. In fact, to our knowledge, no study has yet explored the effects of progesterone and only one study has explored the ­effects of estradiol on social-affiliative processes in a  placebo-controlled administration study (Olsson et al., 2016; reviewed in Bos et al., 2012). Regarding estradiol, baseline levels of this hormone have been reported to be related to unique combinations of attachment style and intimacy motivation in both women and men (Edelstein, Stanton, Henderson, & Sanders, 2010), and attenuated estradiol responses were observed in women with an avoidant attachment style (Edelstein, Kean, & Chopik,  2012). Furthermore, several recent studies suggest that women may be more demanding/less generous toward others at times in their menstrual cycle when estradiol levels are typically elevated, even though estradiol levels were not directly assessed in these studies (e.g., Anderl et al.,  2015; Eisenbruch & Roney,  2016). Whether estradiol may indeed be involved in the regulation of intimacy motivation, interpersonal demandingness, and generosity, as these studies may indicate, should be addressed in future acute estradiol administration studies. Progesterone. Moreover, studies relating baseline progesterone levels/reactivity to social-affiliative behavior, as well as studies observing social-affiliative behavior across the menstrual cycle, support the view that in addition to testosterone (and putatively estradiol), progesterone may be involved in regulating processes relevant to the initiation and maintenance of close social bonds beyond parental (and potentially romantic) relationships. In particular, several studies have provided evidence that at least in women, progesterone concentrations are elevated during or after a variety of situations presumed to increase affiliation arousal, including social inclusion, social rejection, and intimate conversations (e.g., Brown et al.,  2009; Seidel et al.,  2013). Together with ­evidence obtained across the menstrual cycle, these findings have been interpreted as suggesting that progesterone may increase sensitivity to social information to help women consolidate their social support networks during pregnancy, when progesterone levels are high and support from others is particularly needed (Maner & Miller, 2014). However, as noted by Gangestad and Grebe (2016), this interpretation may be challenged by the observation that even though increased support from others would still provide important fitness benefits postpartum,

Endocrinology of Social Rel ationships and Affiliation

l­ actating women’s progesterone levels are in fact low. Furthermore, the ability to recognize other individuals’ facial emotions, which can be interpreted as an indicator of sensitivity to social information and is known to be highly relevant for successful social interactions, was in fact found to be negatively ­associated with naturally fluctuating progesterone levels across several studies (e.g., Derntl et al., 2008; Derntl, Hack, Kryspin-Exner, & Habel,  2013). Whereas these findings also support the idea that, much like testosterone and potentially estradiol, progesterone may be an important modulator of social-affiliative behavior, they might point to a more complex relationship than previously assumed. Future studies will therefore be needed to address the apparent inconsistencies in findings, ideally using a pharmacological manipulation approach.

Cortisol and Human Relationships

Cortisol is a steroid hormone released by the ­adrenal glands. It is typically released during the mammalian stress response—commonly called the “fight or flight” response. Physiologically, this ­response includes activation of the hypothalamicpituitary-adrenal (HPA) axis, so named because it involves a cascade of hormonal messengers including corticotropin-releasing hormone (CRH) by the hypothalamus, followed by adrenocorticotropic hormone (ACTH) by the pituitary gland, and ­finally—as the primary end-product—cortisol by the adrenal glands. In addition to its numerous effects on peripheral organs (e.g., influences on metabolic and immune function), cortisol also exerts central effects by binding to glucocorticoid receptors that are located throughout in the brain. Effects of cortisol on human memory processes have been widely documented (reviewed in Het, Ramlow, & Wolf, 2005; Wolf, 2009; see also Ervin & Choleris, this volume). Of more direct relevance to this chapter, recent research has also begun to explore whether and how cortisol may influence social cognition and affiliative behavior.

Cortisol Assessment Methods

Three methods are commonly used to examine the systematic effects of cortisol on human psychology. Each of the methods has unique strengths and weaknesses, reviewed briefly next. Peripheral assessment of basal levels of cortisol. One widely used cortisol assessment method involves assessing baseline levels of cortisol present in the periphery, typically using saliva, during a particular

time of day (cortisol exhibits a diurnal pattern, rising upon waking and then falling throughout the day). Especially when repeated measurements are taken at different time points, this approach can provide a profile of the individual’s chronic exposure to cortisol in daily life. Chronic elevated exposure to cortisol has been linked to tissue damage and dysregulation of biological systems (reviewed in Miller, Chen, & Zhou, 2007). Like most chronic-assessment methodologies, however, this approach is correlational and is thus limited in its ability to shed light on causal pathways and directionality of effects. In fact, most research in this domain assumes that psychological events (e.g., trauma) precipitate broader changes in the chronic functioning and activation patterns of the HPA axis, rather than the other way around (see Miller et al.,  2007). Time-lagged analyses can ameliorate the inherent limitations of this approach to reveal evidence for the opposite ­direction of causality (i.e., of cortisol on psychology). For example, cortisol levels can be measured at an  earlier time point and used to predict social outcomes at a later time point, while controlling for other variables. Peripheral assessment of acute cortisol reactivity. A second method for studying the effects of cortisol on human psychology involves laboratorybased acute stress induction. The most widely used lab-based psychosocial stress induction method is the Trier Social Stress Task (TSST; Kirschbaum, Pirke, & Hellhammer, 1993), in which the participant gives a speech in front of a panel of judges (see also Dickerson & Kemeny, 2004, for a review of acute stress methodologies). The logic of the stress induction approach is to elicit a natural increase in cortisol levels that can then be linked to downstream consequences on cognition and behavior. As laboratory conditions can be kept highly standardized, this approach provides greater experimental control than daily assessment methods, though it sacrifices a degree of ecological validity. However, it does not provide a fully controlled ­experimental means of investigating the effects of cortisol on cognition and behavior. Stress precipitates other physiological and psychological processes in addition to cortisol release, and these ­additional processes (e.g., release of CRH, ACTH, oxytocin, and other hormones) may also contribute to the downstream consequences of acute stress. Measuring and statistically controlling for such ­additional processes can help to rule out alternate mechanisms of action.

Anderl, Saphire-Bernstein, and Chen

271

Pharmacological manipulation of cortisol action. A third methodology used to investigate the effects of cortisol on psychological processes is exogenous administration. Most commonly, cortisol is administered orally through a caplet containing hydrocortisone, at a dosage that will yield salivary cortisol levels in the upper range of that obtained through physical or psychological stress an hour after ingestion. This method provides greater experimental control than diurnal assessment and stress induction, as it is possible to administer an exact dosage in a placebo-controlled design. However, it is also less ecologically valid than the other methods (precisely because the exogenous cortisol is introduced artificially, or outside the typical context in which it would be naturally released by the body), and similar concerns that have been discussed in the prior sections with respect to other hormones apply to cortisol as well (e.g., ambiguity regarding the proportion of administered hormones that in fact enter the brain/specific brain areas of interest, dynamics of absorption).

Effects of Cortisol on Social Relationships and Affiliation

Here, we review research using all three cortisol assessment methodologies, operating under the assumption that converging evidence from the combination of approaches—rather than any one singly—provides a richer and more nuanced picture about the underlying processes of interest. In contrast to emerging evidence about the social effects of oxytocin and testosterone, but similar to the current state of research on vasopressin, estradiol, and progesterone, there is currently little evidence to suggest that the effects of cortisol are specific to intimate or parental relationships (and potentially friendships). Instead, the effects of cortisol on social cognition and social behavior seem to be broader and perhaps mediated by increased risk tolerance. Therefore, this section is primarily organized by the methodology used in the studies, starting with relevant correlational studies and followed by experimental designs using stress induction and exogenous administration methods. Cortisol and social network activity. A handful of studies suggest that basal levels of cortisol are associated with social activity and status within a social network. For example, in a cohort of largely female students in a competitive nursing program, higher salivary cortisol levels were associated with less social network activity (i.e., fewer self-reported friends; Kornienko, Clemans, Out, & Granger,  2014). In 272

another large mixed-sex social network (a competitive collegiate marching band), higher cortisol levels near the beginning of an academic year in September were associated with greater turnover in friendship ties in the following two months (Kornienko, Schaefer, Weren, Hill, & Granger,  2016). In contrast, a study conducted on children aged 5 to 12 in a small rural village in Dominica revealed no ­relationship between cortisol levels and children’s network size (Ponzi, Muehlenbein, Geary, & Flinn, 2016), though it is worth noting that the sample size in this study (N = 40) was substantially smaller than in the two studies that reported positive associations. However, findings regarding a relationship between basal cortisol and network density (i.e., the extent to which one’s friends are friends with each other) are equivocal, with the study conducted in Dominica suggesting that children with higher basal cortisol had lower network density (Ponzi, Muehlenbein et al.,  2016), whereas the study with nursing students revealed a nonsignificant trend in the opposite direction (Kornienko, Clemans, Out, & Granger, 2013). Overall, the social network literature provides some initial evidence that higher levels of basal cortisol are associated with lower social activity and less stability in friendships. One interpretation of these results is that high basal or chronic levels of cortisol may hinder social activity, consistent with a study showing that in romantically unattached young adults, higher basal cortisol is linked to greater self-reported attachment anxiety (Gordon et al., 2008). Cortisol, impulsivity, and risk taking. Although cortisol administration typically does not cause changes in subjective affect or mood (Putman & Roelofs, 2011), it has been linked to impulsivity and risk-taking behavior—particularly in men—in a number of studies. In one study, young men who received a pill containing 20 mg hydrocortisone showed impaired performance after 45 minutes on the Cognitive Reflection Test, responding in a less deliberative and more intuitive (but incorrect) manner than men who had received placebo (Margittai et al., 2016). In another study, young men showed a greater preference to make risky choices with large potential payoffs in a gambling task after receiving a 40-mg dose of hydrocortisone relative to when they received a placebo (Putman, Antypa, Crysovergi, & van der Does, 2010). Men receiving 10 mg hydrocortisone made more impatient choices 15 minutes later in a temporal discounting task relative to men receiving placebo (the effect was not seen after 195 minutes;

Endocrinology of Social Rel ationships and Affiliation

Cornelisse, van Ast, Haushofer, Seinstra, & Joels, 2013). Men and women who showed a robust cortisol response to the TSST showed greater risk seeking for gains in an economic decision-making task than participants who did not undergo the stressor or who did not show a robust cortisol response to it (Buckert, Schwieren, Kudielka, & Fiebach, 2014). In both men and women, 20 mg of hydrocortisone decreased the acoustic startle response (Buchanan, Brechtel, Sollers, & Lovallo,  2001); indeed, one possible mechanism through which cortisol is thought to increase risk taking is by reducing fear and anxiety (Soravia et al., 2006). Cortisol and sensitivity to social threat. The research reviewed previously suggests that cortisol may reduce fear and promote greater risk taking. How might these cognitive and emotional shifts play out in the domain of social cognition? A small number of studies have tested participants’ response to social stimuli (e.g., photos of emotional faces) after administration of cortisol. Two studies using different methodologies suggest that cortisol administration diminishes preconscious attention to threatening social information in the form of faces displaying angry or fearful expressions (Putman, Hermans, Koppeschaar, van Schijndel, & van Honk,  2007; van Peer, Spinhoven, & Roelofs, 2010). In both male and female participants, cortisol administration was associated with decreased neural responses to angry faces (but also faster reactions to all facial expressions after being provoked in an aggression induction paradigm; Bertsch, Böhnke, Kruk, Richter, & Naumann, 2011). In general, cortisol may reduce sensitivity to social threat, which may in turn either promote aggression (Böhnke, Bertsch, Kruk, Richter, & Naumann, 2010) or—in contrast— promote affiliative tendencies. Cortisol and “tend and befriend” psychology. Research on the effects of cortisol in the social domain suggests that it may promote either aggression or affiliation (see Mogilski et al., this volume), raising the question of when, and for whom, these contrasting effects might be more likely. Traditionally, the ­ HPA axis response to acute stress has been described as a psychological mechanism that prepares the organism to mobilize a rapid response to a threat (i.e., the fight-or-flight response; Cannon,  1932). Behaviorally, this response can be characterized by aggression toward, or defensive withdrawal from, the source of the stress. Taylor et al. (2000) proposed that evolutionary pressures on women in

particular may have promoted the development of a response to stress that contrasts with fight or flight—namely, the tendency to “tend and befriend.” This theory suggests that women might be more inclined than men to engage in protective behaviors toward offspring (“tend”) and assistance seeking (“befriend”) when faced with threat. This pattern of affiliative behavior is assumed to have evolved as an effective strategy to protect offspring and oneself from harm, thereby increasing the likelihood of survival. A classic series of studies (Schachter, 1959) documented women’s tendency to affiliate when stressed (demonstrated by a preference to sit in proximity with strangers while anticipating an experiment that was described as involving electric shocks). More recent research has demonstrated that healthy men show a greater tendency to engage in prosocial ­behavior (trust, trustworthiness, and sharing) after undergoing the TSST (von Dawans, Fischbacher, Kirschbaum, Fehr, & Heinrichs,  2012). Men who ­release more cortisol in response to the TSST subsequently report greater feelings of closeness with a stranger after a get-to-know-you task (Berger, Heinrichs, von Dawans, Way, & Chen, 2016). There may, however, be sex differences in how TSSTinduced cortisol elevations are linked to social cognition (Smeets, Dziobek, & Wolf, 2009). Specifically, there is a positive correlation between the amount of stress-induced cortisol release and subsequent performance on a mindreading task (involving inferring the mental states of characters in a film) among men, whereas women demonstrate the opposite pattern. Open questions. In sum, recent research provides some evidence that cortisol release may promote ­behavioral patterns of response to stress that could be characterized as consistent with fight or flight or tend and befriend. The unifying link between these two apparently opposing patterns of response may be the connection between the presence of cortisol and a higher general tolerance for risk. It is possible that in the social domain, presence of cortisol may encourage a disinhibition of social fears that can promote either aggression or social approach depending on the specific circumstances. Although sex differences in these domains have been documented, the evidence reviewed here also suggests that fightor-flight and tend-and-befriend responses to stress are both readily available as part of the repertoire of men and women. More research incorporating both

Anderl, Saphire-Bernstein, and Chen

273

male and female participants will be necessary to ­examine the conditions that tend to elicit these divergent responses and shed light on remaining open questions about sex differences in this domain. Also, more work will be necessary to disentangle the broader effects of stress (and the entire suite of physiological changes that accompany it) versus the more specific effects of cortisol. Some studies showing an effect of acute stress on cognition or ­behavior report no correlation between the magnitude of observed cognitive or behavioral response and the magnitude of the cortisol response. In these cases, it is possible that a different ­mechanism of action besides cortisol secretion is responsible for the reported changes in cognition or behavior. Another possibility is that some effects of cortisol are dichotomous in nature (i.e., occurring at a certain threshold of cortisol release, which is typically passed in the context of an experimentally induced stressor). A combination of methods, for example comparing the effects of stressinduced cortisol secretion versus exogenous administration of cortisol within the same samples and using the same designs, may be necessary to shed light on this issue. Furthermore, more research that systematically manipulates the timing of cortisol administration and the subsequent assessment of social responses will be necessary to reconcile some of the results that have been documented. Given that cortisol is the primary end product of the HPA response and is part of an inhibitory feedback loop that blocks the secretion of CRH, high levels of cortisol may initially indicate a heightened stress response yet also suggest a more efficient “shutdown” of the same response after a time lag. Just as the effects of cortisol on memory seem to vary substantially ­depending on whether the cortisol administration occurs before or after memory encoding (see Het et al., 2005), the effects of cortisol on social cognition and behavior may vary substantially depending on the social context and timing of administration. Another potentially productive avenue for research in this domain may be to systematically vary the context (social vs. nonsocial) to determine the specificity of some of the effects of cortisol.

Conclusions and Future Directions

Together, the evidence presented in this chapter indicates that the endocrine system is crucially ­involved in regulating human social-affiliative processes, thereby contributing to the initiation and maintenance of close social bonds. At present, the 274

majority of studies investigating possible uni- or bidirectional links between hormones and socialaffiliative processes have focused on the neuropeptide oxytocin and the sex hormone testosterone, whereas evidence on potential associations between vasopressin, estradiol, progesterone, and cortisol with social-affiliative processes remains scarce and clearly needs further study. Although we have mainly highlighted research focusing on each hormone in isolation, a multiplehormone integrative approach is ultimately needed, despite the many challenges standing in the way of a more complete integration of these disparate strands of endocrinological findings. As hormonal systems are highly interdependent, an important direction for future research will be to systematically assess, or manipulate, hormone levels in combination, for example, cortisol and oxytocin (e.g., Heinrichs, Baumgartner, Kirschbaum, & Ehlert, 2003; see also Taylor et al., 2000, for discussion about the possible role of oxytocin in affiliative responses to stress), cortisol and testosterone (e.g., Coates & Herbert, 2008; Sherman, Lerner, Josephs, Renshon, & Gross, 2016), oxytocin and testosterone (van Anders et al., 2011), cortisol and estradiol (Tackett et al., 2015), and estradiol and progesterone (e.g., Derntl et al., 2008). Future research might furthermore benefit from measuring or manipulating hormone levels in combination with assessing hormone receptor genes (see, e.g., Chen et al.,  2015, for OXTR combined with intranasal OT administration; for the androgen receptor gene [AR] CAGn repeat polymorphism and salivary testosterone levels, see Eisenegger, Kumsta, Naef, Gromoll, & Heinrichs, 2017; Roney, Simmons, & Lukaszewski,  2010). A combined approach may not only help to resolve apparent inconsistencies between earlier findings but also help to address which of the effects observed after acute hormone administration are in fact caused by the hormone itself as opposed to their metabolites (cf. Eisenegger et al.,  2011; van Wingen, Ossewaarde, Bäckström, Hermans, & Fernández, 2011). Moreover, existing research does not always suggest a direct correspondence between the cognitive or behavioral effects of chronically elevated hormone levels and the effects of a single dose or a naturally elicited deflection of the same hormone. For example, one study suggested that exogenous cortisol enhances aggressive behavior in women, whereas basal cortisol was negatively associated with aggressive behavior in women (Böhnke et al., 2010). It is likely that acute and chronic processes have different

Endocrinology of Social Rel ationships and Affiliation

implications (e.g., a robust cortisol response to stress may be ­adaptive, whereas chronically elevated cortisol levels may indicate dysregulation). Study designs that incorporate assessment and analysis of both basal and acute hormone levels (and their interactions) may help to shed light on these differences. Finally, it has been established in animal models that hormones can regulate behavior not only by transiently modulating the activity of neural circuits/ hormonal receptors on a functional level but also via long-lasting effects that permanently change the nervous and/or endocrine system during early development, thereby causing differentiation between individuals (reviewed in Berenbaum & Beltz, 2011; Schulz, Molenda-Figueira, & Sisk,  2009; see also Hampson, this volume). At present, research on potential long-term effects of early-life hormone exposure remains limited in humans, but initial evidence in adolescents and young adults supports the notion that developmental hormone levels contribute to changes in brain organization, including limbic areas such as the amygdala and the hippocampus (reviewed in Peper, Pol, Crone, & van Honk, 2011). It may therefore be fruitful to address potential long-term effects of developmental hormone levels (and potentially hormonal disruptors such as hormonal contraceptives) on adult affiliation behavior and how developmental effects leading to stable interindividual differences and acute effects leading to intraindividual variations may interact to produce behavioral phenotypes in future research.

References

Algoe, S.  B., & Way, B.  M. (2014). Evidence for a role of the oxytocin system, indexed by genetic variation in CD38, in  the social bonding effects of expressed gratitude. Social Cognitive and Affective Neuroscience, 9, 1855–1861. Anderl, C., Hahn, T., Notebaert, K., Klotz, C., Rutter, B., & Windmann, S. (2015). Cooperative preferences fluctuate across the menstrual cycle. Judgment and Decision Making, 10(5), 400–406. Apter-Levi, Y., Zagoory-Sharon, O., & Feldman, R. (2014). Oxytocin and vasopressin support distinct configurations of social synchrony. Brain Research, 1580, 124–132. Avinun, R., Ebstein, R. P., & Knafo, A. (2012). Human maternal behaviour is associated with arginine vasopressin receptor 1A gene. Biology Letters, 8, 894–896. Bakermans-Kranenburg, M.  J., & van IJzendoorn, M.  H. (2008). Oxytocin receptor (OXTR) and serotonin transporter (5-HTT) genes associated with observed parenting. Social Cognitive and Affective Neuroscience, 3, 128–134. Barrett, E.  S., Tran, V., Thurston, S., Jasienska, G., Furberg, A.  S., Ellison, P.  T., & Thune, I. (2013). Marriage and motherhood are associated with lower testosterone concentrations in women. Hormones and Behavior, 63(1), 72–79.

Bartz, J. A., & McInnes, L. A. (2007). CD38 regulates oxytocin secretion and complex social behavior. BioEssays, 29, 837–841. Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497–529. Berenbaum, S. A., & Beltz, A. M. (2011). Sexual differentiation of human behavior: Effects of prenatal and pubertal organizational hormones. Frontiers in Neuroendocrinology, 32(2), 183–200. Berger, J., Heinrichs, M., von Dawans, B., Way, B., & Chen, F. S. (2016). Cortisol modulates men’s affiliative responses to acute social stress. Psychoneuroendocrinology, 63, 1–9. Bertsch, K., Böhnke, R., Kruk, M. R., Richter, S., & Naumann, E. (2011). Exogenous cortisol facilitates responses to social threat under high provocation. Hormones and Behavior, 59(4), 428–434. Bisceglia, R., Jenkins, J.  M., Wigg, K.  G., O’Connor, T.  G., Moran, G., & Barr, C.  L. (2012). Arginine vasopressin 1a receptor gene and maternal behavior: Evidence of association and moderation. Genes, Brain and Behavior, 11, 262–268. Böhnke, R., Bertsch, K., Kruk, M. R., Richter, S., & Naumann, E. (2010). Exogenous cortisol enhances aggressive behavior in females, but not in males. Psychoneuroendocrinology, 35(7), 1034–1044. Booth, A., & Dabbs, J.  M. (1993). Testosterone and men’s marriages. Social Forces, 72(2), 463–477. Born, J., Lange, T., Kern, W., McGregor, G. P., Bickel, U., & Fehm, H.  L. (2002). Sniffing neuropeptides: A transnasal approach to the human brain. Nature Neuroscience, 5(6), 514–516. Bos, P. A., Hermans, E. J., Montoya, E. R., Ramsey, N. F., & van Honk, J. (2010). Testosterone administration modulates neural responses to crying infants in young females. Psychoneuroendocrinology, 35(1), 114–121. Bos, P. A., Panksepp, J., Bluthé, R. M., & van Honk, J. (2012). Acute effects of steroid hormones and neuropeptides on human social–emotional behavior: A review of single administration studies. Frontiers in Neuroendocrinology, 33(1), 17–35. Bos, P.  A., Terburg, D., & van Honk, J. (2010). Testosterone decreases trust in socially naive humans. Proceedings of the National Academy of Sciences of the United States of America, 107(22), 9991–9995. Brown, S. L., Fredrickson, B. L., Wirth, M. M., Poulin, M. J., Meier, E.  A., Heaphy, E.  D., . . . Schultheiss, O.  C. (2009). Social closeness increases salivary progesterone in humans. Hormones and Behavior, 56(1), 108–111. Buchanan, T.  W., Brechtel, A., Sollers, J.  J., & Lovallo, W.  R. (2001). Exogenous cortisol exerts effects on the startle reflex independent of emotional modulation. Pharmacology, Biochemistry and Behavior, 68(2), 203–210. Buckert, M., Schwieren, C., Kudielka, B. M., & Fiebach, C. J. (2014). Acute stress affects risk taking but not ambiguity aversion. Frontiers in Neuroscience, 8, 82. Cannon, W. B. (1932). The wisdom of the body. New York, NY: W. W. Norton & Co. Cardoso, C., Kalogeropoulos, C., Brown, C.  A., Orlando, M. A., & Ellenbogen, M. A. (2016). Memory response to oxytocin predicts relationship dissolution over 18 months. Psychoneuroendocrinology, 68, 171–176. Carré, J.  M., Campbell, J.  A., Lozoya, E., Goetz, S.  M., & Welker, K.  M. (2013). Changes in testosterone mediate the  effect of winning on subsequent aggressive behaviour. Psychoneuroendocrinology, 38(10), 2034–2041.

Anderl, Saphire-Bernstein, and Chen

275

Carter, C.  S. (1998). Neuroendocrine perspectives on social attachment and love. Psychoneuroendocrinology, 23, 779–818. Carter, C.  S. (2014). Oxytocin pathways and the evolution of human behavior. Annual Review of Psychology, 65, 17–39. Carter, C. S., Pournajafi-Nazarloo, H., Kramer, K. M., Ziegler, T.  E., White-Traut, R., Bello, D., & Schwertz, D. (2007). Oxytocin: Behavioral associations and potential as a salivary biomarker. Annals of the New York Academy of Sciences, 1098, 312–322. Cents, R.  A.  M., Kok, R., Tiemeier, H., Lucassen, N., Székely,  E., Bakermans-Kranenburg, M.  J., Hofman, A., . . . Lambregtse-van den Berg, M. P. (2014). Variations in maternal 5-HTTLPR affect observed sensitive parenting. Journal of Child Psychology and Psychiatry, 55, 1025–1032. Chang, S. W., Barter, J. W., Ebitz, R. B., Watson, K. K., & Platt, M.  L. (2012). Inhaled oxytocin amplifies both vicarious reinforcement and self reinforcement in rhesus macaques (Macaca mulatta). Proceedings of the National Academy of Sciences of the United States of America, 109(3), 959–964. Chen, F.  S., Kumsta, R., Dvorak, F., Domes, G., Yim, O.  S., Ebstein, R. P., & Heinrichs, M. (2015). Genetic modulation of oxytocin sensitivity: A pharmacogenetics approach. Translational Psychiatry, 5, e664. Coates, J. M., & Herbert, J. (2008). Endogenous steroids and financial risk taking on a London trading floor. Proceedings of the National Academy of Sciences of the United States of America, 105(16), 6167–6172. Cohen-Bendahan, C. C. C., Beijers, R., van Doornen, L. J. P., & de Weerth, C. (2015). Explicit and implicit caregiving interests in expectant fathers: Do endogenous and exogenous oxytocin and vasopressin matter? Infant Behavior & Development, 41, 26–37. Conway, C. A., Jones, B. C., DeBruine, L. M., Welling, L. L. M., Smith, M.  L., Perrett, D.  I., . . . Al-Dujaili, E.  A. (2007). Salience of emotional displays of danger and contagion in faces is enhanced when progesterone levels are raised. Hormones and Behavior, 51(2), 202–206. Cornelisse, S., van Ast, V., Haushofer, J., Seinstra, M., & Joels,  M. (2013). Time-dependent effect of hydrocortisone administration on intertemporal choice. Available at SSRN 2294189. doi:10.2139/ssm.2294189 Creswell, K. G., Wright, A. G. C., Troxel, W. M., Ferrell, R. E., Flory, J. D., & Manuck, S. B. (2015). OXTR polymorphism predicts social relationships through its effects on social temperament. Social Cognitive and Affective Neuroscience, 10, 869–876. Cushing, B.  S., Perry, A., Musatov, S., Ogawa, S., & Papademetriou, E. (2008). Estrogen receptors in the medial amygdala inhibit the expression of male prosocial behavior. Journal of Neuroscience, 28(41), 10399–10403. Derntl, B., Hack, R. L., Kryspin-Exner, I., & Habel, U. (2013). Association of menstrual cycle phase with the core components of empathy. Hormones and Behavior, 63(1), 97–104. Derntl, B., Kryspin-Exner, I., Fernbach, E., Moser, E., & Habel, U. (2008). Emotion recognition accuracy in healthy young females is associated with cycle phase. Hormones and Behavior, 53(1), 90–95. Dickerson, S. S., & Kemeny, M. E. (2004). Acute stressors and cortisol responses: A theoretical integration and synthesis of laboratory research. Psychological Bulletin, 130, 355–391. Ditzen, B., Nater, U. M., Schaer, M., La Marca, R., Bodenmann, G., Ehlert, U., & Heinrichs, M. (2013). Sex-specific effects of intranasal oxytocin on autonomic nervous system and

276

emotional responses to couple conflict. Social Cognitive and Affective Neuroscience, 8, 897–902. Ditzen, B., Schaer, M., Gabriel, B., Bodenmann, G., Ehlert, U., & Heinrichs, M. (2009). Intranasal oxytocin increases positive communication and reduces cortisol levels during couple conflict. Biological Psychiatry, 65, 728–731. Domes, G., Lischke, A., Berger, C., Grossmann, A., Hauenstein, K., Heinrichs, M., & Herpertz, S. C. (2010). Effects of intranasal oxytocin on emotional face processing in women. Psychoneuroendocrinology, 35, 83–93. Eapen, V., Dadds, M., Barnett, B., Kohlhoff, J., Khan, F., Radom, N., & Silove, D.  M. (2014). Separation anxiety, attachment and inter-personal representations: Disentangling the role of oxytocin in the perinatal period. PLoS One, 9, e107745. Ebstein, R.  P., Knafo, A., Mankuta, D., Chew, S.  H., & Lai, P. S. (2012). The contributions of oxytocin and vasopressin pathway genes to human behavior. Hormones and Behavior, 61, 359–379. Edelstein, R. S., Chopik, W. J., Saxbe, D. E., Wardecker, B. M., Moors, A.  C., & LaBelle, O.  P. (2017). Prospective and dyadic associations between expectant parents’ prenatal hormone changes and postpartum parenting outcomes. Developmental Psychobiology, 59, 77–90. Edelstein, R. S., Kean, E. L., & Chopik, W. J. (2012). Women with an avoidant attachment style show attenuated estradiol responses to emotionally intimate stimuli. Hormones and Behavior, 61(2), 167–175. Edelstein, R. S., Stanton, S. J., Henderson, M. M., & Sanders, M. R. (2010). Endogenous estradiol levels are associated with attachment avoidance and implicit intimacy motivation. Hormones and Behavior, 57(2), 230–236. Edelstein, R. S., van Anders, S. M., Chopik, W. J., Goldey, K. L., & Wardecker, B.  M. (2014). Dyadic associations between testosterone and relationship quality in couples. Hormones and Behavior, 65(4), 401–407. Eisenbruch, A.  B., & Roney, J.  R. (2016). Conception risk and  the ultimatum game: When fertility is high, women demand more. Personality and Individual Differences, 98, 272–274. Eisenegger, C., Haushofer, J., & Fehr, E. (2011). The role of testosterone in social interaction. Trends in Cognitive Sciences, 15(6), 263–271. Eisenegger, C., Kumsta, R., Naef, M., Gromoll, J., & Heinrichs, M. (2017). Testosterone and androgen receptor gene polymorphism are associated with confidence and competitiveness in men. Hormones and Behavior, 92, ­93–102. Eisenlohr-Moul, T.  A., DeWall, C.  N., Girdler, S.  S., & Segerstrom, S. C. (2015). Ovarian hormones and borderline personality disorder features: Preliminary evidence for interactive effects of estradiol and progesterone. Biological Psychology, 109, 37–52. Enter, D., Spinhoven, P., & Roelofs, K. (2014). Alleviating social avoidance: Effects of single dose testosterone administration on approach–avoidance action. Hormones and Behavior, 65(4), 351–354. Feldman, R. (2012). Oxytocin and social affiliation in humans. Hormones and Behavior, 61(3), 380–391. Feldman, R., Gordon, I., Schneiderman, I., Weisman, O., & Zagoory-Sharon, O. (2010). Natural variations in maternal and paternal care are associated with systematic changes in oxytocin following parent-infant contact. Psychoneuroendocrinology, 35, 1133–1141.

Endocrinology of Social Rel ationships and Affiliation

Feldman, R., Gordon, I., & Zagoory-Sharon, O. (2011). Maternal and paternal plasma, salivary, and urinary oxytocin and parent-infant synchrony: Considering stress and affiliation components of human bonding. Developmental Science, 14, 752–763. Feldman, R., Monakhov, M., Pratt, M., & Ebstein, R.  P. (2016). Oxytocin pathway genes: Evolutionary ancient system impacting on human affiliation, sociality, and psychopathology. Biological Psychiatry, 79, 174–184. Feldman, R., Weller, A., Zagoory-Sharon, O., & Levine, A. (2007). Evidence for a neuroendocrinological foundation of human affiliation: Plasma oxytocin levels across pregnancy and the postpartum period predict mother-infant bonding. Psychological Science, 18, 965–970. Feldman, R., Zagoory-Sharon, O., Weisman, O., Schneiderman, I., Gordon, I., Maoz, R., . . . Ebstein, R. P. (2012). Sensitive parenting is associated with plasma oxytocin and polymorphisms in the OXTR and CD38 genes. Biological Psychiatry, 72, 175–181. Fleming, A. S., Corter, C., Stallings, J., & Steiner, M. (2002). Testosterone and prolactin are associated with emotional responses to infant cries in new fathers. Hormones and Behavior, 42(4), 399–413. Fleming, A.  S., Ruble, D., Krieger, H., & Wong, P.  Y. (1997).  Hormonal and experiential correlates of maternal responsiveness during pregnancy and the puerperium in human mothers. Hormones and Behavior, 31(2), 145–158. Gangestad, S. W., & Grebe, N. M. (2016). Hormonal systems, human social bonding, and affiliation. Hormones and Behavior, 91, 122–135. Gangestad, S.  W., Haselton, M.  G., Welling, L.  L.  M., Gildersleeve, K., Pillsworth, E.  G., Burriss, R.  P., . . . Puts, D.  A. (2016). How valid are assessments of conception probability in ovulatory cycle research? Evaluations, recommendations, and theoretical implications. Evolution and Human Behavior, 37(2), 85–96. Gangestad, S.  W., Thornhill, R., & Garver-Apgar, C.  E. (2005). Women’s sexual interests across the ovulatory cycle depend on primary partner developmental instability. Proceedings of the Royal Society B: Biological Sciences, 272(1576), 2023–2027. Gettler, L. T., McDade, T. W., Feranil, A. B., & Kuzawa, C. W. (2011). Longitudinal evidence that fatherhood decreases testosterone in human males. Proceedings of the National Academy of Sciences of the United States of America, 108(39), 16194–16199. Gettler, L.  T., & Oka, R.  C. (2016). Aging US males with multiple sources of emotional social support have low testosterone. Hormones and Behavior, 78, 32–42. Glynn, L. M., Davis, E. P., Sandman, C. A., & Goldberg, W. A. (2016). Gestational hormone profiles predict human maternal behavior at 1-year postpartum. Hormones and Behavior, ­85, 19–25. Goetz, S.  M., Tang, L., Thomason, M.  E., Diamond, M.  P., Hariri, A.  R., & Carré, J.  M. (2014). Testosterone rapidly increases neural reactivity to threat in healthy men: A novel two-step pharmacological challenge paradigm. Biological Psychiatry, 76(4), 324–331. Gordon, I., Zagoory-Sharon, O., Leckman, J. F., & Feldman, R. (2010a). Oxytocin and the development of parenting in humans. Biological Psychiatry, 68, 377–382. Gordon, I., Zagoory-Sharon, O., Leckman, J. F., & Feldman, R. (2010b). Prolactin, oxytocin, and the development of

paternal behavior across the first six months of fatherhood. Hormones and Behavior, 58, 513–518. Gordon, I., Zagoory-Sharon, O., Schneiderman, I., Leckman, J. F., Weller, A., & Feldman, R. (2008). Oxytocin and cortisol in romantically unattached young adults: Associations with bonding and psychological distress. Psychophysiology, 45, 349–352. Gossen, A., Hahn, A., Westphal, L., Prinz, S., Schultz, R. T., Gründer, G., & Spreckelmeyer, K.  N. (2012). Oxytocin plasma concentrations after single intranasal oxytocin administration—A study in healthy men. Neuropeptides, 46, 211–215. Gouin, J.-P., Carter, C.  S., Pournajafi-Nazarloo, H., Malarkey, W.  B., Loving  T.  J., Stowell, J., & Kiecolt-Glaser, J.  K. (2012). Plasma vasopressin and interpersonal functioning. Biological Psychology, 91, 270–274. Granger, D. A., Shirtcliff, E. A., Booth, A., Kivlighan, K. T., & Schwartz, E.  B. (2004). The “trouble” with salivary testosterone. Psychoneuroendocrinology, 29(10), 1229–1240. Gray, P. B., Kahlenberg, S. M., Barrett, E. S., Lipson, S. F., & Ellison, P. T. (2002). Marriage and fatherhood are associated with lower testosterone in males. Evolution and Human Behavior, 23(3), 193–201. Gray, P.  B., McHale, T.  S., & Carré, J.  M. (2017). A review of  human male field studies of hormones and behavioral reproductive effort. Hormones and Behavior, 91, 52–67. Grebe, N. M., Kristoffersen, A. A., Grøntvedt, T. V., Thompson, M.  E., Kennair, L.  E.  O., & Gangestad, S.  W. (2017). Oxytocin and vulnerable romantic relationships. Hormones and Behavior, 90, 64–74. Hebbard, P. C., King, R. R., Malsbury, C. W., & Harley, C. W. (2003). Two organizational effects of pubertal testosterone in male rats: Transient social memory and a shift away from long-term potentiation following a tetanus in hippocampal CA1. Experimental Neurology, 182(2), 470–475. Heinrichs, M., Baumgartner, T., Kirschbaum, C., & Ehlert, U. (2003). Social support and oxytocin interact to suppress cortisol and subjective responses to psychosocial stress. Biological Psychiatry, 54(12), 1389–1398. Hermans, E., Putman, P., & van Honk, J. (2006). Testosterone administration reduces empathetic behavior: A facial mimicry study. Psychoneuroendocrinology, 31, 859–866. Het, S., Ramlow, G., & Wolf, O.  T. (2005). A meta-analytic review of the effects of acute cortisol administration on human memory. Psychoneuroendocrinology, 30(8), 771–784. Holt-Lunstad, J., Birmingham, W.  C., & Light, K.  C. (2015). Relationship quality and oxytocin: Influence of stable and modifiable aspects of relationships. Journal of Social and Personal Relationships, 32, 472–490. Jern, P., Westberg, L., & Walum, H. (2016). Failure to replicate a previously reported association between genetic variation in the arginine vasopressin receptor 1a gene and human pair-bonding behavior. Unpublished manuscript. Ketay, S., Welker, K. M., & Slatcher, R. B. (2016). The roles of testosterone and cortisol in friendship formation. Psychoneuroendocrinology, 76, 88–96. Kim, S., Fonagy, P., Koos, O., Dorsett, K., & Strathearn, L. (2014). Maternal oxytocin response predicts mother-to-infant gaze. Brain Research, 1580, 133–142. Kirschbaum, C., Pirke, K.  M., & Hellhammer, D.  H. (1993). The “Trier Social Stress Test”—A tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology, 28, 76–81.

Anderl, Saphire-Bernstein, and Chen

277

Klahr, A. M., Klump, K., & Burt, S. A. (2015). A constructive replication of the association between the oxytocin receptor genotype and parenting. Journal of Family Psychology, 29, 91–99. Kornienko, O., Clemans, K.  H., Out, D., & Granger, D.  A. (2013). Friendship network position and salivary cortisol levels. Social Neuroscience, 8(4), 385–396. Kornienko, O., Clemans, K.  H., Out, D., & Granger, D.  A. (2014). Hormones, behavior, and social network analysis: Exploring associations between cortisol, testosterone, and network structure. Hormones and Behavior, 66(3), 534–544. Kornienko, O., Schaefer, D.  R., Weren, S., Hill, G.  W., & Granger, D. A. (2016). Cortisol and testosterone associations with social network dynamics. Hormones and Behavior, 80, 92–102. Leerkes, E. M., Su, J., Calkins, S., Henrich, V. C., & Smolen, A. (2017). Variation in mothers’ arginine vasopressin receptor 1a and dopamine receptor D4 genes predicts maternal sensitivity via social cognition. Genes, Brain and Behavior, 16, 233–240. Leng, G., & Ludwig, M. (2016). Intranasal oxytocin: Myths and delusions. Biological Psychiatry, 79, 243–250. Mah, B.  L., van IJzendoorn, M.  H., Out, D., Smith, R., & Bakermans-Kranenburg, M.  J. (2017). The effects of intranasal oxytocin administration on sensitive caregiving in mothers with postnatal depression. Child Psychiatry and Human Development, 48, 308–315. Mah, B. L., van IJzendoorn, M. H., Smith, R., & BakermansKranenburg, M. J. (2013). Oxytocin in postnatally depressed mothers: Its influence on mood and expressed emotion. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 40, 267–272. Maher, B.  S., Vladimirov, V.  I., Latendresse, S.  J., Thiselton, D. L., McNamee, R., Kang, M., . . . Vanyukov, M. M. (2011). The AVPR1A gene and substance use disorders: Association, replication, and functional evidence. Biological Psychiatry, 70, 519–527. Maner, J.  K., & Miller, S.  L. (2014). Hormones and social monitoring: Menstrual cycle shifts in progesterone underlie women’s sensitivity to social information. Evolution and Human Behavior, 35(1), 9–16. Marazziti, D., & Canale, D. (2004). Hormonal changes when falling in love. Psychoneuroendocrinology, 29(7), 931–936. Margittai, Z., Nave, G., Strombach, T., van Wingerden, M., Schwabe, L., & Kalenscher, T. (2016). Exogenous cortisol causes a shift from deliberative to intuitive thinking. Psychoneuroendocrinology, 64, 131–135. Marshall, A.  D. (2013). Posttraumatic stress disorder and partner-specific social cognition: A pilot study of sex differences in the impact of arginine vasopressin. Biological Psychology, 93, 296–303. McCall, C., & Singer, T. (2012). The animal and human neuroendocrinology of social cognition, motivation and behavior. Nature Neuroscience, 15(5), 681–688. McClelland, D. C. (1987). Human motivation. New York, NY: Cambridge University Press. McCullough, M. E., Churchland, P. S., & Mendez, A. J. (2013). Problems with measuring peripheral oxytocin: Can the data on oxytocin and human behavior be trusted? Neuroscience and Biobehavioral Reviews, 37, 1485–1492. Mehta, D., Eapen, V., Kohlhoff, J., Diaz, A.  M., Barnett, B., Silove, D., & Dadds, M.  R. (2016). Genetic regulation of

278

maternal oxytocin response and its influence on maternal behavior. Neural Plasticity, 2016, Article ID 5740365. Michalska, K. J., Decety, J., Liu, C., Chen, Q., Martz, M. E., Jacob, S., . . . Lahey, B.  B. (2014). Genetic imaging of the association of oxytocin receptor gene (OXTR) polymorphisms with positive maternal parenting. Frontiers in Behavioral Neuroscience, 8, 21. Mileva-Seitz, V.  R., Bakermans-Kranenburg, M.  J., & van  IJzendoorn, M.  H. (2016). Genetic mechanisms of parenting. Hormones and Behavior, 77, 211–223. Mileva-Seitz, V., Steiner, M., Atkinson, L., Meaney, M.  J., Levitan, R., Kennedy, J.  L.,  .  .  .  Fleming, A.  S. (2013). Interaction between oxytocin genotypes and early experience predicts quality of mothering and postpartum mood. PLoS One, 8, e61443. Miller, G. E., Chen, E., & Zhou, E. S. (2007). If it goes up, must it come down? Chronic stress and the hypothalamicpituitary-adrenocortical axis in humans. Psychological Bulletin, 133, 25–45. Naber, F., Poslawsky, I. E., van IJzendoorn, M. H., van Engeland, H., & Bakermans-Kranenburg, M.  J. (2013). Oxytocin enhances paternal sensitivity to a child with autism: A double-blind within-subject experiment with intranasally administered oxytocin. Journal of Autism and Developmental Disorders, 43, 224–229. Naber, F., van IJzendoorn, M. H., Deschamps, P., van Engeland, H., & Bakermans-Kranenburg, M.  J. (2010). Intranasal oxytocin increases fathers’ observed responsiveness during play with their children: A double-blind within-subject experiment. Psychoneuroendocrinology, 35, 1583–1586. Olsson, A., Kopsida, E., Sorjonen, K., & Savic, I. (2016). Testosterone and estrogen impact social evaluations and vicarious emotions: A double-blind placebo-controlled study. Emotion, 16(4), 515–523. Peper, J. S., Pol, H. H., Crone, E. A., & van Honk, J. (2011). Sex steroids and brain structure in pubertal boys and girls: A minireview of neuroimaging studies. Neuroscience, 191, 28–37. Ponzi, D., Muehlenbein, M. P., Geary, D. C., & Flinn, M. V. (2016). Cortisol, salivary alpha-amylase and children’s perceptions of their social networks. Social Neuroscience, 11(2), 164–174. Ponzi, D., Zilioli, S., Mehta, P. H., Maslov, A., & Watson, N. V. (2016). Social network centrality and hormones: The interaction of testosterone and cortisol. Psychoneuroendocrinology, 68, 6–13. Putman, P., Antypa, N., Crysovergi, P., & van der Does, W. A. J. (2010). Exogenous cortisol acutely influences motivated decision making in healthy young men. Psychopharmacology, 208(2), 257–263. Putman, P., Hermans, E. J., Koppeschaar, H., van Schijndel, A., & van Honk, J. (2007). A single administration of cortisol acutely reduces preconscious attention for fear in anxious young men. Psychoneuroendocrinology, 32(7), 793–802. Putman, P., & Roelofs, K. (2011). Effects of single cortisol administrations on human affect reviewed: Coping with stress through adaptive regulation of automatic cognitive processing. Psychoneuroendocrinology, 36(4), 439–448. Robinson, K.  J., Hazon, N., Lonergan, M., & Pomeroy, P.  P. (2014). Validation of an enzyme-linked immunoassay (ELISA) for plasma oxytocin in a novel mammal species reveals potential errors induced by sampling procedure. Journal of Neuroscience Methods, 226, 73–79.

Endocrinology of Social Rel ationships and Affiliation

Roney, J. R., & Gettler, L. T. (2015). The role of testosterone in human romantic relationships. Current Opinion in Psychology, 1, 81–86. Roney, J.  R., Simmons, Z.  L., & Lukaszewski, A.  W. (2010). Androgen receptor gene sequence and basal cortisol concen­ trations predict men’s hormonal responses to potential mates. Proceedings of the Royal Society B: Biological Sciences, 277, 57–63. Saltzman, W., & Ziegler, T. E. (2014). Functional significance of hormonal changes in mammalian fathers. Journal of Neuroendocrinology, 26(10), 685–696. Schachter, S. (1959). The psychology of affiliation. Stanford, CA: Stanford University Press. Scheele, D., Plota, J., Stoffel-Wagner, B., Maier, W., & Hurlemann, R. (2016). Hormonal contraceptives suppress oxytocin-induced brain reward responses to the partner’s face. Social Cognitive and Affective Neuroscience, 11, 767–774. Scheele, D., Striepens, N., Güntürkün, O., Deutschländer, S., Maier, W., Kendrick, K.  M., & Hurlemann, R. (2012). Oxytocin modulates social distance between males and females. Journal of Neuroscience, 32, 16074–16079. Scheele, D., Wille, A., Kendrick, K.  M., Stoffel-Wagner, B., Becker, B., Güntürkün, O.,  .  .  .  Hurlemann, R. (2013). Oxytocin enhances brain reward system responses in men viewing the face of their female partner. Proceedings of the National Academy of Sciences of the United States of America, 110, 20308–20313. Schneiderman, I., Kanat-Maymon, Y., Ebstein, R.  P., & Feldman, R. (2014). Cumulative risk on the oxytocin receptor gene (OXTR) underpins empathic communication difficulties at the first stages of romantic love. Social Cognitive and Affective Neuroscience, 9, 1524–1529. Schneiderman, I., Kanat-Maymon, Y., Zagoory-Sharon, O., & Feldman, R. (2014). Mutual influences between partners’ hormones shape conflict dialog and relationship duration at the initiation of romantic love. Social Neuroscience, 9, 337–351. Schneiderman, I., Zagoory-Sharon, O., Leckman, J.  F., & Feldman, R. (2012). Oxytocin during the initial stages of  romantic attachment: Relations to couples’ interactive reciprocity. Psychoneuroendocrinology, 37, 1277–1285. Schulz, K. M., Molenda-Figueira, H. A., & Sisk, C. L. (2009). Back to the future: The organizational–activational hypothesis adapted to puberty and adolescence. Hormones and Behavior, 55(5), 597–604. Seidel, E. M., Silani, G., Metzler, H., Thaler, H., Lamm, C., Gur, R. C., . . . Derntl, B. (2013). The impact of social exclusion vs. inclusion on subjective and hormonal reactions in females and males. Psychoneuroendocrinology, 38(12), 2925–2932. Sherman, G. D., Lerner, J. S., Josephs, R. A., Renshon, J., & Gross, J.  J. (2016). The interaction of testosterone and cortisol is associated with attained status in male execu­tives. Journal of Personality and Social Psychology, 110(6), ­921–929. Shirtcliff, E.  A., Granger, D.  A., & Likos, A. (2002). Gender differences in the validity of testosterone measured in saliva by immunoassay. Hormones and Behavior, 42(1), 62–69. Shirtcliff, E. A., Granger, D. A., Schwartz, E. B., Curran, M. J., Booth, A., & Overman, W. H. (2000). Assessing estradiol in biobehavioral studies using saliva and blood spots: Simple radioimmunoassay protocols, reliability, and comparative validity. Hormones and Behavior, 38(2), 137–147. Smeets, T., Dziobek, I., & Wolf, O. T. (2009). Social cognition under stress: Differential effects of stress-induced cortisol

elevations in healthy young men and women. Hormones and Behavior, 55(4), 507–513. Smith, T.  W., Uchino, B.  N., MacKenzie, J., Hicks, A.  M., Campo, R. A., Reblin, M., . . . Light, K. C. (2013). Effects of couple interactions and relationship quality on plasma oxytocin and cardiovascular reactivity: Empirical findings and methodological considerations. International Journal of Psychophysiology, 88, 271–281. Soravia, L.  M., Heinrichs, M., Aerni, A., Maroni, C., Schelling, G., Ehlert, U., . . . de Quervain, D.  J. (2006). Glucocorticoids reduce phobic fear in humans. Proceedings of the National Academy of Sciences of the United States of America, 103, 5585–5590. Storey, A. E., Walsh, C. J., Quinton, R. L., & Wynne-Edwards, K. E. (2000). Hormonal correlates of paternal responsiveness in new and expectant fathers. Evolution and Human Behavior, 21(2), 79–95. Striepens, N., Kendrick, K.  M., Hanking, V., Landgraf, R., Wüllner, U., Maier, W., & Hurlemann, R. (2013). Elevated cerebrospinal fluid and blood concentrations of oxytocin following its intranasal administration in humans. Scientific Reports, 3, 3440. Striepens, N., Kendrick, K.  M., Maier, W., & Hurlemann, R. (2011). Prosocial effects of oxytocin and clinical evidence for its therapeutic potential. Frontiers in Neuroendocrinology, 32, 426–450. Sturge-Apple, M. L., Cicchetti, D., Davies, P. T., & Suor, J. H. (2012). Differential susceptibility in spillover between interparental conflict and maternal parenting practices: Evidence for OXTR and 5-HTT genes. Journal of Family Psychology, 26, 431–442. Szeto, A., McCabe, P. M., Nation, D. A., Tabak, B. A., Rossetti, M.  A., McCullough, M.  E., . . . Mendez, A.  J. (2011). Evaluation of enzyme immunoassay and radioimmunoassay methods for the measurement of plasma oxytocin. Psychosomatic Medicine, 73, 393–400. Tackett, J.  L., Reardon, K.  W., Herzhoff, K., Page-Gould, E., Harden, K.  P., & Josephs, R.  A. (2015). Estradiol and cortisol interactions in youth externalizing psychopathology. Psychoneuroendocrinology, 55, 146–153. Taylor, S. E., Gonzaga, G. C., Klein, L. C., Hu, P., Greendale, G.  A., & Seeman, T.  E. (2006). Relation of oxytocin to psychological stress responses and hypothalamic-pituitaryadrenocortical axis activity in older women. Psychosomatic Medicine, 68, 238–245. Taylor, S.  E., Klein, L.  C., Lewis, B.  P., Gruenewald, T.  L., Gurung, R.  A., & Updegraff, J.  A. (2000). Biobehavioral responses to stress in females: Tend-and-befriend, not fight-or-flight. Psychological Review, 107, 411–429. Taylor, S. E., Saphire-Bernstein, S., & Seeman, T. E. (2010). Are plasma oxytocin in women and plasma vasopressin in men biomarkers of distressed pair-bond relationships? Psychological Science, 21, 3–7. Turner, R. A., Altemus, M., Enos, T., Cooper, B., & McGuinness, T. (1999). Preliminary research on plasma oxytocin in normal cycling women: Investigating emotion and interpersonal distress. Psychiatry, 62, 97–113. Ulmer-Yaniv, A., Avitsur, R., Kanat-Maymon, Y., Schneiderman, I., Zagoory-Sharon, O., & Feldman, R. (2016). Affiliation, reward, and immune biomarkers coalesce to support social synchrony during periods of bond formation in humans. Brain, Behavior, and Immunity, 56, 130–139.

Anderl, Saphire-Bernstein, and Chen

279

van Anders, S.  M., & Goldey, K.  L. (2010). Testosterone and partnering are linked via relationship status for women and “relationship orientation” for men. Hormones and Behavior, 58(5), 820–826. van Anders, S.  M., Goldey, K.  L., & Kuo, P.  X. (2011). The steroid/peptide theory of social bonds: Integrating testosterone and peptide responses for classifying social behavioral contexts. Psychoneuroendocrinology, 36(9), 1265– 1275. van Anders, S. M., & Gray, P. B. (2007). Hormones and human partnering. Annual Review of Sex Research, 18(1), 60–93. van Anders, S. M., Tolman, R. M., & Volling, B. L. (2012). Baby cries and nurturance affect testosterone in men. Hormones and Behavior, 61(1), 31–36. van Anders, S. M., & Watson, N. V. (2007). Testosterone levels in women and men who are single, in long-distance relationships, or same-city relationships. Hormones and Behavior, 51(2), 286–291. van Honk, J., Schutter, D.  J., Bos, P.  A., Kruijt, A.  W., Lentjes, E.  G., & Baron-Cohen, S. (2011). Testosterone administration impairs cognitive empathy in women depending on second-to-fourth digit ratio. Proceedings of National Academy of Sciences of the United States of America, 108(8), 3448–3452. van Peer, J.  M., Spinhoven, P., & Roelofs, K. (2010). Psycho­ physiological evidence for cortisol-induced reduction in early bias for implicit social threat in social phobia. Psycho­ neuroendocrinology, 35(1), 21–32. van Wingen, G.  A., Ossewaarde, L., Bäckström, T., Hermans, E. J., & Fernández, G. (2011). Gonadal hormone regulation of the emotion circuitry in humans. Neuroscience, 191, 38–45. van Wingen, G. A., van Broekhoven, F., Verkes, R. J., Petersson, K.  M., Bäckström, T., Buitelaar, J.  K., & Fernandez, G. (2008). Progesterone selectively increases amygdala reactivity in women. Molecular Psychiatry, 13(3), 325–333. von Dawans, B., Fischbacher, U., Kirschbaum, C., Fehr, E., & Heinrichs, M. (2012). The social dimension of stress reactivity: Acute stress increases prosocial behavior in humans. Psychological Science, 23(6), 651–660.

280

Walum, H., Lichtenstein, P., Neiderhiser, J.  M., Reiss, D., Ganiban, J.  M., Spotts, E.  L.,  .  .  .  Westberg, L. (2012). Variation in the oxytocin receptor gene is associated with pair-bonding and social behavior. Biological Psychiatry, 71, 419–426. Walum, H., Waldman, I. D., & Young, L. J. (2016). Statistical and methodological considerations for the interpretation of intranasal oxytocin studies. Biological Psychiatry, 79, 251–257. Walum, H., Westberg, L., Henningsson, S., Neiderhiser, J. M., Reiss, D., Igl, W.,  .  .  .  Lichtenstein, P. (2008). Genetic variation in the vasopressin receptor 1a gene (AVPR1A) associates with pair-bonding behavior in humans. Proceedings of the National Academy of Sciences of the United States of America, 105, 14153–14156. Weisman, O., Zagoory-Sharon, O., & Feldman, R. (2012). Oxytocin administration to parent enhances infant physiological and behavioral readiness for social engagement. Biological Psychiatry, 72, 982–989. Welling, L. L. M., Moreau, B. J. P., Bird, B. M., Hansen, S., & Carré, J.  M. (2016). Exogenous testosterone increases men’s perceptions of their own physical dominance. Psychoneuroendocrinology, 64, 136–142. Wingfield, J. C., Hegner, R. E., Dufty, A. M., Jr., & Ball, G. F. (1990). The “challenge hypothesis”: Theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. American Naturalist, 136(6), 829–846. Wolf, O. T. (2009). Stress and memory in humans: Twelve years of progress? Brain Research, 1293, 142–154. Wudarczyk, O. A., Earp, B. D., Guastella, A., & Savulescu, J. (2013). Could intranasal oxytocin be used to enhance relation­ ships? Research imperatives, clinical policy, and ethical considerations. Current Opinion in Psychiatry, 26, 474–484. Young, K. A., Gobrogge, K. L., Liu, Y., & Wang, Z. (2011). The neu­robiology of pair bonding: Insights from a socially monog­ amous rodent. Frontiers in Neuroendocrinology, 32, 53–69. Zietsch, B.  P., Westberg, L., Santtila, P., & Jern, P. (2015). Genetic analysis of human extrapair mating: Heritability, between-sex correlation, and receptor genes for vasopressin and oxytocin. Evolution and Human Behavior, 36, 130–136.

Endocrinology of Social Rel ationships and Affiliation

CH A PT E R

16

Hierarchy and Testosterone How Can Testosterone Promote Upward Mobility in Status Hierarchies?

Shawn N. Geniole and Justin M. Carré

Abstract Hierarchy, the relative ranking of individuals with respect to status or other social dimensions, is ubiquitous across human social groups. However, relatively little is known about the biological factors that may promote or inhibit mobility in status hierarchies. Prominent theoretic perspectives suggest that concentrations of testosterone in the bloodstream fluctuate dynamically in anticipation of and in response to social challenges, serving to promote behaviors aimed at gaining/maintaining status. This chapter reviews studies that have directly examined the extent to which endogenous, competitioninduced surges in testosterone predicted subsequent competitiveness and aggressiveness. Studies suggest that testosterone surges promote competitiveness, but that this effect is complex and depends on several factors. More consistent was evidence that testosterone surges predicted subsequent aggressive behavior. Testosterone–behavior links were also specific to men, suggesting that surges in this hormone may serve different functions, or may promote alternative behavioral strategies for gaining and maintaining status, in women. Keywords:  dominance, aggression, testosterone, hierarchy, status

Hierarchy, defined as the ranking of individuals relative to one another on status or other social dimensions (Magee & Galinsky, 2008), is ubiquitous across human social groups (Mazur, 2015), present even in the smallest, least complex, most isolated of human societies (Diamond,  1999; van Vugt, Hogan, & Kaiser, 2008), as well as in groups of children as young as 3 years (Boyce, 2004; Strayer & Strayer,  1976; Thomsen, Frankenhuis, IngoldSmith, & Carey, 2011). The prominence of hierarchy in human social systems has been attributed to the benefits it may provide for group functioning (Darwin, 1871; van Vugt et al., 2008). For example, hierarchy promotes cooperative goal attainment by reducing intragroup conflict (e.g., Ronay, Greenaway, Anicich, & Galinsky, 2012) and enhancing coordination when there are conflicting goals within the group (e.g., De Kwaadsteniet & Van Dijk, 2010). Further, groups that are structured with a clearer hierarchy and chain of command tend to outperform

other, less hierarchically organized, groups (reviewed in Van Vugt et al., 2008; there may be limits to such benefits, however—e.g., see Anderson & Willer, 2014). Hierarchy influences many aspects of social life (e.g., conflict resolution, resource allocation, and mating, reviewed in Cheng, Tracy, Foulsham, Kingstone, & Henrich,  2013), and gaining high status within a hierarchy improves subjective well-­ being (Anderson, Kraus, Galinsky, & Keltner, 2012), access to valued or scarce resources, and health (Ellis, 1994). In short, hierarchy is ubiquitous across social groups, emerges spontaneously, and influences the way we think, feel, and interact with each other. Although the pursuit of status is considered a fundamental human motivation (Anderson, Hildreth, & Howland,  2015; Griskevicius & Kenrick,  2013; Mazur & Booth, 1998), likely because of the aforementioned benefits that it confers, little is known about the situational and individual difference factors that influence upward mobility within status 281

hierarchies. One factor that may be particularly relevant to the pursuit of status is an individual’s circulating testosterone levels in the bloodstream (Carré & Olmstead, 2015; Eisenegger, Haushofer, & Fehr, 2011; Hamilton, Carré, Mehta, Olmstead, & Whitaker,  2015; Knight & Mehta,  2014; Mazur, 2015).1 Testosterone is the most frequently investigated hormone in status-related research (Hamilton et al., 2015; Knight & Mehta, 2014) and has been psychometrically validated as a biological, individual difference variable in humans (Sellers, Mehl, & Josephs, 2007) that may represent one’s motivation to gain and maintain status (reviewed in Eisenegger et al., 2011; Hamilton et al., 2015; Josephs, Sellers, Newman, & Mehta, 2006; Knight & Mehta, 2014; Mazur & Booth, 1998).2

The Challenge Hypothesis and the Biosocial Model of Status

The two most prominent theoretical perspectives that guide most research on testosterone and statusrelated behaviors are the challenge hypothesis and the biosocial model of status. According to the challenge hypothesis (Wingfield, 2012, 2017; Wingfield, Hegner, Dufty, & Ball, 1990), testosterone concen1   The system responsible for the regulation of testosterone secretion is the hypothalamic-pituitary-gonadal (HPG) axis. When gonadotropin-releasing hormone is secreted from the hypothalamus, it stimulates the pituitary gland, which in turn secretes luteinizing and follicle-stimulating hormones. These hormones then function to stimulate the release of testosterone from the gonads (Vadakkadath Meethal & Atwood, 2005). Once released, testosterone binds to androgen receptors either directly or indirectly through its conversion to dihydrotestosterone by the enzyme 5-α reductase (Askew, Gampe, Stanley, Faggart, & Wilson, 2007). Testosterone can also be converted into estradiol by the aromatase enzyme, after which it can bind to estrogen receptors (Ronde & Jong,  2011). Once circulating in the bloodstream, testosterone and its metabolites bind to receptors of target cells to influence gene transcription. Testosterone and its metabolites can also bind to receptors on the hypothalamus and the pituitary, acting as a negative feedback loop and suppressing the release of luteinizing and follicle-stimulating hormones (Hayes, Decruz, Seminara, Boepple, & Crowley,  2001), and ultimately regulating the secretion of testosterone. 2   A strength of investigating status-related motives through testosterone rather than self-report questionnaires is that such questionnaires are often prone to response biases (e.g., socially desirable responding; Johnson, Leedom, & Muhtadie,  2012), whereas the investigation of status-related motivations through testosterone (assessed via saliva or blood sampling) precludes such biases. Additionally, because the strength of one’s motivation to gain and maintain status can vary from situation to situation and also, in some circumstances, operate outside of conscious awareness (e.g., see Schultheiss, Dargel, & Rohde,  2003; Terburg, Aarts, & Van Honk, 2012), testosterone concentrations, which fluctuate dynamically throughout the day, may be better suited to the measurement of these motivations than self-report questionnaires (see Josephs et al., 2006).

282

Hierarchy and Testosterone

trations rise and fall in response to cues of social challenge and threat in the environment, preparing the organism for aggressive encounters over resources important for reproduction and status (e.g., mates, territory). This model was proposed to explain fluctuations in testosterone levels in birds but has since been applied to humans (Archer, 2006). The model posits that testosterone concentrations rise from a low baseline (nonbreeding baseline) to a higher baseline for the breeding season (breeding baseline), an increase that functions to promote spermatogenesis, the development of secondary sexual traits, and reproductive behavior (an increase akin to the male pubertal surge in testosterone in humans; Archer, 2006), but not aggression. More important for antagonistic contests, however, concentrations of testosterone also surge from this breeding baseline to a higher, physiological maximum, promoting the expression of aggressive behavior, which functions to secure mates and territory important for breeding. Relevantly, these fluctuations are thought to support most aggression that serves these goals (acquisition and protection of resources or status valuable for mating) specifically, rather than other goals (e.g., predator deterrence). These effects of testosterone surges on reproductive aggression are posited to be strongest for monogamous (vs. polygamous) birds during periods of social instability, such as when social hierarchies, mates, or territories important for mating are being established or challenged by conspecifics. After the breeding season, testosterone concentrations drop back down to the nonbreeding baseline (a decline that may be akin to the age-related declines seen in older adulthood in humans; e.g., Harman, Metter, Tobin, Pearson, & Blackman,  2001). This flexibility in testosterone is thought to exist in monogamous birds because of trade-offs associated with testosterone; whereas high levels increase aggressiveness (important for securing mates and thus repro­ducing), they also decrease parental care (Wingfield et al., 1990) and immune system function (see, e.g., Folstad & Karter, 2011). The challenge hypothesis has been applied to many species (see meta-analysis and review in Hirschenhauser & Oliveira, 2006), including nonhuman primates (e.g., Muller & Wrangham, 2004) and humans (see review and meta-analyses in Archer, 2006; see also Carré & Olmstead, 2015). When applied to humans, one prediction derived from the challenge hypothesis is that situational cues of competition, especially those related to mating or status, should trigger a surge in testosterone (Archer, 2006). Most studies have not examined

this research question within the context of mating, but instead examine anticipatory surges in testosterone that occur before the start of a competition. In a review and meta-analysis of such anticipation effects, Archer (2006) reported significant increases in testosterone leading up to competitions (after removal of two outliers, mean weighted d = 0.27), which became slightly stronger (mean weighted d = 0.30) when including only studies that involved actual sports contests (rather than contrived laboratory tasks). More recent studies not captured by this meta-analysis also provide some support for an anticipation effect, showing that prematch testosterone concentrations are higher than are prepractice or control day concentrations (e.g., in male basketball players, Arruda, Aoki, Carolina, & Moreira,  2017; male rugby players, Cunniffe, Morgan, Baker, Cardinale,  & Davies, 2015; in female soccer players, there is an effect in some cases, e.g., T. Oliveira, Gouveia, & Oliveira,  2009, but not others, e.g., Casto & Edwards, 2016a). Another prominent model is the biosocial model of status (Mazur, 1985, 2015; Mazur & Booth, 1998). In addition to the anticipation effects outlined previously in the challenge hypothesis, the biosocial model of status posits that concentrations of testosterone fluctuate in response to the outcomes (win vs. loss) of competitive interactions such that winners experience increases in testosterone, promoting future aggressive and assertive behaviors aimed at gaining and maintaining status, whereas losers experience decreases, promoting future deferent and submissive behaviors aimed at avoiding further injury and/or loss of status. Consistent with this idea, an earlier meta-analysis provided some evidence that winners of competitive interactions experience greater increases in testosterone than do losers (mean weighted ds ranged from 0.31 to 0.36, depending on the inclusion or exclusion of an outlier, Archer, 2006; for additional reviews of this literature, see Carré & Olmstead, 2015; G.  A.  Oliveira & Oliveira,  2014). More recently, Geniole, Bird, Ruddick, and Carré (2017) conducted an updated meta-analysis on 60 different effect sizes examining these potential winner–loser differences in testosterone. In short, they found that winners experienced greater increases (or lesser decreases) in testosterone than did losers from pre- to postcompetition, an effect that was of similar magnitude for men (mean weighted d = 0.23) and for women (mean weighted d = 0.22). The nature of these differences was also characterized by examining the specific direction of the pre- to postcompetition changes in winners and losers separately. In men, this winner–loser effect was

driven by winners showing a significant increase in testosterone (mean weighted d = 0.21) and losers showing no meaningful changes (mean weighted d = 0.01). In women, winners and losers showed divergent changes, with winners slightly increasing (mean weighted d = 0.10) and losers slightly decreasing (mean weighted d = –0.12). Therefore, the winner– loser effect in men is driven by pre–post increases in winners and negligible changes in losers, whereas in women it is driven by divergent (albeit weaker) changes, with winners increasing and losers decreasing. It was also discovered that this winner–loser effect was strongest in studies conducted in the field, rather than in the lab, possibly because the outcomes of competitions in the field are more meaningful and salient to athletes/participants than are those in contrived laboratory tasks (Geniole, Bird, et al., 2017). Other moderators not examined in the meta-analysis but that are likely important and account for some of the heterogeneity include one’s implicit power motives (e.g., Schultheiss et al.,  2005; Schultheiss, Campbell, & McClelland,  1999; Vongas & Al Hajj,  2017), self-construal (Welker et al., 2017), anxiety (although evidence is mixed: see Maner, Miller, Schmidt, & Eckel,  2008 vs. Norman, Moreau, Welker, & Carré, 2014), concentrations of other hormones (e.g., cortisol, Zilioli & Watson, 2012), whether the opponents are friends or members of an out-group (e.g., Flinn, Ponzi, & Muehlenbein,  2012), whether the outcome is ­attributed to skill versus luck (e.g., van Anders & Watson, 2007), and whether the victory was close or decisive (e.g., Zilioli, Mehta, & Watson,  2014; see also Wu, Eisenegger, Zilioli, Watson, & Clark, 2017; reviewed in G. A. Oliveira & Oliveira, 2014).

The Function of Competition-Induced Surges in Testosterone

An additional prediction that can be derived from the biosocial model of status is that competitioninduced fluctuations in testosterone play a functional role in promoting subsequent behaviors aimed at gaining and maintaining status. The challenge ­hypothesis also posits that such fluctuations should promote male-to-male aggression aimed at securing mates and territory important for mating. Neverthe­ less, this aspect of these models—that competitioninduced fluctuations in testosterone should function to promote behaviors that help the organism secure mates and status (e.g., competitiveness and aggression)—was not often investigated in earlier studies in humans (Archer, 2006; Carré, McCormick, & Hariri, 2011). Instead, most of the earlier work on Geniole and Carré

283

testosterone and aggression focused on baseline or resting-level concentrations of the hormone, which share rather weak associations with aggression (r = 0.08 in meta-analysis by Archer, Graham-Kevan, & Davies, 2005). Based on the hypothesis that testosterone surges function to promote upward mobility in status hierarchies (or the maintenance of high-status positions), we would predict that fluctuations in concentrations of this hormone drive a variety of subsequently measured behaviors and decisions that are important for climbing status hierarchies. According to the dual-strategies theory, there are two primary routes through which people can gain and maintain their standings in a social hierarchy (Cheng et al., 2013; Henrich & Gil-White, 2001; for recent review, see Case & Maner, 2016): through dominance, which involves intimidation, coercion, and aggression and functions by eliciting fear and deference in others; and through prestige, which involves the use of skills or knowledge critical for group advancement and success and functions by helping the individual gain respect and admiration from others. Given the centrality of competition to both the biosocial model of status and the challenge hypothesis, and the importance of aggression in the challenge hypothesis, here we focus on the extent to which testosterone surges motivate competitiveness and aggressiveness, behaviors more consistent with the first route to high status: dominance.3 We also focus, primarily, on studies involving natural, endogenous fluctuations in testosterone, rather than studies employing pharmacological manipulations that modulate testosterone experimentally (for reviews of studies that employed pharmacological manipulations, see Bos, Panksepp, Bluthé, & van Honk, 2012; for a recent review of literature on endogenous and exogenous testosterone, see Zilioli & Bird, 2017).

Endogenous Testosterone Surges

One limitation to this research on endogenous fluctuations in testosterone and competition and aggression is that because testosterone and behavior are proposed to share reciprocal links (i.e., surges in testosterone promote competition and aggression, and competitiveness and aggressiveness promote surges in testosterone), studies that measure testosterone changes concurrently with behavior are rather ambiguous with respect to the direction of causality. A better 3   Nevertheless, there is emerging evidence that testosterone also promotes prosocial behavior, which may be more consistent with the prestige route to high status. We also close this chapter with a brief review of this emerging literature.

284

Hierarchy and Testosterone

approach is to measure surges in testosterone and determine if these surges predict future behavior on subsequent tasks. Although this approach does not solve the third variable problem—that another variable may be causing both the increase in testosterone and the subsequent competitiveness or aggressiveness—the temporal ordering of the measures provides some reassurance that the subsequent, future behavior did not cause the preceding surge in testosterone. Here, we review evidence from studies that used such designs.

Do Competition-Induced Surges in Testosterone Predict Subsequent Decisions to Compete?

The first investigation of the relationship between testosterone fluctuations and subsequent behavior in humans involved a laboratory competition in which participants’ outcomes (win vs. loss) were predetermined through random assignment (Mehta  & Josephs,  2006). The researchers had participants compete against one another in a number-tracing task. To rig the outcomes of the competition, the researchers gave half of the participants an easy version of the task (win condition) and the other half of participants a difficult version of the task (loss condition) such that, in each testing session, one participant of the competition was likely to lose and the other was likely to win. Testosterone concentrations were determined through saliva samples obtained before and after the competitive task. After the task and postcompetition saliva sample, participants were asked if they wished to compete again against the same participant (competitive choice) or to complete an unrelated questionnaire (noncompetitive choice). If testosterone surges function to promote status-related behaviors (e.g., competitiveness aimed at gaining additional status), then changes in testosterone throughout the task should predict one’s willingness to compete in a second task. The results were partially consistent with this prediction: Changes in testosterone predicted the subsequent decision to compete again against the competitor, but only among the participants randomly assigned to the loss condition (Mehta & Josephs, 2006); losers (but not winners) who experienced an increase in testosterone were more likely to compete again than were losers who experienced a decrease in testosterone. Therefore, this study provided some initial evidence for the hypothesis that fluctuations in testosterone function to promote subsequent behaviors aimed at gaining and maintaining status, and also highlighted the importance of considering whether the

testosterone surge was accompanied or driven by a preceding win or a loss, which may be especially important for predicting participants’ decisions to have a rematch against the same opponent. In such rematches, testosterone-induced competitiveness may not be as functional for winners, who only stand to lose the increased rank they just gained (i.e., they have little to gain and everything to lose against the same opponent), compared to losers, who can potentially regain their lost status (i.e., have little to lose but everything to gain against the same opponent). The asymmetry in these costs and benefits across winners and losers may explain the differential role of testosterone in this study. What about situations in which it is advantageous for winners to compete again? One possibility is that if winners can compete against a new opponent, rather than the same one, they may have an additional opportunity to enhance their status further and, as such, testosterone may more strongly promote competition. The decision to compete would still, however, depend on the individual’s chances of winning a subsequent competition, something that may be determined, in part, by the decisiveness of his or her last victory. For example, if the individual barely won his or her last competition, he or she may perceive his or her status as unstable, and competing again may be more risky than beneficial. If, however, the individual decisively won, he or she may perceive his or her status as more stable and enduring, increasing the perceived benefit of competing again and potentially gaining additional status. In such situations, testosterone may more strongly promote competitiveness. A more recent study conducted by Mehta, Snyder, Knight, and Lassetter (2015) examined these possibilities using the same number-tracing task described earlier. Rather than manipulating whether participants won or lost in this task, the researchers instead manipulated the ease with which participants won. In the task, participants were asked to compete against a research confederate posed as another participant. During each round of the task, participants were asked to shout “done” once they had finished each trial of the task. In the close victory condition, the research confederate acted as if he or she had finished the task only moments after the participant, shouting “done” one to three seconds after the participant. In the decisive victory condition, the research confederate acted as if he or she had finished much later than the participant, instead shouting “done” 6 to 10 seconds after the participant. After the rigged competition, participants could compete again against the

same or a new opponent or complete an unrelated, noncompetitive task. The decisiveness of the preceding victory interacted with testosterone surges to promote competition: Among those who won decisively, testosterone surges predicted a greater likelihood of competing again,4 whereas among those who won narrowly, testosterone surges predicted (albeit nonsignificantly) a decreased likelihood of competing again. Therefore, testosterone–behavior links may be functionally tuned to information about the previous competition and the subsequent competition, increasing competitiveness only when such decisions are advantageous for climbing a social hierarchy.5 In certain contests, however, there are no clear winners or losers. How does testosterone promote subsequent competitiveness when the outcomes of a preceding competition are ambiguous? Another study (Carré & McCormick, 2008) investigated testosterone fluctuations and subsequent competitive behavior using a competitive computer game in which participants, throughout the task, rapidly pressed a key to earn points that were later exchangeable for money. Participants were told, however, that at random intervals their point counter may decrease by 1 point, indicating that another participant had stolen a point from them (a salient provocation or challenge). Participants could ignore the provocations and continue earning or switch keys to either protect their points for a variable amount of time or steal points back from the other player (although such behavior is costly as participants are told they do not get to keep points they steal). This task, known as the Point Subtraction Aggression Paradigm (PSAP, originally developed by Cherek, 1981), induces highly variable testosterone changes, making it well suited to the investigation of links between competitioninduced testosterone fluctuations and subsequent behavior (for review, see Geniole, MacDonell, & 4   Given that so few participants chose to compete against the same opponent (6.5 percent), the authors collapsed across these two “compete” groups and examined the extent to which testosterone surges during close versus decisive victories predicted willingness to choose one of the two competitive options in general (competing against the same or a new opponent) versus the noncompetitive option. The same pattern of results was obtained when the authors restricted the analysis to those who chose to compete against a new opponent. The authors also examined whether victory type interacted with cortisol concentrations to predict decisions to compete again but found no evidence for such interactions or main effects of baseline or changes in cortisol. 5  This process is likely operating outside of conscious awareness, however: When the researchers asked participants to predict their likelihood of winning the subsequent competition, this variable was associated with neither testosterone fluctuations nor their decision to compete again in a subsequent competition.

Geniole and Carré

285

McCormick,  2017). During the task, participants could see their own point counter, but not the point counter of the competitor, and the outcome was never announced to the participants. Instead, after the task, participants were simply asked if they wished to compete again in a second but different task against the same opponent (puzzle completion task) or to help the experimenter validate a different program (noncompetitive choice). Carré and McCormick (2008) found that participants who experienced greater increases in testosterone during the task were more likely to choose to compete again in the subsequent task than were participants who experienced lesser increases (or decreases) in testosterone. Although the outcome was neither announced nor manipulated, participants may still have formed beliefs about how they performed relative to the competitor. The authors also asked participants about these beliefs but found no associations between this measure and testosterone changes, nor did the beliefs interact with testosterone to predict competitiveness. Therefore, testosterone surges during competitions in which the outcome is ambiguous appear to promote subsequent competitiveness. It should be noted, however, that this association between testosterone fluctuations and subsequent competitiveness may have still depended on certain characteristics of the PSAP competition; for example, in other versions of the task in which participants were not provoked and/or retaliatory stealing was beneficial (rather than costly), links between testosterone fluctuations and decisions to compete were weaker/reversed (Carré, Gilchrist, Morrissey, & McCormick, 2010). Together, these studies provide some evidence that endogenous surges in testosterone, which occur during competitive interactions, may function to influence subsequent competitive behavior. Whereas both the biosocial model of status and the challenge hypothesis posit that surges in testosterone promote subsequent competitive or aggressive behaviors, this relationship does not appear to be so straightforward. Instead, testosterone surges may interact with other situational factors such as the challenge, competitiveness, or outcome of the preceding competition that induced the testosterone fluctuations and, if the outcome involved a victory, the decisiveness of the victory. Although losers who experience increases in testosterone appear more willing to compete again, winners who show similar increases only compete again if their preceding victory was decisive. If their preceding victory was narrow, surges in testosterone may instead promote a decrease in the willingness to 286

Hierarchy and Testosterone

compete again. Testosterone surges after narrow victories may thus promote status maintenance (or status defense, the avoidance of future opportunities to lose status) rather than future attempts at further status enhancement (e.g., Mehta, Snyder, et al., 2015).

Testosterone Fluctuations and Aggressive and Antagonistic Behavior

In addition to driving the willingness to compete, testosterone surges may also promote the expression of aggressive behavior, which can assist the individual in climbing a status hierarchy, primarily through the dominance route of the dual-strategies model of status outlined previously. Researchers define aggression as any behavior intended to cause harm to another individual who would rather avoid such treatment, with the behavior not necessarily being physical but at least being aversive to the recipient (Baron & Richardson, 1994). Several studies have assayed aggression using the PSAP (described earlier). Recall that when participants are provoked in this task (i.e., the other player steals a point from them), they can (1) continue earning points, (2) protect their points, or (3) steal points from the other participant. Because the points are exchangeable for money (monetary loss is aversive) and because participants do not financially benefit from retaliating and stealing points in the task (they are specifically told they do not get to keep any of the points they steal from the fictitious opponent), steal presses represent a behavioral measure of aggression. Indeed, the task has been well validated, with steal presses higher among clinical and forensic populations known for violent and aggressive tendencies and those who self-report being more aggressive, compared to control populations and those who self-report being less aggressive (reviewed in Geniole, MacDonell, et al., 2017). Men who steal more (vs. less) in the task are more likely to report using retaliatory steals as a means to protect their status (Geniole, Cunningham, Keyes, Busseri, & McCormick, 2015). In one study (Carré, Putnam, & McCormick, 2009), participants competed against one another in a number-tracing task in which the outcomes were rigged (half of participants won and half lost). After the task, participants played against the same individual in the PSAP. Men (but not women) who  experienced greater increases in testosterone during the number-tracing task were more aggressive against their opponent in the PSAP (stole more points) than were those who experienced lesser increases (or greater decreases). Although there was no interaction between these changes in testosterone

and the outcome of the number-tracing contest (win vs. loss), the effect was nonetheless driven by men assigned to the loss condition. Conversely, among men who were assigned to the win condition, the association between changes in testosterone and subsequent behavior was not significant and was instead moderated by an additional personality factor (see section on moderators later). Vongas and Al Hajj (2017) also investigated how testosterone fluctuations during the numbertracing task would predict subsequent aggressive behavior in men, this time using both the Hot Sauce paradigm (Lieberman, Solomon, Greenberg, & McGregor,  1999) and the PSAP. Testosterone increases during the number-tracing task predicted greater aggression during both tasks, but neither of the associations was significant. Although directionally consistent with the study reviewed earlier (Carré et al., 2009), the effects may have been weaker because of differences in methodology; in Carré and colleagues (2009), points earned in the PSAP were converted to money at the end of the study (thus, steals represented a form of financial aggression), whereas in Vongas and Al Hajj (2017), the points were not exchangeable for money. Without financial consequences associated with stealing, provocations may not have been as meaningful and aversive, and retaliatory stealing may not have tapped into the construct of aggression as much as it does in the standard version of the task. In another study (Carré, Campbell, Lozoya, Goetz,  & Welker,  2013), researchers examined changes in testosterone that occurred during an interactive video game in which participants either fought or played volleyball against computer opponents. In the task, participants had to use their actual body movements to control the avatars, which better simulates competitive interactions that occur in the real world, compared to other more standard lab tasks. Results indicated that male winners were more aggressive and demonstrated a larger increase in testosterone concentrations related to losers. Moreover, the effect of competition outcome on subsequent aggression was statistically mediated by changes in testosterone concentrations. Specifically, winners tended to be more aggressive than losers were, in part because they demonstrated a larger increase in testosterone relative to losers. Again, there was no significant association between testosterone surges and subsequent PSAP aggression in women. If testosterone surges during competition promote subsequent aggressive behavior, at least in men, could interventions designed to reduce aggression

operate through the dampening of testosterone reactivity to competition? One study (Carré, Iselin, Welker, Hariri,  & Dodge,  2014) examined the ­efficacy of a 10-year aggression intervention program, which started when children were in kindergarten. At 26 years of age, participants were contacted for a follow-­up experiment in which they played three rounds of the PSAP against a fictitious opponent. Participants assigned to the intervention program showed less testosterone reactivity to competition during the first round of the PSAP and less aggressive responding in the second and third rounds of the PSAP. Consistent with the idea that testosterone reactivity may mediate the effect of intervention on reduced aggression, when testosterone surges during the first round were entered into the statistical model, the effect of intervention on the second and third round aggression levels was no longer significant. In other words, the intervention was successful at reducing aggression in rounds 2 and 3 because it reduced testosterone reactivity during the task. Thus, intervention programs aimed at reducing aggression may function by dampening testosterone reactivity to provocation, at least in men for whom testosterone– aggression links appear more consistent. Recall that the PSAP is also used to induce variability in testosterone responses and to examine whether such variability predicts competitiveness (e.g., Carré et al., 2010; Carré & McCormick, 2008) or aggressiveness on subsequent tasks. In one study employing such a design (Geniole, Busseri,  & McCormick,  2013), participants completed the PSAP against a fictitious opponent and were then asked to decide the honorarium of the opponent (participants could give as much or as little of a $5 honorarium to their opponent; giving lower amounts can be considered a form of financial aggression). Testosterone increases during the PSAP predicted lower allocations to the opponents, an effect that became stronger when controlling for concurrent changes in cortisol and estradiol, which may have suppressed the main effects of testosterone (changes in testosterone were associated positively with changes in estradiol, but such estradiol changes were correlated positively, albeit marginally, with the amount of money given, see Table 3 in Geniole et al., 2013). Therefore, testosterone surges during competitions appear to promote subsequent aggressive behavior across different measures (the PSAP: Carré et al., 2009, 2013, 2014; a money allocation paradigm: Geniole et al., 2013), effects that may function to assist the individual in climbing a status hierarchy through the dominance route outlined in the dual Geniole and Carré

287

strategies model of status (Cheng et al.,  2013; Henrich & Gil-White, 2001; for recent review, see Case & Maner, 2016).

Important Moderators of the Links Between Testosterone, Competition, and Aggression

A number of studies suggest that testosterone does not share straightforward links with competitive and aggressive behavior, but instead interacts with sex and situational and individual difference factors. The studies reviewed previously suggest that competitioninduced fluctuations in testosterone are a stronger and more consistent predictor of subsequent competition and aggression in men than in women, with several of the studies testing associations in both sexes but findings effects only in men (e.g., Carré et al., 2009, 2013; Geniole et al., 2013). This discrepancy may be related to differences in the sensitivity of the testosterone determination methods often used (e.g., salivary testosterone assessed using enzymelinked immunosorbent assays), with testosterone concentrations in women being much lower than men’s and thus detected less reliably, introducing additional error in female versus male testosterone samples. Effects in women may also be obscured by additional variability related to menstrual cycle or the use of hormonal contraceptives, both of which may influence aggression (e.g., Geniole et al., 2013) or hormones directly (e.g., Arslan et  al.,  2008; Sowers, Beebe, McConnell, Randolph, & Jannausch, 2001) or interact with testosterone (Dougherty, Bjork, Moeller, & Swann, 1997) to influence aggression. It is also possible that such sex differences in the link between testosterone and aggression and between testosterone and competition exist because testosterone serves different functions in men and in women (although see Hahn, Fisher, Cobey, DeBruine, & Jones, 2016). Whereas competitioninduced surges in testosterone in men may promote subsequent competitiveness, aggression, and antagonistic behavior, such surges in women may promote less direct forms of aggression (e.g., ostracism, rumor spreading, gossiping), not often investigated with respect to testosterone. These behaviors, however, may be more common expressions of dominance and aggression in women (see review in Archer, 2009). Another possibility is that testosterone promotes other, more prosocial behaviors aimed at gaining and maintaining status, such as reconciliation after a conflict or competition in women (Casto & Edwards, 2016b). Additional studies would benefit from examining how testosterone fluctuations in 288

Hierarchy and Testosterone

both men and women (as in Carré et al., 2009, 2013; Geniole et al., 2013) predict both types of behaviors (prosocial vs. aggressive and competitive). One exciting possibility is that testosterone promotes high status in both men and women, but through different routes (e.g., the dominance route in men and the prestige route in women), the examination of which will be important in future studies. Situational factors are also important. Some of the evidence reviewed earlier suggests that competitioninduced fluctuations in testosterone may have different effects depending on the outcome of the competition (e.g., win vs. loss, see Carré & McCormick, 2008; Mehta & Josephs, 2006; for effects in samples of winners only, see Geniole et al., 2013; Study 1 of Norman et al., 2014) and its decisiveness (close vs. decisive outcomes, Mehta, Snyder, et al., 2015). Testosterone may also share differential links with subsequent competitive behavior depending on the types of future competitions that are actually available, promoting competitiveness only when a future victory helps one climb the hierarchy (e.g., when against a new opponent, as discussed in Mehta & Josephs, 2006; Mehta, Snyder, et al., 2015). The inducer of testosterone may also be important; cues or challenges that signal the potential for violence (e.g., a handgun) may promote testosterone surges that function to increase aggression (e.g., Klinesmith, Kasser, & McAndrew, 2006), whereas cues or challenges that signal the potential for nurturance (e.g., a baby crying, as in van Anders, Tolman, & Jainagaraj, 2014) may not. Future studies may benefit from testing this possibility directly. Other moderators of the testosterone–competition and testosterone–aggression link may include more stable, individual difference factors. Some research in rodents and humans, for example, suggests that anxiety may be an important factor for understanding links between testosterone and behavior. Social challenges elicit greater aggression and testosterone responses among male rats bred for low versus high anxiety (Veenema, Torner, Blume, Beiderbeck,  & Neumann, 2007), and in humans, there is some evidence that the winner–loser effect is moderated by social anxiety (such that losing men show decreases, but only to the extent that they are high in social anxiety, Maner et al., 2008). In a more recent study (Study 1, Norman et al.,  2014), testosterone surges during a victory predicted greater aggression in a subsequent task, the PSAP, but only among individuals low in trait anxiety (in fact, testosterone surges predicted marginal decreases in aggressive behavior among those high in trait anxiety). In a follow-up

investigation (Study 2, Norman et al.,  2014), the authors reanalyzed an archival dataset, finding the same pattern of results in men (but not women) who both won or lost a preceding competition. Specifically, those with greater testosterone increases during the competition (irrespective of the outcome) were more aggressive in a subsequent task, but only if they were also low in trait anxiety.6 Another important moderator of the testosterone– aggression link may be one’s level of ­independent versus interdependent self-construal. Those high in independent self-construal view their identity as separate or independent from a larger social group, valuing and striving for this independence and autonomy. Conversely, those higher in interdependent selfconstrual see themselves as part of a larger social group and value and are cognizant of their role within the group (Markus & Kitayama, 1991). Because of this value placed on one’s role within a larger social group, individuals high in interdependent self-construal may avoid using aggression (especially direct/overt forms of aggression) as it may jeopardize such relationships, whereas those higher in independent self-construal may see aggression as an opportunity to establish or solidify their position within a social hierarchy (for review, see Cross & Madson, 1997). Consistent with this idea, interdependent self-construal is negatively associated with meanness, a psychopathic personality trait reflecting one’s ability to callously exploit others (Blagov, Patrick, Oost, Goodman, & Pugh, 2016). Therefore, higher scores on this trait, and lower scores on independent self-construal, may buffer against the effects of testosterone and aggression. Indeed, in a recent set of studies, Welker and colleagues (2017) found that testosterone changes during a rigged boxing game in men predicted subsequent aggressive behavior in the PSAP, but only among men high in ­independent versus interdependent self-construal. In a reanalysis of Carré and colleagues (2013), the authors again found the same effect (irrespective of whether participants won or lost the preceding competition), but only among men. Individual differences in dominance may also play a role. Testosterone surges in male winners predicted subsequent aggression, but only among male winners high in trait dominance (Carré et al., 2009). In another study, however, a psychopathic personality factor involving a combination of low anxiety, fear, and high social dominance (fearless dominance) did 6   The authors suggested that this interaction may have emerged because of differences in the interpretation of, or ability to suppress anger in response to, provocation during the PSAP.

not moderate the association between testosterone fluctuations and subsequent antagonistic behavior (Geniole et al., 2013). One possibility for this discrepancy may be the different tasks used to measure subsequent aggressive/antagonistic behavior. In Carré and colleagues (2009), the postcompetition m ­ easure of aggression (PSAP) also involved competition and a dynamic interaction between the participant and the (fictitious) opponent, during which counterretaliation was possible and aggression was costly to personal earnings. Conversely, in Geniole and colleagues (2013), the postcompetition measure of aggressive/antagonistic behavior was a one-shot decision in which aggressiveness (giving lower amounts) was not costly and the opponent could not retaliate. One possibility is that the moderating role of dominance only applies to aggressive behaviors that are costly to the actor and/or may elicit retaliation. More recently, using pharmacological manipulations, researchers have shown that the effects of exogenous testosterone on competition and aggression are strongest among men (Carré et al.,  2017) and women (Mehta, van Son, et al., 2015) high in trait dominance. Self-control also emerged as a moderator in Carré and colleagues (2017), buffering against the effects of testosterone on aggression. In other words, for those low in self-control, testosterone promoted aggression. For those high in self-control, however, this effect was not significant. Although more work is needed, it appears that testosterone may promote the impulse to aggress, which manifests in behavioral displays of aggression especially among those high in the desire to outcompete and impose their will on others (i.e., those high in trait dominance) and among those who find it difficult to exercise self-control over impulses in general (e.g., those low in self-control).

Conclusion

The research reviewed in this chapter suggests that testosterone surges during competitive interactions influence subsequent status-related behaviors. First, there is evidence that testosterone surges during competition function to promote subsequent attempts at additional status enhancement (through increased willingness to compete). Importantly, however, this relationship depends on several factors: Rather than promoting a willingness to compete across all contexts, testosterone surges appear to drive this willingness among individuals who lost a preceding competition or decisively (vs. narrowly) won and could compete against a new opponent (and further elevate status), in particular. Geniole and Carré

289

In addition to increasing the willingness to c­ompete, testosterone fluctuations also appear to drive aggressive behavior. Men who experienced larger increases in testosterone, across various tasks (e.g., a number-tracing task, the PSAP, and a more ecologically relevant fight simulation), were more aggressive toward their competitors on a subsequent task than were men who experienced lesser increases or decreases in testosterone. Further, in addition to promoting aggression in the PSAP, men who experienced greater increases in testosterone were more likely to allocate a smaller honorarium to participants after a competition than those who experienced lesser increases or decreases. Interventions that reduced aggression appeared to act by suppressing these testosterone surges in response to competition. The most consistent finding in these studies is that whereas many of these effects were significant for men, none of the studies reported significant associations between competition-induced surges in testosterone and subsequent competitive or aggressive behavior in women. Thus, testosterone may serve different functions in men and in women. On the other hand, it is also possible that testosterone promotes behaviors aimed at gaining and maintaining status, but the expression of these behaviors is qualitatively different for women than for men. For example, rather than direct forms of aggression, women may use more indirect and subtle forms to climb status hierarchies and maintain their position within social groups. Little research, however, has investigated this possibility directly, presenting an important opportunity for subsequent studies on the investigation of testosterone surges and how these promote behaviors aimed at gaining and maintaining status. Thus, testosterone may function to influence upward mobility in status hierarchies through competitiveness and aggressiveness. It is important to note, however, that findings from the status and hierarchy literature suggest that these behaviors may not always be beneficial for one’s status (reviewed in Anderson & Killduff, 2009; Cheng et al., 2013; van Vugt et al., 2008). Further, there is emerging evidence that testosterone also promotes prosocial behaviors in some circumstances, such as when reciprocating prosocial behavior (Boksem et al.,  2013; Dreher, Dunne, Pazderska, Frodl, & Nolan, 2016), when interacting with in- versus out-group members (e.g., Diekhof, Wittmer, & Reimers, 2014; Reimers & Diekhof, 2015), or when used as a strategy to avoid potential losses of status or resources (Eisenegger, 290

Hierarchy and Testosterone

Naef, Snozzi, Heinrichs, & Fehr, 2010, 2012; Van Honk, Montoya, Bos, van Vugt, & Terburg, 2012; for other evidence of prosocial effects, see Wibral, Dohmen, Klingmüller, Weber, & Falk,  2012). Accordingly, researchers have proposed a model of  testosterone in which its actions are posited to guide behaviors more generally aimed at status attainment and maintenance, which can be either prosocial or antisocial, depending on the context (Eisenegger et al., 2011). Nevertheless, the circumstances (i.e., situational factors) in which and people (i.e., individual difference factors) for whom testosterone flexibly promotes prosocial versus antisocial behaviors aimed at gaining and maintaining status remain poorly understood and an important area for future research (Hamilton et  al., 2015; Knight & Mehta, 2014).

References

Anderson, C., Hildreth, J. A. D., & Howland, L. (2015). Is the desire for status a fundamental human motive? A review of the empirical literature. Psychological Bulletin, 141, 574–601. http://doi.org/10.1037/a0038781 Anderson, C., & Killduff, G. J. (2009). The pursuit of status in social groups. Psychological Science, 18(5), 295–298. Anderson, C., Kraus, M. W., Galinsky, A. D., & Keltner, D. (2012). The local-ladder effect: Social status and subjective well-being. Psychological Science, 23(7), 764–771. http://doi. org/10.1177/0956797611434537 Anderson, C., & Willer, R. (2014). Do status hierarchies benefit groups? A bounded functionalist account of status. In J. T. Cheng, J. L. Tracy, & C. Anderson (Eds.), The psychology of social status (pp. 47–70). New York, NY: Springer. Archer, J. (2006). Testosterone and human aggression: An evaluation of the challenge hypothesis. Neuroscience & Biobehavioral Reviews, 30(3), 319–345. http://doi.org/10.­1016/j.­neubiorev.2004.12.007 Archer, J. (2009). Does sexual selection explain human sex differences in aggression? Behavioral and Brain Sciences, 32(3– 4), 249–311. http://doi.org/10.1017/S0140525X09990951 Archer, J., Graham-Kevan, N., & Davies, M. (2005). Testosterone and aggression: A reanalysis of Book, Starzyk, and Quinsey’s (2001) study. Aggression and Violent Behavior, 10(2), 241–261. http://doi.org/10.1016/j.avb.2004.01.001 Arruda, A. F. S., Aoki, M. S., Carolina, A., & Moreira, A. (2017). Salivary steroid response and competitive anxiety in elite basketball players: Effect of opponent level. Physiology & Behavior, 177, 291–296. http://doi.org/10.1016/j.phys­ beh.2017.05.017 Arslan, A.  A., Gu, Y., Zeleniuch-Jacquotte, A., Koenig, K.  L., Liu, M., Velikokhatnaya, L., . . . Lokshin, A.  E. (2008). Reproducibility of serum pituitary hormones in women. Cancer Epidemiology Biomarkers & Prevention, 17(8), 1880– 1883. http://doi.org/10.1158/1055-9965.EPI-08-0103 Askew, E.  B., Gampe, R.  T., Stanley, T.  B., Faggart, J.  L., & Wilson, E.  M. (2007). Modulation of androgen receptor activation function 2 by testosterone and dihydrotestosterone. Journal of Biological Chemistry, 282(35), 25801–25816. http://doi.org/10.1074/jbc.M703268200

Baron, R. A., & Richardson, D. R. (1994). Human aggression (2nd ed.). New York, NY: Plenum Press. Blagov, P.  S., Patrick, C.  J., Oost, K.  M., Goodman, J.  A., & Pugh, A. T. (2016). Triarchic psychopathy measure: Validity in relation to normal-range traits, personality pathology, and psychological adjustment. Journal of Personality Disorders, 30(1), 71–81. Boksem, M. A. S., Mehta, P. H., Van den Bergh, B., van Son, V., Trautmann, S.  T., Roelofs, K., . . . Sanfey, A.  G. (2013). Testosterone inhibits trust but promotes reciprocity. Psychological Science, 24(11), 2306–2314. http://doi.org/10. 11­77/0956797613495063 Bos, P. A., Panksepp, J., Bluthé, R. M., & Van Honk, J. (2012). Acute effects of steroid hormones and neuropeptides on human social-emotional behavior: A review of single administration studies. Frontiers in Neuroendocrinology, 33(1), 17–35. http:// doi.org/10.1016/j.yfrne.2011.01.002 Boyce, W. T. (2004). Social stratification, health, and violence in the very young. Annals of the New York Academy of Sciences, 1036, 47–68. http://doi.org/10.1196/annals.1330.003 Carré, J. M., Campbell, J. A., Lozoya, E., Goetz, S. M. M., & Welker, K.  M. (2013). Changes in testosterone mediate the effect of winning on subsequent aggressive behaviour. Psychoneuroendocrinology, 38, 2034–2041. http://doi.org/­ 10.1016/j.psyneuen.2013.03.008 Carré, J. M., Geniole, S. N., Ortiz, T. L., Bird, B. M., Videto, A., & Bonin, P. L. (2017). Exogenous testosterone rapidly increases aggressive behavior in dominant and impulsive men. Biological Psychiatry, 82(4), 249–256. http://doi.org/10.1016/j.bio­ psych.2016.06.009 Carré, J. M., Gilchrist, J. D., Morrissey, M. D., & McCormick, C. M. (2010). Motivational and situational factors and the relationship between testosterone dynamics and human aggression during competition. Biological Psychology, 84(2), 346–353. http://doi.org/10.1016/j.biopsycho.2010.04.001 Carré, J. M., Iselin, A. R., Welker, K. M., Hariri, A. R., & Dodge, K. A. (2014). Testosterone reactivity to provocation mediates the effect of early intervention on aggressive behavior. Psychological Science, 25(5), 1140–1146. http://doi.org/10.­ 1177/0956­797614525642 Carré, J. M., & McCormick, C. M. (2008). Aggressive behavior and change in salivary testosterone concentrations predict willingness to engage in a competitive task. Hormones and Behavior, 54(3), 403–409. http://doi.org/10.1016/j.yhbeh.­ 2008.04.008 Carré, J. M., McCormick, C. M., & Hariri, A. R. (2011). The social neuroendocrinology of human aggression. Psychoneuroendo­ crinology, 36(7), 935–944. http://doi.org/10.1016/j.psyn­ euen.2011.02.001 Carré, J. M., & Olmstead, N. A. (2015). Social neuroendocrinology of human aggression: Examining the role of competitioninduced testosterone dynamics. Neuroscience, 286, 171–186. http://doi.org/10.1016/j.neuroscience.2014.11.029 Carré, J.  M., Putnam, S.  K., & McCormick, C.  M. (2009). Testosterone responses to competition predict future aggressive behaviour at a cost to reward in men. Psychoneuroendocrinology, 34(4), 561–570. http://doi.org/10.1016/j.psyneuen.­2008.­ 10.­018 Case, C.  R., & Maner, J.  K. (2016). Dominance and prestige: Dual strategies for navigating social hierarchies. Advances in Experimental Social Psychology, 54, 129–180. http://doi.org/­ 10.1016/bs.aesp.2016.02.001

Casto, K. V., & Edwards, D. A. (2016a). Before, during, and after: How phases of competition differentially affect testosterone, cortisol, and estradiol levels in women athletes. Adaptive Human Behavior and Physiology, 2(June), 11–25. http://doi. org/10.1007/s40750-015-0028-2 Casto, K.  V., & Edwards, D.  A. (2016b). Testosterone and reconciliation among women: After-competition testosterone predicts prosocial attitudes towards opponents. Adaptive Human Behavior and Physiology, 2, 220–233. http://doi.org/­ 10.1007/s40750-015-0037-1 Cheng, J. T., Tracy, J. L., Foulsham, T., Kingstone, A., & Henrich, J. (2013). Two ways to the top: Evidence that dominance and prestige are distinct yet viable avenues to social rank and influence. Journal of Personality and Social Psychology, 104(1), 103–125. http://doi.org/10.1037/a0030398 Cherek, D.  R. (1981). Psychopharmacology effects of smoking different doses of nicotine on human aggressive behavior. Psychopharmacology, 75, 339–345. Cross, S.  E., & Madson, L. (1997). Models of the self: Selfconstruals and gender. Psychological Bulletin, 122, 5–37. Cunniffe, B., Morgan, K. A., Baker, J. S., Cardinale, M., & Davies, B. (2015). Home versus away competition: Effect on psycho­physiological variables in elite rugby union. International Journal of Sports Physiology and Performance, 10, 687–694. http://doi.org/10.1123/ijspp.2014–0370 Darwin, C. (1871). The descent of man. London, UK: Appleton. De Kwaadsteniet, E. W., & Van Dijk, E. (2010). Social status as a cue for tacit coordination. Journal of Experimental Social Psychology, 46(3), 515–524. http://doi.org/10.1016/j.jesp.­ 2010.01.005 De Ronde, W., & De Jong, F. H. (2011). Aromatase inhibitors in men: Effects and therapeutic options. Reproductive Biology and Endocrinology, 9(1), 93. http://doi.org/10.1186/14777827-9-93 Diamond, J. (1999). Guns, germs and steel: The fates of human societies. New York. New York, NY: W. W. Norton & Company. Diekhof, E.  K., Wittmer, S., & Reimers, L. (2014). Does competition really bring out the worst? Testosterone, social distance and inter-male competition shape parochial altruism in human males. PLoS One, 9(7), e98977. http://doi.org/­ 10.1371/journal.pone.0098977 Dougherty, D., Bjork, J., Moeller, F., & Swann, A. (1997). The influence of menstrual-cycle phase on the relationship between testosterone and aggression. Physiology & Behavior, 62(2), 431–435. http://doi.org/10.1016/S0031-9384(97)88991-3 Dreher, J., Dunne, S., Pazderska, A., Frodl, T., & Nolan, J. J. (2016). Testosterone causes both prosocial and antisocial status-enhancing behaviors in human males. Proceedings of  the National Academy of Sciences of the United States of America, 113, 11633–11638. http://doi.org/10.1073/pnas. 1608085113 Eisenegger, C., Haushofer, J., & Fehr, E. (2011). The role of testosterone in social interaction. Trends in Cognitive Sciences, 15(6), 263–271. http://doi.org/10.1016/j.tics.2011.04.008 Eisenegger, C., Naef, M., Snozzi, R., Heinrichs, M., & Fehr, E. (2010). Prejudice and truth about the effect of testosterone on human bargaining behaviour. Nature, 463(7279), 356–359. http://doi.org/10.1038/nature08711 Eisenegger, C., Naef, M., Snozzi, R., Heinrichs, M., & Fehr, E. (2012). New evidence on testosterone and cooperation: Eisenegger et al. reply. Nature, 485. http://doi.org/10.1038/ nature11137

Geniole and Carré

291

Ellis, L. (1994). Social stratification and socioeconomic inequality, Vol. 2: Reproductive and interpersonal aspects of dominance and status. Westport, CT: Praeger. Flinn, M. V., Ponzi, D., & Muehlenbein, M. P. (2012). Hormonal mechanisms for regulation of aggression in human coalitions. Human Nature (Hawthorne, N.Y.), 23(1), 68–88. http://doi. org/10.1007/s12110-012-9135-y Folstad, I., & Karter, A. J. (2011). Parasites, bright males, and the immunocompetence handicap. American Naturalist, 139(3), 603–622. Geniole, S. N., Bird, B. M., Ruddick, E. L., & Carré, J. M. (2017). Effects of competition outcome on testosterone concentrations in humans: An updated meta-analysis. Hormones and Behavior, 92, 37–50. http://doi.org/10.1016/j.yhbeh.2016.10.002 Geniole, S.  N., Busseri, M.  A., & McCormick, C.  M. (2013). Testosterone dynamics and psychopathic personality traits independently predict antagonistic behavior towards the perceived loser of a competitive interaction. Hormones and Behavior, 64, 790–798. http://doi.org/10.1016/j.yhbeh.­ 2013.­09.005 Geniole, S. N., Cunningham, C. E., Keyes, A. E., Busseri, M. A., & McCormick, C. M. (2015). Costly retaliation is promoted by threats to resources in women and threats to status in men. Aggressive Behavior, 41(6), 515–525. http://doi.org/10.1002/ ab.21589 Geniole, S. N., MacDonell, E. T., & McCormick, C. M. (2017). The Point Subtraction Aggression Paradigm as a laboratory tool for investigating the neuroendocrinology of aggression and competition. Hormones and Behavior, 92, 103–116. http://doi.org/10.1016/j.yhbeh.2016.04.006 Griskevicius, V., & Kenrick, D. T. (2013). Fundamental motives: How evolutionary needs influence consumer behavior. Journal of Consumer Psychology, 23(3), 372–386. http://doi. org/10.1016/j.jcps.2013.03.003 Hahn, A. C., Fisher, C. I., Cobey, K. D., DeBruine, L. M., & Jones, B. C. (2016). A longitudinal analysis of women’s salivary testosterone and intrasexual competitiveness. Psychoneuro­ endocrinology, 64, 117–122. Hamilton, L. D., Carré, J. M., Mehta, P. H., Olmstead, N., & Whitaker, J. D. (2015). Social neuroendocrinology of status: A review and future directions. Adaptive Human Behavior and Physiology. http://doi.org/10.1007/s40750-015-0025-5 Harman, S. M., Metter, E. J., Tobin, J. D., Pearson, J. A. Y., & Blackman, M.  R. (2001). Longitudinal effects of aging on serum total and free testosterone levels in healthy men. Journal of Clinical Endocrinology & Metabolism, 86(2), 724–731. Hayes, F. J., Decruz, S., Seminara, S. B., Boepple, P. A., & Crowley, W.  F., Jr. (2001). Differential regulation of gonadotropin secretion by testosterone in the human male: Absence of a negative feedback effect of testosterone on follicle-stimulating hormone secretion. Journal of Clinical Endocrinology and Metabolism, 86(1), 53–58. http://doi.org/10.1210/jc.86.1.53 Henrich, J., & Gil-White, F. J. (2001). The evolution of prestige: Freely conferred deference as a mechanism for enhancing the benefits of cultural transmission. Evolution and Human Behavior, 22, 165–196. Hirschenhauser, K., & Oliveira, R. F. (2006). Social modulation of androgens in male vertebrates: Meta-analyses of the challenge hypothesis. Animal Behaviour, 71(2), 265–277. http://doi.org/10.1016/j.anbehav.2005.04.014 Johnson, S.  L., Leedom, L.  J., & Muhtadie, L. (2012). The dominance behavioral system and psychopathology: Evidence from self-report, observational, and biological studies.

292

Hierarchy and Testosterone

Psychological Bulletin, 138(4), 692–743. http://doi.org/10.1037/ a0027503 Josephs, R. A., Sellers, J. G., Newman, M. L., & Mehta, P. H. (2006). The mismatch effect: When testosterone and status are at odds. Journal of Personality and Social Psychology, 90(6), 999–1013. http://doi.org/10.1037/0022-3514.90.6.999 Klinesmith, J., Kasser, T., & McAndrew, F.  T. (2006). Guns, testosterone, and aggression: An experimental test of a mediational hypothesis. Psychological Science, 17(7), 568–571. http://doi.org/10.1111/j.1467–9280.2006.01745.x Knight, E. L., & Mehta, P. H. (2014). Hormones and hierarchies. In J. T. Cheng, J. L. Tracy, & C. Anderson (Eds.), The psychology of social status (pp. 269–302). New York, NY: Springer. Lieberman, J. D., Solomon, S., Greenberg, J., & McGregor, H. A. (1999). A hot new way to measure aggression. Aggressive Behavior, 25(June), 331–348. Magee, J. C., & Galinsky, A. D. (2008). Social hierarchy: The selfreinforcing nature of power and status. Academy of Management Annals, 2(1), 351–398. http://doi.org/10.1080/1941652080 22­11628 Maner, J. K., Miller, S. L., Schmidt, N. B., & Eckel, L. A. (2008). Submitting to defeat: Social anxiety, dominance threat, and decrements in testosterone. Psychological Science, 19(8), 764–768. http://doi.org/10.1111/j.1467–9280.2008.02154.x Markus, H.  R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psycho­ logical Review, 98(2), 224–253. Mazur, A. (1985). A biosocial model of status in face-to-face primate groups. Social Forces, 64(2), 377–402. Mazur, A. (2015). A biosocial model of status in face-to-face groups. In V. Zeigler-Hill, L. L. M. Welling, & T. K. Shackelford (Eds.), Evolutionary perspectives on social psychology (pp.  303–315). Switzerland: Springer International. http:// doi.org/10.1007/978-3-319-12697-5 Mazur, A., & Booth, A. (1998). Testosterone and dominance in men. Behavioral and Brain Sciences, 21(3), 353–363. Mehta, P. H., & Josephs, R. A. (2006). Testosterone change after losing predicts the decision to compete again. Hormones and Behavior, 50(5), 684–692. http://doi.org/10.1016/j.yhbeh.­ 2006.07.001 Mehta, P. H., Snyder, N. A., Knight, E. L., & Lassetter, B. (2015). Close versus decisive victory moderates the effect of testosterone change on competitive decisions and task enjoyment. Adaptive Human Behavior and Physiology, 1, 291–311. http://doi. org/10.1007/s40750-014-0014-0 Mehta, P.  H., van Son, V., Welker, K.  M., Prasad, S., Sanfey, A. G., Smidts, A., & Roelofs, K. (2015). Exogenous testosterone in women enhances and inhibits competitive decision-making depending on victory—Defeat experience and trait dominance. Psychoneuroendocrinology, 60, 224–236. Muller, M.  N., & Wrangham, R.  W. (2004). Dominance, aggression and testosterone in wild chimpanzees: A test of the “challenge hypothesis.” Animal Behavior, 67, 113–123. http:// doi.org/10.1016/j.anbehav.2003.03.013 Norman, R. E., Moreau, B. J. P., Welker, K. M., & Carré, J. M. (2014). Trait anxiety moderates the relationship between testosterone responses to competition and aggressive behavior. Adaptive Human Behavior and Physiology, 1, 312–324. http:// doi.org/10.1007/s40750-014-0016-y Oliveira, G. A., & Oliveira, R. F. (2014). Androgen responsiveness to competition in humans: The role of cognitive variables. Neuroscience and Neuroeconomics, 3(August), 19–32. http:// doi.org/10.2147/NAN.S55721

Oliveira, T., Gouveia, M. J., & Oliveira, R. F. (2009). Testosterone responsiveness to winning and losing experiences in female soccer players. Psychoneuroendocrinology, 34(7), 1056–1064. http://doi.org/10.1016/j.psyneuen.2009.02.006 Reimers, L., & Diekhof, E. K. (2015). Testosterone is associated with cooperation during intergroup competition by enhancing parochial altruism. Frontiers in Neuroscience, 9, 183. http://doi. org/10.3389/fnins.2015.00183 Ronay, R., Greenaway, K., Anicich, E.  M., & Galinsky, A.  D. (2012). The path to glory is paved with hierarchy: When hierarchical differentiation increases group effectiveness. Psychological Science, 23(6), 669–677. http://doi.org/10.1177/­ 095­6797611433876 Schultheiss, O. C., Campbell, K. L., & McClelland, D. C. (1999). Implicit power motivation moderates men’s testosterone responses to imagined and real dominance success. Hormones and Behavior, 36(3), 234–241. http://doi.org/10.1006/hbeh.­ 1999.1542 Schultheiss, O.  C., Dargel, A., & Rohde, W. (2003). Implicit motives and gonadal steroid hormones: Effects of menstrual cycle phase, oral contraceptive use, and relationship status. Hormones and Behavior, 43(2), 293–301. http://doi.org/­ 10.1016/S0018-506X(03)00003-5 Schultheiss, O.  C., Wirth, M.  M., Torges, C.  M., Pang, J.  S., Villacorta, M. A., & Welsh, K. M. (2005). Effects of implicit power motivation on men’s and women’s implicit learning and testosterone changes after social victory or defeat. Journal of Personality and Social Psychology, 88(1), 174–188. http://doi. org/10.1037/0022-3514.88.1.174 Sellers, J. G., Mehl, M. R., & Josephs, R. A. (2007). Hormones and personality: Testosterone as a marker of individual differences. Journal of Research in Personality, 41(1), 126–138. http://doi. org/10.1016/j.jrp.2006.02.004 Sowers, M., Beebe, J., McConnell, D., Randolph, J., & Jannausch, M. (2001). Testosterone concentrations in women aged 25–50 years: Associations with lifestyle, body composition, and ovarian status. American Journal of Epidemiology, 153(3), 256–264. http://doi.org/10.1093/aje/153.3.256 Strayer, F. F., & Strayer, J. (1976). An ethological analysis of social agonism and dominance relations among preschool children. Child Development, 47(4), 980–989. http://doi.org/­ 10.­ 2307/­1128434 Terburg, D., Aarts, H., & Van Honk, J. (2012). Testosterone affects gaze aversion from angry faces outside of conscious awareness. Psychological Science, 23(5), 459–463. http://doi. org/10.1177/0956797611433336 Thomsen, L., Frankenhuis, W. E., Ingold-Smith, M., & Carey, S. (2011). Big and mighty: Preverbal infants mentally represent social dominance. Science, 331(6016), 477–480. http://doi.org/­ 10.­1126/science.1199198 Vadakkadath Meethal, S., & Atwood, C. S. (2005). The role of hypothalamic-pituitary-gonadal hormones in the normal structure and functioning of the brain. Cellular and Molecular Life Sciences, 62, 257–270. http://doi.org/10.1007/s00018004-4381-3 van Anders, S.  M., Tolman, R.  M., & Jainagaraj, G. (2014). Examining how infant interactions influence men’s hormones, affect, and aggression using the Michigan infant

nurturance simulation paradigm. Fathering, 12(2), 143–160. http://doi.org/10.3149/fth.1202.143 van Anders, S. M., & Watson, N. V. (2007). Effects of ability- and chance-determined competition outcome on testosterone. Physiology & Behavior, 90(4), 634–642. http://doi.org/10.­ 1016/­j.physbeh.2006.11.017 Van Honk, J., Montoya, E. R., Bos, P. A., van Vugt, M., & Terburg, D. (2012). New evidence on testosterone and cooperation. Nature, 485(7399), E4–E5. http://doi.org/10.1038/nature11136 van Vugt, M., Hogan, R., & Kaiser, R. B. (2008). Leadership, followership, and evolution: Some lessons from the past. American Psychologist, 63(3), 182–196. http://doi.org/10.1037/­ 0003-066X.63.3.182 Veenema, A.  H., Torner, L., Blume, A., Beiderbeck, D.  I., & Neumann, I. (2007). Low inborn anxiety correlates with high intermale aggression: Link to ACTH response and neuronal activation of the hypothalamic paraventricular nucleus. Hormones and Behavior, 51, 11–19. http://doi.org/10.1016/j. yhbeh.2006.07.004 Vongas, J. G., & Al Hajj, R. (2017). The effects of competition and implicit power motive on men’s testosterone, emotion recognition, and aggression. Hormones and Behavior, 92, 57–71. http://doi.org/10.1016/j.yhbeh.2017.04.005 Welker, K.  M., Norman, R.  E., Goetz, S., Moreau, B.  J.  P., Kitayama, S., & Carré, J. M. (2017). Preliminary evidence that testosterone’s association with aggression depends on selfconstrual. Hormones and Behavior, 92, 117–127. http://doi.org/­ 10.1016/j.yhbeh.2016.10.014 Wibral, M., Dohmen, T., Klingmüller, D., Weber, B., & Falk, A. (2012). Testosterone administration reduces lying in men. PLoS One, 7, e46774. Wingfield, J.  C. (2012). The challenge hypothesis: Behavioral ecology to neurogenomics. Journal of Ornithology, 153(153), S85–S96. http://doi.org/10.1007/s10336-012-0857-8 Wingfield, J. C. (2017). The challenge hypothesis: Where it began and relevance to humans. Hormones and Behavior, 92, 9–12. http://doi.org/10.1016/j.yhbeh.2016.11.008 Wingfield, J. C., Hegner, R. E., Dufty, A. M., & Ball, G. F. (1990). The “challenge hypothesis”: Theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. American Naturalist, 136(6), 829–846. Wu, Y., Eisenegger, C., Zilioli, S., Watson, N. V., & Clark, L. (2017). Comparison of clear and narrow outcomes on testosterone levels in social competition. Hormones and Behavior, 92, 51–56. http://doi.org/10.1016/j.yhbeh.­ 2016.­ 05.­016 Zilioli, S., & Bird, B. M. (2017). Functional significance of men’s testosterone reactivity to social stimuli. Frontiers in Neuro­ endocrinology. http://doi.org/10.1016/j.yfrne.2017.­06.­002 Zilioli, S., Mehta, P. H., & Watson, N. V. (2014). Losing the battle but winning the war: Uncertain outcomes reverse the usual effect of winning on testosterone. Biological Psychology, 103, 54–62. http://doi.org/10.1016/j.biopsycho.2014.07.022 Zilioli, S., & Watson, N. V. (2012). The hidden dimensions of the competition effect: Basal cortisol and basal testosterone jointly predict changes in salivary testosterone after social victory in men. Psychoneuroendocrinology, 37(11), 1855–1865. http://doi. org/10.1016/j.psyneuen.2012.03.022

Geniole and Carré

293

CH A PT E R

17

Competition, Dominance, and Social Hierarchy

Kathleen V. Casto and Pranjal H. Mehta

Abstract This chapter reviews literature having to do with the social-behavioral neuroendocrinology of competition, dominance, and status hierarchies in humans. After defining these terms, their importance, and everyday relevance, the chapter discusses the major research findings that suggest a bidirectional influence between these social behaviors and the steroid hormones cortisol and testosterone. Specifically, the association between cortisol and social rank and cortisol’s reactivity to social challenges are discussed. Further, this chapter discusses research that tests the predictions that basal testosterone is related to status-motivated behavior, that testosterone levels are transiently altered during contests for status, and that these changes function to promote subsequent status-seeking behavior. Noting the nuance of these findings, the personality and context factors that appear to moderate testosterone–status relationships are highlighted. Finally, this chapter includes both a new theoretical model for the testosterone–social status relationship that captures this complexity and, in closing, summarizes promising areas of future research. Key words:  competition, dominance, status, social hierarchy, testosterone, cortisol, dual-hormone hypothesis

In ancient Egyptian society, formalized social ­hierarchy was manifested in the ritualistic burial of the dead. Social status was symbolized in almost every aspect of the mortuary arrangement—how close in proximity one’s tomb was to the king, height from the ground, size of the tomb, level of artistic decoration, and material accoutrements (Romano,  1990; Taylor,  2001). Male members of the elite class would begin constructing a tomb at the peak of their career, even setting aside an endowment for upkeep and offerings after death (Baines & Lacovara, 2002). These preemptive structures served as status symbols for the living as the “tomb was a  central vehicle for peer competition (Baines & Lacovara, 2002, p. 10). For Egyptians, death was as important as life; it was imperative that social status attained in life be maintained in death. Social hierarchy remains a pervasive and fundamental framework of modern human society and

social relationships across cultural boundaries (Diefenbach,  2013; Sidanius & Pratto,  1999). Formalized and highly stratified hierarchies exist within political and government structures, such as  traditional monarchies, dictatorships, capitalist systems, and resulting socioeconomic status (SES). There are also less large-scale yet pervasive social hierarchical systems: class or grade in school, workplace title or position, finish place in a competition, veteran or rookie status on a professional sports team, the lead versus backup role in a play or dance assemble, and even the naming system for increasingly luxurious seats on an airplane (first class, business class, and main cabin). These systematic rankings within everyday life are representative of the fundamental human need to formally classify social position based on level of distinction, wealth, talent or ability, and assumed power and privilege within each role. Some evolutionary psychologists even 295

suggest that it was the emergence of increasingly complex social dominance interactions that provided the selection pressure for the more sophisticated human intelligence and capacity for language (Alexander, 1989; Flinn, Geary, & Ward, 2005). Importantly, society-level social hierarchies result in stepwise differences among the upper and increas­ ingly lower echelons (Dasgupta, 2015; Sidanius & Pratto, 1999) in access to resources directly linked to survival, such as food, water, housing, land, and health care (e.g., “food deserts” common in lowincome neighborhoods, lack of access to potable water in certain developing nations; Montgomery & Elimelech, 2007; Walker, Keane, & Burke, 2010). In review of the adverse physiological effects of stress resulting from “dominance hierarchies” in primates, Sapolsky (2005, p. 652) states that despite the complexities of human social structure, “the SES gradient of health among Westernized humans is a robust example of social inequalities predicting patterns of disease.” Indeed, there appears to be a strong positive relationship in primates, including humans, between both objective and subjective social status ranking and psychological stress, health, and, ultimately, the quality of life and survivability of individuals (Demakakos, Nazroo, Breeze, & Marmot, 2008; Sapolsky, 2004). Thus, aside from the basic science motivation to discover patterns of human behavior and related causes, there is also a perhaps more important ethical responsibility to understand the psychological and interacting biological processes involved in social dominance behavior and resulting social hierarchies. Not all dominance-motivated social ordering is easily recognized. There are subtler hierarchies that individuals form naturally among social groups in various contexts (e.g., spontaneously emerging leader–follower relationships and popularity among peers). Everyday acts of dominance and deference (e.g., eye contact, the firmness of a handshake, the giving and taking of verbal directives) are used to attain and maintain social standing (Tiedens & Fragale,  2003). People also go to great lengths to advertise status ranking with luxury goods like clothes, shoes, jewelry, cars, and houses (a term known as conspicuous consumption; O’Cass & McEwen, 2004); we even have objects that serve no other purpose than to advertise status (e.g., trophies, diplomas, and award ribbons). Even more subtle, social media platforms provide individuals far-­ reaching opportunity (70 percent of Americans polled used some form of social media; Pew Research 296

Center, 2017) to attempt to outrank their peers in quality of life, personal achievements, travel, number of friends, and perceived happiness (Chua & Chang,  2016). Though seemingly nonconsequential, these small and sometimes subtle everyday gestures may serve to reinforce larger scale hierarchical social systems and represent the pervasiveness of the fundamental human striving for status. Like other basic psychological drives, such as thirst, hunger, and sexual desire, the motivation for status has biological underpinnings in the form of a cascade of bidirectional brain–body interactions communicated via chemical messengers, specifically steroid hormones. Testosterone was first recognized for its long-term impact on the development of male secondary sexual characteristics and sexual behavior (August, Grumbach, & Kaplan,  1972; Phoenix, Goy, Gerall, & Young, 1959; Phoenix, Slob, & Goy, 1973). Given the more physically dominant nature of male-typical behavior in male primates, including humans, early research on social hierarchies proposed that testosterone would increase with status rank and social dominance (Ehrenkranz, Bliss, & Sheard, 1974; Purifoy & Koopmans, 1979; Rose, Bernstein, & Gordon, 1975; Sapolsky, 1982). Cortisol, well known for its link to physical and psychological stress, was also linked in earlier research to rank within the social hierarchy (Sapolsky,  1982). Subsequent research in the decades that followed has revealed a complex relationship between testosterone, cortisol, and dominance rank and related behaviors (Casto & Edwards,  2016a; Hamilton, Carré, Mehta, Olmstead, & Whitaker,  2015; Mazur & Booth, 1998). This chapter will review literature having to do with the social-behavioral neuroendocrinology of competition, dominance, and status hierarchies, first defining these terms and then discussing the major research findings that have emerged in this field. There is a vast research literature on these concepts pertaining to nonhuman animal behavior (e.g., Gleason, Fuxjager, Oyegbile, & Marler, 2009); however, this chapter focuses primarily on the ­research involving human participants with implications for human behavior. The literature discussed here lies at the crossroads of social-­ personality psychology and behavioral endocrinology (two otherwise quite distinct fields), wherein the complexity of human socially, culturally, and contextually embedded behavior is predicted by elegant, yet primitive, and evolutionarily adaptive hormonal fluctuations.

Competition, Dominance, and Social Hierarchy

Defining Social Hierarchy, Dominance, and Competition

That valuable resources (e.g., food, water, land, sexual partners) are limited stratifies social order: Some (dominants/superiors) have more access to these resources than others (subordinates). Those who command these resources do so because they possess qualities that are attractive to or more dominant than others, qualities that confer greater attention, influence, or power over those that are lower ranking (Anderson, Hildreth, & Howland, 2015; Berger, Cohen, & Zelditch,  1972; Berger, Rosenholtz, & Zelditch, 1980; Kemper, 1990). The compilation of these relationships within social group results in social hierarchy, the rank ordering “of individuals or groups on a valued social dimension” (Magee & Galinsky, 2008, p. 354; as cited by Cheng, Tracy, Foulsham, Kingstone, & Henrich, 2013). A higher rank order indicates that an individual or group has higher social status (Berger et al., 1972, 1980; Ellyson & Dovidio, 1985). An individual’s actual rank is not necessarily fixed— some hierarchies can be rather unstable—and is often context dependent (e.g., at work vs. among peers or family). Attaining a higher position within a social hierarchy can predict increases in subsequent status-seeking behavior in an effort to maintain this high status and attain even higher positions in the future (e.g., Zilioli & Watson, 2014). Because of its notable survival advantages, direct reward value, universality, and long-term impact on psychological well-being, social status is considered a fundamental human motive (Anderson et al., 2015). Indeed, social hierarchies are a universal and evolutionarily conserved characteristic of most human societies (Chiao, 2010; Diefenbach, 2013; Gledhill, Bender, & Larsen, 1988). Additionally, neuroscientists have revealed distinct neural networks in the brain for the perception and maintenance of social hierarchy (Chiao, 2010). Status is often decided through competition, “a social interaction in which access to something valued is contested between individuals and groups” (Casto & Edwards, 2016a, p. 21). The “something valued” could be the resource that is in limited supply or the simple feeling that one possesses rank or power over others. Prevailing over one’s opponent signals dominance, whereas failing to prevail or conceding defeat signals deference (Mazur,  1985). In an everyday sense, the motivation for engaging in competition may be the simple joy of winning— undoubtedly resulting from the inherent reward of

status gained by the demonstration of dominance. However, the precise definition of dominance and related terms appears to depend on the specific literature within which it is discussed—social psychology, personality psychology, sociobiology, or evolutionary psychology (Cheng et al.,  2013). One can possess dominance (i.e., have physical or psychological qualities that others defer to or admire, such as strength or competence), be considered dominant (i.e., rank higher), behave dominantly (e.g., verbal and nonverbal signaling of dominance, acting as a leader, refusal to submit), or dominate an opponent (soundly defeat, outperform, or display considerably greater strength than an opponent; Burgoon, Johnson, & Koch, 1998; Cheng et al., 2013; Cheng, Tracy, & Henrich, 2010; Cheng, Tracy, Ho, & Henrich, 2016; Ellyson & Dovidio, 1985). Additionally, dominance is often used to describe a general style of relating to others that expresses the explicit and implicit motivation for status and ­status-seeking behavior more broadly (Anderson & Kilduff, 2009): A social hierarchy may be a “status hierarchy” or a “dominance hierarchy” (Chase, Tovey, Spangler-Martin, & Manfredonia, 2002; see Sapolsky, 2005; Mazur, Welker, & Peng, 2015). Figure 17.1 visually demonstrates the relationships between the terms competition, dominance, deference, social hierarchy, status ranking, and statusseeking behavior. Although “dominance” is typically used as the catch-all concept for general behavior involved in the attainment and maintenance of high status, recent research has determined that there are at least two distinct, yet equally viable, cognitive and behavioral strategies for gaining influence—dominance and prestige (Cheng et al., 2010, 2013; Cheng & Tracy,  2014). In this view, dominance refers to a more coercive, physical or psychologically aggressive, intimidating, and conflict-based style in which an individual attains power by demanding deference. Eminence (Kemper, 1990) or prestige (Cheng et al., 2013) is a style of achieving high status through the demonstration of competence, likability, and prosociality and is characterized by the voluntary deference, attention, and respect of others. Given this distinction, continued research on the complexities of competition- and status-related behavior, including the underlying biological influence, will expand understanding of these important and consequential social interactions. Due to the more broad interpretation of “dominance” (meaning statusseeking behavior) in the social neuroendocrinology Casto and Mehta

297

e nc

re

efe D

Social Status

D om in an ce

Competition

Sorting into the Social Hierarchy

Status-Seeking Figure 17.1  Visual description of terms related to status and social hierarchy.

literature, the remainder of this chapter will operate under this generalized definition.

Cortisol, Stress, the Social Hierarchy, and Competition

Cortisol is well known for its positive relationship with stress, both psychological and physical. Acute psychological experiences of stress, particularly social-evaluative stress combined with a perceived lack of control over one’s environment and outcomes, produces reliable and transient increases in cortisol (reviewed by Dickerson & Kemeny, 2004). Shortterm elevations in cortisol, secreted as an end-product of the hypothalamic-pituitary-adrenal axis and via rapid and direct sympathetic stimulation of the adrenal gland (Engeland & Arnhold, 2005; Tsigos & Chrousos,  2002), provide immediate advantages for escaping or managing a stressful stimulus (e.g., mobilization and breakdown of glucose). Long-term activation or dysregulation of this system is energetically costly and can result in deleterious effects on physical and psychological health and immune system functioning (Cohen et al., 2012; McEwen, 1998; Whitworth, Williamson, Mangos, & Kelly,  2005)—a process known as “­ allostatic load” (McEwen, 1998, 2000, 2004). Thus, individual differences in basal cortisol are interpreted as a reliable predictor of allostatic load (e.g., Goymann & Wingfield, 2004).

Basal Cortisol and Social Rank

Social hierarchy rank is thought to be inversely ­related to basal cortisol due to the increasing life adversity and resulting allostatic load experienced by increasingly lower ranking individuals (Knight 298

& Mehta,  2014; Sapolsky,  2005). Early studies with nonhuman animals supported this notion; for example, Sapolsky (1982) showed that high-ranking males demonstrated significantly lower baseline cortisol in baboons than their more subordinate counterparts. One of the first studies in humans corroborated this finding: Among a relatively small sample of male Dominican villagers, lower cortisol levels were related to higher peer ratings of likability and influence (Decker,  2000). At the society level, there is a reliable and robust negative relationship between basal cortisol levels and SES (Cohen, Doyle, & Baum,  2006; reviewed by Knight & Mehta, 2014). Recent research in humans has expanded on the cortisol–status relationship. In a relatively large sample of individuals enrolled in an executive ­education program, Sherman et al. (2012) showed that leaders (those responsible for managing others) had significantly lower cortisol and anxiety than nonleaders. In a follow-up study among leaders, leadership level (having and managing more subordinates and with greater authority) produced the same effect. Additionally, sense of control (generated using a measure of personal sense of power) mediated the relationship between leadership level and cortisol, as well as anxiety. However, studies of collegiate athletes have shown no direct relationship between cortisol and peer ratings of leadership ability, a proxy for status (Edwards & Casto, 2013; Edwards, Wetzel, & Wyner,  2006). Using social network analysis, Kornienko, Clemans, Out, and Granger (2014) demonstrated that among a competitive pool of first-year nursing students, high cortisol levels were associated with low gregariousness, the

Competition, Dominance, and Social Hierarchy

number of perceived friends within the network of nursing students, but not popularity, the number of network members who perceived them as friends. In a follow-up study among members of a large mixed-sex collegiate marching band, higher basal cortisol was also related to an inability to maintain friendships within the network (i.e., greater turnover in friendships over a two-month period; Kornienko, Schaefer, Weren, Hill, & Granger, 2016).

Cortisol Reactivity to a Status Challenge

The direction and strength of the cortisol response to stress depends on a multitude of psychological and contextual factors (Dickerson & Kemeny, 2004; Knight & Mehta, 2014; Kudielka, Hellhammer, & Wüst,  2009). Cortisol increase in response to a stressor could be considered adaptive (e.g., benefiting social status) or maladaptive (e.g., a sign of dysregulation or overreactivity) depending on the context (e.g., competition) and the timing and magnitude of the response (Aschbacher et al., 2013; Shirtcliff, Peres, Dismukes, Lee, & Phan, 2014). Social rank appears to be one important factor for predicting patterns of cortisol reactivity to stress (e.g., Hellhammer, Buchtal, Gutberlet, & Kirschbaum,  1997; Sapolsky,  1982), with higher social status producing what researchers interpret as more adaptive responses, depending on the task and context (Akinola & Mendes, 2014; Shirtcliff et al., 2014). In one study, women whose cortisol levels did not habituate after repeated exposure to a stressor (i.e., a maladaptive response) subjectively rated themselves lower in SES (Adler, Epel, Castellazzo, & Ickovics, 2000). Under the condition of socialevaluative threat, men and women with high subjective social status (perceived rank among dormitory peers) showed significantly larger cortisol increases in a single session of the Trier Social Stress Test (TSST) than men and women who rated themselves low in status (Gruenewald, Kemeny, & Aziz, 2006). Cortisol increase, in this sense, could reflect a greater mobilization of energy and activity required to defend one’s status when status is indeed at stake, with a nonincrease in cortisol reflecting a more maladaptive (i.e., blunted) physiological response threat. However, the adaptive function of elevated cortisol would likely depend on the social context. For example, increased cortisol in response to social stressors, in some studies, appears to predict subsequent risky decision making (van den Bos, Harteveld, & Stoop, 2009; reviewed by Starcke & Brand, 2012), a behavior that could be beneficial in

contexts where risk taking is advantageous (i.e., choosing to fight rather than flee when status is relatively high), but could be deleterious in other contexts (e.g., choosing to fight rather than flee when status is relatively low). However, there is some evidence in men and women dyads that transiently increased cortisol levels predict subsequent prosocial behaviors and cognitive states such as affiliation (in women dog handlers, Sherman, Rice, Jin, Jones, & Josephs, 2017) and feelings of interpersonal closeness (in men participating in the TSST, Berger, Heinrichs, von Dawans, Way, & Chen, 2016) that effectively buffer the negative psychological and physiological effects of stress (Cohen & McKay, 1984). Situational factors also appear to moderate the relationship between status and cortisol reactivity. In at least two studies, individuals who were high on indices of dominance motivation (basal testosterone; implicit power motivation) showed greater cortisol increases across a competition for status, but only when that competition resulted in defeat (Mehta, Jones, & Josephs, 2008; Wirth, Welsh, & Schultheiss, 2006). Directly testing the moderating effects of social context on the status–cortisol reactivity relationship, Knight and Mehta (2017) manip­ ulated both the status position and hierarchy stability of men and women competing in a mock job interview. Participants assigned to a high-status role (“manager”) showed reduced cortisol reactivity and better performance compared to participants assigned to the low-status role (“builder”), but only if the assigned roles were fixed (i.e., a stable hierarchy). When told that the assigned roles could change based on performance in the interview (i.e., an unstable hierarchy), high status increased cortisol reactivity and did not result in better performance compared to low status. In real-world athletic competition, where status is formally contested, cortisol significantly increases over the match period in men and women (e.g., Casto, Elliott, & Edwards,  2014; Edwards et al., 2006; Edwards & Kurlander,  2010; reviewed by Casto & Edwards,  2016a), an effect that is likely influenced, to some degree, by the physical stress of exercise (Copeland, Consitt, & Tremblay, 2002; M. S. Tremblay, Copeland, & van Helder, 2005; Viru et al., 2010). Contrary to the aforementioned study where high rank predicted a higher cortisol increase across the TSST, competition losers appear to show relatively higher increases in cortisol compared to winners (e.g., Bateup, Booth, Shirtcliff, & Granger, 2002; Jiménez, Aguilar, & Alvero-Cruz, 2012). Casto and Mehta

299

Testosterone, Status, and Status-Seeking Behavior

Several decades of research have revealed a generally positive and bidirectional relationship between testosterone, social status, and status-related behavior. Initial observationally and empirically based models of this relationship, the biosocial model of status (Mazur,  1985) and the challenge hypothesis (Wingfield, Hegner, Dufty, & Ball,  1990), have provided a theoretical basis for this research. Detailed historical accounts and an extended description of these models are published elsewhere (biosocial model of status—Mazur & Booth,  1998; Mazur, 2017; challenge hypothesis—Archer,  2006; Wingfield, 2017). The challenge hypothesis asserts that baseline levels of testosterone in monogamously mating birds regulate reproductive development during breeding season, increase with periods of territorial aggression under conditions of social instability (in response to male–male contests for status and sexual partners), and decrease with the expression of parental care (Wingfield, 2017). This hypothesis has been extended to humans to explain, more generally, testosterone increases in response to “competitive situations between young men” (Archer, 2006, p. 322). The biosocial model of status, initially informed by empirical studies with nonhuman primates and men engaged in athletic competition, proposes that stable baseline levels of testosterone predict statusrelated behavior. Furthermore, this model proposes that testosterone should increase in response to status gained and decrease in response to status lost. Specifically, high-testosterone individuals are expected to behave more dominantly and, as a result of status gained from this dominance, demonstrate rising levels of testosterone to promote future competitive behavior. Likewise, low-testosterone individuals are expected to behave more submissively and, as a result of status lost from this deference, demonstrate falling levels of testosterone. That is, transient shifts in testosterone represent the physiological mechanism underlying each stage of the competition, status-sorting, and status-seeking processes depicted in Figure  17.1. Although originally thought to be an acute mechanism regulating even subtle dominance and deference signals in everyday interactions with others (Mazur,  1985), evidence of substantial testosterone fluctuations resulting from status contests may be specific to more formal competitive contexts (Mazur et al., 2015). The decades that followed the dissemination of the biosocial model for status and the challenge hypothesis have produced an abundance of empirical 300

research testing specific predictions from these models in the context of human competition. There are several recent comprehensive reviews and metaanalyses that summarize the findings from this literature (Carré & Olmstead, 2015; Casto & Edwards, 2016a; Geniole, Bird, Ruddick, & Carré,  2017; Hamilton et al., 2015; G. A. Oliveira & Oliveira, 2014). Table 17.1 provides a list of specific hypotheses that can be derived from the biosocial model of status and challenge hypothesis regarding the relationships between testosterone and status ranking, status-seeking motivation, and status-seeking behavior. Some of these hypotheses have been tested empirically more than others (e.g., #1 and #2 more than #5). Among the specific hypotheses that have been well tested, nearly all of them have been supported by some studies but also not supported by  others (#1: Burnham,  2007; Cashdan,  2003; Dabbs & Morris, 1990; Josephs, Sellers, Newman, & Mehta,  2006; Mehta, DesJardins, van Vugt, & Josephs,  2017; R.  E.  Tremblay et al.,  1998; van Bokhoven et al., 2006; Wirth & Schultheiss, 2007; #2—Carré, Putnam, & McCormick,  2009; Grant & France,  2001; Schultheiss,  2007; Sellers, Mehl, & Josephs,  2007; Welker & Carré,  2015; #3—Cashdan, 1995; Purifoy & Koopmans, 1979; Zyphur, Narayanan, Koh, & Koh, 2009; Apicella, Dreber, & Mollerstrom, 2014; Bateup et al., 2002; #4—Carré, Campbell, Lozoya, Goetz, & Welker, Table 17.1.  Hypotheses Derived from the Biosocial Model for Status and the Challenge Hypothesis 1. High-testosterone (T) individuals behave more dominantly than low-T individuals; low-T individuals behave more submissively. 2. High-T individuals are high in dominance motivation (dominant personality, motivated for status); low-T ­individuals are low in dominance motivation. 3. Individuals with high status have higher T than individuals with relatively lower status. 4. Winning a competition increases T; losing a ­competition decreases T. 5. Behaving dominantly increases T; behaving submissively decreases T. 6. Engaging in a competition (a threat/challenge to status) increases T. 7. Individuals high in dominant personality/motivation for status show increases in T across a competition; ­individuals low in dominant personality/motivation for status show decreases in T across a competition. 8. Increased T promotes an increase in subsequent ­status-seeking behavior or cognitive states that would benefit social status seeking.

Competition, Dominance, and Social Hierarchy

2013; Costa & Salvador, 2012; Jiménez et al., 2012; Norman, Moreau, Welker, & Carré,  2015; #5— Peters, Hammond, Reis, & Jamieson,  2016; #6—Apicella et al.,  2014; Carré & McCormick, 2008; Casto & Edwards, 2016b; Casto et al., 2014; Edwards et al., 2006; Guezennec, Lafarge, Bricout, Merino, & Serrurier,  1995; Steiner, Barchard, Meana, Hadi, & Gray, 2010; van der Meij, Buunk, Almela, & Salvador,  2010; #7—Schultheiss, Campbell, & McClelland,  1999; Schultheiss & Rohde,  2002; Schultheiss et al.,  2005; #8—Bos, Hermans, Ramsey, & Van Honk,  2012; Carré, Baird-Rowe, & Hariri, 2014; Carré & McCormick, 2008; Hermans, Ramsey, & van Honk,  2008; Mehta & Josephs,  2006; Mehta, van Son, et al., 2015; Welling, Moreau, Bird, Hansen, & Carré, 2016). To clarify general concepts that have emerged from the research testing these hypotheses, we summarize the three main empirically supported predictions from this literature next.

Prediction #1: Basal Testosterone Is Related to Status-Motivated and Dominance Behavior

One of the original and more simplistic predictions of the biosocial model of status is that individuals in high-status positions in a social hierarchy would have relatively higher baseline testosterone levels. Higher testosterone would not only motivate behaviors to achieve such a high-status position but also increase further as a result of its attainment. Although some earlier research supported this prediction in humans and other species (Dabbs & Morris, 1990; Purifoy & Koopmans, 1979; Sapolsky, 1982), subsequent research has shown that hierarchical social ranks such as socioeconomic status and peer-rated ranking are not directly related to testosterone levels in men and women (Cashdan, 1995; Edwards et al., 2006; Mehta & Josephs, 2010; earlier work reviewed by Mazur & Booth, 1998; Newman, Sellers, & Josephs, 2005). However, those who want status may not always have it, and the complexities (e.g., context ­dependence) of a social hierarchy may make it difficult to directly link social status and absolute testosterone levels. Indeed, a recent study of male employees working for corporate businesses revealed that testosterone levels among these men were positively related to self-reported authoritarian leadership style, but only for those who were not in management (leadership) positions (van der Meij, Schaveling, & van Vugt,  2016). Consistent with meta-analytic data on the relationship between actual leadership position and testosterone (van der

Meij et al., 2016), managers did not have higher testosterone levels on average than their subordinate workers. Rather than directly predicting status position, basal testosterone appears to predict how individuals respond to shifts in status. That is, high-testosterone individuals respond negatively (e.g., poorer performance on a spatial or verbal test) to a drop in status, whereas low-testosterone individuals respond neutrally or negatively to a rise in status (Josephs, Newman, Brown, & Beer, 2003; Josephs et al., 2006; Mehta et al., 2008; Newman et al., 2005). Due to the complexities of achieving high-status or leadership positions, basal testosterone may better predict personality characteristics and behaviors motivated toward achieving status (whether or not those efforts are successful). Indeed, basal testosterone is considered an important “personality variable” that predicts dominance behaviors in various contexts (Newman & Josephs, 2009; Sellers et al., 2007; Mehta et al.,  2008). In competition, testosterone levels have been positively related to perceptions of one’s personal success (Casto, Rivell, & Edwards, 2017), competitive decision making and subsequent task confidence (Eisenegger, Kumsta, Naef, Gromoll, & Heinrichs, 2017, but see Apicella et al., 2011), and competitive task persistence (Welker & Carré, 2015). However, testosterone has been found to be negatively related to the ability to accurately judge the thoughts and feelings of others (empathic accuracy) in laboratory and real-world settings, an aspect of cognition that may have negative consequences on other’s perceptions of one’s leadership ability (Ronay & Carney, 2013). A recent study employing an economic decision-making task in which dyadic status relationships are determined during play (dominant–submissive, dominant–dominant, and submissive–submissive; the hawk-dove game) showed that baseline testosterone was positively correlated to taking a dominant position (Mehta et al., 2017). Although there are mixed reports on the direct relationship between testosterone and selfreported dominance (e.g., Cobey, Nicholls, Leongómez, & Roberts,  2015; Grant & France, 2001; Neave, Laing, Fink, & Manning, 2003), trait dominance has emerged as an important moderator of testosterone’s relationship to dominant, competitive, or aggressive behavior (Mehta, van Son, et al., 2015; for review, Carré & Archer, 2017). For example, in one study of men competing for the affection of a woman, high-testosterone individuals displayed more dominant behaviors, but the relationship was specific to those who self-identified as dominant Casto and Mehta

301

(Slatcher, Mehta, & Josephs,  2011). Furthermore, the relationship between testosterone and social status or dominance behaviors appears to also be moderated by basal levels of cortisol (see the section on the dual-hormone hypothesis).

Prediction #2: Testosterone Levels Are Transiently Altered During Contests for Status, More Often in the Positive Direction for Winners Than Losers

A second main prediction of the biosocial model of status is that gaining status through winning a dominance contest should increase testosterone levels from baseline, whereas losing should decrease testosterone levels (i.e., the winner–loser effect, also referred to as the winner effect; reviewed by Casto & Edwards,  2016a). Support for this prediction has been found in studies of laboratory competition, where testosterone levels increased across competition for those who won but decreased for those who lost (e.g., Apicella et al., 2014; Carré et al., 2013; Costa & Salvador,  2012; Norman et al.,  2015). However, other studies have found that testosterone increases across competition regardless of the outcome (Carré & McCormick,  2008; Henry, ­ Sattizahn, Norman, Beilock, & Maestripieri, 2017; Steiner et al.,  2010; van der Meij et al.,  2010). Summarizing the extant literature on the winner– loser effect, Carré and Olmstead (2015) concluded that a number of studies have reported that male winners had elevated testosterone levels relative to losers, but that a nearly equal number of studies have failed to find such an effect. For women, less studied in general, the proportion of studies that show null findings is even greater (Carré & Olmstead, 2015). However, a recent meta-analysis of 60 effect sizes on the winner–loser effect (Geniole et al.,  2017) determined that winners do in fact show an increase in testosterone compared to losers, who on average experience no change in testosterone. Although the effect was not moderated by sex, the effect was only significant in men (men Cohen’s d = 0.23; women d = 0.14). However, the winner–loser effect in this metaanalysis appears to depend on important contextual factors; average effect sizes were moderately large only when the studies were conducted in nonlaboratory testing locations (e.g., athletic competitions), when the outcome (win or loss) was determined naturally (not contrived or manipulated), when the competition duration was greater than 15 minutes, and when the precompetition saliva sample was taken more than 10 minutes before competition 302

(Geniole et al.,  2017). Even though each of these moderators ­independently accounted for effect size differences in the winner–loser effect, they are not mutually exclusive factors; studies of the hormonal response to naturally occurring athletic competition are always outside the laboratory, where the competition outcome cannot be manipulated, and usually last longer than 15 minutes. Thus, the winner–loser effect is less well supported in experimental designs outside of formal athletic competition. As a matter of convenience and limited access to athletes immediately prior to and following a match, participants in studies of athletic competition are often asked to  give their precompetition sample more than 10 minutes before the match begins and more than 10 minutes after the competition has ended (e.g., Jiménez et al., 2012). These studies have been more likely to show dramatic win–loss differences in testosterone change. For studies that acquired samples immediately prior to and following competition (mostly with women athletes), testosterone increases significantly across the competition period regardless of outcome (e.g., Casto et al.,  2014,  2017; Casto & Edwards,  2016b; Edwards et al.,  2006; Edwards & Kurlander, 2010). The issue of lab versus nonlab competition in determining the emergence of the winner–loser effect is currently unresolved. What the laboratory context gains in experimental control, it perhaps loses in being able to sufficiently activate competitive motivation and in being able to create a more realistic social setting where actual status is at stake. What field studies gain in ecological validity, they lose in experimental control. One of the most important potential confounds in the hormonal response to athletic competition is physical exertion, a factor that can elevate testosterone levels independent of competition and the psychological experience of gaining or losing social status (Copeland et al., 2002; M. S. Tremblay et al., 2005; Viru et al., 2010). However, efforts to quantify physical exertion in studies of the testosterone response to athletic competition (e.g., blood lactate, number of minutes played, self-reported physical exertion, observer-rated physical exertion) have not found any significant correlations between these metrics of exertion and testosterone (Aguilar, Jiménez, & Alvero-Cruz, 2013; Casto & Edwards,  2016b; T.  Oliveira, Gouveia, & Oliveira,  2009; Suay et al.,  1999). Furthermore, one of us (KVC) has recently collected saliva samples from trained men and women rifle shooters competing in intra- and intersquad rifle competition, a sport that requires the athletes

Competition, Dominance, and Social Hierarchy

to remain as still as possible. As with other sporting competitions that require more movement, rifle competition results in significant elevations in testosterone on average and for the majority of the athletes sampled over the course of competition (+6 to 41 percent change, Casto & Edwards, unpublished).

Prediction #3: Transient Increases in Testosterone (Associated With a Status Challenge) Promote Future or Subsequent Status-Seeking Behavior

More recently, researchers have begun to explore the functional significance of transient increases in testosterone, that is, the adaptive consequence on subsequent cognition and behavior (for initial review, see Carré, McCormick, & Hariri, 2011). If, in fact, testosterone increases in some individuals and under certain contexts, does this change serve a beneficial purpose for attaining or maintaining status in the future? The desire to compete again may reflect elevated dominance motivation when the individual has the potential to improve status rank (i.e., when the potential benefits outweigh the risks). In men, increases in testosterone during competition predict the subsequent decision to compete again against the same opponent in losers (Mehta & Josephs, 2006) and aggressive individuals (Carré & McCormick, 2008), or a different opponent in decisive winners (Mehta, Snyder, Knight, & Lasseter, 2015). Testosterone increase during competition may also predict subsequent aggressive behavior in men (Carré et al.,  2009), an effect that appears to be moderated by trait anxiety (Norman et al., 2015). Also in men, testosterone increases associated with monetary wins and losses relate to future financially risky choices (Apicella et al., 2014). Increased testosterone has also been related to a more positive subsequent performance in athletic contexts (Cook & Crewther, 2012a, 2012b). In one study of women soccer players, the higher an athlete’s testosterone remained within the 30 minutes after competition, the greater her willingness to reconcile with her opponent—a prosocial strategy for status maintenance (Casto & Edwards,  2016c). Extending beyond correlational evidence, studies of exogenously administered testosterone have shown that, for periods of time up to four hours after administration, participants administered testosterone demonstrate altered cognitions and behaviors that may promote status seeking or aid in status maintenance in competitive contexts (e.g., Bos, Terberg, van Honk, 2010; Eisenegger, Naef, Snozzi, Heinrichs, & Fehr, 2010; Hermans, Putman, Baas, Koppeschaar, &

van Honk,  2006; Mehta, van Son, et al.,  2015; Radke et al., 2015; van Honk, Montoya, Bos, Van Vugt, & Terburg,  2012; reviewed by Eisenegger, Haushofer, & Fehr, 2011). For example, a series of studies by van Honk and colleagues suggests that testosterone administration, after several hours, increases threat vigilance and reduces fear-potentiated startle, responses that are matched, in some cases, with activation of brain areas associated with emotional reactivity (Hermans et al., 2006, 2008; van Honk et al., 1999; for review, see Carré & Olmstead, 2015). Testosterone administration also may reduce cognitive reflection in men (Nave, Nadler, Zava, & Camerer,  2017) and behaviors thought to reflect empathy in women (Hermans et al.,  2006; van Honk & Schutter,  2007; van Honk et al.,  2011; Wright et al.,  2012). Additionally, other research has shown that after receiving a dose of exogenous testosterone, men appear to perceive themselves as more physically dominant (Welling et al., 2016).

Testosterone and Cortisol Interact to Predict Social Status: The Dual-Hormone Hypothesis

Early predictions that testosterone should directly and positively predict social status have failed to garner unanimous empirical support (e.g., Neave et al.,  2003; review by Mazur & Booth,  1998; Mehta & Josephs,  2010). Evidence that cortisol inhibits, suppresses, or otherwise antagonizes testosterone secretion and action at target tissues (Burnstein, Maiorino, Dai, & Cameron,  1995; Chen, Wang, Yu, Liu, & Pearce,  1997; Johnson, Kamilaris, Chrousos, & Gold,  1992) combined with initial behavioral evidence that these two hormones might interact to predict aggression (Dabbs, Jurkovic, & Frady,  1991; Popma et al.,  2007) suggests that a more integrative approach to the hormone–social status relationship was necessary. Following these early indications, Mehta and Josephs (2010) proposed that “only at low levels of cortisol should higher testosterone encourage higher status” (p. 898)—a statement that serves as the basis of the dual-hormone hypothesis. Testing this hypothesis, Mehta and Josephs (2010) indeed showed that only if an individual was relatively low in cortisol did testosterone positively relate to dominance behaviors when instructed to act as a leader (Study 1, in men and women) and when deciding to compete again ­following a defeat in a rigged puzzle competition (Study 2, in men). Several novel replications of the dual-hormone effect have been published (for an earlier review, see Casto and Mehta

303

Mehta & Prasad, 2015). In varsity women athletes, Edwards and Casto (2013) showed that testosterone positively related to actual status as ranked by teammates, but only if the athlete had relatively low levels of cortisol. As with the original studies conducted by Mehta and Josephs, the relationship trended toward an inverse relationship between testosterone and status for those with high levels of cortisol. In a sample of male business executives, testosterone positively predicted the number of subordinates over which an executive had authority (but not income or education level) only if the individual had relatively low cortisol (Sherman, Lerner, Josephs, Renshon, & Gross,  2016). Using social network analysis, Ponzi, Zilioli, Mehta, Maslov, and Watson (2016) showed that among professional male rugby players, participants with low cortisol (but not those with high cortisol) demonstrated higher network centrality (a proxy for social status) in measures of “betweenness” and popularity. Recently published research has even extended the dual-hormone hypothesis to collective hormone profiles in groups. Akinola, Page-Gould, Mehta, and Lu (2016) measured basal testosterone and cortisol in a large sample of MBA students organized into (diverse and mixed-sex) groups of three to six members and asked to compete against other groups in a business-related decision-making task. Collectively high testosterone was significantly and positively related to group performance, but only if that group had relatively low collective cortisol. Although null findings are less likely to be published, there have been other studies that have also failed to support the dual-hormone hypothesis (e.g., Geniole et al., 2013; Mehta et al., 2017). A high-testosterone/low-cortisol profile has also been found to relate to antisocial attitudes (Sollberger, Bernauer, & Ehlert,  2016) and externalizing psychopathology in adolescents (Tackett, Herzhoff, Harden, Page-Gould, & Josephs, 2014), factors that would seem to be detrimental to social status in modern society. However, other studies have shown that a high-testosterone/high-cortisol profile (the inverse of the original dual-hormone effect) predicts deviant or psychopathic traits (Welker, Lozoya, Campbell, Neumann, & Carré, 2014), reactive aggression, and self-reported feelings of dominance (Denson, Mehta, & Ho Tan, 2013). However, at least one study has shown that the interaction between testosterone and cortisol has no relationship to antisocial, socially deviant behavior in a large sample (4,462) of male U.S. Army veterans (Mazur & Booth, 2014). 304

How a high-testosterone/low-cortisol individual (or group) behaves in everyday interactions with others to result in higher status is not yet fully ­understood. This hormone profile may manifest in a style of interacting that balances a desire for status with a desire to affiliate and promote social bonding among others in the group or network. Or these individuals could have a personality that balances status and power motivation with relaxed confidence and low anxiety—a personality profile that is likable, and therefore more likely to garner the support of  others required to attain and maintain status. Perhaps high-testosterone/high-cortisol individuals interact with others with high anxiety, low confidence, aggression, or desperation, effectively thwarting attempts to actually achieve status. Future ­research should explore, more in depth, personality correlates and behavioral interaction styles of hightestosterone/low-cortisol and high-testosterone/ high-cortisol individuals. It is also possible that the mechanism explaining the dual-hormone effect is an interaction between the negative effects of chronically elevated cortisol and physiological processes involved in testosterone’s ability to drive statusseeking behavior. That is, perhaps allostatic load– related stress and resulting high basal cortisol dampens or inhibits mechanisms for status motivation and related behaviors (i.e., a high-testosterone– status relationship). Future research should expand the theoretical basis and practical applications of the dual-hormone hypothesis. Additionally, although this hypothesis originally concerned the interaction between basal testosterone and basal cortisol, how testosterone and cortisol changes interact to predict status and performance-related behavior (e.g., Mehta, Mor, Yap, Prasad, 2015) is a topic of interest for future research.

Moderators of the Relationships Between Competition and Testosterone

Recent research has exposed increasingly complex qualifiers, moderators, and extenuating circumstances that impact both the effect of competition on ­testosterone change and the effect of testosterone change on subsequent behavior. These moderating variables fall under two broad categories: (1) person and (2) context factors (reviewed by Carré et al., 2010; Casto & Edwards,  2016a; Hamilton et al., 2015; G. A. Oliveira & Oliveira, 2014, Salvador & Costa, 2009). Table 17.2 provides a categorical list of the person and context factors that have been studied and appear to be important for influencing either the relationship between basal testosterone

Competition, Dominance, and Social Hierarchy

Table 17.2.  Potential Moderators of the Relationship Between Testosterone and Status or Dominance Person

Context

Physiological Basal cortisol1 2D:4D digit ratio2 Sex3 Personality traits Implicit power motivation4 Dominance5 Aggressiveness6 Competitiveness7 Social anxiety8 Self-construal9 Psychological/cognitive states Mood10 Task-related self-efficacy11 Cognitive appraisal of performance or opponent12 Task enjoyment13

Performance related Win/loss14 Status relations History of wins and losses against same opponent15 Closeness or decisiveness of win/loss16 Situational aspects of competition design Intra- or intergroup nature of a competition17 Group versus individual-based competition18 Outcome is determined by chance or ability19 Geographical/territorial Home versus away location of competition20

Edwards and Casto (2013); Mehta and Josephs (2010); Ponzi et al. (2016); Sherman et al. (2016); Wu et al. (2017). van Honk et al. (2012). Carré et al. (2013). 4 Schultheiss and Rohde (2002); Schultheiss et al. (2005). 5 Mehta, van Son, et al. (2015). 6 Carré and McCormick (2008). 7 Costa and Salvador (2012). 8 Maner et al. (2008); Norman et al. (2015). 9 Welker et al. (2017). 10 Mazur and Lamb (1980); Mehta and Josephs (2006); Mazur et al. (1997); Zilioli and Watson (2013). 11 Costa, Serrano, and Salvador (2016); Salvador and Costa (2009). 12 Casto et al. (2017); Gonzalez-Bono et al. (1999); Oliveira et al. (2014). 13 Mehta, Snyder, et al. (2015). 14 Apicella et al. (2014); Carré et al. (2013); Costa and Salvador (2012); Jiménez et al. (2012); Norman et al. (2015); Zilioli and Watson (2014). 15 Zilioli and Watson (2014). 16 Mehta, Snyder, et al. (2015); Zilioli, Mehta, and Watson (2014). 17 Oxford, Ponzi, and Geary (2010); Wagner, Flinn, and England (2002). 18 Mehta, Wuehrmann, and Josephs (2009). 19 van Anders and Watson (2007). 20 Carré (2009); Carré et al. (2006); Neave and Wolfson (2003). 1 2 3

and dominance behavior, the testosterone response to competition, or the effect of testosterone change on subsequent status-seeking behavior. Early studies of the testosterone response to competition included mood as an additional variable based on the notion that testosterone should increase when dominance is achieved through winning, but only if the individual experienced high positive emotions regarding the win (Gladue, Boechler, & McCaul, 1989; Mazur & Lamb, 1980; McCaul, Gladue, & Joppa, 1992; Mazur, Susman, & Edelbrock, 1997). In the first study published on the testosterone response to status enhancement and competition in humans, Mazur and Lamb (1980) stated that “when a man achieves a rise in status through his own efforts, and he has an elation of mood over the achievement, then he is likely to

have a rise in testosterone” (p. 236). Although subsequent research has shown the importance of mood as an intervening factor (Mehta & Josephs, 2006; Zilioli & Watson,  2013), state-level mood may serve only as a proxy for a more important personality factor: dominance motivation. Arguably, only those who have a strong motivation for dominance would experience elation upon achieving it  through competition. Indeed, Schultheiss et al. (1999) wrote, “It would seem reasonable to assume that personality factors may moderate individuals’ testosterone responses to succeeding or failing at a dominance contest” (p. 234). Implicit power motivation is the degree with which an individual derives reward from “having physical, and mental or emotional impact” on others (Stanton & Schultheiss, 2009, p. 942). Those Casto and Mehta

305

higher in implicit power motivation tend to be more likely to show an increase in testosterone (and cortisol) in response to competition, particularly under the context of a win (Schultheiss & Rohde, 2002; Schultheiss et al., 1999, 2005; Wirth et al., 2006). However, this relationship appears to depend on sex—with stronger, more consistent relationships found for men than for women (Stanton & Schultheiss,  2007). Although implicit power ­motivation has received the most empirical attention, self-reported trait dominance (e.g., Carré et al., 2009; Mehta, van Son, et al., 2015) and competitiveness (Casto,  2016; Costa & Salvador,  2012) may also play an equally important role in regulating testosterone–status relationships and testosterone response to competition. In Mehta, van Son, et al. (2015), testosterone administration in women ­resulted in increased competitive decision making (i.e., a greater percentage of the trials in which the competitor chose to compete again afterward), but only if the participant scored high on a personality measure of dominance motivation and also won the competition. For participants who lost the competition, testosterone administration decreased competitive decision making regardless of individual differences in dominance motivation. Social context also moderates the relationship between testosterone and competitive performance. In one of the first demonstrations of this, Newman et al. (2005) showed that high-testosterone individuals placed in a high-status position performed well on a spatial and verbal task, but high-testosterone individuals placed in a low-status position performed poorly in comparison. This effect, dubbed the mismatch effect (Josephs et al.,  2006), explains how situational constraints that contrast with selfperceptions hamper status-seeking efforts or competitive performance when status is threatened. The mismatch effect is also relevant for group-level performance. Among college students assigned to work in a group on class assignments for an entire semester, the greater the mismatch between testosterone and status rank within the group (i.e., the more negative the relationship), the lower the group’s collective self-efficacy (i.e., the lower their shared confidence in ability to succeed), a measure that is indicative of group functioning and performance (Zyphur et al., 2009). Another social context factor that could affect that relationship between testosterone and status is hierarchy stability, the degree with which one’s status could readily be changed. Zilioli, Mehta, and Watson (2014) proposed that under conditions of 306

status instability, there should be a reversal of the winner–loser effect, whereby winners should decrease in testosterone and losers should increase. This amendment to the biosocial model of status, termed the status instability hypothesis, is based on the notion that if the function of a testosterone increase is to promote future status-seeking behavior, then winners who just barely won should be less motivated to compete again because of a high chance of moving down in status. However, losers who just barely lost should be more motivated to compete again because of a high chance of moving up in status. Testing this hypothesis with women, Zilioli et al. (2014) showed that women competing in a number-tracing task who won by only a small margin (given feedback of their narrow win throughout multiple trials in the task) decreased in testosterone from before to after the task, whereas women who lost by a narrow margin slightly increased in testosterone (Study 1). In a second companion study, under conditions of relative performance uncertainty, women competing in Tetris showed greater competition-related declines in testosterone than women who lost (Study 2). A simultaneous study in men competing in Tetris on two consecutive days (Zilioli & Watson, 2014) found evidence that day 2 testosterone increased significantly more across the competition period for those whose status reversed from the day 1 competition (i.e., a day 1 win followed by a day 2 loss, or vice versa; an unstable hierarchy) compared to men who won both days or lost both days (a stable hierarchy). When wins and losses where manipulated to be either “clear” or “narrow,” Wu, Eisenegger, Zilioli, Watson, and Clark (2017) showed that testosterone levels decreased ­significantly for narrow winners, but only if their basal cortisol levels were relatively high. Despite some ­empirical support in these studies for the status instability hypothesis with regard to the direction of testosterone change, no studies have demonstrated the reverse win–lose effect pattern ­consistent with this hypothesis nor have any studies shown that this pattern of testosterone change predicts subsequent motivation to compete again—a main tenet of the hypothesis. To make more i­nformed inferences about the functional significance of testosterone ­increases and decreases r­esulting from status gained and lost within stable and unstable social hierarchies, future research should attempt to corroborate testosterone reactivity with postcontest motivational states related to social status. The wide array of moderating factors complicates attempts to understand relationships between

Competition, Dominance, and Social Hierarchy

Social Context

Biological Sex

Basal Testosterone

Subsequent competitive motivation & cognitive states

Power Motivation & Trait Dominance

Psychological & Cognitive State

Gender Socialization, Genetics, & life Experiences

Dominance & status-seeking behavior

Testosterone change in response to a status challenge

Basal Cortisol Social Status

Figure 17.2  Theoretical model for the testosterone–social status relationship.

hormones and competition, but is rightly indicative of the complexity of human nature and the intricate social context of status striving. Figure 17.2 displays an updated theoretical framework for future studies testing both basal and dynamic testosterone–status relationships. For a more complete understanding of the role of person and context factors, future research will require relatively large sample sizes to properly test moderation and other complex interactions.

Additional Future Directions

The field of social-behavioral neuroendocrinology is a burgeoning area of scientific inquiry. Although the last three decades have produced foundational discoveries, there are many unanswered questions and new directions for future research.

The Role of Trait Competitiveness in Predicting Testosterone Response to Competition

According to social comparison theory, the drive to compete is derived from the basic human need to reduce uncertainty between one’s own performance and the performance of others in order to maintain superior relative position (Festinger, 1954; Garcia, Tor, & Schiff, 2013). For some, this drive is sufficiently strong to prompt greater efforts to engage and succeed in situations where relative judgments about performance are made. Because comparison to others through competition is how relative social

status is determined, individual differences in competitiveness, the “desire to win in interpersonal situations” (Smither & Houston,  1992, p. 408), may be a direct predictor of relative social status or the motivation to acquire it. Given the apparent connection between status seeking and testosterone, competition-related changes in testosterone levels may depend on individual differences in trait competitiveness. Welker and Carré (2015) recently reported that basal testosterone in men correlates with ­ persistence in attempting to solve puzzles made intentionally unsolvable by the experimenters. Although conceptually different than competitiveness per se, task persistence is a core quality of highly competitive individuals. Future research should consider including competitiveness as a person factor regulating hormonal response to competition and explore conceptual and statistical relationships ­between competitiveness and trait dominance, as well as power motivation.

Group Dynamics and Social Network Analysis

Real-world and everyday contests for social status occur in the context of complex social networks involving multilayered social group interactions with others. Despite the importance of group ­dynamics in status attainment and maintenance, previous research on the social neuroendocrinology of dominance has largely focused on the individual—testing participants in solitary rooms with Casto and Mehta

307

little to no interaction with their “opponent.” It is reasonable to assume that status lost or gained in contexts where participants have no current or future potential interaction with each other would have, at best, a minor impact on one’s physiology or behavior (i.e., relative standing to an unknown person may have little relevance to perceptions of social status). Indeed, individuals who are more characteristically similar and closer in a social network typically have amplified social comparison and competitiveness with each other (e.g., sibling rivalry; Garcia et al.,  2013). Thus, behaviors driven by status motivation are most effectively employed when relevant to others in a social group, one that the individual identifies with. Recognizing the role of group processes and social influence on competitive efforts and hormonal responses will be important for a comprehensive understanding of the social neuroendocrinology of competition. Status motivation and underlying hormonal correlates should be considered within the context of social groups (e.g., Oxford, Ponzi, & Geary, 2010) with reference to factors such as an individual’s level of group identification, status ranking within the group, interand intragroup competition, and need for affiliation. Social neuroendocrinologists could consider implementing group social psychology techniques (e.g., Cheng et al.,  2010,  2013; Ronay, Greenaway, Anicich, & Galinsky, 2012) to explore relationships among group members and group-level interactional characteristics of social hierarchies. Advanced statistical techniques could also prove helpful in implementing more comprehensive social group dynamics (e.g., social network analysis; Kornienko et al., 2014, 2016).

The Functional Significance of Transient Elevations in Testosterone

Another fruitful area of future research involves increasing our understanding of the adaptive purpose of testosterone reactivity in status-relevant contexts. Thus, it is important to expand research on the immediate and subsequent benefits of rapid and transient elevations in testosterone during competitive behavior and as a result of status shifts. As discussed earlier, previous research has shown that testosterone elevations, under certain contexts, increase subsequent willingness to compete again and appear to alter cognitions in ways that could benefit status seeking. Going further, researchers could explore testosterone-related competitive decision making when manipulating aspects about one’s opponent (e.g., photos of potential opponents could be shown 308

with aspects such as gender and dominance-related facial features, and information about that opponent’s skill level and history of success in competition could be manipulated). Another area of interest involves moving beyond just the categorical choice of willingness to compete again or not and devising a quantifiable measure of competitive effort following testosterone elevation (i.e., a subsequent task of competitive persistence). As researchers expand the  postcompetition repertoire of status-related behaviors, both antisocial and prosocial means of status maintenance should be considered, as these are opposing yet equally viable strategies (Cheng et al., 2013). It is also important to understand the potential benefit of testosterone change, in relation to both physiology and psychology, with the purpose of expanding the relevance, reach, and applicability of social neuroendocrinology more generally. That short-term testosterone elevations can advantageously alter future behaviors and cognitions suggests that there could be some marketable use in developing evidence-based interventions, tasks that could reliably produce endogenous testosterone increases in contexts where this response could prove beneficial for performance. Researchers in other fields, policymakers, and organizations interested in how social groups function (e.g., corporate business and sports teams) would better understand the value of implementing social neuroendocrinology methods with proper tools for taking advantage of the testosterone–status relationship.

Other Areas of Future Research

Other future directions include pursuing a better understanding of hormonal modulation of reward circuitry in the brain (e.g., Bless, McGinnis, Mitchell, Hartwell, & Mitchell,  1997; Frye, Rhodes, Rosellini, & Svare, 2002; Hermans et al., 2010; Montoya, Bos, Terburg, Rosenberger, & van Honk, 2014; Packard, Cornell, & Alexander, 1997; Salvador & Costa,  2009). Although researchers have a growing understanding of how testosterone and other sex steroids relate to behavior, less is known about the neural mechanisms of this relationship. Additionally, social neuroendocrinology, discipline-wide, is challenged with the vital goal of improving assay methodologies to increase the validity of hormone measurement (Granger, Shirtcliff, Booth, Kivlighan, & Schwartz,  2004; Granger et al., 2007; Welker et al., 2016). Perhaps there is no more pressing task than this given that the accuracy of hormone data is the basis of all other assumptions about hormone–behavior rela-

Competition, Dominance, and Social Hierarchy

tionships. Finally, future research should make greater efforts to explore how the research discussed in this chapter is dependent on sex, gender, gender identity, and gender socialization and expand theoretical models to include a more comprehensive understanding of the endocrinology of status motivation and social hierarchies among women (Casto & Prasad, 2017).

Conclusion

Competition, dominance, and social hierarchy are aspects of human behavior, past and present, that represent a fundamental human need to attain and maintain social status. The core of these behaviors is driven and can be explained by complex endocrinological processes interacting with the social environment. Understanding these relationships helps explain the causes and consequences of social stratification—why, for example, those born in conditions of poverty face harsher realities and have more barriers to educational achievement, job advancement, and health care. Recognizing how our biology influences the struggle to have and be more than others at the society level, and the individual differences of this drive, is perhaps one vital step in altering perceptions and creating greater awareness for the disparate situations of others. Through the lens of individual and group functioning and performance, there is also great opportunity. By seeking out the best of human potential, hormonal–status relationships can be used to discover interventions beneficial to performance in competitive tasks and functioning within social networks.

References

Adler, N. E., Epel, E. S., Castellazzo, G., & Ickovics, J. R. (2000). Relationship of subjective and objective social status with psychological and physiological functioning: Preliminary data in healthy white women. Health Psychology, 19, 586–592. Aguilar, R., Jiménez, M., & Alvero-Cruz, J.  R. (2013). Testosterone, cortisol and anxiety in elite field hockey players. Physiology & Behavior, 119, 38–42. Akinola, M., & Mendes, W.  B. (2014). It’s good to be king: Neurobiological benefits of higher social standing. Social Psychological and Personality Science, 5, 43–51. Akinola, M., Page-Gould, E., Mehta, P. H., & Lu, J. G. (2016). Collective hormonal profiles predict group performance. Proceedings of the National Academy of Sciences, 113, 9774–9779. Alexander, R.  D. (1989). Evolution of the human psyche. In P.  Mellars & C.  Stringer (Eds.), The human revolution: Behavioural and biological perspectives on the origins of modern humans (pp. 455–513). Princeton, NJ: Princeton University Press. Anderson, C., Hildreth, J. A. D., & Howland, L. (2015). Is the desire for status a fundamental human motive? A review of the empirical literature. Psychological Bulletin, 141, 574–601.

Anderson, C., & Kilduff, G. J. (2009). The pursuit of status in social groups. Current Directions in Psychological Science, 18, 295–298. Apicella, C.  L., Dreber, A., Gray, P.  B., Hoffman, M., Little, A.  C., & Campbell, B.  C. (2011). Androgens and competitiveness in men. Journal of Neuroscience, Psychology, and Economics, 1, 54–62. Apicella, C.  L., Dreber, A., & Mollerstrom, J. (2014). Salivary testosterone change following monetary wins and losses predicts future financial risk-taking. Psychoneuroendocrinology, 39, 58–64. Archer, J. (2006). Testosterone and human aggression: An evaluation of the challenge hypothesis. Neuroscience & Biobehavioral Reviews, 30, 319–345. Aschbacher, K., O’Donovan, A., Wolkowitz, O.  M., Dhabhar, F. S., Su, Y., & Epel, E. (2013). Good stress, bad stress and oxidative stress: Insights from anticipatory cortisol reactivity. Psychoneuroendocrinology, 38, 1698–1708. August, G.  P., Grumbach, M.  M., & Kaplan, S.  L. (1972). Hormonal changes in puberty: III. Correlation of plasma testosterone, LH, FSH, testicular size, and bone age with male pubertal development. Journal of Clinical Endocrinology & Metabolism, 34, 319–326. Baines, J., & Lacovara, P. (2002). Burial of the dead in ancient Egyptian society. Journal of Social Archaeology, 2, 5–36. Bateup, H.  S., Booth, A., Shirtcliff, E.  A., & Granger, D.  A. (2002). Testosterone, cortisol, and women’s competition. Evolution and Human Behavior, 23, 181–192. Berger, J., Cohen, B.  P., & Zelditch, M. (1972). Status characteristics and social interaction. American Sociological Review, 37, 241–255. Berger, J., Heinrichs, M., von Dawans, B., Way, B. M., & Chen, F. S. (2016). Cortisol modulates men’s affiliative responses to acute social stress. Psychoneuroendocrinology, 63, 1–9. Berger, J., Rosenholtz, S.  J., & Zelditch, M. (1980). Status organizing processes. Annual Review of Sociology, 6, 479–508. Bless, E. P., McGinnis, K. A., Mitchell, A. L., Hartwell, A., & Mitchell, J.  B. (1997). The effects of gonadal steroids on brain stimulation reward in female rats. Behavioral Brain Research, 82, 235–244. Bos, P. A., Hermans, E. J., Ramsey, N. F., & Van Honk, J. (2012). The neural mechanisms by which testosterone acts on interpersonal trust. NeuroImage, 61(3), 730–737. Bos, P.  A., Terburg, D., & van Honk, J. (2010). Testosterone decreases trust in socially naive humans. Proceedings of the National Academy of Sciences, 107(22), 9991–9995. Burgoon, J. K., Johnson, M. L., & Koch, P. T. (1998). The nature and measurement of interpersonal dominance. Commu­ nications Monographs, 65, 308–335. Burnham, T. (2007). High-testosterone men reject low ultimatum game offers. Proceedings of the Royal Society B, 274, 2327–2330. Burnstein, K., Maiorino, C., Dai, J., & Cameron, D. (1995). Androgen and glucocorticoid regulation of androgen receptor cDNA expression. Molecular and Cellular Endocrinology, 115, 177–186. Carré, J. (2009). No place like home: Testosterone responses to victory depend on game location. American Journal of Human Biology, 21, 392–394. Carré, J.  M., & Archer, J. (2017). Testosterone and human behavior: The role of individual and contextual variables. Current Opinion in Psychology, 19, 149–153. Carré, J.  M., Baird-Rowe, C., & Hariri, A.  R. (2014). Testosterone responses to competition predict decreased trust

Casto and Mehta

309

ratings of emotionally neutral faces. Psychoneuroendocrinology, 49, 79–83. Carré, J.  M., Campbell, J.  A., Lozoya, E., Goetz, S.  M., & Welker, K. M. (2013). Changes in testosterone mediate the effect of winning on subsequent aggressive behaviour. Psychoneuroendocrinology, 38, 2034–2041. Carré, J. M., Gilchrist, J. D., Morrissey, M. D., & McCormick, C. M. (2010). Motivational and situational factors and the relationship between testosterone dynamics and human aggression during competition. Biological Psychology, 84, 346–353. Carré, J. M., & McCormick, C. M. (2008). Aggressive behavior and change in salivary testosterone concentrations predict willingness to engage in a competitive task. Hormones and Behavior, 54, 403–409. Carré, J.  M., McCormick, C.  M., & Hariri  A.  R. (2011). The social neuroendocrinology of human aggression. Psychoneuro­ endocrinology, 36, 935–944. Carré, J., Muir, C., Belanger, J., & Putnam, S. K. (2006). Precompetition hormonal and psychological levels of elite hockey players: Relationship to the “home advantage.” Physiology & Behavior, 89, 392–398. Carré, J.  M., & Olmstead, N.  A. (2015). Social neuro­ endocrinology of human aggression: Examining the role of competition-induced testosterone dynamics. Neuroscience, 286, 171–186. Carré, J.  M., Putnam, S.  K., & McCormick, C.  M. (2009). Testosterone responses to competition predict future aggressive behavior at a cost to reward in men. Psychoneuroendocrinology, 34, 561–570. Cashdan, E. (1995). Hormones, sex, and status in women. Hormones and Behavior, 29, 354–366. Cashdan, E. (2003). Hormones and competitive aggression in women. Aggressive Behavior, 29, 107–115. Casto, K. V. (2016). The role of status motivation and social context in predicting competitive will and hormonal response to competition (Doctoral dissertation). Retrieved from ProQuest and Emory University’s Electronic Thesis and Dissertation Repository, http://pid.emory.edu/ark:/25593/rrqrg Casto, K. V., & Edwards, D. A. (2016a). Testosterone, cortisol, and human competition. Hormones and Behavior, 82, 21–37. Casto, K.  V., & Edwards, D.  A. (2016b). Before, during, and after: How phases of competition differentially affect testosterone, cortisol, and estradiol levels in women athletes. Adaptive Human Behavior and Physiology, 2, 11–25. Casto, K.  V., & Edwards, D.  A. (2016c). Testosterone and reconciliation among women: After-competition testosterone predicts prosocial attitudes towards opponents. Adaptive Human Behavior and Physiology, 2, 220–233. Casto, K.  V., Elliott, C.  M., & Edwards, D.  A. (2014). Intercollegiate cross country competition: Effects of warm-up and racing on salivary levels of cortisol and testosterone. International Journal of Exercise Science, 7, 318–328. Casto, K.  V., & Prasad, S. (2017). Commentary to the special issue, Hormones and Human Competition: Recommen­ dations for the study of women and sex effects in hormones and competition research. Hormones and Behavior, 92, 190–194. Casto, K. V., Rivell, A. & Edwards, D. A. (2017). Competitionrelated testosterone, cortisol, and perceived personal success in recreational women athletes. Hormones and Behavior, 92, 29–36. Chase, I. D., Tovey, C., Spangler-Martin, D., & Manfredonia, M. (2002). Individual differences versus social dynamics in

310

the formation of animal dominance hierarchies. Proceedings of the National Academy of Sciences of the United States of America, 99, 5744–5749. Chen, S., Wang, J., Yu, G., Liu, W., & Pearce, D. (1997). Androgen and glucocorticoid receptor heterodimer formation. A possible mechanism for mutual inhibition of transcriptional activity. Journal of Biological Chemistry, 272, 14087–14092. Cheng, J. T., & Tracy, J. L. (2014). Toward a unified science of hierarchy: Dominance and prestige are two fundamental pathways to human social rank. In J.  Cheng, J.  Tracy, & C. Anderson (Eds.), The psychology of social status (pp. 3–27). New York, NY: Springer. Cheng, J.  T., Tracy, J.  L., Foulsham, T., Kingstone, A., & Henrich, J. (2013). Two ways to the top: Evidence that dominance and prestige are distinct yet viable avenues to social rank and influence. Journal of Personality and Social Psychology, 104, 103–125. Cheng, J. T., Tracy, J. L., & Henrich, J. (2010). Pride, personality, and the evolutionary foundations of human social status. Evolution and Human Behavior, 31, 334–347. Cheng, J.  T., Tracy, J., Ho, S., & Henrich, J. (2016). Listen, follow me: Dynamic vocal signals of dominance predict emergent social rank in humans. Journal of Experimental Psychology, 145, 536–547. Chiao, J. Y. (2010). Neural basis of social status hierarchy across species. Current Opinion in Neurobiology, 20, 803–809. Chua, T.  H.  H., & Chang, L. (2016). Follow me and like my beautiful selfies: Singapore teenage girls’ engagement in selfpresentation and peer comparison on social media. Computers in Human Behavior, 55, 190–197. Cobey, K. D., Nicholls, M., Leongómez, J. D., & Roberts, S. C. (2015). Self-reported dominance in women: Associations with hormonal contraceptive use, relationship status, and testosterone. Adaptive Human Behavior and Physiology, 1, 449–459. Cohen, S., Doyle, W.  J., & Baum, A. (2006). Socioeconomic status is associated with stress hormones. Psychosomatic Medicine, 68, 414–420. Cohen, S., Janicki-Deverts, D., Doyle, W.  J., Miller, G.  E., Frank, E., Rabin, B.  S., & Turner, R.  B. (2012). Chronic stress, glucocorticoid receptor resistance, inflammation, and disease risk. Proceedings of the National Academy of Sciences, 109, 5995–5999. Cohen, S., & McKay, G. (1984). Social support, stress, and the buffering hypothesis: A theoretical analysis. In A.  Baum, S.  E.  Taylor, & J.  E.  Singer (Eds.), Handbook of psychology and health (pp. 253–267). Hillsdale, NJ: Erlbaum. Cook, C.  J., & Crewther, B.  T. (2012a). Changes in salivary testosterone concentrations and subsequent voluntary squat performance following the presentation of short video clips. Hormones and Behavior, 61, 17–22. Cook, C. J., & Crewther, B. T. (2012b). The effects of different pre-game motivational interventions on athlete free hormonal state and subsequent performance in professional rugby union matches. Physiology & Behavior, 106, 683–688. Copeland, J.  L., Consitt, L.  A., & Tremblay, M.  S. (2002). Hormonal responses to endurance and resistance exercise in females aged 19–69 years. Journals of Gerontology Series A, Biological and Medical Sciences, 57, 158–165. Costa, R., & Salvador, A. (2012). Associations between success and failure in a face-to-face competition and psychobiological parameters in young women. Psychoneuroendocrinology, 37, 1780–1790.

Competition, Dominance, and Social Hierarchy

Costa, R., Serrano, M. A., & Salvador, A. (2016). Importance of self-efficacy in psychoendocrine responses to competition and performance in women. Psicothema, 28, 66–70. Dabbs, J.  M., Jurkovic, G.  J., & Frady, R.  L. (1991). Salivary testosterone and cortisol among late adolescent male offenders. Journal of Abnormal Child Psychology, 19, 469–478. Dabbs, J. M., & Morris, R. (1990). Testosterone, social class, and antisocial behavior in a sample of 4,462 men. Psychological Science, 1, 209–211. Dasgupta, K. (2015). Introducing social stratification: The causes and consequences of inequality. Boulder, CO: Lynne Rienner Publishers. Decker, S. A. (2000). Salivary cortisol and social status among Dominican men. Hormones and Behavior, 38, 29–38. Demakakos, P., Nazroo, J., Breeze, E., & Marmot, M. (2008). Socioeconomic status and health: The role of subjective social status. Social Science & Medicine, 67, 330–340. Denson, T. F., Mehta, P. H., & Tan, D. H. (2013). Endogenous testosterone and cortisol jointly influence reactive aggression in women. Psychoneuroendocrinology, 38, 416–424. Dickerson, S. S., & Kemeny, M. E. (2004). Acute stressors and cortisol responses: A theoretical integration and synthesis of laboratory research. Psychological Bulletin, 130, 355–391. Diefenbach, T. (2013). Hierarchy and organization: Toward a general theory of hierarchical social systems. New York, NY: Routledge. Edwards, D. A., & Casto, K. V. (2013). Women’s intercollegiate athletic competition: Cortisol, testosterone, and the dualhormone hypothesis as it relates to status among teammates. Hormones and Behavior, 64, 153–160. Edwards, D.  A., & Kurlander, L.  S. (2010). Women’s intercollegiate volleyball and tennis: Effects of warm-up, competition, and practice on saliva levels of cortisol and testosterone. Hormones and Behavior, 58, 606–613. Edwards, D.  A., Wetzel, K., & Wyner, D.  R. (2006). Intercollegiate soccer: Saliva cortisol and testosterone are elevated during competition, and testosterone is related to status and social connectedness with teammates. Physiology and Behavior, 87, 135–143. Ehrenkranz, J., Bliss, E., & Sheard, M.  H. (1974). Plasma testosterone: Correlation with aggressive behavior and social dominance in men. Psychosomatic Medicine, 36, 469–475. Eisenegger, C., Haushofer, J., & Fehr, E. (2011). The role of testosterone in social interaction. Trends in Cognitive Sciences, 15, 263–271. Eisenegger, C., Kumsta, R., Naef, M., Gromoll, J., & Heinrichs, M. (2017). Testosterone and androgen receptor gene polymorphism are associated with confidence and competitiveness in men. Hormones and Behavior, 92, 93–102. Eisenegger, C., Naef, M., Snozzi, R., Heinrichs, M., & Fehr, E. (2010). Prejudice and truth about the effect of testosterone on human bargaining behaviour. Nature, 463, 356–359. Ellyson, S. L., & Dovidio, J. F. (1985). Power, dominance, and nonverbal behavior: Basic concepts and issues. In S. L. Ellyson & J.  F.  Dovidio (Eds.), Power, dominance, and nonverbal behavior (pp. 1–28). New York, NY: Springer-Verlag. Engeland, W. C., & Arnhold, M. M. (2005). Neural circuitry in the regulation of adrenal corticosterone rhythmicity. Endocrine, 28, 325–331. Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117–140. Flinn, M. V., Geary, D. C., & Ward, C. V. (2005). Ecological dominance, social competition, and coalitionary arms races:

Why humans evolved extraordinary intelligence. Evolution and Human Behavior, 26, 10–46. Frye, C. A., Rhodes, M. E., Rosellini, R., & Svare, B. (2002). The nucleus accumbens as a site of action for rewarding properties of testosterone and its 5α-reduced metabolites. Pharmacology Biochemistry and Behavior, 74, 119–127. Garcia, S. M., Tor, A., & Schiff, T. M. (2013). The psychology of competition a social comparison perspective. Perspectives on Psychological Science, 8, 634–650. Geniole, S.  N., Bird, B.  M., Ruddick, E.  L., & Carré, J.  M. (2017). Effects of competition outcome on testosterone concentrations in humans: An updated meta-analysis. Hormones and Behavior, 92, 37–50. Geniole, S.  N., Busseri, M.  A., & McCormick, C.  M. (2013). Testosterone dynamics and psychopathic personality traits independently predict antagonistic behavior towards the perceived loser of a competitive interaction. Hormones and Behavior, 64, 790–798. Gladue, B. A., Boechler, M., & McCaul, K. D. (1989). Hormonal response to competition in human males. Aggressive Behavior, 15, 409–422. Gleason, E. D., Fuxjager, M. J., Oyegbile, T. O., & Marler, C. A. (2009). Testosterone release and social context: When it occurs and why. Frontiers in Neuroendocrinology, 30, 460–469. Gledhill, J., Bender, B., & Larsen, M. T. (1988). State and society: The emergence and development of social hierarchy and political centralization. Boston, MA: Unwin Hyman. Gonzalez-Bono, E., Salvador, A., Serrano, M. A., & Ricarte, J. (1999). Testosterone, cortisol, and mood in a sports team competition. Hormones and Behavior, 35, 55–62. Goymann, W., & Wingfield, J. C. (2004). Allostatic load, social status and stress hormones: The cost of social status matter. Animal Behaviour, 67, 591–602. Granger, D. A., Kivlighan, K. T., Fortunato, C., Harmon, A. G., Hibel, L. C., Schwartz, E. B., & Whembolua, G. L. (2007). Integration of salivary biomarkers into developmental and behaviorally-oriented research: Problems and solutions for collecting specimens. Physiology & Behavior, 92, 583–590. Granger, D. A., Shirtcliff, E. A., Booth, A., Kivlighan, K. T., & Schwartz, E.  B. (2004). The “trouble” with salivary testosterone. Psychoneuroendocrinology, 29, 1229–1240. Grant, V., & France, J. T. (2001). Dominance and testosterone in women. Biological Psychology, 58, 41–47. Gruenewald, T. L., Kemeny, M. E., & Aziz, N. (2006). Subjective social status moderates cortisol responses to social threat. Brain, Behavior, and Immunity, 20, 410–419. Guezennec, C. Y., Lafarge, J. P., Bricout, V. A., Merino, D., & Serrurier, B. (1995). Effect of competition stress on tests used to assess testosterone administration in athletes. International Journal of Sports Medicine, 16, 368–372. Hamilton, L. D., Carré, J. M., Mehta, P. H., Olmstead, N., & Whitaker, J. D. (2015). Social neuroendocrinology of status: A review and future directions. Adaptive Human Behavior and Physiology, 1, 202–230. Hellhammer, D. H., Buchtal, J., Gutberlet, I., & Kirschbaum, C. (1997). Social hierarchy and adrenocortical stress reactivity in men. Psychoneuroendocrinology, 22, 643–650. Henry, A., Sattizahn, J.  R., Norman, G.  J., Beilock, S.  L., & Maestripieri, D. (2017). Performance during competition and competition outcome in relation to testosterone and cortisol among women. Hormones and Behavior, 92, 82–92. Hermans, E.  J., Bos, P.  A., Ossewaarde, L., Ramsey, N.  F., Fernández, G., & van Honk, J. (2010). Effects of exogenous

Casto and Mehta

311

testosterone on the ventral striatal BOLD response during reward anticipation in healthy women. Neuroimage, 52, 277–283. Hermans, E. J., Putman, P., Baas, J. M., Koppeschaar, H. P., & van Honk, J. (2006). A single administration of testosterone reduces fear-potentiated startle in humans. Biological Psychiatry, 59, 872–874. Hermans, E.  J., Ramsey, N.  F., & van Honk, J. (2008). Exogenous testosterone enhances responsiveness to social threat in the neural circuitry of social aggression in humans. Biological Psychiatry, 63, 263–270. Jiménez, M., Aguilar, R., & Alvero-Cruz, J. R. (2012). Effects of victory and defeat on testosterone and cortisol response to competition: Evidence for same response patterns in men and women. Psychoneuroendocrinology, 37, 1577–1581. Johnson, E. O., Kamilaris, T. C., Chrousos, G. P., & Gold, P. W. (1992). Mechanisms of stress: A dynamic overview of hormonal and behavioral homeostasis. Neuroscience and Biobehavioral Reviews, 16, 115–130. Josephs, R.  A., Newman, M.  L., Brown, R.  P., & Beer, J.  M. (2003). Status, testosterone, and human intellectual performance: Stereotype threat as status concern. Psychological Science, 14, 158–163. Josephs, R. A., Sellers, J. G., Newman, M. L., & Mehta, P. H. (2006). The mismatch effect: When testosterone and status are at odds. Journal of Personality and Social Psychology, 90, 999–1013. Kemper, T. D. (1990). Social structure and testosterone: Explorations of the socio-bio-social chain. New Brunswick, NJ: Rutgers University Press. Knight, E. L., & Mehta, P. H. (2014). Hormones and hierarchies. In J. Cheng, J. Tracy, & C. Anderson (Eds.), The psychology of social status (pp. 269–302). New York, NY: Springer. Knight, E.  L., & Mehta, P.  H. (2017). Hierarchy stability moderates the effect of status on stress and performance in humans. Proceedings of the National Academy of Sciences of the United States of America, 114, 78–83. Kornienko, O., Clemans, K.  H., Out, D., & Granger, D.  A. (2014). Hormones, behavior, and social network analysis: Exploring associations between cortisol, testosterone, and network structure. Hormones and Behavior, 66, 534–544. Kornienko, O., Schaefer, D. R., Weren, S., Hill, G. W., & Granger, D. A. (2016). Cortisol and testosterone associations with social network dynamics. Hormones and Behavior, 80, 92–102. Kudielka, B.  M., Hellhammer, D.  H., & Wüst, S. (2009). Why do we respond so differently? Reviewing determinants of  human salivary cortisol responses to challenge. Psychoneuroendocrinology, 34, 2–18. Magee, J. C., & Galinsky, A. D. (2008). Social hierarchy: The self- reinforcing nature of power and status. Academy of Management Annals, 2, 351–398. Maner, J. K., Miller, S. L., Schmidt, N. B., & Eckel, L. A. (2008). Submitting to defeat: Social anxiety, dominance threat, and decrements in testosterone. Psychological Science, 19, 764–768. Mazur, A. (1985). A biosocial model of status in face-to-face primate groups. Social Forces, 64, 377–402. Mazur, A. (2017). Testosterone in biosociology: A memoir. Hormones and Behavior, 92, 3–8. Mazur, A., & Booth, A. (1998). Testosterone and dominance in men. Behavioral and Brain Sciences, 21, 353–397. Mazur, A., & Booth, A. (2014). Testosterone is related to deviance in male army veterans, but relationships are not moderated by cortisol. Biological Psychology, 96, 72–76.

312

Mazur, A., & Lamb, T. A. (1980). Testosterone, status, and mood in human males. Hormones and Behavior, 14, 236–246. Mazur, A., Susman, E. J., & Edelbrock, S. (1997). Sex difference in testosterone response to a video game contest. Evolution and Human Behavior, 18, 317–326. Mazur, A., Welker, K. M., & Peng, B. (2015). Does the biosocial model explain the emergence of status differences in conversations among unacquainted men? PLoS One, 10, e0142941. McCaul, K. D., Gladue, B. A., & Joppa, M. (1992). Winning, losing, mood, and testosterone. Hormones and Behavior, 26, 486–504. McEwen, B. (1998). Stress, adaptation, and disease: Allostasis and allostatic load. Annals of the New York Academy of Sciences, 840, 33–44. McEwen, B. (2000). Allostasis and allostatic load: Implications for neuropsychopharmacology. Neuropsychopharmacology, 22, 108–124. McEwen, B. (2004). Protection and damage from acute and chronic stress: Allostasis and allostatic overload and relevance to the pathophysiology of psychiatric disorders. Annals of the New York Academy of Sciences, 1032, 1–7. Mehta, P. H., DesJardins, N. M. L., van Vugt, M., & Josephs, R.  A. (2017). Hormonal underpinnings of status conflict: Testosterone and cortisol are related to decisions and satisfaction in the hawk-dove game. Hormones and Behavior, 92, 141–154. Mehta, P. H., Jones, A. C., & Josephs, R. A. (2008). The social endocrinology of dominance: Basal testosterone predicts cortisol changes and behavior following victory and defeat. Journal of Personality and Social Psychology, 94, 1078–1093. Mehta, P. H., & Josephs, R. A. (2006). Testosterone change after losing predicts the decision to compete again. Hormones and Behavior, 50, 684–692. Mehta, P. H., & Josephs, R. A. (2010). Testosterone and cortisol jointly regulate dominance: Evidence for a dual-hormone hypothesis. Hormones and Behavior, 58, 898–906. Mehta, P.  H., Mor, S., Yap, A.  J., & Prasad, S. (2015). Dualhormone changes are related to bargaining performance. Psychological Science, 26, 866–876. Mehta, P. H., & Prasad, S. (2015). The dual-hormone hypothesis: A brief review and future research agenda. Current Opinions in Behavioral Science, 3, 163–168. Mehta, P. H., Snyder, N. A., Knight, E. L., & Lasseter, B. (2015). Close versus decisive victory moderates the effect of testosterone change on competitive decisions and task enjoyment. Adaptive Human Behavior and Physiology, 1, 291–311. Mehta, P.  H, van Son, V., Welker, K.  M., Prasad, S., Sanfey, A.  G., Smidts, A., & Roelofs, K. (2015). Exogenous testosterone in women enhances and inhibits competitive decision-making depending on victory-defeat experience and trait dominance. Psychoneuroendocrinology, 60, 224–236. Mehta, P.  H., Wuehrmann, E.  V., & Josephs, R.  A. (2009). When are low testosterone levels advantageous? The moderating role of individual versus intergroup competition. Hormones and Behavior, 56, 158–162. Montgomery, M.  A., & Elimelech, M. (2007). Water and sanitation in developing countries: Including health in the equation. Environmental Science & Technology, 41, 17–24. Montoya, E. R., Bos, P. A., Terburg, D., Rosenberger, L. A., & van Honk, J. (2014). Cortisol administration induces global down-regulation of the brain’s reward circuitry. Psychoneuroendocrinology, 47, 31–42.

Competition, Dominance, and Social Hierarchy

Nave, G., Nadler, A., Zava, D., & Camerer, C. (2017). Single dose testosterone administration impairs cognitive reflection in men. Psychological Science, 28(10), 1398–1407. Neave, N., Laing, S., Fink, B., & Manning, J.  T. (2003). Second to fourth digit ratio, testosterone and perceived male dominance. Proceedings of the Royal Society B, 270, 2167–2172. Neave, N., & Wolfson, S. (2003). Testosterone, territoriality, and the “home advantage.” Physiology & Behavior, 78, 269–275. Newman, M.  L., & Josephs, R.  A. (2009). Testosterone as a personality variable. Journal of Research in Personality, 43, 258–259. Newman, M.  L., Sellers, J.  G., & Josephs, R.  A. (2005). Testosterone, cognition, and social status. Hormones and Behavior, 47, 205–211. Norman, R.  E., Moreau, B.  J., Welker, K.  M., & Carré, J.  M. (2015). Trait anxiety moderates the relationship between testosterone responses to competition and aggressive behavior. Adaptive Human Behavior and Physiology, 1, 312–324. O’Cass, A., & McEwen, H. (2004). Exploring consumer status and conspicuous consumption. Journal of Consumer Behaviour, 4, 25–39. Oliveira, G. A., & Oliveira, R. F. (2014). Androgen responsiveness to competition in humans: The role of cognitive variables. Neuroscience and Neuroeconomics, 3, 19–32. Oliveira, G.  A., Uceda, S., Oliveira, T.  F., Fernandes, A.  C., Garcia-Marques, T., & Oliveira, R.  F. (2014). Testosterone response to competition in males is unrelated to opponent familiarity or threat appraisal. Frontiers in Psychology, 5, 1240. Oliveira, T., Gouveia, M. J., & Oliveira, R. F. (2009). Testosterone responsiveness to winning and losing experiences in female soccer players. Psychoneuroendocrinology, 34, 1056–1064. Oxford, J., Ponzi, D., & Geary, D.  C. (2010). Hormonal responses differ when playing violent video games against an ingroup and outgroup. Evolution and Human Behavior, 31, 201–209. Packard, M.  G., Cornell, A.  H., & Alexander, G.  M. (1997). Rewarding affective properties on intra-nucleus accumbens injections of testosterone. Behavioral Neuroscience, 111, 219–224. Peters, B. J., Hammond, M. D., Reis, H. T., & Jamieson, J. P. (2016). The consequences of having a dominant romantic partner on testosterone responses during a social interaction. Psychoneuroendocrinology, 74, 308–315. Pew Research Center. (2017, January 12). Social media fact sheet. Retrieved from http://www.pewinternet.org/fact-sheet/socialmedia/ Phoenix, C.  H., Goy, R.  W., Gerall, A.  A., & Young, W.  C. (1959). Organizing action of prenatally administered testos­ terone propionate on the tissues mediating mating behavior in the female guinea pig. Endocrinology, 65, 369–382. Phoenix, C.  H., Slob, A.  K., & Goy, R.  W. (1973). Effects of castration and replacement therapy on sexual behavior of adult male rhesuses. Journal of Comparative and Physiological Psychology, 84, 472–481. Ponzi, D., Zilioli, S., Mehta, P.  H., Maslov, A., & Watson, N.  V. (2016). Social network centrality and hormones: The  interaction of testosterone and cortisol. Psychoneuro­ endocrinology, 68, 6–13. Popma, A., Vermeiren, R., Geluk, C.  A., Rinne, T., van den Brink, W., Knol, D. L., . . . Doreleijers, T. A. (2007). Cortisol moderates the relationship between testosterone and aggression in delinquent male adolescents. Biological Psychiatry, 61, 405–411.

Purifoy, F.  E., & Koopmans, L.  H. (1979). Androstenedione, testosterone, and free testosterone concentration in women of various occupations. Social Biology, 26, 179–188. Radke, S., Volman, I., Mehta, P., van Son, V., Enter, D., Sanfey, A., . . . Roelofs, K. (2015). Testosterone biases the amygdala toward social threat approach. Science Advances, 1, e1400074. Romano, J. F. (1990). Death, burial, and afterlife in ancient Egypt. Pittsburgh, PA: Carnegie Museum of Natural History. Ronay, R., & Carney, D.  R. (2013). Testosterone’s negative relationship with empathic accuracy and perceived leadership ability. Social Psychological and Personality Science, 4, 92–99. Ronay, R., Greenaway, K., Anicich, E.  M., & Galinsky, A.  D. (2012). The path to glory is paved with hierarchy: When hierarchical differentiation increases group effectiveness. Psychological Science, 23, 669–677. Rose, R.  M., Bernstein, I.  S., & Gordon, T.  P. (1975). Consequences of social conflict on plasma testosterone levels in rhesus monkeys. Psychosomatic Medicine, 37, 50–61. Salvador, A., & Costa, R. (2009). Coping with competition: Neuroendocrine responses and cognitive variables. Neuro­ science & Biobehavioral Reviews, 33, 160–170. Sapolsky, R. M. (1982). The endocrine stress-response and social status in the wild baboon. Hormones and Behavior, 16, 279–292. Sapolsky, R.  M. (2004). Social status and health in humans and  others animals. Annual Review of Anthropology, 33, 393–418. Sapolsky, R.  M. (2005). The influence of social hierarchy on primate health. Science, 308, 648–652. Schultheiss, O.  C. (2007). A biobehavioral model of implicit power motivation: Arousal, reward, and frustration. In E. Harmon-Jones & P. Winkielman (Eds.), Social neuroscience: Integrating biological and psychological explanations of social behavior (pp. 176–196). New York, NY: Guilford Press. Schultheiss, O. C., Campbell, K. L., & McClelland, D. C. (1999). Implicit power motivation moderates men’s testosterone responses to imagined and real dominance success. Hormones and Behavior, 36, 234–241. Schultheiss, O. C., & Rohde, W. (2002). Implicit power motivation predicts men’s testosterone changes in implicit learning in a contest situation. Hormones and Behavior, 41, 195–202. Schultheiss, O.  C., Wirth, M.  M., Torges, C.  M., Pang, J.  S., Villacorta, M. A., & Welsh, K. M. (2005). Effects of implicit power motivation on men’s and women’s implicit learning and testosterone changes after social victory or defeat. Journal of Personality and Social Psychology, 88, 174–188. Sellers, J. G., Mehl, M. R., & Josephs, R. A. (2007). Hormones and personality: Testosterone as a marker of individual differences. Journal of Research in Personality, 41, 126–138. Sherman, G. D., Jooa, J. L., Cuddy, A. J. C., Renshon, J., Oveis, C., Gross, J. J., & Lerner, J. S. (2012). Leadership is associated with lower levels of stress. PNAS, 109, 17903–17907. Sherman, G.  D., Lerner, J.  S., Josephs, R.  A., Renshon, J., & Gross, J.  J. (2016). The interaction of testosterone and cortisol is associated with attained status in male executives. Journal of Personality and Social Psychology, 110, 921–929. Sherman, G. D., Rice, L. K., Jin, E. S., Jones, A. S., & Josephs, R.  A. (2017). Sex differences in cortisol’s regulation of affiliative behavior. Hormones and Behavior, 92, 20–28. Shirtcliff, E. A., Peres, J. C., Dismukes, A. R., Lee, Y., & Phan, J. M. (2014). Riding the physiological roller coaster: Adaptive significance of cortisol stress reactivity to social contexts. Journal of Personality Disorders, 28, 40–51.

Casto and Mehta

313

Sidanius, J., & Pratto, F. (1999). Social dominance: An intergroup theory of social hierarchy and oppression. New York, NY: Cambridge University Press. Slatcher, R. B., Mehta, P. H., & Josephs, R. A. (2011). Testosterone and self-reported dominance interact to influence human mating behavior. Social Psychological and Personality Science, 2, 531–539. Smither, R. D., & Houston, J. M. (1992). The nature of compet­ itiveness: The development and validation of the compet­ itiveness index. Educational and Psychological Measurement, 52, 407–418. Sollberger, S., Bernauer, T., & Ehlert, U. (2016). Salivary testos­ terone and cortisol are jointly related to pro-environmental behavior in men. Social Neuroscience, 11, 553–566. Stanton, S. J., & Schultheiss, O. C. (2007). Basal and dynamic relationships between implicit power motivation and estradiol in women. Hormones and Behavior, 52, 571–580. Stanton, S.  J., & Schultheiss, O.  C. (2009). The hormonal correlates of implicit power motivation. Journal of Research in Personality, 43, 942–949. Starcke, K., & Brand, M. (2012). Decision making under stress: A selective review. Neuroscience & Biobehavioral Reviews, 36, 1228–1248. Steiner, E.  T., Barchard, K.  A., Meana, M., Hadi, F., & Gray, P.  B. (2010). The deal on testosterone responses to poker competition. Current Psychology, 29, 45–51. Suay, F., Salvador, A., González-Bono, E., Sanchıs, C., Martınez, M., Martınez-Sanchis, S., . . . Montoro, J. B. (1999). Effects of competition and its outcome on serum testosterone, cortisol and prolactin. Psychoneuroendocrinology, 24, 551–566. Tackett, J. L., Herzhoff, K., Harden, K. P., Page-Gould, E., & Josephs, R. A. (2014). Personality × hormone interactions in adolescent externalizing psychopathology. Personality Disorders: Theory, Research, and Treatment, 5, 235. Taylor, J.  H. (2001). Death and the afterlife in ancient Egypt. Chicago, IL: University of Chicago Press. Tiedens, L.  Z., & Fragale, A.  R. (2003). Power moves: Complementarity in dominant and submissive nonverbal behavior. Journal of Personality and Social Psychology, 84, 558. Tremblay, M.  S., Copeland, J.  L., & van Helder, W. (2005). Influence of exercise duration on post-exercise steroid responses in trained males. European Journal of Applied Physiology, 94, 505–513. Tremblay, R.  E., Schaal, B., Boulerice, B., Arseneault, L., Soussignan, R.  G., Paquette, D., & Laurent, D. (1998). Testosterone, physical aggression, dominance, and physical development in early adolescence. International Journal of Behavioral Development, 22, 753–777. Tsigos, C., & Chrousos, G. P. (2002). Hypothalamic-pituitaryadrenal axis, neuroendocrine factors and stress. Journal of Psychosomatic Research, 53, 865–871. van Anders, S. M., & Watson, N. V. (2007). Effects of abilityand chance-determined competition outcome on testosterone. Physiology & Behavior, 90, 634–642. van Bokhoven, I., van Goozen, S. H., van Engeland, H., Schaal, B., Arseneault, L., Séguin, J. R., . . . Tremblay, R. E. (2006). Salivary testosterone and aggression, delinquency, and social dominance in a population-based longitudinal study of adolescent males. Hormones and Behavior, 50, 118–125. van den Bos, R., Harteveld, M., & Stoop, H. (2009). Stress and decision-making in humans: Performance is related to cortisol reactivity, albeit differently in men and women. Psychoneuroendocrinology, 34, 1449–1458.

314

van der Meij, L., Buunk, A.  P., Almela, M., & Salvador, A. (2010). Testosterone responses to competition: The opponent’s psychological state makes it challenging. Biological Psychology, 84, 330–335. van der Meij, L., Schaveling, J., & van Vugt, M. (2016). Basal testosterone, leadership and dominance: A field study and meta-analysis. Psychoneuroendocrinology, 72, 72–79. van Honk, J., Montoya, E.  R., Bos, P.  A., Van Vugt, M., & Terburg, D. (2012). New evidence on testosterone and cooperation. Nature, 485, E4–E5. van Honk, J., & Schutter, D.  J.  L.  G. (2007). Testosterone reduces conscious detection of signals serving social correction: Implications for antisocial behavior. Psychological Science, 18, 663–667. van Honk, J., Schutter, D. J., Bos, P. A., Kruijt, A. W., Lentjes, E. G., & Baron-Cohen, S. (2011). Testosterone administration impairs cognitive empathy in women depending on secondto-fourth digit ratio. Proceedings of the National Academy of Sciences, 108, 3448–3452. van Honk, J., Tuiten, A., Verbaten, R., van den Hout, M., Koppeschaar, H., Thijssen, J., & de Haan, E. (1999). Corre­ lations among salivary testosterone, mood, and selective attention to threat in humans. Hormones and Behavior, 36(1), 17–24. Viru, M., Hackney, A. C., Karelson, K., Janson, T., Kuus, M., & Viru, A. (2010). Competition effects on physiological responses to exercise: Performance, cardiorespiratory and hormonal factors. Acta Physiologica Hungarica, 97, 22–30. Wagner, J. D., Flinn, M. V., & England, B. G. (2002). Hormonal response to competition among male coalitions. Evolution and Human Behavior, 23, 437–442. Walker, R. E., Keane, C. R., & Burke, J. G. (2010). Disparities and access to healthy food in the United States: A review of food deserts literature. Health & Place, 16, 876–884. Welker, K.  M., & Carré, J.  M. (2015). Individual differences in testosterone predict persistence in men. European Journal of Personality, 29, 83–89. Welker, K.  M., Lassetter, B., Brandes, C.  M., Prasad, S., Koop,  D.  R., & Mehta, P.  H. (2016). A comparison of salivary testosterone measurement using immunoassays and tandem mass spectrometry. Psychoneuroendocrinology, 71, 180–188. Welker, K. M., Lozoya, E., Campbell, J. A., Neumann, C. S., & Carré, J. M. (2014). Testosterone, cortisol, and psycho­pathic traits in men and women. Physiology & Behavior, 129, 230–236. Welker, K.  M., Norman, R.  E., Goetz, S., Moreau, B.  J.  P., Kitayama, S., & Carré, J.  M. (2017). Preliminary evidence that testosterone’s association with aggression depends on self-construal. Hormones and Behavior, 92, 117–127. Welling, L. L. M., Moreau, B. J. P., Bird, B. M., Hansen, S., & Carré, J. M. (2016). Exogenous testosterone increases men’s perceptions of their own dominance. Psychoneuroendocrinology, 64, 136–142. Whitworth, J. A., Williamson, P. M., Mangos, G., & Kelly, J. J. (2005). Cardiovascular consequences of cortisol excess. Vascular Health and Risk Management, 1, 291–299. Wingfield, J.  C. (2017). The challenge hypothesis: Where it began and relevance to humans. Hormones and Behavior, 92, 9–12. Wingfield, J. C., Hegner, R. E., Dufty, A. M., Jr., & Ball, G. F. (1990). The “challenge hypothesis”: Theoretical implications for testosterone secretion, mating systems, and breeding strategies. American Naturalist, 136, 829–846.

Competition, Dominance, and Social Hierarchy

Wirth, M. M., & Schultheiss, O. C. (2007). Basal testosterone moderates responses to anger faces in humans. Physiology & Behavior, 90, 496–505. Wirth, M.  M., Welsh, K.  M., & Schultheiss, O.  C. (2006). Salivary cortisol changes in humans after winning or losing a dominance contest depend on implicit power motivation. Hormones and Behavior, 49, 346–352. Wright, N. D., Bahrami, B., Johnson, E., Di Malta, G., Rees, G., Frith, C.  D., & Dolan, R.  J. (2012). Testosterone disrupts human collaboration by increasing egocentric choices. Proceedings of the Royal Society B, 279, 2275–2280. Wu, Y., Eisenegger, C., Zilioli, S., Watson, N. V., & Clark, L. (2017). Comparison of clear and narrow outcomes on testosterone levels in social competition. Hormones and Behavior, 92, 51–56.

Zilioli, S., Mehta, P.  H., & Watson, N.  V. (2014). Losing the battle but winning the war: Uncertain outcomes reverse the usual effect of winning on testosterone. Biological Psychology, 103, 54–62. Zilioli, S., & Watson, N.  V. (2013). Winning isn’t everything: Mood and testosterone regulate the cortisol response in competition. PLoS One, 8, e52582. Zilioli, S., & Watson, N. V. (2014). Testosterone across successive competitions: Evidence for a “winner effect” in humans? Psychoneuroendocrinology, 47, 1–9. Zyphur, M.  J., Narayanan, J., Koh, G., & Koh, D. (2009). Testosterone-status mismatch lowers collective efficacy in groups: Evidence from a slope-as-predictor multilevel structural equation model. Organizational Behavior and Human Decision Processes, 110, 70–79.

Casto and Mehta

315

CH A PT E R

18

Oxytocin An Evolutionary Framework

Nicholas M. Grebe and Steven W. Gangestad

Abstract Substantial excitement surrounds the mammalian peptide hormone oxytocin (OT) due to its potential to be a “hormone of love”—and more generally, a biological foundation for the diverse classes of intimate social bonds. Yet, theoretical models have struggled to absorb inconsistent, even contradictory, findings. Evolutionary theory will guide a coherent functional interpretation of the OT system. This chapter focuses on life history theory, a branch of theoretical biology that seeks to identify how natural selection shapes organisms’ efforts to optimally allocate limited resources. Endocrine hormones are important mediators of this process. A review of the psychological and physiological literature regarding OT suggests a number of possible trade-offs negotiated by oxytocinergic activity. This chapter proposes a provisional life history model in which OT is central to the regulation of important but vulnerable social relationships. It outlines implications of this model, addresses a number of caveats, and suggests directions for future research. Keywords:  oxytocin, life history theory, trade-offs, behavioral endocrinology, evolutionary biology

Evolutionary psychologists, in the tradition of Tinbergen’s (1963) “four questions,” are interested in the physiological underpinnings of psychological mechanisms. Hormones have become a point of focus as theoretically powerful tools for building a  conceptually coherent evolutionary framework linking physiology to behavior. Life history theory, a branch of theoretical biology, identifies hormones as components of a system that evolved to solve contingent allocation problems—that is, they are solutions to demands that organisms adaptively ­allocate energetic and other resources differently depending on conditions, as assessed by cues external or internal to the organism (e.g., Finch & Rose, 1995). Though some hormonal systems have been analyzed in this manner (e.g., Bribiescas,  2001; Ellison,  2003), oxytocin (OT) has not. Here, we ­advance a conceptualization of the OT system derived from life history theory, as a demonstration of the significance this approach has for evolutionary behavioral endocrinologists.

OT is a mammalian protein hormone. It evolved within a larger family of structurally similar “-tocin” peptides (e.g., vasotocin, mesotocin) that is observed in birds, reptiles, and invertebrates, which likely debuted 600+ million years ago (Gruber, 2014). OT is  well known within medical circles for its functions in parturition and lactation, and it has well-­ established effects on maternal behavior (e.g., Numan, 2017). Since the discovery that OT may also be crucial to the formation of sexual pair-bonds in certain species (e.g., Williams, Insel, Harbaugh, & Carter, 1994) in ways similar to how it affects mother–­ offspring bonds (Nelson & Panksepp, 1998), scientists and laypeople alike have been fascinated by the possibility that this biological messenger is integral to the most meaningful, “close,” social relationships in our lives. Indeed, the hormone has been labeled, variously, as the “love,” “cuddle,” or “trust” hormone (e.g., Uvnäs-Moberg,  2003; Zak, 2012). About 15 years ago, means of experimentally administering OT via nasal spray were developed, and today over 317

500 experimental administration studies of OT can be found in the literature. Contrary to some initial expectations, OT does not uniformly promote ­prosocial behavior and feelings of warmth. Mixed findings have led to a proliferation of attempts to conceptualize OT’s many and contrasting effects within a single, coherent framework (see, e.g., Bethlehem, Baron-Cohen, van Honk, Auyeung, & Bos, 2014; Churchland & Winkielman, 2012; Crespi, 2016; Numan & Young,  2016). New perspectives continue to appear and evolve (e.g., Hurlemann & Scheele, 2016), and no consensually agreed-upon resolution has been reached. We review the major conceptual frameworks for understanding OT, but, more fundamentally, we step back and try to view OT and its manifestations with a wider lens. The OT system is an evolved one. Though this system has a range of psychological effects, it also controls physiological outcomes that are nonpsychological in nature via effects on a wide array of bodily tissues. Psychological and nonpsychological manifestations of the OT system likely have effects coordinated to serve its functions. Hence, a full, coherent explanation of the functions of the OT system should also account for OT’s nonpsychological effects. In early mammalian species, these functions likely regulated particular relationships, notably the mother–offspring one; only more recently did it acquire functions that promote other relationships. A broader, evolutionary view suggests the OT system coordinates a host of effects, which likely evolved initially in the context of certain relationships—this, we feel, guides a coherent interpretation. In addition to ­reviewing the major psychological conceptualizations put forward to date, we develop and emphasize the argument for this framework. Our chapter has four sections. First, we start with some foundational notions—specifically, a broad life history framework for understanding the evolved functions of endocrine hormones. This framework offers reasons that, when seeking to explain the function of any particular hormonal effect, other effects are pertinent. It also integrates phylogenetic perspectives that pertain to how endocrine systems evolve. Second, we give an overview of the literature on OT’s psychological effects. We begin with well-­ established neuromodulatory effects in nonhuman mammals and end with a special focus on humans. Third, we attempt to place OT’s neuromodulatory effects in the broader evolutionary framework emphasizing hormonal modulation of resource ­allocation. That is, we use a life history framework 318 Oxy tocin

to propose an integrative model for understanding how OT functions to affect resource allocations. Our proposals are necessarily preliminary, but, we think, proposals of the kind we offer are needed for further progress toward conceptual integration; our proposals illustrate this broader point. Fourth, we discuss caveats, unanswered questions, and future directions of theory and research.

A Life History Framework for Understanding the Function of Endocrine Hormones Life History Theory

Organisms allocate energy harvested from the environment to fitness-enhancing activities. Energy allocated to one kind of activity is not available for ­allocation to other kinds. Organisms hence inevitably face allocation problems. As some systematic ways of allocating energy promote a particular organism’s ­fitness, given its circumstances, better than others, solutions to these allocation problems evolve through natural selection. In other words, selection sifts through the myriad possible allocation “strategies” that organisms within a species utilize; solutions that make it through selection’s sieve tend to be ones that, relative to others, maximize fitness. Other limited resources such as micronutrient building blocks, time, and neural or other tissue-specific resources also give rise to allocation problems. Life history theory seeks to identify how selection shapes organisms’ solutions to these allocation problems (e.g., Charnov,  1993; Del Giudice, Gangestad, & Kaplan, 2015). How an organism can best use limited resources depends on life circumstances. Optimal allocation solutions hence embody contingencies. For instance, an organism’s relative body size, current state of pathogen load, level of imminent threat from a predator or conspecific, and immediate opportunities to potentially mate may all affect how the organism should best allocate resources. Because energy and other resources are limited, any decision to allocate additional energy toward certain activities (e.g., growth, immune function, defense or flight, mating display or intrasexual competition) inevitably requires, simultaneously, decisions to draw energy away from alternative current or future activities.

Endocrine Hormones Within a Life History Framework

Endocrine hormones are released by a gland (e.g., the gonads, the pituitary) into the circulatory system. They then travel to and bind to receptors located in multiple bodily tissues. Binding initiates chains of

reactions that affect cellular activity, typically in a tissue-specific manner. Changes in cellular activity then produce phenotypic changes. A hormone’s phenotypic manifestations may be numerous and diverse across the many tissues it affects, a phenomenon referred to as hormonal pleiotropy (e.g., Flatt, Tu, & Tatar, 2005). In essence, then, hormones are chemicals that relay messages to multiple cellular “recipients.” This essential character of hormones gives rise to a straightforward conceptualization of what endocrine systems have been shaped by selection to do: These systems were shaped to coordinate simultaneous shifts of energetic and other limited resources from one set of activities to another, contingent on life circumstances (e.g., Ellison, 2017; Finch & Rose,  1995; Ketterson & Nolan, 1999; Lancaster & Sinervo, 2011). That is, endocrine systems were designed to ­mediate allocation decisions. We can systematically identify the components of an endocrine system shaped by selection to achieve optimal allocations. Subject to selection are (1) the mechanisms that dictate the circumstances under which a hormone will be released (and then “shut off ”), (2) the distribution of receptors that receive signals, and (3) how tissues respond to receipt of signals. Two additional concepts flesh out this life history view. The first is phenotypic integration (e.g., Ketterson, Atwell, & McGlothlin, 2009). The multiple phenotypic changes that hormones coordinate have been selected to “work together” to produce benefits. For example, Ketterson and Nolan (1999) found that, during dark-eyed juncos’ mating season, male testosterone increases both attractiveness to females and range size, each of which foster mating success. The second is allocation trade-offs. To “pay for” up-regulated effort toward some ends, organisms must down-regulate effort toward others. Some phenotypic changes induced by hormones, then, are beneficial not because they directly produce benefits, but rather because they pay the costs for other beneficial changes. For instance, once again in dark-eyed juncos, male testosterone leads to decrements in parental efforts (e.g., feeding of offspring), not because there is inherent value in doing so, but rather because other efforts—for example, toward finding and attracting mates—are prioritized (Ketterson & Nolan, 1999).

Neuromodulation as Part of an Adaptive Complex of Modulatory Responses

Receptors for a host of hormones reside within the  brain, such that hormonal “messages” regulate

allocation of neural resources (e.g., allocation of attention to competing stimuli, the appraisal of ­ those stimuli, the potency of particular rewards and punishments). OT, like some other hormones, is projected directly into brain regions, where it acts as a neurotransmitter that up-regulates or down-regulates specific neural networks.

Conditional Responsivity and Internal Regulatory Variables

Once again, allocation solutions should be sensitive to circumstances perceived and interpreted by the organism. The concept of “internal regulatory variables” (Del Giudice et al., 2015; Tooby, Cosmides, Sell, Lieberman, & Sznycer, 2008) reflects appraisals regarding the timing and magnitude of allocation modifications (e.g., of environmental predictability, exogenous mortality risk, the state of social relationships, and potentially many other conditions). Organismal systems are designed to secrete hormones in response to these appraisals, which then coordinate reallocations.

Functional Reverse Engineering of Hormonal Coordination

Hormonal effects are phenotypically integrated to yield particular benefits. Hence, nonpsychological effects of hormones, as well as psychological effects, critically inform an understanding of a hormone’s specific way of achieving functions. That is, both constrain the range of plausible interpretations of how selection shaped the endocrine system to modulate allocations.

Phylogeny and Adaptation

Phylogenetic perspectives complement adaptationist ones. A hormonal system present now also existed, in some form, in deep evolutionary time. But most likely, it did not coordinate the exact same suite of coordinated outcomes, activated by the exact same circumstances. Co-option has occurred when a hormonal system gains a new benefit that did not exist in previous species in a lineage. Co-opted outcomes have been shaped by previous benefits and must serve the new benefit sufficiently well. Secondary adaptation occurs when the new function (or benefit) leads to modification of the system in ways that serve the new function better while maintaining previous functionality (see Gould & Vrba, 1982). OT illustrates these points well. OT-like peptides had functions in ancestors common to vertebrates and some invertebrates (e.g., mollusks, arthropods, annelids; Gruber, 2014), often pertaining to control Grebe and Gangestad

319

of specialized smooth muscle contractions. In an early vertebrate, both OT and vasopressin homologs emerged and evolved to acquire distinct, but sometimes overlapping, functions (e.g., Hoyle, 1999). OT per se debuted in placental mammals ~250 million years ago (Donaldson & Young, 2008; but see also Gwee, Amemiya, Brenner, & Venkatesh, 2008), in which it induces uterine contractions (e.g., Bell, 1909; Dale, 1906), reduces the severity of postpartum hemorrhage (e.g., Elbourne, Prendiville, & Chalmers, 1988; Weitzman, Glatz, & Fisher, 1978), and activates the milk letdown reflex during lactation (e.g., McNeilly, Robinson, Houston, & Howie, 1983; Wakerley & Lincoln, 1973). These events set the stage for an important co-option. In particular, the presence of OT in mothers with newborns, owing to its peripheral reproductive functions facilitating parturition and lactation, may have led it to acquire functions coordinating behavioral aspects of maternal care through OT projections in the central nervous system (Carter,  2014; Crespi, 2016; Feldman, Monakhov, Pratt, & Ebstein, 2016). OT is now well established as a key mediator of the mother–offspring bond forged during parturition (Carter,  2014), lactation (Crowley & Armstrong, 1992), emotionally “warm” touch (Feldman, Gordon, Schneiderman, Weisman, & Zagoory-Sharon, 2010), and responses to infants’ cries (Riem et al., 2011; see also Elmadih et al., 2014). Classic work on rodents demonstrates roles for OT in both establishing and maintaining maternal behavior via particular brain regions (e.g., the nucleus accumbens; e.g., Numan, 2017). OT plays a role in infant responses to mothers too, and the success or failure of this bond’s formation early in an infant’s life may have long-lasting effects on the OT system (e.g., Fries, Ziegler, Kurian, Jacoris, & Pollak, 2005). OT likely regulates responses to partners in interdependent social relationships aside from the mother–infant one (e.g., Carter, 2014; Crespi, 2016; Numan & Young, 2016). These effects reflect additional co-options. Crucially, OT’s effects on social behavior likely stem from the “biological prototype” for mammalian sociality: the mother–infant bond (e.g., Carter, 2014; Numan & Young, 2016). Most notably, OT has been co-opted for pairbonding. Seminal studies focused on voles. Prairie voles form enduring sexual pair-bonds. Montane and meadow voles do not. OT receptors are much denser in members of the monogamous species, in which OT production during mating affects preferences for pair-bond partners (Cho, DeVries, Williams, & Carter,  1999; Insel & Shapiro,  1992; 320 Oxy tocin

Williams et al., 1994; see also Young & Wang, 2004). This work led to more recent work on human pairbonded couples. OT administration leads to more engaged, constructive communication about relationship conflicts (Ditzen et al., 2009); more intense orgasms; and greater contentment with a partner following intercourse (Behnia et al.,  2014). OT levels predict success of emotional support relationship interventions (Holt-Lunstad, Birmingham, & Light, 2008) and overall relationship satisfaction (Holt-Lunstad, Birmingham, & Light,  2015; but see T. W. Smith et al., 2013). New lovers have elevated OT compared to singles, and OT levels at a ­relationship’s outset predict its success half a year later (Schneiderman, Zagoory-Sharon, Leckman, & Feldman, 2012). (For related effects in marmosets, see Seltzer & Ziegler, 2007; T. W. Smith et al., 2010).1 OT perhaps has been co-opted to play roles in yet other interdependent social relationships, such as those between kin, friends, and in-group members (e.g., Crespi, 2016). The OT homologs in birds appear to have acquired some such social functions (Goodson, 2013). And OT may play such roles in some mammals as well (Anacker & Beery,  2013). Although such roles in humans are possible, we offer a cautionary note: As emphasized previously, co-option of OT should not affect social relationships in ways that would markedly compromise its role in the mother–infant relationship. We believe it is more likely that co-option occurred for pairbonding than for other kinds of social relationships. More work is needed along these lines.

Oxytocin’s Psychological Effects and Correlates

A vast literature examining the psychological effects of OT has accumulated over the last 15 years. Although the literature on the associations between levels of OT and behavior or context (outside of maternal behavior) is defined by perhaps a couple dozen studies, over 500 published intranasal administration studies (including some clinical trials) have appeared in scientific journals (see Bos, Panksepp, Bluthé, & van Honk, 2012). The former, observational studies typically speak to the causes or 1  The OT homolog present in passerine birds (mesotocin) may or may not play similar roles in pair-bonding. Although some research suggests it does (Klatt & Goodson, 2013; Lowrey & Tomaszycki, 2014; Pedersen & Tomaszycki, 2012; for effects of vasotocin, see also Baran, Tomaszycki, & Adkins-Regan, 2016), whether mesotocin’s role in either maternal care or pair-bonding is widespread across bird species remains unknown.

concomitants of OT production, whereas the latter, experimental studies speak to effects of increased OT.

The Maternal Brain in Rats

Before we review the literature on human OT, we briefly describe classic work on the neurobiology of maternal behavior in rats, as it offers a foundation for understanding proximate mechanisms of OT’s psychological effects. Whereas nulliparous female rats actively avoid rat pups, new mothers engage in well-established maternal behaviors with them (e.g., nursing, licking, retrieval from the nest), even when their own pups are experimentally removed and replaced with another mother’s pups. Various hormones (e.g., estradiol, progesterone, prolactin) present at birth alter maternal brain structures and lead to these effects (see Numan, 2017). OT too plays a crucial role in the neurobiological alterations behind maternal behavior. OT projections to the medial preoptic area (MPOA) and ventral tegmental area (VTA) modulate neural input to the nucleus accumbens (NAc). When cooccurring with mesolimbic dopaminergic (DA) activity, OT inhibits NAc activity, which has the effect of releasing inhibition of the ventral pallidum (VP). Resultant VP activity promotes maternal behavior (partly via sensory processing of offspring stimuli, such as scent cues; see Numan et al., 2005; Numan & Stolzenberg,  2009; Stolzenberg & Numan, 2011; Numan, 2017; for similar effects in sheep, see Numan & Young,  2016). These neurobiological ­networks fundamentally regulate motivation. OT affects pair-bonding in prairie voles through nearidentical networks (e.g., once again, OT and DA work together to inhibit NAc control of VP; e.g., Aragona, Liu, Curtis, Stephan, & Wang, 2003; Liu & Wang, 2003; see also Numan & Young, 2016). OT not only establishes maternal behavior but  also maintains it. Though maternal behavior ­persists even if OT is withdrawn, variations in OT levels postestablishment affect its quality (e.g., Numan, 2017; for correlational work on humans, see Elmadih et al.,  2014). OT likely also affects, downstream, cortical systems regulating attentional, cognitive, or emotional processes.

She Loves Me, She Loves Me Not: The OT Paradox

One popularized view is that OT is, metaphorically, the neurobiological “cement” that bonds an individual to another, whether it be mother to infant or one pair-bond partner to another. The OT system has been characterized as one that fosters interest in

finding social connection between people—a “calm and connect” system (Uvnäs-Moberg,  2003). As search for connection purportedly promotes prosocial motivation (e.g., trust and trustworthiness; Kosfeld, Heinrichs, Zak, Fischbacher, & Fehr, 2005; Zak, Kurzban, & Matzner, 2005), OT has also been proclaimed to be a “moral molecule” (Zak, 2012). OT may indeed shape a neurobiological system to promote positive valence of an infant or romantic partner (though, as noted earlier, other hormones promote nurturing behavior too; Numan, 2017). The view that OT generically or unconditionally promotes social connection and prosociality, however, is simply no longer viable. A recent meta-analysis of the effects of OT administration on trust in experimental trust games found no robust overall effect (Nave, Camerer, & McCullough, 2015). Hence, while found to increase trust in some experimental games (Kosfeld et al., 2005), OT has been reported to decrease it in others (Bartz et al., 2011). OT administration has sometimes been found to disrupt social connection, for example, by prompting envy and gloating (Shamay-Tsoory et al., 2009), leading men to perceive less warmth in faces (Hoge et al., 2014), promoting self-interested moral judgments in men (Scheele et al., 2014), and derogating out-groups (De Dreu et al., 2011). Relatedly, some ­correlational studies report OT to be positively associated with relationship distress (Taylor, Saphire-Bernstein, & Seeman, 2010) and anxiety about r­omantic relationships (Marazziti et al., 2006; Weisman, ZagoorySharon, Schneiderman, Gordon, & Feldman, 2013). After women were asked to think about relational distress, their OT levels increased (Tabak, McCullough, Szeto, Mendez, & McCabe, 2011). The mixed nature of findings regarding OT’s ­effects on social bonding and connection gives rise to the oxytocin paradox (Bethlehem et al.,  2014). How can a unified, coherent conceptualization of OT explain its contrasting effects?

Toward Resolving the Oxytocin Paradox

A number of views about what OT does, fundamentally, to produce its psychological effects have been put forward. OT is fundamentally anxiolytic. In one view, OT dampens reactivity of the hypothalamic-pituitaryadrenal (HPA) axis to threats and, hence, suppresses cortisol responses to stress (e.g., Windle, Shanks, Lightman, & Ingram,  1997). Relatedly, OT may reduce amygdala activity and, thus, suppress fear responses (e.g., Kirsch et al., 2005). Fundamentally, Grebe and Gangestad

321

in this view, its psychological effects—for example, on interpersonal trust, perceived trustworthiness of others, willingness to cooperate with in-group members, and interest in affiliation with others—all stem from generalized dampening of stress responses (e.g., Churchland & Winkielman, 2012; Neumann & Slattery,  2016). Influenced by OT, people see other people as less threatening (for details, see Churchland & Winkielman, 2012). Although OT does suppress anxiety and fear in some circumstances, other findings challenge this view. First, once again, OT administration sometimes leads to diminished trust (see Bethlehem et al., 2014). Second, OT administration may not always suppress stress responses; in some research paradigms, it has promoted episodic memory for aversive events (Striepens et al., 2012), anxiety responses to unpredictable events (Grillon et al., 2013), emotional intensity in response to conflict with partners in men (Ditzen et al., 2012), and Pavlovian fear conditioning (Eckstein et al.,  2015). To explain mixed findings, Eckstein et al. (2015) propose that OT has targeted anxiolytic effects—specifically, to promote extinction of stress responses (see also Neumann & Slattery, 2016). In absence of conditions conducive to extinction (e.g., when social support is absent), OT may have no effect (e.g., Heinrichs, Baumgartner, Kirschbaum, & Ehlert,  2003) or even potentiate threat responses. OT promotes social salience. Another view is that OT heightens the salience of social cues and information (e.g., Bartz et al., 2011; Striepens et al., 2012), largely via effects on dopaminergic reactivity, especially in mesolimbic regions (see Shamay-Tsoory & Abu-Akel, 2016). In this view, OT’s effect on perceived safety or threat is context dependent. When social threats are minimal, OT administration should bolster a sense of safety. However, in the presence of social threats, OT should augment perception of those threats. Although the social salience hypothesis expects that OT will only selectively promote prosociality, it does not uniformly render correct predictions (see Bethlehem et al., 2014). Hence, for instance, whereas OT has been found to promote in-group favoritism, regardless of how the in-group and out-group are appraised (De Dreu et al., 2011), the social salience hypothesis would seem to predict that group appraisals matter. OT enhances self-referential processing. Yet another view argues that OT enhances emotional interoception—awareness of one’s own subjective emotional 322 Oxy tocin

state (Hurlemann & Scheele, 2016). Like the social salience hypothesis, this view proposes that OT’s  ­effects build off of pre-existing psychological appraisals. The difference is that, in this view, self-­ referential processing, not simply situational construal, is key. Hurlemann and Scheele argue that representations of the self often include close social partners (see Aron & Fraley,  1999). Accordingly, people may be especially sensitive to social information emanating from close social partners, who are seen as part of the “self.” OT modifies reward sensitivity. Neurobiological studies of rats, once again, establish that OT receptors populate dopaminergic mesolimbic structures, ­notably the VTA and NAc, which are known to ­importantly regulate reward sensitivity: detection of rewarding stimuli, tracking and reinforcement of behaviors leading to reinforcing consequences, and attention to discriminative cues upon which reinforcement is contingent. OT, then, may exert its major psychological effects through modification of these reward circuits (see Bethlehem et al.,  2014; Numan & Young, 2016). A cascade of effects. Naturally, OT need not have psychological effects through just one route. Indeed, at a neurobiological level, OT affects, directly and through downstream influences, multiple structures and systems (Numan, 2017). Perhaps the most sensible view, then, is that OT affects motivation and cortisol responsivity and attention (both to external stimuli and interoceptively). Of course, this view, while likely, requires that key questions be answered: Through what processes does it do so? How are these effects functionally coordinated? How does the OT system “work”? Answers to these questions, in our minds, require an understanding of the evolved (and likely current) contexts in which OT exerted its important psychological effects.

The Mother–Infant Prototype Revisited

Once again, OT’s effects on social behavior in mammals likely debuted evolutionarily in the context of maternal–infant interactions. And, although its psychological effects may have been co-opted and modified for other social contexts, modification of the OT system should not seriously disrupt its functionality within the maternal–infant relationship. That is, OT’s psychological effects likely evolved because they promoted maternal care for and protection of an infant, and, in a maternal–infant relationship, they should still do so.

With this assumption in mind, let us reflect on  proposed conceptualizations about how OT ­affects psychological processing. The social salience hypothesis argues that OT renders social cues salient and nonsocial cues less salient. A related view claims that OT potentiates social rewards and de-potentiates nonsocial ones. Would these effects promote maternal care for an infant in the context of a maternal– infant relationship? OT may facilitate proficient maternal care if it were to render salient social cues emitted by the infant or promote the social reward value of cues of infant responsiveness and well-being. Yet generalized salience of social cues or potentiation of social rewards could be expected to degrade the quality of maternal care (e.g., if mothers attended to other social partners rather than their infants). Similarly, one can ask about how OT would promote pair-bonding in these views. Although attention to social cues emitted by a partner or potentiation of rewards in the context of the pair-bond may facilitate bond formation, attention to social cues emitted by others or potentiation of rewards garnered from other social relationships could disrupt pair-bonding. OT’s effects, then, may well be expected to target particular relationships. The salience of social cues or modification of social rewards should be “tagged” to specific partners. Indeed, one might expect attention to social cues emitted by some individuals to be diminished. Consistent with this notion, experimental work on rats examining OT’s effects on maternal behavior finds that maternal motivations are specially affected: It facilitates maternal care (e.g., pup carrying, attentiveness to pups, feeding of pups, maternal protective aggression; see, e.g., Numan & Young, 2016).

Targeting of Oxytocin-Facilitated Behavior

How might targeted social attention and motivation emerge? In a species in which OT has been co-opted to function in multiple social contexts (e.g., pairbonding, as well as maternal care), OT’s effects should not be tagged to specific phenotypic cues of one class of social targets, such as the faces of infants. Such a system would not permit OT’s effects to be tagged to another class of individuals, such as adult pair-bond partners. A more sensible design might tag OT’s effects on social relationships that otherwise exert potent motivational effects within the context in which an OT response occurs. For instance, in the context of maternal care, the potent relationship figure is the infant; in the context of pair-bonding, it is the acquired or potential pairbond partner.

Two important implications follow. First, the specific effects of OT will depend on the exact nature of the circumstances in which the OT response occurs (e.g., differ across mother–infant and pairbond relationships). Second, we cannot fully understand how OT functions without knowing the contexts in which OT increases are naturally experienced. The second implication derives from the first: If OT’s specific effects are contingent on the context in which an OT response occurs, then OT’s functionality is tied to the contexts in which the OT system is designed to activate.2

Natural Circumstances Producing Oxytocin Responses in Romantic Relationships

What, then, are the natural circumstances in which an OT response is produced? For instance, what ­circumstances in romantic relationships lead to OT responses? Once again, multiple proposals exist. The calm-and-connect model (Uvnäs-Moberg, 2003) expects positive associations between OT levels and romantic relationship quality. Some research finds that relationship bonding (expressed by touch or social cues) leads to OT production, which may reinforce nurturing behaviors and thereby facilitate connection between partners (e.g., Grewen, Girdler, Amico, & Light, 2005). By contrast, the tend-andbefriend model (see also Mogilski et al., this volume) proposes that distress or anxiety within relationships leads to OT release, which in turn increases “appetite” for social affiliation outside of the distressful bond (Taylor,  2006). Some studies report associations between OT and relationship distress (e.g., Taylor, 2006; Taylor et al., 2010; see also Tabak et al., 2011). (See also T. W. Smith et al., 2013, who found no support for one model over the other.) 2   Methodologically, this means that the functionality of the OT system cannot be understood through experimental adminis­ tration studies alone. In administration experiments, OT’s effects are examined within contexts chosen by the experimenters, not contexts in which OT secretions and projections naturally occur. Furthermore, OT administration outside the presence of a natural circumstance that would produce OT secretion may evoke a response, but one that bears little similarity to the response produced in any natural circumstance leading to OT secretion. Other methodological limitations of OT administration studies have recently received attention. The typical administration study is woefully underpowered to detect small to moderate effects (power ~15 percent; Walum, Waldman, & Young, 2016). Moreover, intranasal administration results in massive increases in peripheral levels, but only a small amount is absorbed centrally, such that some effects may be downstream behavioral outcomes arising from peripheral effects (e.g., on heart rate; Leng & Ludwig, 2016).

Grebe and Gangestad

323

In collaboration with others, we propose a third model, the identify-and-invest model (Grebe et al., 2017). It follows from the notions we laid out previously. In this view, events that prompt motivation to attend and respond to a relationship partner lead to OT release—most notably, cues that a valued ­relationship is threatened—which then functions to reorient psychological resources toward the relationship. Romantic relationship threat may arise when individuals themselves are highly invested in those relationships yet their partners are less invested or attentive to them. Grebe et al. (2017) asked romantically involved women and men to think about ways their relationship partners were responsive to them or not and measured the rise or fall of OT as a function of this task. We also administered a battery of measures of relationship involvement to both members of the couple. We regressed individuals’ OT response to the task on both self- and partner reports of relationship involvement. Analyses revealed positive associations between self-reports of relationship involvement and OT, yet negative associations between an individual’s OT and his or her partner’s reports of investment. Consistent with the identify-and-invest model, peripheral OT release was predicted by a discrepancy between an individual’s own and his or her partner’s relationship involvement. Discrepancy prompts motivation to attend to the relationship, which OT purportedly functions to do. In our view, the identify-and-invest model is compatible with the idea that OT-regulated maternal care constitutes the foundation from which the OT system was co-opted to function to regulate other relationships, such as pair-bonds. OT should function to lead mothers to attend to their infants’ needs and protect them in the face of threats. It may make sense, then, that OT’s effects co-opted to pair-bonding should lead to attention to the threatened relationship (and not the desire to forge new relationships, as the tend-and-befriend model proposes).

Nonprosocial and Aggressive Responses

The perspective that OT’s prosocial effects should be targeted to specific relationships also facilitates an understanding of OT’s nonprosocial and aggressive effects. If and when third parties represent threats to infant well-being, adaptive prosocial ­responses with regard to the infant may constitute defensive or aggressive responses to third parties. Indeed, in some nonhuman species, OT is widely recognized to promote maternal aggression to protect 324 Oxy tocin

infants against threats (e.g., predators, threatening conspecifics; e.g., Bosch, Meddle, Beiderbeck, Douglas, & Neumann, 2005). OT may promote maternal aggression through its anxiolytic effects (specifically, via inhibition of the effect of corticotropin-releasing factor [CRF] in the amygdala on generalized fear responses; Numan, 2017). As defense should be targeted to threats to an infant, however, it seems likely that OT’s effects on maternal aggression are also mediated through motivation to protect the infant and, hence, motivated attentional monitoring of threats to the infant. This line of thinking suggests OT may similarly motivate protective responses (and aggressive responses to threatening outside parties) in the context of pairbonding, fostered by attentional monitoring to relationship threats. The identify-and-invest model fits well with the view that one major function of OT within close relationships is the protection against threats. Circumstances that threaten a relationship are precisely those in which attention to those threats and action to defend against them are called for.3 Some scholars propose that OT has been co-opted in humans to regulate social relationships. For instance, OT may facilitate prosocial behavior toward in-group members and defensive behavior toward out-group members (De Dreu et al.,  2010, 2011). This view similarly interprets OT’s antisocial behavioral effects toward specific parties in terms of its protective or prosocial effects toward other parties. When individuals were presented with mixed praise and criticism from others, Gao et al. (2016) found that OT administration enhanced salience of praise in women, but fostered salience of criticism in men. They propose that the sex difference may be rooted in different parental roles that women and men played ancestrally and, hence, stem from basic evolved functions of OT. Whereas women’s parental duties may have more centrally involved direct caregiving, men may have played a more protective role, guarding against potential harm and threat. (For similar sex differences, see Hoge et al., 2014; Scheele et al., 2014.) 3   Earlier, we emphasized a limitation of OT administration studies: OT may be administered in contexts in which OT is never naturally projected centrally (at levels attained). Though OT may have effects in such contexts, their effects may not be  understandable in terms of adaptive response to particular circumstances. The point may be illustrated by antisocial effects: These effects may actually be prosocial or protective with regard to other relationships, but if OT administration occurs in the absence of such a relationship, the study does not readily lend itself to such an interpretation.

Toward a Life History Model of the Oxytocin System: Integrating Psychological and Physiological Effects

As laid out earlier, within a life history perspective, hormonal systems have fundamentally been shaped by selection to carry out resource allocation decisions; when activated, they lead energy and other resources to be directed toward certain fitness-enhancing ­activities, while shunting them away from other ­activities. The many simultaneous effects are phenotypically integrated, working together to promote a benefit or set of benefits in certain conditions— those in which the adjustments paid off ancestrally when the system was shaped. As we furthermore discussed, hormonal systems often adjust both psychological and nonpsychological physiological ­ phenotypes. Even when one is interested in understanding the function of psychological outcomes, nonpsychological effects can inform an understanding of a hormonal system’s design. Having discussed OT’s psychological effects, we now turn to physiological effects of peripherally circulating OT.

Physiological Effects of Oxytocin

Energy intake and expenditure. OT reduces energy intake and, possibly, increases expenditure. OT administration to rodents, centrally or peripherally, decreases food intake (see Blevins & Baskin, 2015, for an extensive review). The inactivation of OT, either via antagonist administration or receptor blocking, promotes caloric consumption (e.g., Arletti, Benelli, & Bertolini, 1989; Kublaoui, Gemelli, Tolson, Wang, & Zinn, 2008; G. Zhang & Cai, 2011). The neural mediators implicate many of the same brain structures that mediate outcomes on maternal motivation produced by OT: Hypothalamic projections affect the VTA and subsequently, with dopamine, the NAc (e.g., Lawson et al., 2015; Sabatier, Leng, & Menzies, 2013). Increases in central OT selectively affect consumption of carbohydrate-rich foods; they do not affect appetite for fat-rich foods (e.g., Herisson, Brooks, Waas, Levine, & Olszewski, 2014). In male and nonlactating female rats, circulating insulin and glucose activate this system, resulting in diminished rates of feeding (Sladek, Stevens, Song, Johnson, & MacLean, 2016). OT sensitivity to insulin and glucose is blunted in pregnant and lactating female rats, who maintain eating behavior (functionally, to nourish offspring; Sladek et al.,  2016; Olszewski et al., 2016). Researchers have proposed that OT administration may promote weight loss in humans (H. Zhang

et al., 2013; see also Blevins et al., 2015). Indeed, a single-dose intranasal OT administration decreased men’s caloric intake during a meal by an average of 122 kcal (Lawson et al., 2015). Another study found effects specific to consumption of carbohydrate-rich foods (Ott et al., 2013). At the same time, OT may increase energy expenditure. OT knockout mice develop obesity, even when nutrient intake is constant (Takayanagi et al.,  2008; see also Deblon et al., 2011). Cardiac muscle cells in particular utilize more glucose-fueled energy in response to OT exposure (Gutkowska & Jankowski, 2012). OT administration to rhesus monkeys increases resting energy expenditure too (Blevins et al.,  2015). Little research has examined metabolic changes as a function of OT administration in humans. H. Zhang et al. (2013) reported no evidence for changes in resting energy expenditure, though more research is needed. OT appears to stimulate lipolysis specifically. Rats centrally infused with OT, compared to controls, have greater expression of several enzymes involved in fat metabolism, produce more oleoylethanolamide (a lipid that itself increases fatty acid oxidization), and have lower respiratory exchange ratios, indicating a shift toward fat—rather than carbohydrate—metabolism (Deblon et al.,  2011). OT might act directly upon adipocytes, which contain OT receptors (Yi et al., 2015; also reviewed in Blevins & Baskin, 2015; Chaves, Tilelli, Brito, & Brito, 2013). Taken together, evidence clearly indicates that, outside of pregnancy and lactation, OT negatively affects energy balance. In this sense, then, OT is similar to cortisol, though the sources of mobilized energy differ (lipolysis of triglycerides stored in fat cells vs. gluconeogenesis of energy stored in the liver). In any event, OT releases energy to be used toward some end. Immunological modulation. In animal models, OT administration reduces inflammation in peripheral tissues (e.g., Nation et al., 2010; Szeto et al., 2013), and it has similar effects on human cells in vitro (Szeto et al., 2008). Yet OT may modulate immune function rather than suppress it (see review in Li et al., 2017; Wang et al., 2015). OT may facilitate wound healing (Detillion, Craft, Glasper, Prendergast, & DeVries,  2004), which in turn results primarily from repair mechanisms operating in the absence of inflammation (Guo & DiPietro, 2010). Immuno­ logists contrast “­resistance” functions, such as inflammation, which operate to eradicate pathogens, with “tolerance” functions, such as repair, which Grebe and Gangestad

325

clean up harm done by pathogens (Schneider & Ayres, 2008). Resistance is immediately more energetically costly than tolerance (and hence the former may be accompanied by “sickness behavior” that energetically “pays for” increased allocation to immune function; Shattuck & Muehlenbein, 2015). OT may shift effort away from resistance and toward tolerance and repair, perhaps to limit immediate allocation of energy toward immune function. Indeed, OT reduces sickness behavior in rats injected with lipopolysaccharide (LPS; Reyes-Lagos et al., 2016). Such modulation could be adaptive if OT release occurred in circumstances that demanded energy allocation to other activities. These effects parallel effects of cortisol on costly inflammatory responses; cortisol similarly shifts immune function from resistance functions to tolerance functions (e.g., Garbers et al., 2012), perhaps because it often functions to facilitate adaptive responses to external threats. Cardiac function. OT receptors richly populate heart tissue. In research on injured and healthy nonhuman animals, OT perfusions into the heart slow heart rate, decrease the force of cardiac myocyte contraction, and suppress blood pressure (Gutkowska & Jankowski, 2012; Hicks et al., 2014). One interpretation is that OT protects and maintains cardiac function through these actions (e.g., by reducing oxidative stress in the heart; Gutkowska & Jankowski,  2012). At the same time, under some conditions, OT administration in lab animals may increase heart rate and blood pressure (reviewed in Blevins & Baskin, 2015). An alternative conceptualization emerges from recent studies. Gamer and Büchel (2012) administered OT to men who were presented with neutral, positively valenced (happy), or negatively valenced (fearful) faces and asked to perform an emotional classification task. The effect of OT depended on the emotional valence of stimuli; it increased heart rate especially when men were confronted with fearful faces. Gamer and Büchel (2012) propose that OT potentiates the significance of stimuli that evoke approach or avoidance responses. OT administration enhances attention to others’ eye region (e.g., Gamer, Zurowski, & Büchel,  2010) and performance in recognizing emotional expression in others, especially avoidancerelated (Feeser et al.,  2014) or negative emotions (e.g., Striepens et al., 2012; see also Schulze et al., 2011). When OT leads to reductions in heart rate, it may do so through increased parasympathetic control rather than reduced sympathetic control (e.g., Tattersall & Hockey, 1995; see also Norman 326 Oxy tocin

et al.,  2011; Prehn et al.,  2013); the former may function to enhance vigilant attention (Kassam, Koslov, & Mendes, 2009). In fact, the findings that OT perfusions reduce heart rate, strength of contraction, and blood pressure in rodents and dogs are mediated, at least in part, by atrial natriuretic peptide (ANP; e.g., Gutkowska et al., 1997)—which, when released in heart muscle, sensitizes it to parasympathetic control (Atchison & Ackermann,  1990). All in all, OT’s ­effects on heart rate may reflect its role in active, motivated information processing and vigilance for indications of threat, especially in the social environment, an interpretation that contrasts with it reflecting calmness and a sense of safety. This role may be especially important if the natural circumstances that lead to peripheral release of OT call for active, motivated vigilance (e.g., for emotionally significant events). In particular, OT-induced parasympathetic control of heart rate may permit rapid modulation of heart rate in response to stimuli pertinent to threats. As we have already argued, the OT system may respond to potential threats within a relationship context, thereby attuning other systems to detect these potential threats, as well as prepare systems to adaptively respond to threats that are detected by making energy available for utilization, suppressing costly immune function, and rendering the cardiac system ready to respond quickly to emergent events that demand attention and, potentially, action. Insulin production and sensitivity. In rodents and humans, OT increases the rate of insulin secretion in response to glucose (e.g., Björkstrand, Eriksson, & Uvnäs-Moberg,  1996; Chiodera et al.,  1984; Klement et al., 2017) and boosts insulin sensitivity, thereby enhancing glucose uptake by skeletal and heart muscle (e.g., Camerino, 2009; Deblon et al., 2011; Florian, Jankowski, & Gutkowska, 2010; Lee et al., 2008; H. Zhang et al., 2013). Cortisol has opposite effects in this regard; it induces insulin resistance. Arguably, it does so to maintain levels of circulating glucose needed to fuel the brain during fight or flight; unlike muscles, the brain’s uptake of glucose is not insulin dependent (e.g., Ellison,  2017). OT may respond to threats that do not immediately require energetic exertion. Possibly, then, it prepares the body to be able to have energy available for action, should the need arise. HPA axis responsivity. We noted that OT may suppress HPA responses to threats and other stressors

and resulting increases in plasma cortisol levels, mediated by blunted CRF-stimulated release by adrenocorticotropic hormone (ACTH) into circulation (e.g., A. S. Smith et al., 2016). Yet OT’s effect on the HPA axis appears to be context specific (Yee et al., 2016). In prairie voles pretreated with OT, corticosterone response to a stressor (walking in shallow water) was not dampened by OT. Rather, OT pretreatment decoupled hypothalamic paraventricular nucleus activity to the stressor from the resulting rise in corticosterone levels and coupled it to autonomic nervous system responses. There, OT permits a glucocorticoid and autonomic response to a stressor when an energetically expensive response (fight or flight) is called for. OT, then, is not simply a “de-stressor.” It may promote vigilance to possible threats and, should an external threat emerge, permit a robust HPA response to it. That might well be an attunement that would promote, for instance, maternal protection against intruders (Bosch et al., 2005).4 The OT system conceptualized as an alternative threatsensitive system. Based on OT’s psychological effects, the circumstances in which OT is released peripherally, and OT’s physiological effects, we have provisionally proposed a conceptualization of how the OT system has been shaped by selection to carry out resource allocation decisions (Grebe & Gangestad, unpublished). Specifically, we propose that the OT system is responsive to particular kinds of threats or potential threats, and is designed to (1)  monitor those threats and (2) respond to them. Hence, we argue, the OT system is itself a threat-sensitive system, much like the cortisol system is a threatsensitive system. Yet the OT system is distinct from 4   In addition to these major physiological effects, OT has several others, which we mention briefly. Most notably, it has effects consistent with its fundamental role as a smooth muscle contractor (Altura & Altura,  1977)—during labor, lactation, and orgasm/ejaculation, OT binds to receptors on the respective muscle fibers, inducing calcium ion mobilization and subsequent contraction (Tahara et al.,  2000; see also Conklin, Smith, & Olson, 1999). These effects, however, need not be coordinated with the physiological effects we describe. OT’s well-established role in parturition (indeed, its name derives from Greek roots for “quick birth”) results partly from a rapid and transitory proliferation of OT receptors in uterine tissue, where OT is locally synthesized (at least in primates; Gimpl & Fahrenholz, 2001). OT is produced in male testes too, where it is involved in sperm transport and androgen synthesis (Gimpl & Fahrenholz, 2001). Naturally, OT produces milk letdown through smooth muscle control. As already noted, some effects of OT (e.g., decrease in eating) are blocked during lactation and, hence, are decoupled from other coordinated outcomes.

the cortisol system, as the kinds of threats these ­systems respond to are different. The OT system responds to potential threats to a valued relationship or relationship partner. We argued this claim, captured by the identify-and-invest model, earlier. The original context in which the OT system evolved to affect behavior is the mother–infant relationship. From the mother’s perspective, the infant is a valued relationship partner. The infant is, by its nature, vulnerable to threats—threats that stem from outside agents (e.g., predators, conspecifics), but also threats owing to the fact that the infant’s well-being and viability depend on maternal care and responsiveness. In this context, the OT system up-regulates (1) vigilance for signs of potential threats to this relationship and relationship partner (e.g., outside agents, infant need states), and attention to those threats, and (2) motivations to respond to these threats. This system was co-opted to operate in the context of pair-bonding. In that context, the OT system operates to protect a valued relationship with a pair-bond partner and, accordingly, is responsive to conditions that threaten the relationship (e.g., when a highly valued partner’s relationship involvement is low or uncertain; Grebe et al., 2017). The psychological consequences of OT under naturally occurring conditions are multiple and depend on the precise context eliciting an OT response. OT likely affects motivational states through neurobiological networks that it fundamentally affects. At the same time, it likely also influences vigilance and attention downstream, sensitizing individuals to evidence regarding the presence or absence of potential threats. Additionally, it may promote perceived integration of an important other into a sense of self. Yet, we argue, these effects should be defined by the context in which an OT response occurs: Mothers should attend to information pertinent to the safety of their infants, and threatened relationship partners should attend to that relationship. Figure  18.1 depicts the routes through which OT may affect psychological processes, as provisionally proposed within our model. Physiological effects are key components of the OT system for modulating allocation trade-offs. Physiological effects of OT importantly inform our proposal that the OT system is a threat-sensitive system. OT mobilizes energy for utilization. It reduces allocation of energy to energetically expensive ­ immune function. It promotes vigilance to threatening stimuli while readying the cardiac, insulin, Grebe and Gangestad

327

key brain regions

mPOA

VTA

NAc

VP Threatened, valued relationships

OT

Motivation Competing motivations (e.g., eating, establishing new relationships)

Attentional/ cognitive attunement to threats

HPA modulation Parasympathetic cardiac control Insulin secretion and sensitivity

Lipolysis

Immunomodulation (tolerance)

Physiology of threat perception

other physiological effects

Figure 18.1  A schematic representation of oxytocin’s (OT’s) functions, with an emphasis on motivational circuits. OT is conceptualized as a system that responds to actual or potential threats to certain classes of relationships via its proximate effects on motivational circuits. These motivational circuits are roughly divided into (1) key regions originally identified as part of the “maternal brain” (see Numan, 2017) and (2) pathways that regulate the physiology of threat perception. Both of these circuits, initiated by OT activity, are proposed to (1) attune motivation toward investment in threatened, valued social bonds and (2) increase vigilance to certain present or future threats. The OT system has other physiological manifestations that may function in part to “free up” resources allocated to the relationship demands responsible for engagement of the OT system. mPOA, medial preoptic area; VTA, ventral tegmental area; NAc, nucleus accumbens; VP, ventral pallidum.

and glucocorticoid systems for rapid responses to demands for action. To be clear, in suggesting that the OT system is an “alternative threat-sensitive” system, we do not mean to suggest that, similar to an HPA reaction to a threat, an OT response often leads to a state of active readiness to act, alertness, and subjective feelings of tenseness and anxiety. To the contrary, consistent with others’ suggestions, OT may often dampen subjective feelings of anxiety or tenseness. Our proposal that the OT system is a threat-sensitive system is a functional design argument: We argue that the system is designed to support adaptive responses to potential threats of particular kinds (e.g., potential threats to the viability of a vulnerable infant) at the cost of reduced capacity to deal with other problems (e.g., reduced inflammatory responses). 328 Oxy tocin

Other Outstanding Issues Additional Roles

The OT system may well have specialized design features to deal with particular kinds of problems, while still playing other roles. The HPA system is a threat-sensitive system, but more functions as a metabolic hormone (e.g., Ellison, 2017). Similarly, the OT system may regulate basic metabolic function (de Jong et al., 2015; Hew-Butler, Noakes, Soldin, & Verbalis, 2008; Onaka, Takayanagi, & Yoshida, 2012; see also Grebe & Gangestad, unpublished).

The Range of Relationships to Which the Oxytocin System Has Been Co-opted to Respond

In mammals and perhaps birds, OT has been adapted to function within the context of parent–offspring

and pair-bond relationships. But some scholars have argued that it has been co-opted and secondarily modified to function adaptively in many other relationship contexts as well: with friends, kin, in-group members, out-group members, and strangers. Some evidence is consistent with these proposals. Hence, for instance, Feldman et al. (2010) found that young children’s reciprocity with friends positively covaried with their OT levels, themselves predicted by maternal hormone levels and behavior. At the same time, in our view there currently exists little evidence for (or, for that matter, against) OT playing a critical role in these relationships. Much more research is needed to ­establish its precise role and range of involvement.

Interactions With Other Hormones

OT has well-known interactions with other hormones. In rats, both estradiol and progesterone cause a proliferation of OT receptors in certain brain ­regions (e.g., hypothalamus) and thereby enhance the effect of infused OT on sexual receptivity (McCarthy, 1995; Schumacher, Coirini, Frankfurt, & McEwen, 1989; Schumacher, Coirini, Pfaff, & McEwen, 1990). Griffin and Flanagan-Cato (2011) note that estradiol and progesterone have different structural effects on the ventromedial nucleus in rats, such that the latter should especially enhance OT-primed sexual receptivity. Testosterone has been claimed to have effects that oppose those of OT (e.g., Crespi,  2016), but some evidence showing positive covariation between these hormones in men might lead one to wonder about this conceptualization (Jaeggi, Trumble, Kaplan, & Gurven, 2015). Interactions between testosterone and OT are also of interest. Finally, OT may interact with opioids to affect social outcomes (see review in Gangestad & Grebe, 2017).

Variability in Responsivity of the Oxytocin System

Not all mothers demonstrate the same responsivity of the OT system. For instance, maternal depression predicts lower responsivity of the OT system both prenatally and during lactation (for a review, see Moura, Canavarro, & Figueiredo-Bragas, 2016). Mothers who have fewer resources, both physical and social in nature, may evidence lower OT responsivity to their infants. Although the literature to date proposes that the OT system of depressed mothers exhibits “dysfunction” (Moura et al., 2016, p. 561), from a life history perspective this variation may reflect adaptation, with maternal OT

regulating a trade-off between dedicated attention to an infant and attention to competing demands. Ancestrally, mothers with greater resources and social support may have benefited from (or, in effect, been able to afford) greater attention to infants. The OT system may have accordingly evolved to be sensitive to information pertaining to resources and social support.

Conclusion

OT is now a hot topic within social endocrinology and psychology, more generally. Hundreds of published empirical studies have examined its effects. At the same time, consensus on an overarching perspective—how empirical findings should be organized under an umbrella of how the OT ­ system fundamentally works—remains elusive. In this chapter, we have reviewed the primary proposals that have been put forward. We have also argued that a perspective rooted in evolutionary biology, in which OT is understood as a mediator of life history trade-offs, offers one potentially fruitful way forward. We offer one possible interpretation along these lines and discuss evidence consistent with it. Nonetheless, key questions remain unanswered. Perhaps most foundationally, a life history perspective’s fundamental strength is that it seeks to understand any particular hormonal system in ways guided by what is known about the evolutionary forces that give rise to and shape the design of hormonal systems. Ultimately, an adequate social endocrinological understanding of OT will be an evolutionary social endocrinological understanding.

References

Altura, B. M., & Altura, B. T. (1977). Vascular smooth muscle and neurohypophyseal hormones. Federation Proceedings, 36(6), 1853–1860. Anacker, M. J., & Beery, A. K. (2013). Life in groups: The roles of oxytocin in mammalian sociality. Frontiers in Behavioral Neuroscience, 7, 185. Aragona, B.  J., Liu, Y., Curtis, J.  T., Stephan, F.  K., & Wang, Z. J. (2003). A critical role for nucleus accumbens dopamine in partner-preference formation in male prairie voles. Neuroscience, 23, 3483–3490. Arletti, R., Benelli, A., & Bertolini, A. (1989). Influence of oxytocin on feeding behavior in the rat. Peptides, 10(1), 89–93. Aron, A., & Fraley, B. (1999). Relationship closeness as including other in the self: Cognitive underpinnings and measures. Social Cognition, 17(2), 140–160. Atchison, D.  J., & Ackermann, U. (1990). Influence of atrial natriuretic factor on autonomic control of heart rate. American Journal of Physiology, 258, R718–R723. Baran, N. M., Tomaszycki, M. L., & Adkins-Regan, E. (2016). Early life manipulations of the nonapeptide system alter pair

Grebe and Gangestad

329

maintenance behaviors and neural activity in adult male zebra finches. Frontiers in Behavioral Neuroscience, 10, 58. Bartz, J., Simeon, D., Hamilton, H., Kim, S., Crystal, S., Braun, A., . . . Hollander, E. (2011). Oxytocin can hinder trust and cooperation in borderline personality disorder. Social Cognitive and Affective Neuroscience, 6(5), 556–563. Behnia, B., Heinrichs, M., Bergmann, W., Jung, S., Germann, J., Schedlowski, M., . . . Kruger, T.  H. (2014). Differential effects of intranasal oxytocin on sexual experiences and partner interactions in couples. Hormones and Behavior, 65(3), 308–318. Bell, W. B. (1909). The pituitary body and the therapeutic value of the infundibular extract in shock, uterine atony, and intestinal paresis. British Medical Journal, 2(2553), 1609. Bethlehem, R. A., Baron-Cohen, S., van Honk, J., Auyeung, B., & Bos, P.  A. (2014). The oxytocin paradox. Frontiers in Behavioral Neuroscience, 8, 48. Björkstrand, E., Eriksson, M., & Uvnäs-Moberg, K. (1996). Evidence of a peripheral and a central effect of oxytocin on pancreatic hormone release in rats. Neuroendocrinology, 63, 377–383. Blevins, J.  E., & Baskin, D.  G. (2015). Translational and therapeutic potential of oxytocin as an anti-obesity strategy: insights from rodents, nonhuman primates and humans. Physiology & Behavior, 152, 438–449. Blevins, J.  E., Graham, J.  L., Morton, G.  J., Bales, K.  L., Schwartz, M.  W., Baskin, D.  G., & Havel, P.  J. (2015). Chronic oxytocin administration inhibits food intake, increases energy expenditure, and produces weight loss in fructose-fed obese rhesus monkeys. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, 308(5), R431–R438. Bos, P. A., Panksepp, J., Bluthé, R. M., & van Honk, J. (2012). Acute effects of steroid hormones and neuropeptides on human social–emotional behavior: A review of single administration studies. Frontiers in Neuroendocrinology, 33(1), 17–35. Bosch, O. J., Meddle, S. L., Beiderbeck, D. I., Douglas, A. J., & Neumann, I.  D. (2005). Brain oxytocin correlates with maternal aggression: Link to anxiety. Journal of Neuroscience, 25(29), 6807–6815. Bribiescas, R. G. (2001). Reproductive ecology and life history of the human male. American Journal of Physical Anthropology, 116(S33), 148–176. Camerino, C. (2009). Low sympathetic tone and obese phenotype in oxytocin-deficient mice. Obesity, 17, 980–984. Carter, C.  S. (2014). Oxytocin pathways and the evolution of human behavior. Annual Review of Psychology, 65, 17–39. Charnov, E. L. (1993). Life history invariants: Some explorations of symmetry in evolutionary ecology (Vol. 6). Oxford, UK: Oxford University Press. Chaves, V. E., Tilelli, C. Q., Brito, N. A., & Brito, M. N. (2013). Role of oxytocin in energy metabolism. Peptides, 45, 9–14. Chiodera, P., Coiro, V., Camellini, L., Rossi, G., Pignatti, D., Volpi, R., & Roti, E. (1984). Effect of pharmacological doses of oxytocin on insulin response to glucose in normal man. Hormone Research in Paediatrics, 20(2), 150–154. Cho, M.  M., DeVries, A.  C., Williams, J.  R., & Carter, C.  S. (1999). The effects of oxytocin and vasopressin on partner preferences in male and female prairie voles (Microtus ochrogaster). Behavioral Neuroscience, 113(5), 1071–1079. Churchland, P. S., & Winkielman, P. (2012). Modulating social behavior with oxytocin: How does it work? What does it mean? Hormones and Behavior, 61(3), 392–399.

330 Oxy tocin

Conklin, D. J., Smith, M. P., & Olson, K. R. (1999). Pharma­ cological characterization of arginine vasotocin vascular smooth muscle receptors in the trout (Oncorhynchus mykiss) in vitro. General and Comparative Endocrinology, 114(1), 36–46. Crespi, B.  J. (2016). Oxytocin, testosterone, and human social cognition. Biological Reviews, 91(2), 390–408. Crowley, W.  R., & Armstrong, W.  E. (1992). Neurochemical regulation of oxytocin secretion in lactation. Endocrine Reviews, 13(1), 33–65. Dale, H.  H. (1906). On some physiological actions of ergot. Journal of Physiology, 34(3), 163–206. De Dreu, C. K., Greer, L. L., Handgraaf, M. J., Shalvi, S., Van Kleef, G.  A., Baas, M., . . . Feith, S.  W. (2010). The neuro­ peptide oxytocin regulates parochial altruism in intergroup conflict among humans. Science, 328(5984), 1408–1411. De Dreu, C.  K., Greer, L.  L., Van Kleef, G.  A., Shalvi, S., & Handgraaf, M.  J. (2011). Oxytocin promotes human ethnocentrism. Proceedings of the National Academy of Sciences of the United States of America, 108(4), 1262–1266. de Jong, T. R., Menon, R., Bludau, A., Grund, T., Biermeier, V., Klampfl, S. M., . . . Neumann, I. D. (2015). Salivary oxytocin concentrations in response to running, sexual selfstimulation, breastfeeding and the TSST: The Regensburg Oxytocin Challenge (ROC) study. Psychoneuroendocrinology, 62, 381–388. Deblon, N., Veyrat-Durebex, C., Bourgoin, L., Caillon, A., Bussier Peteut, A. L., Petrosino, S., . . . Rohner-Jeanrenaud, F. (2011). Mechanisms of the anti-obesity effects of oxytocin in diet-induced obese rats. PLoS One, 6(9), e25565. Del Giudice, M., Gangestad, S. W., & Kaplan, H. S. (2015). Life history theory and evolutionary psychology. In D. M. Buss (Ed.), The handbook of evolutionary psychology (pp. 88–114). Hoboken, NJ: John Wiley & Sons. Detillion, C. E., Craft, T. K., Glasper, E. R., Prendergast, B. J., & DeVries, A.  C. (2004). Social facilitation of wound healing. Psychoneuroendocrinology, 29(8), 1004–1011. Ditzen, B., Nater, U. M., Schaer, M., La Marca, R., Bodenmann, G., Ehlert, U., & Heinrichs, M. (2012). Sex-specific effects of intranasal oxytocin on autonomic nervous system and emotional responses to couple conflict. Social Cognitive and Affective Neuroscience, 8(8), 897–902. Ditzen, B., Schaer, M., Gabriel, B., Bodenmann, G., Ehlert, U., & Heinrichs, M. (2009). Intranasal oxytocin increases positive communication and reduces cortisol levels during couple conflict. Biological Psychiatry, 65(9), 728–731. Donaldson, Z. R., & Young, L. J. (2008). Oxytocin, vasopressin, and the neurogenetics of sociality. Science, 322(5903), 900–904. Eckstein, M., Scheele, D., Patin, A., Preckel, K., Becker, B., Walter, A., . . . Hurlemann, R. (2015). Oxytocin facilitates Pavlovian fear learning in males. Neuropsychopharmacology, 41(4), 932–939. Elbourne, D., Prendiville, W., & Chalmers, I. (1988). Choice of oxytocic preparation for routine use in the management of the third stage of labour: An overview of the evidence from controlled trials. British Journal of Obstetrics and Gynaecology, 95(1), 17–30. Ellison, P. T. (2003). Energetics and reproductive effort. American Journal of Human Biology, 15(3), 342–351. Ellison, P. T. (2017). Endocrinology, energetics, and human life his­ tory: A synthetic model. Hormones and Behavior, 91, 97–106. Elmadih, A., Wan, M. W., Numan, M., Elliott, R., Downey, D., & Abel, K. M. (2014). Does oxytocin modulate variation in

maternal caregiving in healthy new mothers? Brain Research, 1580, 143–150. Feeser, M., Fan, Y., Weigard, A., Hahn, A., Gärtner, M., Aust, S., . . . Grimm, S. (2014). The beneficial effect of oxytocin on avoidance-related facial emotion recognition depends on early life stress experience. Psychopharmacology, 231(24), 4735–4744. Feldman, R. (2012). Oxytocin and social affiliation in humans. Hormones and Behavior, 61(3), 380–391. Feldman, R., Gordon, I., Schneiderman, I., Weisman, O., & Zagoory-Sharon, O. (2010). Natural variations in maternal and paternal care are associated with systematic changes in oxytocin following parent–infant contact. Psychoneuroendocrinology, 35(8), 1133–1141. Feldman, R., Monakhov, M., Pratt, M., & Ebstein, R. P. (2016). Oxytocin pathway genes: Evolutionary ancient system impact­ ing on human affiliation, sociality, and psychopathology. Biological Psychiatry, 79(3), 174–184. Finch, C.  E., & Rose, M.  R. (1995). Hormones and the physiological architecture of life history evolution. Quarterly Review of Biology, 70(1), 1–52. Flatt, T., Tu, M. P., & Tatar, M. (2005). Hormonal pleiotropy and the juvenile hormone regulation of Drosophila development and life history. Bioessays, 27(10), 999–1010. Florian, M., Jankowski, M., & Gutkowska, J. (2010). Oxytocin increases glucose uptake in neonatal rat cardiomyocytes. Endocrinology, 151(2), 482–491. Fries, A. B. W., Ziegler, T. E., Kurian, J. R., Jacoris, S., & Pollak, S. D. (2005). Early experience in humans is associated with changes in neuropeptides critical for regulating social behavior. Proceedings of the National Academy of Sciences of the United States of America, 102(47), 17237–17240. Gamer, M., & Büchel, C. (2012). Oxytocin specifically enhances valence-dependent parasympathetic responses. Psychoneuro­ endocrinology, 37(1), 87–93. Gamer, M., Zurowski, B., & Büchel, C. (2010). Different amygdala subregions mediate valence-related and attentional effects of oxytocin in humans. Proceedings of the National Academy of Sciences of the United States of America, 107(20), 9400–9405. Gangestad, S. W., & Grebe, N. M. (2017). Hormonal systems, human social bonding, and affiliation. Hormones and Behavior, 91, 122–135. Gao, S., Becker, B., Luo, L., Geng, Y., Zhao, W., Yin, Y., . . . Kendrick, K. (2016). Oxytocin, the peptide that bonds the sexes also divides them. Proceedings of the National Academy of Sciences of the United States of America, 113(27), 7650–7654. Garbers, C., Hermanns, H. M., Schaper, F., Müller-Newen, G., Grötzinger, J., Rose-John, S., & Scheller, J. (2012). Plasticity and cross-talk of interleukin 6-type cytokines. Cytokine & Growth Factor Reviews, 23(3), 85–97. Gimpl, G., & Fahrenholz, F. (2001). The oxytocin receptor system: Structure, function, and regulation. Physiological Reviews, 81(2), 629–683. Goodson, J. L. (2013). Deconstructing sociality, social evolution and relevant nonapeptide functions. Psychoneuroendocrinology, 38(4), 465–478. Gould, S. J., & Vrba, E. S. (1982). Exaptation—A missing term in the science of form. Paleobiology, 8(1), 4–15. Grebe, N. M., & Gangestad, S. W. Oxytocin and Life History Trade-Offs: Its Neuromodulatory Effects Understood Within a Broad Evolutionary Framework. Unpublished Manuscript. Grebe, N. M., Kristoffersen, A. A., Grontvedt, T. V., Thompson, M.  E., Kennair, L.  E.  O., & Gangestad, S.  W. (2017).

Oxytocin and vulnerable romantic relationships. Hormones and Behavior, 90, 64–74. Grewen, K. M., Girdler, S. S., Amico, J., & Light, K. C. (2005). Effects of partner support on resting oxytocin, cortisol, norepinephrine, and blood pressure before and after warm partner contact. Psychosomatic Medicine, 67(4), 531–538. Griffin, G. D., & Flanagan-Cato, L. M. (2011). Ovarian hormone action in the hypothalamic ventromedial nucleus: Remodelling to regulate reproduction. Journal of Neuroendocrinology, 23, 465–471. Grillon, C., Krimsky, M., Charney, D. R., Vytal, K., Ernst, M., & Cornwell, B. (2013). Oxytocin increases anxiety to unpredictable threat. Molecular Psychiatry, 18(9), 958–960. Gruber, C. W. (2014). Physiology of invertebrate oxytocin and vasopressin neuropeptides. Experimental Physiology, 99, 55–61. Guo, S. A., & DiPietro, L. A. (2010). Factors affecting wound healing. Journal of Dental Research, 89(3), 219–229. Gutkowska, J., & Jankowski, M. (2012). Oxytocin revisited: Its role in cardiovascular regulation. Journal of Neuroendocrinology, 24(4), 599–608. Gutkowska, J., Jankowski, M., Lambert, C., Mukaddam-Daher, S., Zingg, H.  H., & McCann, S.  M. (1997). Oxytocin releases atrial natriuretic peptide by combining with oxytocin receptors in the heart. Proceedings of the National Academy of Sciences of the United States of America, 94(21), 11704–11709. Gwee, P.  C., Amemiya, C.  T., Brenner, S., & Venkatesh, B. (2008). Sequence and organization of coelacanth neurohypophysial hormone genes: Evolutionary history of the vertebrate neurohypophysial hormone gene locus. BMC Evolutionary Biology, 8(1), 93. Heinrichs, M., Baumgartner, T., Kirschbaum, C., & Ehlert, U. (2003). Social support and oxytocin interact to suppress cortisol and subjective responses to psychosocial stress. Biological Psychiatry, 54, 1389–1398. Herisson, F.  M., Brooks, L.  L., Waas, J.  R., Levine, A.  S., & Olszewski, P.  K. (2014). Functional relationship between oxytocin and appetite for carbohydrates versus saccharin. Neuroreport, 25(12), 909–914. Hew-Butler, T., Noakes, T.  D., Soldin, S.  J., & Verbalis, J.  G. (2008). Acute changes in endocrine and fluid balance markers during high-intensity, steady-state, and prolonged endurance running: Unexpected increases in oxytocin and brain natriuretic peptide during exercise. European Journal of Endocrinology, 159(6), 729–737. Hicks, C., Ramos, L., Reekie, T., Misagh, G. H., Narlawar, R., Kassiou, M., & McGregor, I.  S. (2014). Body temperature and cardiac changes induced by peripherally administered oxytocin, vasopressin and the non-peptide oxytocin receptor agonist WAY 267,464: A biotelemetry study in rats. British Journal of Pharmacology, 171, 2868–2887. Hoge, E. A., Anderson, E., Lawson, E. A., Bui, E., Fischer, L. E., Khadge, S. D., . . . Simon, N. M. (2014). Gender moderates the effect of oxytocin on social judgments. Human Psycho­ pharmacology: Clinical and Experimental, 29(3), 299–304. Holt-Lunstad, J., Birmingham, W. A., & Light, K. C. (2008). Influence of a “warm touch” support enhancement inter­ vention among married couples on ambulatory blood pressure, oxytocin, alpha amylase, and cortisol. Psychosomatic Medicine, 70(9), 976–985. Holt-Lunstad, J., Birmingham, W.  C., & Light, K.  C. (2015). Relationship quality and oxytocin: Influence of stable and modifiable aspects of relationships. Journal of Social and Personal Relationships, 32(4), 472–490.

Grebe and Gangestad

331

Hoyle, C. H. (1999). Neuropeptide families and their receptors: Evolutionary perspectives. Brain Research, 848(1), 1–25. Hurlemann, R., & Scheele, D. (2016). Dissecting the role of oxytocin in the formation and loss of social relationships. Biological Psychiatry, 79(3), 185–193. Insel, T.  R., & Shapiro, L.  E. (1992). Oxytocin receptor distribution reflects social organization in monogamous and polygamous voles. Proceedings of the National Academy of Sciences of the United States of America, 89(13), 5981–5985. Jaeggi, A.  V., Trumble, B.  C., Kaplan, H.  S., & Gurven, M. (2015). Salivary oxytocin increases concurrently with testosterone and time away from home among returning Tsimane’ hunters. Biology Letters, 11, 20150058. Kassam, K. S., Koslov, K., & Mendes, W. B. (2009). Decisions under distress: Stress profiles influence anchoring and adjustment. Psychological Science, 20(11), 1394–1399. Ketterson, E.  D., Atwell, J.  W., & McGlothlin, J.  W. (2009). Phenotypic integration and independence: Hormones, performance, and response to environmental change. Integrative and Comparative Biology, 49(4), 365–379. Ketterson, E. D., & Nolan, V., Jr. (1999). Adaptation, exaptation, and constraint: A hormonal perspective. American Naturalist, 154(S1), S4–S25. Kirsch, P., Esslinger, C., Chen, Q., Mier, D., Lis, S., Siddhanti, S., . . . Meyer-Lindenberg, A. (2005). Oxytocin modulates neural circuitry for social cognition and fear in humans. Journal of Neuroscience, 25(49), 11489–11493. Klatt, J. D., & Goodson, J. L. (2013). Sex-specific activity and function of hypothalamic nonapeptide neurons during nestbuilding in zebra finches. Hormones and Behavior, 64(5), 818–824. Klement, J., Ott, V., Rapp, K., Brede, S., Piccinini, F., Cobelli, C., . . . Hallschmid, M. (2017). Oxytocin improves β-cell responsivity and glucose tolerance in healthy men. Diabetes, 66(2), 264–271. Kosfeld, M., Heinrichs, M., Zak, P. J., Fischbacher, U., & Fehr, E. (2005). Oxytocin increases trust in humans. Nature, 435(7042), 673–676. Kublaoui, B. M., Gemelli, T., Tolson, K. P., Wang, Y., & Zinn, A.  R. (2008). Oxytocin deficiency mediates hyperphagic obesity of Sim1 haploinsufficient mice. Molecular Endocrinology, 22(7), 1723–1734. Lancaster, L.  T., & Sinervo, B. (2011). Epistatic social and endocrine networks and the evolution of life history tradeoffs and plasticity. In T.  Flatt & F.  Heyland (Eds.), Mechanisms of life history evolution. The genetics and physiology of life history traits and trade-offs (pp. 329–348). New York, NY: Oxford University Press. Lawson, E. A., Marengi, D. A., DeSanti, R. L., Holmes, T. M., Schoenfeld, D. A., & Tolley, C. J. (2015). Oxytocin reduces caloric intake in men. Obesity, 23(5), 950–956. Lee, E. S., Uhm, K. O., Lee, Y. M., Kwon, J., Park, S. H., & Soo, K. H. (2008). Oxytocin stimulates glucose uptake in skeletal muscle cells through the calcium–CaMKK–AMPK pathway. Regulatory Peptides, 151(1), 71–74. Leng, G., & Ludwig, M. (2016). Intranasal oxytocin: Myths and delusions. Biological Psychiatry, 79(3), 243–250. Li, T., Wang, P., Wang, S. C., & Wang, Y. F. (2017). Approaches mediating oxytocin regulation of the immune system. Frontiers in Immunology, 7, 693. Liu, Y., & Wang, Z. X. (2003). Nucleus accumbens oxytocin and dopamine interact to regulate pair bond formation in female prairie voles. Neuroscience, 121(3), 537–544.

332 Oxy tocin

Lowrey, E.  M., & Tomaszycki, M.  L. (2014). The formation and  maintenance of social relationships increases nonapeptide mRNA in zebra finches of both sexes. Behavioral Neuroscience, 128(1), 61. Marazziti, D., Dell’Osso, B., Baroni, S., Mungai, F., Catena, M., Rucci, P., . . . Dell’Osso, L. (2006). A relationship between oxytocin and anxiety of romantic attachment. Clinical Practice and Epidemiology in Mental Health, 2(1), 28. McCarthy, M. M. (1995). Estrogen modulation of oxytocin and its relation to behavior. Advances in Experimental Medicine and Biology, 395, 235–245. McNeilly, A.  S., Robinson, I.  C., Houston, M.  J., & Howie, P. W. (1983). Release of oxytocin and prolactin in response to suckling. British Medical Journal, 286(6361), 257–259. Moura, D., Canavarro, M. C., & Figueiredo-Bragas, M. (2016). Oxytocin and depression in the perinatal period—A systematic review. Archives of Women’s Mental Health, 19, 561–570. Nation, D. A., Szeto, A., Mendez, A. J., Brooks, L. G., Zaias, J., Herderick, E.  E., . . . McCabe, P.  M. (2010). Oxytocin attenuates atherosclerosis and adipose tissue inflammation in socially isolated ApoE−/− mice. Psychosomatic Medicine, 72(4), 376–382. Nave, G., Camerer, C., & McCullough, M. (2015). Does oxytocin increase trust in humans? A critical review of research. Perspectives on Psychological Science, 10(6), 772–789. Nelson, E. E., & Panksepp, J. (1998). Brain substrates of infant– mother attachment: Contributions of opioids, oxytocin, and norepinephrine. Neuroscience & Biobehavioral Reviews, 22(3), 437–452. Neumann, I. D., & Slattery, D. A. (2016). Oxytocin in general anxiety and social fear: A translational approach. Biological Psychiatry, 79(3), 213–221. Norman, G.  J., Cacioppo, J.  T., Morris, J.  S., Karelina, K., Malarkey, W. B., DeVries, A. C., & Berntson, G. G. (2011). Selective influences of oxytocin on the evaluative processing of social stimuli. Journal of Psychopharmacology, 25(10), 1313–1319. Numan, M. (2017). Parental behavior. In Reference module in neuro­science and biobehavioral psychology. Elsevier, https:// www.elsevier.com/solutions/sciencedirect/content/referencemodules/neuroscience-and-biobehavioral-psychology-module. Numan, M., Numan, M.  J., Schwarz, J.  M., Neuner, C.  M., Flood, T. F., & Smith, C. D. (2005). Medial preoptic area interactions with the nucleus accumbens–ventral pallidum circuit and maternal behavior in rats. Behavioural Brain Research, 158(1), 53–68. Numan, M., & Stolzenberg, D. S. (2009). Medial preoptic area interactions with dopamine neural systems in the control of the onset and maintenance of maternal behavior in rats. Frontiers in Neuroendocrinology, 30(1), 46–64. Numan, M., & Young, L.  J. (2016). Neural mechanisms of mother–infant bonding and pair bonding: Similarities, differences, and broader implications. Hormones and Behavior, 77, 98–112. Olszewski, P. K., Klockars, A., & Levine, A. S. (2016). Oxytocin: a conditional anorexigen whose effects on appetite depend on the physiological, behavioural and social contexts. Journal of Neuroendocrinology, 28(4). Onaka, T., Takayanagi, Y., & Yoshida, M. (2012). Roles of oxytocin neurones in the control of stress, energy metabolism, and social behaviour. Journal of Neuroendocrinology, 24(4), 587–598.

Ott, V., Finlayson, G., Lehnert, H., Heitmann, B., Heinrichs, M., Born, J., & Hallschmid, M. (2013). Oxytocin reduces reward-driven food intake in humans. Diabetes, 62(10), 3418–3425. Pedersen, A., & Tomaszycki, M. L. (2012). Oxytocin antagonist treatments alter the formation of pair relationships in zebra finches of both sexes. Hormones and Behavior, 62(2), 113–119. Prehn, K., Kazzer, P., Lischke, A., Heinrichs, M., Herpertz, S. C., & Domes, G. (2013). Effects of intranasal oxytocin on pupil dilation indicate increased salience of socioaffective stimuli. Psychophysiology, 50(6), 528–537. Reyes-Lagos, J.  J., Hadamitzky, M., Peña-Castillo, M.  Á., Echeverría, J.  C., Bösche, K., Lückemann, L., . . . PachecoLópez, G., (2016). Exogenous oxytocin reduces signs of sickness behavior and modifies heart rate fluctuations of endotoxemic rats. Physiology & Behavior, 165, 223–230. Riem, M. M., Bakermans-Kranenburg, M. J., Pieper, S., Tops, M., Boksem, M.  A., Vermeiren, R.  R., . . . Rombouts, S.  A. (2011). Oxytocin modulates amygdala, insula, and inferior frontal gyrus responses to infant crying: A randomized controlled trial. Biological Psychiatry, 70(3), 291–297. Sabatier, N., Leng, G., & Menzies, J. (2013). Oxytocin, feeding, and satiety. Frontiers in Endocrinology, 4, 35. Scheele, D., Striepens, N., Kendrick, K.  M., Schwering, C., Noelle, J., Wille, A., . . . Hurlemann, R. (2014). Opposing effects of oxytocin on moral judgment in males and females. Human Brain Mapping, 35(12), 6067–6076. Schneider, D.  S., & Ayres, J.  S. (2008). Two ways to survive infection: What resistance and tolerance can teach us about treating infectious diseases. Nature Reviews Immunology, 8(11), 889–895. Schneiderman, I., Zagoory-Sharon, O., Leckman, J.  F., & Feldman, R. (2012). Oxytocin during the initial stages of romantic attachment: Relations to couples’ interactive reciprocity. Psychoneuroendocrinology, 37(8), 1277–1285. Schulze, L., Lischke, A., Greif, J., Herpertz, S. C., Heinrichs, M., & Domes, G. (2011). Oxytocin increases recognition of masked emotional faces. Psychoneuroendocrinology, 36(9), 1378–1382. Schumacher, M., Coirini, H., Frankfurt, M., & McEwen, B. S. (1989). Localized actions of progesterone in hypothalamus involve oxytocin. Proceedings of the National Academy of Sciences, 86, 6798–6801. Schumacher, M., Coirini, H., Pfaff, D.  W., & McEwen, B.  S. (1990). Behavioral effects of progesterone associated with rapid modulation of oxytocin receptors. Science, 250, 691–694. Seltzer, L. J., & Ziegler, T. E. (2007). Non-invasive measurement of small peptides in the common marmoset (Callithrix jacchus): A radiolabeled clearance study and endogenous excretion under varying social conditions. Hormones and Behavior, 51(3), 436–442. Shamay-Tsoory, S. G., & Abu-Akel, A. (2016). The social salience hypothesis of oxytocin. Biological Psychiatry, 79(3), 194–202. Shamay-Tsoory, S. G., Fischer, M., Dvash, J., Harari, H., PerachBloom, N., & Levkovitz, Y. (2009). Intranasal administration of oxytocin increases envy and schadenfreude (gloating). Biological Psychiatry, 66(9), 864–870. Shattuck, E. C., & Muehlenbein, M. P. (2015). Human sickness behavior: Ultimate and proximate explanations. American Journal of Physical Anthropology, 157(1), 1–18. Sladek, C.  D., Stevens, W., Song, Z., Johnson, G.  C., & MacLean, P. S. (2016). The “metabolic sensor” function of rat supraoptic oxytocin and vasopressin neurons is attenuated

during lactation but not in diet-induced obesity. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, 310(4), R337–R345. Smith, A. S., Ågmo, A., Birnie, A. K., & French, J. A. (2010). Manipulation of the oxytocin system alters social behavior and attraction in pair-bonding primates, Callithrix penicillata. Hormones and Behavior, 57(2), 255–262. Smith, A.  S., Tabbaa, M., Lei, K., Butler, M.  J., Linton, L., Altshuler, R., . . . Wang, Z. (2016). Local oxytocin tempers anxiety by activating GABA A receptors in the hypothalamic paraventricular nucleus. Psychoneuroendocrinology, 63, 50–58. Smith, T.  W., Uchino, B.  N., MacKenzie, J., Hicks, A.  M., Campo, R. A., Reblin, M., . . . Light, K. C. (2013). Effects of couple interactions and relationship quality on plasma oxytocin and cardiovascular reactivity: Empirical findings and methodological considerations. International Journal of Psychophysiology, 88(3), 271–281. Stolzenberg, D.  S., & Numan, M. (2011). Hypothalamic interaction with the mesolimbic DA system in the control of the maternal and sexual behaviors in rats. Neuroscience & Biobehavioral Reviews, 35(3), 826–847. Striepens, N., Scheele, D., Kendrick, K. M., Becker, B., Schäfer, L., Schwalba, K.,  .  .  .  Hurlemann, R. (2012). Oxytocin facilitates protective responses to aversive social stimuli in males. Proceedings of the National Academy of Sciences of the United States of America, 109(44), 18144–18149. Szeto, A., Nation, D. A., Mendez, A. J., Dominguez-Bendala, J., Brooks, L. G., Schneiderman, N., & McCabe, P. M. (2008). Oxytocin attenuates NADPH-dependent superoxide activity and IL-6 secretion in macrophages and vascular cells. American Journal of Physiology-Endocrinology and Metabolism, 295(6), E1495–E1501. Szeto, A., Rossetti, M.  A., Mendez, A.  J., Noller, C.  M., Herderick, E. E., Gonzales, J. A., . . . McCabe, P. M. (2013). Oxytocin administration attenuates atherosclerosis and inflammation in Watanabe heritable hyperlipidemic rabbits. Psychoneuroendocrinology, 38(5), 685–693. Tabak, B. A., McCullough, M. E., Szeto, A., Mendez, A. J., & McCabe, P.  M. (2011). Oxytocin indexes relational distress following interpersonal harms in women. Psychoneuro­ endocrinology, 36(1), 115–122. Tahara, A., Tsukada, J., Tomura, Y., Wada, K. I., Kusayama, T., Ishii, N., . . . & Tanaka, A. (2000). Pharmacologic charac­ terization of the oxytocin receptor in human uterine smooth muscle cells. British journal of Pharmacology, 129(1), 131–139. Takayanagi, Y., Kasahara, Y., Onaka, T., Takahashi, N., Kawada, T., & Nishimori, K. (2008). Oxytocin receptor-deficient mice developed late-onset obesity. Neuroreport, 19(9), 951–955. Tattersall, A.  J., & Hockey, G.  R.  J. (1995). Level of operator control and changes in heart rate variability during simulated flight maintenance. Human Factors, 37, 682–698. Taylor, S. E. (2006). Tend and befriend biobehavioral bases of affiliation under stress. Current Directions in Psychological Science, 15(6), 273–277. Taylor, S. E., Saphire-Bernstein, S., & Seeman, T. E. (2010). Are plasma oxytocin in women and plasma vasopressin in men biomarkers of distressed pair-bond relationships? Psychological Science, 21(1), 3–7. Tinbergen, N. (1963). On aims and methods of ethology. Zeitschrift für Tierpsychologie, 20(4), 410–433. Tooby, J., Cosmides, L., Sell, A., Lieberman, D., & Sznycer, D. (2008). 15 Internal regulatory variables and the design of human motivation: A computational and evolutionary

Grebe and Gangestad

333

approach. In A.  Elliot (Ed.), Handbook of approach and avoidance motivation (pp. 252–271). Mahwah, NJ: Lawrence Erlbaum. Uvnäs-Moberg, K. (2003). The oxytocin factor: Tapping the hormone of calm, love, and healing. Boston, MA: Da Capo Press. Wakerley, J.  B., & Lincoln, D.  W. (1973). The milk-ejection reflex of the rat: A 20-to 40-fold acceleration in the firing of paraventricular neurones during oxytocin release. Journal of Endocrinology, 57(3), 477–493. Walum, H., Waldman, I. D., & Young, L. J. (2016). Statistical and methodological considerations for the interpretation of intranasal oxytocin studies. Biological Psychiatry, 79, 251–257. Wang, P., Yang, H. P., Tian, S., Wang, L., Wang, S. C., Zhang, F., & Wang, Y.  F. (2015). Oxytocin-secreting system: A major part of the neuroendocrine center regulating immunologic activity. Journal of Neuroimmunology, 289, 152–161. Weitzman, R. E., Glatz, T. H., & Fisher, D. A. (1978). The effect of hemorrhage and hypertonic saline upon plasma oxytocin and arginine vasopressin in conscious dogs. Endocrinology, 103(6), 2154–2160. Weisman, O., Zagoory-Sharon, O., Schneiderman, I., Gordon, I., & Feldman, R. (2013). Plasma oxytocin distributions in a large cohort of women and men and their gender-specific associations with anxiety. Psychoneuroendocrinology, 38(5), 694–701. Williams, J. R., Insel, T. R., Harbaugh, C. R., & Carter, C. S. (1994). Oxytocin administered centrally facilitates formation of a partner preference in female prairie voles (Microtus ochrogaster). Journal of Neuroendocrinology, 6, 247–250.

334 Oxy tocin

Windle, R.  J., Shanks, N., Lightman, S.  L., & Ingram, C.  D. (1997). Central oxytocin administration reduces stressinduced corticosterone release and anxiety behavior in rats. Endocrinology, 138(7), 2829–2834. Yee, J. R., Kenkel, W. M., Friling, J. L., Dohdia, J. S., Onishi, K. G., Tovar, S., . . . Carter, C. S. (2016). Oxytocin promotes functional coupling between paraventricular nucleus and both sympathetic and parasympathetic cardioregulatory nuclei. Hormones and Behavior, 80, 82–91. Yi, K. J., So, K. H., Hata, Y., Suzuki, Y., Kato, D., Watanabe, K., . . . Roh, S. G. (2015). The regulation of oxytocin receptor gene expression during adipogenesis. Journal of Neuroendo­ crinology, 27(5), 335–342. Young, L.  J., & Wang, Z. (2004). The neurobiology of pair bonding. Nature Neuroscience, 7(10), 1048–1054. Zak, P.  J. (2012). The moral molecule: The source of love and prosperity. New York, NY: Random House. Zak, P. J., Kurzban, R., & Matzner, W. T. (2005). Oxytocin is associated with human trustworthiness. Hormones and Behavior, 48, 522–527. Zhang, G., & Cai, D. (2011). Circadian intervention of obesity development via resting-stage feeding manipulation or oxytocin treatment. American Journal of Physiology-Endocrinology and Metabolism, 301(5), E1004–E1012. Zhang, H., Wu, C., Chen, Q., Chen, X., Xu, Z., Wu, J., & Cai, D. (2013). Treatment of obesity and diabetes using oxytocin or analogs in patients and mouse models. PloS One, 8, e61477.

CH A PT E R

19

Social Bond Paradoxes

Lynea R. Witczak, Trenton C. Simmons, and Karen L. Bales

Abstract An emergent literature illustrates some mechanisms by which social bonds are maintained throughout the animal kingdom. Examples of social bonds include parent–offspring bonds, monogamous pair-bonds, and friendships. Each type of bond involves a set of behaviors that can impose paradoxical physiological consequences. Testosterone levels, for example, are generally low during displays of parental care and high for infant defense, two behaviors that are necessary for maintaining parent–offspring bonds. This chapter details instances where social bonds may force physiological trade-offs in oxytocin, vasopressin, testosterone, and cortisol signaling. Specifically, it examines the endocrine trade-offs necessary for balancing mating and parenting behaviors; affiliation with mating and parenting; group stability among dominant and subordinate individuals; and multiple strong, simultaneous bonds across individuals. The synthesis provides evidence of physiological trade-offs as an explanation for social paradoxes and provides a framework for further exploration into the understanding of hormone mediation of social bond maintenance. Keywords:  social bond paradoxes, vasopressin, oxytocin, testosterone, cortisol, social behavior, physiological trade-offs

Social Bonds

A social bond is a close, selective, enduring, interpersonal relationship that is maintained by physiological and behavioral mechanisms (Feldman, 2012b; Lim & Young, 2006; Numan, 2015). Bowlby (1977a,b) argued that humans display attachment behaviors (e.g., seeking proximity with preferred individuals) throughout their lives. Bonded individuals often display synchronous activation of these two mechanisms, and the degree of this activation has previously been shown to be predictive of bonding strength (Feldman, 2007, 2012a, 2012b). Various types of social bonds exist across species, including mother–offspring bonds, pair-bonds, and friendships. Ainsworth (1985, 1989) examined these various bonds, which develop throughout one’s entire life, through the lens of attachment theory. Mother–offspring bonds, for example, are com­ mon across mammal species and have been studied

extensively. Like any other social bond, mother– offspring bonds depend on both neurological and physiological mechanisms, such as individual recognition and selective attachment (Numan, 2015). Oxytocin (OT) may have a particularly important role in regulating close social bonds. Species-specific affiliative behaviors, such as mutual gaze in humans and chimpanzees (Feldman, 2007; Tomonaga et al., 2004) and licking in rodents (Rilling & Young, 2014; Van Anders, Goldey, & Kuo, 2011), are necessary for the maintenance of parent–offspring bonds. It has been hypothesized that adult pair-bonds may have evolved from the physiological mechanisms mediating parent–offspring bonds (Carter,  1998; Fernandez-Duque, Valeggia, & Mendoza, 2009) but modified to include additional systems that promote reproductive behaviors (Van Anders et al., 2011). A monogamous pair-bond is a bond between two adults, exemplified by a preference for close physical 335

proximity to a specific individual (Carp et al., 2015), aggression toward unfamiliar intruders (Winslow, Hastings, Carter, Harbaugh, & Insel, 1993), distress upon separation from the pair-bonded individual (Mendoza & Mason,  1986a), and the ability of partners to buffer each other from stressors (Cohen & Wills,  1985). Mating also plays a large role in facilitating and maintaining strong affiliative bonds between adults (Lim & Young, 2006; Numan, 2015). Studies on monogamy in prairie voles and New World primates, such as tamarins, marmosets, and titi monkeys, have demonstrated the importance of OT and arginine vasopressin (AVP) for pair-bonding (e.g., Bales, Mason, Catana, Cherry, & Mendoza, 2007; Carter, 1998; Smith, Ågmo, Birnie, & French, 2010; Young, Wang, & Insel, 1998). OT and AVP are also clearly involved in social attachment in humans (Insel & Young, 2001). Bonds between siblings are also regulated by physiological mechanisms. In cooperatively breeding species, older offspring help parents raise younger siblings (Riedman,  1982). Administration of exogenous OT has been found to enhance offspring guarding and feeding in meerkats (Madden & Clutton-Brock, 2010) and food sharing in marmosets (Saito & Nakamura, 2011). OT similarly has been found to promote alloparental care of younger siblings (Bales, van Westerhuyzen, et al.,  2007). Correspondingly, administration of an OT antagonist inhibits alloparental care in male prairie voles (Bales, Kim, Lewis-Reese, & Carter, 2004). Friendship is a strong, rewarding bond to unrelated individuals that can incorporate bidirectional links between social cognition, emotions, behaviors, and goal orientations (Bigelow, 1977; Bigelow & La Gaipa, 1975; Güroğlu et al., 2008). The difference between a friendship and a stronger attachment (e.g., parent–offspring bond, pair-bond) is the duration of the bond (affectional bonds are typically enduring, whereas friendship may or may not last long), the interchangeability of individuals within a friendship dyad (whereas affectional bonds tend to be more specific to a single individual), and the history of interactions between friends, which may include components not related to friendship (Ainsworth, 1989). A functional magnetic resonance imaging (fMRI) study found increased activity in the ventromedial prefrontal cortex, hippocampus, amygdala, and nucleus accumbens following interactions between friends (Güroğlu et al., 2008). All of these brain areas have been linked to reward processing and emotion 336

Social Bond Paradoxes

r­ egulation. These findings suggest a role for dopamine and serotonin in promoting positive social development (Güroğlu et al., 2008).

Social Paradoxes

Coordination of physiology and behavior is necessary for the maintenance of social bond stability. Given the complexity of sociality, there are often times where the costs and benefits of one behavior supplant those of another behavior. Balancing behaviors that conflict with each other may help maintain an overarching social bond. For example, parent– offspring bonds are maintained by engaging in both infant defense and infant care, two behaviors that are marked by low blood testosterone levels. But, although it may seem paradoxical for two competing behaviors to elicit similar hormone activation profiles, they may actually reflect trade-offs that are physiologically mediated by multiple hormones. Trade-offs occur when the benefits of one behavior outweigh the costs of another. Using the example of infant defense and infant care, the benefits of protecting offspring when the occasion arises may outweigh the costs of a temporary reduction in the quality of other forms of parental care. Although both behaviors are necessary for offspring survival, certain contexts force the individual to make a choice. Context and previous experience are also important to consider when examining correlations between hormones and behavior. For example, early-life adversity, low baseline AVP (Veenema, 2009), and high AVP reactivity in response to social stressors (Veenema, Blume, Niederle, Buwalda, & Neumann, 2006) predict increased aggression in animals. For primiparous female rats, high AVP during the early lactation period is correlated with decreased aggression, and low AVP later in the lactation period predicts increased aggression (Nephew & Bridges, 2008; Nephew, Byrnes, & Bridges, 2010). Females bred for high-anxiety-related behaviors, though, exhibit positive correlations between AVP and maternal aggression (Bosch & Neumann,  2010; Bosch, ­ Pförtsch, Beiderbeck, Landgraf, & Neumann, 2010). These differences in hormone functioning may reflect trade-offs that are evolutionarily beneficial under varying social and environmental contexts.

Chapter Outline

In this chapter, we describe how levels of OT, AVP, testosterone, and cortisol are mediated to support and suppress certain behaviors to uphold multiple bonds. First, we examine the endocrine trade-offs

necessary for balancing mating and parenting ­behaviors. In particular, we outline the challenge and steroid/peptide hypotheses and the role of testosterone in infant care versus infant defense. Second, we describe the roles of OT and AVP in harmonizing affiliation with mating and parenting. Examples ­include balancing a pair’s need to mate during postpartum estrus with the need to provide for offspring, as well as maintaining an affiliative pair-bond while exhibiting agonistic behaviors such as pair-mate restraint. Third, we discuss how context and individuals’ perception of events determines whether high or low cortisol and testosterone levels predict social dominance. Finally, we identify examples of how social bonds within individuals can be in competition with each other. Specifically, we investigate why some bonds are stronger than others in certain species, and how humans are able to maintain strong attachments to multiple individuals simultaneously.

Mating Versus Parenting Overlap in Mechanism, Trade-Offs in Function

Hormones are co-opted across social contexts to ­accomplish specific behavioral goals. Such is the case with the maintenance of pair-bonds and parent– offspring bonds, which may overlap evolutionarily in aspects of function and mechanism (van Anders et al., 2011). Both bond styles share aspects of attachment and intimacy, which likely explains the overlapping hormonal circuitry employed by these bonds (Carter, 1998). Given such overlap and the likelihood that parent–offspring bonds evolved first in mammalian species, the manifestation of pair-bonding behavior may depend in part on the neuroendocrine system supplied by parent–offspring bonds (Carter, 1998; Fernandez-Duque et al., 2009). This idea is supported by the fact that species that engage in pair-bonding behavior are also more likely to exhibit biparental care (Kleiman,  1977; Wittenberger & Tilson, 1980). Indeed, the nurturance and defense of offspring often co-occurs with the nurturance and defense of a pair-bonded mate, but certain aspects of the two social bonds may not entirely overlap. Despite the existence of similarities between pair and parent–offspring bonds, pair-bonds involve the attachment and intimacy aspects of parent–offspring bonds while still giving space for sexual facilitation and reproductive potential. Given the co-occurrence of pair-bonding behavior and biparental care, there may be circumstances where the needs of one system take precedence over the needs of the other.

In other words, sexual intimacy may come at the cost of attenuated parenting behavior. To better understand these trade-offs, testosterone emerges as a key component due to its role in both sexuality and nurturance (van Anders & Watson, 2006a).

The Challenge Hypothesis

In exploring seasonal variation in male bird testosterone levels, Wingfield, Hegner, Dufty, and Ball (1990) theorized that testosterone increases may only be predictive of behaviors that emerge in response to direct challenges to reproductive fitness. These challenges may come in the form of intrasexual rivals, which necessitate aggressive behaviors like mate guarding. In the absence of such challenges, testosterone levels would be expected to decrease to promote the maintenance of pair-bonds and parent– offspring bonds. These observations collectively form the challenge hypothesis, which has been used to explain the complicated role testosterone plays in modulating social contexts. Over the past few decades, the challenge hypothesis has been studied extensively, and there is a growing body of supportive evidence (Wingfield, 2017). However, certain nuanced behaviors seem to be at odds with the challenge hypothesis, depicting a somewhat paradoxical role for testosterone in mating and parenting. A few of the most widely debated deviations include offspring defense, aggression, and sexual intimacy. Although parental care would be expected to decrease testosterone levels, engaging in offspring defense has been shown to increase testosterone levels (Teichroeb & Sicotte, 2008; van Anders, Tolman, & Volling, 2012). Similarly, testosterone has been shown to promote aggression when beneficial to reproductive competition while concurrently stimulating certain aspects of prosocial behavior, which are two opposing behavioral states. Testosterone has also been shown to stimulate sexual intimacy while inhibiting pairbonding behavior, which complicates the fact that sexual intimacy is often necessary for both the formation and maintenance of pair-bonding behavior across species (Carter, 1998; Snowdon, Ziegler, Schultz-Darken, & Ferris, 2006).

Paradoxes as Trade-Offs

Although these paradoxes may simply represent the nuanced effects of social bond tradeoffs, it is more likely that they represent nuanced behaviors that fall outside the standards of the challenge hypothesis. van Anders and colleagues (2011) proposed the Witcz ak, Simmons, and Bales

337

steroid/peptide theory of social bonds (S/P theory), building on the challenge hypothesis to resolve such nuanced behaviors by subdividing aggression and intimacy into specific categories. According to S/P theory, and in relation to social bonds, aggression can either be antagonistic or protective in nature. Intimacy, on the other hand, can be either sexual or nurturant (van Anders et al., 2011). Using this framework, both types of aggression and sexual intimacy would be expected to increase testosterone, whereas nurturant intimacy should decrease testosterone. Partitioning aggression and intimacy by context facilitates explanations for the seemingly paradoxical role of testosterone in certain behaviors. Although the challenge hypothesis holds that all parental behaviors belong in a low-testosterone category (Kuzawa, Gettler, Muller, McDade, & Feranil, 2009; Kuzawa, Gettler, Huang, & McDade, 2010; E. S. Barrett et al., 2013), S/P theory maintains that low testosterone might be better predictive of only certain aspects of parental behavior rather than of parenting in general. Given the presence of a challenge that solicits offspring defense behavior, it would make sense that testosterone activation might be employed to help prepare the body for aggressive rivals by stimulating muscle hypertrophy and protective aggressive responses. Subsequently, any situation that presents a challenge in the context of parenting might be associated with high testosterone, whereas any nurturant behaviors should be associated with low testosterone. Although higher testosterone levels may not be conducive to nurturant responses to infant stimuli, the trade-off manifests in the increased ability to react appropriately to challenging stimuli.

Challenge Hypothesis Sexual Intimacy

Infant Defense

Steroid/Peptide Theory of Social Bonds Sexual Intimacy

Protective Aggression

OT

AVP

Nurturant Intimacy

Infant care

Antagonistic Aggression

Higher plasma hormone levels Lower plasma hormone levels

T AVP

OT

AVP

OT

T

T

Nurturant Intimacy

As for the contrasting roles of testosterone in mutually exclusive behaviors like aggression and sociality, S/P theory postulates that the differential results may be explained by the presence or absence of coreleased peptides (van Anders et al.,  2011). Like testosterone, AVP contributes to infant defense and aggressive behavior (Bosch, 2011). Both of these behaviors could situationally fall under the category of protective aggression, a category that should increase both AVP and testosterone levels. However, antagonistic aggression would only be expected to increase testosterone levels. The presence of AVP thereby could help explain how increases in testosterone could contextually be prosocial if facilitating defensive behaviors (Figure 19.1). As mentioned before, the challenge hypothesis positions sexual activity as a high-testosterone behavior and pair-bonds in a low-testosterone category despite pair-bonding behavior often necessitating sexual intimacy. As with aggression, the presence of coreleased peptides and the salience of the context might help explain the paradoxical testosterone signaling. As predicted by S/P theory, both sexual and nurturant intimacies increase OT levels, but only sexual intimacy would be expected to increase testosterone. Given the importance of sexual intimacy for reproductive fitness, there are likely stronger selective pressures for a testosterone response to sexual intimacy (van Anders et al.,  2011). Therefore, the ability of sexual intimacy to increase testosterone supplants the ability of nurturant intimacy to decrease testosterone. Alternatively, testosterone and sexual activity usually go down over the course of a pair-bond as affiliation goes up, which may also

AVP

OT

Social Withdrawal

Figure 19.1  Comparison of behavioral outcomes predicted by concentrations of testosterone (T), arginine vasopressin (AVP), and oxytocin (OT) based on the challenge hypothesis and the steroid/peptide (S/P) Theory.

338

Social Bond Paradoxes

explain the differential testosterone levels (Gray et al., 2004; Prior et al., 2016).

Affiliation Versus Parenting and Mating Maintenance of Affiliative Bonds Across Social Contexts

Several hormones have been implicated in the maintenance of affiliation. OT in particular has been associated with the formation of close social bonds (Gimpl & Fahrenholz,  2001; Insel,  1997; Ross & Young, 2009). AVP has also been found to be ­involved with certain aspects of sociality, like paternal care (Bester-Meredith, Young, & Marler,  1999; Goodson & Bass, 2001; Wang, Ferris, & De Vries, 1994). Additionally, low testosterone has also been tied to nurturing behaviors in human mothers (Fleming, Ruble, Krieger, & Wong, 1997; Kuzawa et al., 2010), human fathers (Kuzawa et al., 2009), and monogamous male nonhuman primates (Prudom et al., 2008). To maintain an overall affiliative bond with offspring and mates, individuals must also display aggressive behaviors such as mate guarding (Fisher-Phelps et al.,  2015) and infant defense (Teichroeb & Sicotte, 2008), which are coordinated with changes in peptide and steroid hormone activity (Van Anders et al., 2011). Thus, physiological tradeoffs underlying aggressive behaviors facilitate and maintain strong affiliative bonds.

Affiliative Versus Aggressive Behaviors in Parenting

OT plays a large role in maternal behaviors (Feldman, 2007; Finkenwirth, van Schaik, Ziegler, & Burkart, 2015; Kendrick, 2000). In particular, studies that manipulated OT levels have demonstrated that ­elevated OT facilitates motivation to engage in maternal behaviors in rats (Fahrbach, Morrell, & Pfaff, 1984; Pedersen, Caldwell, Walker, Ayers, & Mason,  1994) and sheep (Kendrick, Keverne, & Baldwin, 1987). AVP has also been found to be a regulator of maternal care, as demonstrated by chronic and acute intracerebroventricular (ICV) administration of AVP in lactating rodents (Bosch & Neumann, 2008; Kessler, Bosch, Bunck, Landgraf, & Neumann, 2011). Mater­nal care is reduced when AVP V1a receptor (V1aR) activity is acutely blocked with selective antagonists (Bosch & Neumann, 2008). V1aR activity in the paraventricular nucleus (PVN) may be particularly important for nursing behavior, but not maternal motivation (Bayerl, Hönig, & Bosch,  2016). V1bR activation in the PVN, however, does not seem to play a role in maternal care (Bayerl et al., 2016).

OT and AVP play an important role in facilitating paternal behavior as well (Feldman, Gordon, Schneiderman, Weisman, & Zagoory-Sharon, 2010; Zimmermann-Peruzatto, Lazzari, de Moura, Almeida, & Giovenardi, 2015). Acute administration of AVP in male prairie voles increases licking and grooming behavior, and blockade of V1aR in the lateral septum blocks this form of paternal care (Wang et al., 1994). Genetic variation of the V1aR gene may predict increased activation of reward neural circuitry in the left hemisphere anterior prefrontal cortex (APFC) in human fathers (Nishitani et al., 2017). Specifically, male AVPR1ARSC-non-334 carriers showed greater APFC activation than male AVPR1ARSC-334 carriers when viewing video clips of their own child smiling (Nishitani et al., 2017). Physiology and behavior can exert bidirectional effects on each other (Cohen & Wills, 1985). Hor­ mone manipulation studies in humans have observed an increase in peripheral OT levels in infants interacting with fathers who had been treated with exogenous OT (Feldman, Gordon, & Zagoory-Sharon, 2011; Weisman, Zagoory-Sharon, & Feldman, 2012). In addition to exhibiting physiological synchrony, parents and offspring also display coordination of behaviors (Feldman et al., 2007; Feldman, 2012b). This coordination of affiliative behaviors and increased OT may reflect a positive feedback mechanism underlying parent–offspring bonds (Crockford et al., 2013). In addition to affiliative parental behaviors, aggressive behaviors are necessary to successfully rear offspring (Bosch, 2011; Numan & Insel, 2003). This phenomenon is observed across the animal kingdom, including fish (Taborsky & Limberger, 1981), birds (Lynn, Hayward, Benowitz-Fredericks, & Wingfield, 2002), and mammals (Trainor & Marler,  2001). Both OT and AVP are involved in aggressive parental behavior. Studies by Bosch, Krömer, Brunton, and Neumann (2004) and Bosch, Meddle, Beiderbeck, Douglas, and Neumann (2005) demonstrated a rise in OT release from the hypothalamic paraventricular nucleus (PVN) and the central amygdala when female rats display aggression toward intruders. Local manipulations of V1aR show a positive correlation between AVP and maternal aggression in lactating rats bred for high-anxiety-related behaviors (Bosch & Neumann, 2010; Bosch et al., 2010); this relationship may be affected by lactation state. For example, ICV administration of AVP decreased maternal aggression on day 5 of lactation, whereas V1aR antagonists increase maternal aggression on day 15 (Nephew & Bridges, 2008; Nephew et al., 2010). Low AVP Witcz ak, Simmons, and Bales

339

during lactation may suppress attack behaviors to facilitate increased maternal care during early infancy (Nephew et al., 2010). van Anders and colleagues (2011) describe the “aggression paradox” as the knowledge gap that fails to explain why peptides like AVP and OT, which are positively correlated with affiliative behaviors, are also linked to aggressive behaviors. They argue that coordination between steroid hormones and peptides may explain this paradox. Specifically, OT and AVP may rise in response to a need for protective aggressive behaviors, whereas testosterone activity may be positively correlated with antagonistic aggression (van Anders et al., 2011). Findings from Trainor and Marler (2001) support this theory, demonstrating that affiliative aspects of paternal care were positively correlated with aggressive infant defense in a monogamous rodent, independent of testosterone levels. At first glance, it appears that increases in AVP and OT, but not testosterone, may simultaneously promote nurturing care of infants and defense behaviors against intruders. However, Trainor and Marler (2002) followed up on these findings and found that the aromatization of testosterone to estrogen was essential for facilitation of paternal care in this species. California mice, however, do seem to be an exception to many of the usual findings regarding testosterone and parenting, and it would be worthwhile to study all three hormones in concert in other species.

Postpartum Estrus Versus Infant Care

AVP, OT, and testosterone may also play a role in coordinating mating behaviors and infant care while females are in postpartum estrus. In some species, females begin sexual cycling while still lactating and caring for infants (Gubernick, 1988; Kholkute, 1984). As previously discussed, OT and AVP are positively correlated with infant care in males and females. Testosterone, conversely, is negatively linked to nurturing behaviors and positively linked to sexual behaviors (Bales, French, McWilliams, Lake, & Dietz,  2006; van Anders & Watson,  2006b; Ziegler, Prudom, Zahed, Parlow, & Wegner, 2009). Van Anders et al. (2011) argue that sexual behavior is a more salient stimulus for testosterone; therefore, sexual intimacy will correlate with a rise in testosterone and a decrease in infant care. Males in biparental species need to balance the need to reproduce with the need to care for new infants in the early postpartum period. It has been proposed that the neuroendocrine pathways that mediate maternal behavior may be the same as those 340

Social Bond Paradoxes

driving paternal care (Wynne-Edwards & Timonin, 2007). For example, prolactin has been found to be positively correlated with paternal care in rodents (Brown, Murdoch, Murphy, & Moger,  1995; Gubernick & Nelson, 1989; Sakaguchi et al., 1996) and nonhuman primates (Dixson & George, 1982; Ziegler, Wegner, Carlson, Lazaro-Perea, & Snowdon, 2000). In several species, testosterone is positively correlated with mating behavior and negatively correlated with paternal behavior and prolactin (Nunes, Fite, & French, 2000; Van Roo, Ketterson, & Sharp, 2003). This trade-off in testosterone and prolactin activity is particularly true for seasonally breeding birds, where breeding and parenting behaviors do not co-occur. Wynne-Edwards and Timonin (2007) suggest that the timing of fluctuations of various hormones may be important for facilitating coordination of postpartum mating and offspring care. For tamarins, males will care for infants at the same time that ­females are ovulating and therefore able to conceive (Ziegler, Bridson, Snowdon, & Eman, 1987). Ziegler et al. (2000) did not find any hormonal changes in male tamarins immediately following the birth of their infant. They suggest this may be because hormonal changes occur immediately prior to infant birth to facilitate the onset of infant care. There did not appear to be a trade-off between prolactin and testosterone in tamarin males, suggesting that testosterone levels were at reproductively functional levels without interfering with paternal care behaviors (Ziegler et al., 2000). It is possible that no trade-off may be necessary between parental and mating effort if the two behaviors are compatible, for instance, in species with low levels of mating competition (Stiver & Alonzo, 2009). For example, defensive behaviors like infant defense and mate guarding involve the same suite of behaviors driven by the same hormones, whereby the only difference is who is being defended (Stiver & Alonzo, 2009). In species like the California mouse (Peromyscus californicus), aggressive attacks on intruders lead to both successful mating and infant rearing (Trainor & Marler, 2001). Within a species, individual differences in mating and parenting strategies may result in differing hormone profiles (Stiver & Alonzo,  2009). One exemplary species is the striped mouse (Rhabdomys pumillio). Male striped mice display three discrete mating and parenting tactics: (1) males who remain in breeding groups to care for young but do not mate, (2) males who hold smaller territories and mate and rear offspring with partners, and (3) males who sneak copulations and

do not raise offspring (Schradin, Scantlebury, Pillay, & König, 2009). Males who exclusively care for offspring have the lowest levels of endogenous testosterone, whereas males who sneak copulations have the highest levels, and those who mate and care for offspring have intermediate levels of testosterone (Schradin et al.,  2009). Further investigation is required to better understand the coordination of mating and parenting behaviors.

Pair-Bonding Versus Mating

Sexual activity can strengthen pair-bonds in monogamous species (Carter, 1998; Snowdon et al., 2006; van Anders et al.,  2007). However, testosterone, which is positively correlated with sexual activity, has been found to inhibit pair-bonds (van Anders & Watson, 2006a). Physiological trade-offs are therefore necessary to maintain pair-bonds while facilitating mating behaviors. Indeed, human couples who are involved in nurturant pair-bonds exhibit lower levels of testosterone than single individuals (van Anders & Gray, 2007; van Anders, 2009). Van Anders et al. (2011) theorize that trade-offs in testosterone levels may coordinate affiliative behaviors between pair-mates and reproductive behaviors. Specifically, testosterone may be aromatized to estradiol through peptide activities to promote nurturant behaviors (van Anders et al.,  2011) and may be positively correlated with sexual contexts due to its link to competitive behavioral contexts. Other hormones, including OT, are also involved in sexual behavior. OT has also been found to be positively correlated with sexual intimacy (Carter, 1992; Insel, Winslow, Wang, & Young, 1998; Young, Liu, & Wang, 2008; Snowdon et al., 2010). The S/P theory proposed by van Anders et al. (2011) posits that increased OT in conjunction with increased testosterone may facilitate sexual intimacy, whereas high OT and low testosterone may facilitate nurturant intimacy. Context, therefore, is essential for understanding the coordination of hormones driving pair-bonding and mating (van Anders et al., 2011).

Pair-Bonding Versus Mate Guarding

OT and AVP underlie pair-bonding in monogamous mammals (Bales et al.,  2013; Insel & Hulihan, 1995; Winslow et al., 1993). Although OT is sensitive to estrogens and AVP to androgens, both hormones are necessary for facilitating and maintaining monogamy in both sexes (Cho, DeVries, Williams, & Carter, 1999). These neuropeptides can bind to each other’s receptors, and their coordinated activation has been found to develop the

adult pair-bonds observed in prairie voles (Microtus ochrogaster; Ross et al., 2009). In conjunction with neurobiological interactions, mating behaviors, and affiliative behaviors, evolved agonistic behaviors such as mate guarding have been hypothesized to help maintain pair-bonds in monogamous species (Fisher-Phelps et al., 2015). In the wild, extra-pair copulations occur occasionally in monogamous titi monkeys (Callicebus moloch), and males have been observed restraining their mates to prevent females from mating with other males (Mason, 1966). Based on results from live intruder tests, males are more responsive to perceived samesex competitors than females, displaying aggressive behaviors such as tail lashing and back arching (Fernandez-Duque, Valeggia, & Mason,  2000; Mendoza & Mason, 1986a). Mate guarding could lead to additional partner-directed agonistic behaviors such as mate restraint. AVP, in particular, has been associated with aggression in non-human animals (Delville, Mansour, & Ferris,  1996) and humans (Marshall,  2013). Because AVP has dual roles in facilitating aggressive behaviors toward unfamiliar conspecifics (Numan, 2015) and maintaining persistent selective attraction toward a familiar pair-mate (Donaldson, Spiegel, & Young,  2010), it is essential to take into account context when measuring plasma AVP. Gouin et al. (2010) ­measured plasma OT and AVP following a 30-minute session where couples were asked to describe the history of their relationship. Couples with higher baseline AVP and OT showed more positive interactions than those with low baseline AVP and OT (Gouin et al., 2010). Conversely, individuals with low baseline AVP displayed a greater number of negative interactions with their partners. When examining AVP responses to social stressors, though, plasma AVP is elevated in pair-bonded men experiencing distress (Taylor, Saphire-Bernstein, & Seeman, 2010). This positive relationship between AVP and social stress has also been observed in rodents (Carter, Grippo, Pournajafi-Nazarloo, Ruscio, & Porges, 2008). Based on these findings, low AVP in the absence of social stressors and high AVP in the presence of relationship distress predict greater agonistic partner interactions. Increased testosterone activity may up-regulate AVP, due to the permissive effects of testosterone on AVP (Carter, 2007; van Anders et al., 2011), and result in an increase in aggressive responses to social stimuli. Positive correlations between testosterone and AVP may facilitate competitive behaviors that may be important for maintaining pair-bonds (e.g., mate Witcz ak, Simmons, and Bales

341

guarding, territory defense, rejection of unfamiliar conspecifics). Therefore, although the behaviors correlated with changes in testosterone and AVP are categorized as aggressive, the outcome of the behavior is the maintenance of an affiliative pair-bond.

In-Group Versus Out-Group

Part of maintaining a strong, selective affiliative bond with an individual includes exclusion of unfamiliar individuals. This is seen in monogamous rodents and primates, which reject strangers and prefer to spend more time in close social proximity to familiar individuals (e.g., Aragona et al., 2006; Gubernick & Nordby, 1993; Koenig & Rothe, 1994). The differentiation of in-group versus out-group also exists on a larger scale in nonmonogamous social structures. Physiological trade-offs may exist that facilitate ­affiliation with in-group members and rejection of out-group members. OT is positively correlated with trust, empathy, and cooperation (Heinrichs, von Dawans, & Domes, 2009; Israel et al., 2009; Rodrigues, Saslow, Garcia, John, & Keltner,  2009). Several human studies have found that OT administration increased ­empathy toward in-group members, but not outgroup members (De Dreu et al., 2010; De Dreu, Greer, Van Kleef, Shalvi, & Handgraaf,  2011; Abu-Akel, Fischer-Shofty, Levkovitz, Decety, & Shamay-Tsoory, 2014). In one study, it was demonstrated that increased exogenous OT led to increased protection of in-group members and distrust of out-group members, but not increased aggression toward out-group members (De Dreu et al., 2010). A follow-up study by De Dreu and colleagues (2011) supported these findings and postulated that OT functioning may serve an evolved need to give preferential treatment to one’s own in-group. They warn that this bias toward in-group members may promote intergroup aggression and conflict (De Dreu et al., 2011). Testosterone has been shown to mediate ingroup and out-group dynamics. Flinn, Ponzi, and Muehlenbein (2012) found that precompetition testosterone levels were lower in males who were competing against close friends compared to when they were competing against out-group members. Additionally, they found that testosterone levels increased only after males defeated individuals whom they did not consider to be friends. Hence, friendship, or perception of an individual being part of one’s in-group, may modulate testosterone reactivity to facilitate switches between aggressive and affiliative behaviors. 342

Social Bond Paradoxes

Rejection of out-group members, though, seems to be fluid and can change based on social context, physiological state, social status, and sex of conspecifics (French, 1986; Koenig & Rothe, 1994). As individuals become more familiar, they may be more readily admitted to the group, as demonstrated in nonhuman primate studies (Hubrecht, 1985; Koenig & Rothe, 1994). Mechanistically, trade-offs between OT and testosterone may facilitate this switch from aggressive to affiliative conspecific-directed behaviors as individuals become more familiar.

Dominance Versus Submission Defining Social Dominance

In many species, social groups are organized into status hierarchies. Individuals who obtain and maintain high status within a hierarchy may receive many benefits, including power and status (Magee & Galinsky, 2008). Central to the idea of power is the preferential access to important resources, while status is defined as a large sphere of influence over others (Magee & Galinsky,  2008). However, to achieve a high social rank, individuals must first display dominance. Across animal species, dominance rank is typically defined as the ability to successfully control social interactions with other individuals on both a quantitative and qualitative scale. Conse­ quently, there is a linear relationship between the number of defeated opponents and the individual’s subsequent rank (Sapolsky, Alberts, & Altmann, 1997). Qualitatively speaking, displays of dominance are often assertive and self-assured, and have been associated with higher status across both animal and human groups (Anderson & Kilduff, 2009; Sapolsky, 2005). Thus, understanding the biological basis for dominance is key to explaining the motivational forces that define a hierarchical society.

Dual Hormones of Dominance

A large body of literature implicates testosterone and cortisol as two of the primary factors that drive displays of dominance. Testosterone release is stimulated through the actions of the hypothalamicpituitary-gonadal (HPG) axis, and its recruitment is most commonly associated with sexual behavior in males. Testosterone stimulates muscle hypertrophy, aggressive behavior, negative feedback on the endocrine system, and spermatogenesis (Wingfield, Lynn, & Soma, 2001). However, the benefits of acute testosterone activation can be lost with chronic stimulation. For example, the high energetic costs of prolonged testosterone levels have been associated with reduced fat stores, increased injury and

mortality, suppression of the immune system, and interference with pair-bonding and parental behavior (Wingfield et al., 2001). Therefore, optimal testosterone activation depends on context, with higher testosterone acutely helpful for challengerelated contexts such as territorial and reproductive aggression (Wingfield et al., 1990). Cortisol, on the other hand, is the primary hormonal output of the hypothalamic-pituitary­ adrenocortical (HPA) axis, and its activation ­embodies the dualistic nature of the stress response (Sapolsky, 1990). On the one side, cortisol activation is vital for surviving physical stressors, playing a role in ­mobilizing energy, increasing cardiovascular tone, and suppressing unessential anabolic activities. An excess, however, has been implicated in a variety of pathological diseases like steroid diabetes (Björntorp, Holm, & Rosmond, 1999), myopathy (Golding, Murray, Pearce, & Thompson, 1961), hypertension (Fraser et al., 1999), and suppression of reproductive (Ziegler, Scheffler, & Snowdon, 1995) and immune systems (McEwen,  2004; see also Mogilski et al., this volume). Thus, optimal HPA functionality must involve a robust activation during stressors and marginal activation at baseline (Sapolsky, 1990), although depending on social context, an individual’s physiological response may be forced to make trade-offs. If it is more probable for an individual to encounter physical stressors, chronic activation of the stress response may be useful as an ­anticipatory response. But although this individual may be better prepared to react to a stressor at any given time, the trade-off manifests in an increased likelihood of acquiring the detrimental effects of cortisol excess. Given the benefits and risks (Maestripieri & Georgiev, 2016; Muehlenbein & Watts 2010) of testosterone and cortisol activation, it is likely that these two hormones operate under a delicate balance to regulate dominance interactions (Carré & Mehta, 2011; Mehta & Josephs, 2010; Pfattheicher, 2017; Turan, Tackett, Lechtreck, & Browning,  2015). Robust activation upon acute stimulation likely optimizes the body’s response to a stressor while minimizing long-term health impairments. For dominant individuals, the stressor often manifests in the form of a competitive challenge that provides an opportunity for a change in status. Humans, for example, who won competitive interactions (e.g., sports) had increased basal testosterone levels, whereas the opposite was seen in those who lost the interactions (Aguilar, Jiménez, & Alvero-Cruz,  2013; Zilioli & Watson,  2012,  2013). The opposite was seen

with cortisol; winners were more likely to experience decreased cortisol, whereas losers experienced an increase. Yet, despite the wealth of information our current body of literature has regarding these two hormones, it still would be difficult to predict social behavior across species using only cortisol and testosterone samples. For example, testosterone has been positively correlated with aggressive and dominant behaviors in many animal species (Archer, 1991, 2006; Carré & Olmstead, 2015). However, several other studies have failed to replicate these findings (Alonso, Honji, Moreira, & Pandolfi, 2012). Similar inconsistencies have been found in the cortisol literature, with traditional studies supporting a negative relationship with dominance (Shively, Laber-Laird, & Anton, 1997), others finding no association (Proctor, Freeman, & Brown,  2010; Maestripieri, Klimczuk, Seneczko, Traficonte, & Wilson,  2013; Dong-Dong et al., 2013), and several others suggesting a positive relationship (G. M. Barrett, Shimizu, Bardi, Asaba, & Mori, 2002; Carlson et al., 2004). The idea that two individuals of greatly differing social rank can have identical endocrine profiles seems paradoxical. Recent research, however, confirms that in regard to social hierarchies, both testosterone and cortisol activation may depend on the context in which a certain rank occurs and the social tactics employed.

An Interplay of Context and Character

It is worth explicitly stating that there is no current consensus on whether dominant or subordinate animals are more “stressed.” Studies on olive baboons suggest that although physiological reactions arise from rank, they are sensitive to the rank and social setting in which the rank occurs (Sapolsky, 1990). Generally speaking, dominant phenotypes experience the highest amount of psychological stress in species with transient periods of major rank instability and where dominant individuals must physically reassert their dominance or employ cooperative breeding strategies (Bartos, Schams, Bubenik, Kotrba, & Tománek, 2010). On the contrary, subordinate phenotypes experience the most psychological stress when hierarchies stabilize and in species where higher ranks are maintained through nonphysical intimidation or with low availability for social support (Bartos et al., 2010). Despite these trends that exist across many ­species, research continues to add support to the importance of individual variation. Although cortisol profiles may be explained by the context in which dominance occurs, they may be better explained by Witcz ak, Simmons, and Bales

343

an individual’s social tactics in coping with stressful contexts. For example, rank instability may generally be anxiogenic for a dominant individual, but the stress might depend on whether or not the male is rising or declining in the hierarchy (Sapolsky, 1992). Rank stability, therefore, could also potentially be anxiogenic for a dominant individual that does not have the social fortitude to differentiate between winning and losing fights (Sapolsky, 1990). Similarly, subordinate individuals may also employ a set of social tactics that alter their perception of a stressful context. Subordinate baboons, for example, which displace aggression after losing fights, affiliate the most with nonestrus females, engage in the highest rates of overt consortship, and have lower basal cortisol levels than other subordinates who do not have these traits (Virgin & Sapolsky, 1997). These same individuals were more likely to be in the dominant cohort in coming years and to have elevated basal testosterone (Virgin & Sapolsky, 1997). It is possible that this subset of males was beginning the successful transition into dominance. Given the interplay between context and individual variation, there can be considerable overlap in the frequency of dominant and subordinate individuals experiencing the same trends for testosterone and cortisol activation. Congruent cortisol profiles can be expected for dominant and subordinate individuals when stressors or social tactics are shared (Abbott et al., 2003; Pavlidis, Sundvik, Chen, & Panula, 2011). Corresponding testosterone profiles, on the other hand, might be expected when individuals maintain similar perceptions of agonistic challenges. Therefore, individuals might experience low cortisol and high testosterone if the perceived stress of an agonistic encounter is low. Bartos et al. (2010) ­provided evidence for this idea by observing the agonistic interactions of red deer. Control and experimental groups of adult red deer were sequestered and allowed to establish stable dominance hierarchies while hormone concentrations and behaviors were ­measured. Once stability had been achieved, younger conspecifics were added to the experimental group. The addition of subordinate deer gave the more dominant, adult deer increased opportunities to redirect their aggression without introducing social instability. Prior to adding the younger deer, dominant deer in the experimental group had lower testosterone and higher cortisol. After the addition, these same dominant deer experienced higher testosterone and lower cortisol than control deer. The social environment, therefore, was the strongest predictor 344

Social Bond Paradoxes

of testosterone and cortisol activation (Feng et al., 2016; Maruska, 2015).

Social Bonds Versus Social Bonds Maintenance of Multiple Social Attachments

Social networks across human societies can be extremely complex and include multiple types of relationships and attachments simultaneously, including parent–child bonds, pair-bonds between partners, and friendships (Flinn et al.,  2012). Forming and maintaining social bonds can have survival and reproductive benefits (Ainsworth,  1989; Axelrod & Hamilton, 1981; Baumeister & Leary, 1995). The alliance hypothesis posits that maintaining relationships with nonkin in addition to kin may result in a greater sense of protection and act as a buffer from attacks from those outside of one’s alliance (DeScioli & Kurzban, 2009). Although nonhuman animals may also form friendships and alliances, the difference for humans is that they are capable of supporting several strong, selective social bonds simultaneously. For example, humans can maintain bonds with their partners, parents, siblings, and close friends. This capability of humans to maintain multiple strong bonds is exemplified by the propensity of humans to form social bonds rather than dissolve existing bonds, as demonstrated by the tendency for humans in all cultures to respond to dissolutions of social bonds with distress (Baumeister & Leary, 1995; Hazan & Shaver, 1994). In addition to the number and strength of social bonds in humans, they are also uniquely motivated to maintain these bonds long term, which allows for the existence of simultaneous strong ­attachments. But, although cognitive approaches have been used since the 1930s to study the maintenance of multiple bonds (Baumeister & Leary, 1995), the physiological mechanisms driving these complex social networks are greatly understudied (Flinn et al., 2012). As detailed in the previous sections, OT and AVP systems may support bonding behaviors, whereas androgens such as testosterone may detract from them. Low levels of testosterone may be important for facilitating affiliation among in-group members, whereas high testosterone concentrations are positively correlated with antisocial behaviors (Flinn et al., 2012). In certain contexts, however, testosterone can promote social-emotional processing, and plays an important role in enhancing sympathetic arousal in  response to angry faces (Eisenegger, Haushofer, & Fehr, 2011). High OT may similarly play a role

in promoting affiliative interactions among in-group members and exclusion of outsiders (De Dreu et al., 2010). Additionally, the coordinated activation of OT and AVP is important for driving both affiliative and aggressive behaviors necessary for maintaining monogamous pair-bonds (Bales et al.,  2004). According to Flinn et al. (2012), all three of these hormones may be necessary to maintain coalitionary relationships in humans. More research is necessary to elucidate how trade-offs between OT, AVP, and testosterone facilitate multiple strong social attachments in humans. Although humans are capable of maintaining several strong bonds simultaneously, many other animals appear capable only of maintaining one strong bond at a time. For example, the primary attachment figure for monogamous titi monkey (Callicebus spp.) infants is their father, as demonstrated by differences in behavioral and physiological responses during a separation study (Hoffman, Mendoza, Hennessy, & Mason, 1995). Infants show increased stress response behaviors and elevated cortisol following separation from their father. Separation from the mother, on the other hand, elicited no such r­ esponse. However, the attachment relationship ­between infant and father titi monkeys is unidirectional, with fathers failing to reciprocate either a behavioral or physiological response to distress from the infant. For the father, the primary attachment figure is his mate (Mendoza & Mason, 1986b). Both male and female titi monkeys show a robust stress response when separated from their pair-mate (Mendoza & Mason, 1986b). Differences in neurobiological activity may explain this lack of ability to maintain multiple strong attachments.

Conclusion

In this chapter, we examined the hormonal trade-offs necessary for balancing (1) mating and parenting behaviors, (2) affiliation with mating and parenting, (3) group stability among dominant and subordinate individuals, and (4) multiple strong, simultaneous bonds across individuals. Hormones important for maintaining social stability include testosterone, cortisol, OT and AVP. The S/P theory of social bonds posits that trade-offs between testosterone, OT, and AVP may facilitate sexual intimacy, nurturant intimacy, antagonistic aggression, and protective aggression. This coordination of hormone levels may allow for the maintenance of aggressive and affiliative behaviors necessary for parent–offspring bonds and adult pair-bonds. Changes in hormone levels

in  response to social cues may also allow for the maintenance of affiliative bonds while facilitating aggressive behaviors necessary for maintaining those bonds. Social dominance is also dependent on context, and high testosterone and cortisol, as well as low testosterone and cortisol, can be related to dominance status. The perception of another individual being part of one’s in-group or out-group also alters hormone functioning underlying interactions between individuals. Finally, humans appear relatively unique among primates in their ability to  balance multiple social bonds simultaneously. Further investigation is necessary to fully understand the mechanisms that cause a switch in hormone levels to facilitate different behaviors necessary for maintaining social bonds.

References

Abbott, D. H., Keverne, E. B., Bercovitch, F. B., Shively, C. A., Mendoza, S. P., Saltzman, W., . . . Sapolsky, R. M. (2003). Are subordinates always stressed? A comparative analysis of rank differences in cortisol levels among primates. Hormones and Behavior, 43(1), 67–82. Abu-Akel, A., Fischer-Shofty, M., Levkovitz, Y., Decety, J., & Shamay-Tsoory, S. (2014). The role of oxytocin in empathy to the pain of conflictual out-group members among patients with schizophrenia. Psychological Medicine, 44(16), 3523–3532. Aguilar, R., Jiménez, M., & Alvero-Cruz, J. R. (2013). Testosterone, cortisol and anxiety in elite field hockey players. Physiology & Behavior, 119, 38–42. Ainsworth, M.  D. (1985). Attachments across the life span. Bulletin of the New York Academy of Medicine, 61(9), 792. Ainsworth, M. D. (1989). Attachments beyond infancy. American Psychologist, 44(4), 709. Alonso, F., Honji, R. M., Moreira, R. G., & Pandolfi, M. (2012). Dominance hierarchies and social status ascent opportunity: Anticipatory behavioral and physiological adjustments in a neotropical cichlid fish. Physiology & Behavior, 106(5), 612–618. Anderson, C., & Kilduff, G.  J. (2009). Why do dominant personalities attain influence in face-to-face groups? The competence-signaling effects of trait dominance. Journal of Personality and Social Psychology, 96(2), 491. Aragona, B. J., Liu, Y., Yu, Y. J., Curtis, J. T., Detwiler, J. M., Insel, T. R., & Wang, Z. (2006). Nucleus accumbens dopamine differentially mediates the formation and maintenance of monogamous pair bonds. Nature Neuroscience, 9(1), 133–139. Archer, J. (1991). The influence of testosterone on human aggression. British Journal of Psychology, 82(1), 1–28. Archer, J. (2006). Testosterone and human aggression: An evalu­ ation of the challenge hypothesis. Neuroscience & Biobehavioral Reviews, 30(3), 319–345. Axelrod, R., & Hamilton, W.  D. (1981). The evolution of cooperation. Science, 211, 1390–1396. Bales, K. L., French, J. A., McWilliams, J., Lake, R. A., & Dietz, J. M. (2006). Effects of social status, age, and season on andro­ gen and cortisol levels in wild male golden lion tamarins

Witcz ak, Simmons, and Bales

345

(Leontopithecus rosalia). Hormones and Behavior, 49(1), 88–95. Bales, K. L., Kim, A. J., Lewis-Reese, A. D., & Carter, C. S. (2004). Both oxytocin and vasopressin may influence alloparental behavior in male prairie voles. Hormones and Behavior, 45(5), 354–361. Bales, K.  L., Mason, W.  A., Catana, C., Cherry, S.  R., & Mendoza, S. P. (2007). Neural correlates of pair bonding in a monogamous primate. Brain Research, 1184, 245–253. Bales, K.  L., Perkeybile, A.  M., Conley, O.  G., Lee, M.  H., Guoynes, C. D., Downing, G. M., . . . Mendoza, S. P. (2013). Chronic intranasal oxytocin causes long-term impairments in partner preference formation in male prairie voles. Biological Psychiatry, 74(3), 180–188. Bales, K. L., van Westerhuyzen, J. A., Lewis-Reese, A. D., Grotte, N. D., Lanter, J. A., & Carter, C. S. (2007). Oxytocin has dose-dependent developmental effects on pair bonding and alloparental care in female prairie voles. Hormones and Behavior, 52(2), 274–279. Barrett, E. S., Tran, V., Thurston, S., Jasienska, G., Furberg, A. S., Ellison, P. T., & Thune, I. (2013). Marriage and motherhood are associated with lower testosterone concentrations in women. Hormones and Behavior, 63(1), 72–79. Barrett, G. M., Shimizu, K., Bardi, M., Asaba, S., & Mori, A. (2002). Endocrine correlates of rank, reproduction, and female-directed aggression in male Japanese macaques (Macaca fuscata). Hormones and Behavior, 42(1), 85–96. Bartoš, L., Schams, D., Bubenik, G. A., Kotrba, R., & Tománek, M. (2010). Relationship between rank and plasma testosterone and cortisol in red deer males (Cervus elaphus). Physiology & Behavior, 101(5), 628–634. Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497. Bayerl, D. S., Hönig, J. N., & Bosch, O. J. (2016). Vasopressin V1a, but not V1b, receptors within the PVN of lactating rats mediate maternal care and anxiety-related behaviour. Behavioural Brain Research, 305, 18–22. Bester-Meredith, J. K., Young, L. J., & Marler, C. A. (1999). Species differences in paternal behavior and aggression in Peromyscus and their associations with vasopressin immunoreactivity and receptors. Hormones and Behavior, 36(1), 25–38. Bigelow, B.  J. (1977). Children’s friendship expectations: A cognitive-developmental study. Child Development, 48, 246–253. Bigelow, B. J., & La Gaipa, J. J. (1975). Children’s written descrip­ tions of friendship: A multidimensional analysis. Developmental Psychology, 11(6), 857. Björntorp, P., Holm, G., & Rosmond, R. (1999). Hypothalamic arousal, insulin resistance and type 2 diabetes mellitus. Diabetic Medicine, 16(5), 373–383. Bosch, O. J. (2011). Maternal nurturing is dependent on her innate anxiety: The behavioral roles of brain oxytocin and vasopressin. Hormones and Behavior, 59, 202–212. Bosch, O. J., Krömer, S. A., Brunton, P. J., & Neumann, I. D. (2004). Release of oxytocin in the hypothalamic paraventricular nucleus, but not central amygdala or lateral septum in lactating residents and virgin intruders during maternal defense. Neuroscience, 124(2), 439–448. Bosch, O. J., Meddle, S. L., Beiderbeck, D. I., Douglas, A. J., & Neumann, I. D. (2005). Brain oxytocin correlates with maternal aggression: link to anxiety. Journal of Neuroscience, 25(29), 6807–6815.

346

Social Bond Paradoxes

Bosch, O. J., & Neumann, I. D. (2008). Brain vasopressin is an important regulator of maternal behavior independent of dams’ trait anxiety. Proceedings of the National Academy of Sciences of the United States of America, 105(44), 17139–17144. Bosch, O.  J., & Neumann, I.  D. (2010). Vasopressin released within the central amygdala promotes maternal aggression. European Journal of Neuroscience, 31(5), 883–891. Bosch, O.  J., Pförtsch, J., Beiderbeck, D.  I., Landgraf, R., & Neumann, I. D. (2010). Maternal behaviour is associated with vasopressin release in the medial preoptic area and bed nucleus of the stria terminalis in the rat. Journal of Neuroendocrinology, 22(5), 420–429. Bowlby, J. (1977a). The making and breaking of affectional bonds: I. Aetiology and psychopathology in the light of attachment theory. The British Journal of Psychiatry, 130(3), 201–210. Bowlby, J. (1977b). The Making and Breaking of Affectional Bonds: II. Some Principles of Psychotherapy: The Fiftieth Maudsley Lecture (expanded version). The British Journal of Psychiatry, 130(5), 421–431. Brown, R. E., Murdoch, T., Murphy, P. R., & Moger, W. H. (1995). Hormonal responses of male gerbils to stimuli from their mate and pups. Hormones and Behavior, 29(4), 474–491. Carlson, A.  A., Young, A.  J., Russell, A.  F., Bennett, N.  C., McNeilly, A.  S., & Clutton-Brock, T. (2004). Hormonal correlates of dominance in meerkats (Suricata suricatta). Hormones and Behavior, 46(2), 141–150. Carp, S.  B., Rothwell, E.  S., Bourdon, A., Freeman, S.  M., Ferrer, E., & Bales, K. L. (2015). Development of a partner preference test that differentiates between established pair bonds and other relationships in socially monogamous titi monkeys (Callicebus cupreus). American Journal of Primatology, 78(3), 326–339. Carré, J. M., & Mehta, P. H. (2011). Importance of considering testosterone–cortisol interactions in predicting human aggres­ sion and dominance. Aggressive Behavior, 37(6), 489–491. Carré, J. M., & Olmstead, N. A. (2015). Social neuroendocrinology of human aggression: Examining the role of competitioninduced testosterone dynamics. Neuroscience, 286, 171–186. Carter, C. S. (1992). Oxytocin and sexual behavior. Neuroscience & Biobehavioral Reviews, 16(2), 131–144. Carter, C. S. (1998). Neuroendocrine perspectives on social attach­ ment and love. Psychoneuroendocrinology, 23(8), 779–818. Carter, C. S. (2007). Sex differences in oxytocin and vasopressin: Implications for autism spectrum disorders? Behavioural Brain Research, 176(1), 170–186. Carter, C.  S., Grippo, A.  J., Pournajafi-Nazarloo, H., Ruscio, M. G., & Porges, S. W. (2008). Oxytocin, vasopressin and sociality. Progress in Brain Research, 170, 331–336. Cho, M.  M., DeVries, A.  C., Williams, J.  R., & Carter, C.  S. (1999). The effects of oxytocin and vasopressin on partner preference in male and female prairie voles (Microtus ochrogaster). Behavioral Neuroscience, 133, 1071–1079. Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98(2), 310. Crockford, C., Wittig, R. M., Langergraber, K., Ziegler, T. E., Zuberbühler, K., & Deschner, T. (2013). Urinary oxytocin and social bonding in related and unrelated wild chimpanzees. Proceedings of the Royal Society Series B, 280(1755), 20122765. De Dreu, C. K., Greer, L. L., Handgraaf, M. J., Shalvi, S., Van Kleef, G. A., Baas, M., . . . Feith, S. W. (2010). The neuropeptide oxytocin regulates parochial altruism in intergroup conflict among humans. Science, 328(5984), 1408–1411.

De Dreu, C.  K., Greer, L.  L., Van Kleef, G.  A., Shalvi, S., &  Handgraaf, M.  J. (2011). Oxytocin promotes human ethnocentrism. Proceedings of the National Academy of Sciences of the United States of America, 108(4), 1262–1266. Delville, Y., Mansour, K. M., & Ferris, C. F. (1996). Testosterone facilitates aggression by modulating vasopressin receptors in the hypothalamus. Physiology & Behavior, 60(1), 25–29. DeScioli, P., & Kurzban, R. (2009). The alliance hypothesis for human friendship. PloS One, 4(6), e5802. Dixson, A.  F., & George, L. (1982). Prolactin and parental behaviour in a male New World primate. Nature, 299(5883), 551–553. Donaldson, Z.  R., Spiegel, L., & Young, L.  J. (2010). Central vasopressin V1a receptor activation is independently necessary for both partner preference formation and expression in socially monogamous male prairie voles. Behavioral Neuroscience, 124(1), 159–163. Dong-Dong, Q. I. N., Rizak, J. D., Xiao-Li, F. E. N. G., Xun-Xun, C. H. U., Shang-Chuan, Y. A. N. G., Chun-Lu, L. I., . . . XinTian, H.  U. (2013). Social rank and cortisol among female rhesus macaques (Macaca mulatta). Zoological Research, 34, E42–E49. Eisenegger, C., Haushofer, J., & Fehr, E. (2011). The role of testosterone in social interaction. Trends in Cognitive Sciences, 15(6), 263–271. Fahrbach, S. E., Morrell, J. I., & Pfaff, D. W. (1984). Oxytocin induction of short-latency maternal behavior in nulliparous, estrogen-primed female rats. Hormones and Behavior, 18(3), 267–286. Feldman, R. (2007). Parent-infant synchrony and the construction of shared timing: Physiological precursors, developmental outcomes, and risk conditions. Journal of Child Psychology and Psychiatry, 48, 329–354. Feldman, R. (2012a). Bio-behavioral synchrony: A model for integrating biological and microsocial behavioral processes in the study of parenting. Parenting, 12(2–3), 154–164. Feldman, R. (2012b). Oxytocin and social affiliation in humans. Hormones and Behavior, 61, 380–391. Feldman, R., Gordon, I., Schneiderman, I., Weisman, O., & Zagoory-Sharon, O. (2010). Natural variations in maternal and paternal care are associated with systematic changes in oxytocin following parent–infant contact. Psychoneuroendocrinology, 35(8), 1133–1141. Feldman, R., Gordon, I., & Zagoory-Sharon, O. (2011). Maternal and paternal plasma, salivary, and urinary oxytocin and parent– infant synchrony: Considering stress and affiliation components of human bonding. Developmental Science, 14(4), 752–761. Feldman, R., Weller, A., Zagoory-Sharon, O., & Levine, A. (2007). Evidence for a neuroendocrinological foundation of human affiliation plasma oxytocin levels across pregnancy and the postpartum period predict mother-infant bonding. Psychological Science, 18(11), 965–970. Feng, X., Wu, X., Morrill, R. J., Li, Z., Li, C., Yang, S., . . . Hu, X. (2016). Social correlates of the dominance rank and long-term cortisol levels in adolescent and adult male rhesus macaques (Macaca mulatta). Scientific Reports, 6, 25431. Fernandez-Duque, E., Valeggia, C. R., & Mason, W. A. (2000). Effects of pair bond and social context on male-female interac­ tions in captive titi monkeys (Callicebus moloch, Primates: Cebidae). Ethology, 106(12), 1067–1082. Fernandez-Duque, E., Valeggia, C. R., & Mendoza, S. P. (2009). The biology of paternal care in human and nonhuman primates. Annual Review of Anthropology, 38, 115–130.

Finkenwirth, C., van Schaik, C., Ziegler, T. E., & Burkart, J. M. (2015). Strongly bonded family members in common marmo­ sets show synchronized fluctuations in oxytocin. Physiology & Behavior, 151, 246–251. Fisher-Phelps, M.  L., Mendoza, S.  P., Serna, S., Griffin, L.  L., Schaefer, T.  J., Jarcho, M.  R., . . . Bales, K.  L. (2015). Laboratory simulations of mate-guarding as a component of the pair bond in male titi monkeys, Callicebus cupreus. American Journal of Primatology, 78(5), 573–582. Fleming, A. S., Ruble, D., Krieger, H., & Wong, P. Y. (1997). Hormonal and experiential correlates of maternal respon­ siveness during pregnancy and puerperium in human mothers. Hormones and Behavior, 31, 145–158. Flinn, M. V., Ponzi, D., & Muehlenbein, M. P. (2012). Hormonal mechanisms for regulation of aggression in human coalitions. Human Nature, 23(1), 68–88. Fraser, R., Ingram, M.  C., Anderson, N.  H., Morrison, C., Davies, E., & Connell, J. M. (1999). Cortisol effects on body mass, blood pressure, and cholesterol in the general population. Hypertension, 33(6), 1364–1368. French, J. A. (1986). Encounters with unfamiliar conspecifics in the laboratory: What do species and context differences reveal about callitrichid social structure. Primate Report, 14, 75. Gimpl, G., & Fahrenholz, F. (2001). The oxytocin receptor system: Structure, function, and regulation. Physiological Reviews, 81, 629–683. Golding, D. N., Murray, S. M., Pearce, G. W., & Thompson, M. (1961). Corticosteroid myopathy. Rheumatology, 6(4), 171–177. Goodson, J. L., & Bass, A. H. (2001). Social behavior functions and related anatomical characteristics of vasotocin/vasopressin systems in vertebrates. Brain Research. Brain Research Reviews, 35(3), 246–265. Gouin, J. P., Carter, C. S., Pournajafi-Nazarloo, H., Glaser, R., Malarkey, W. B., Loving, T. J., . . . Kiecolt-Glaser, J. K. (2010). Marital behavior, oxytocin, vasopressin, and wound healing. Psychoneuroendocrinology, 35(7), 1082–1090. Gray, P. B., Chapman, J. F., Burnham, T. C., McIntyre, M. H., Lipson, S.  F., & Ellison, P.  T. (2004). Human male pair bonding and testosterone. Human Nature, 15, 119–131. Gubernick, D. J. (1988). Reproduction in the California mouse, Peromyscus californicus. Journal of Mammalogy, 69(4), 857–860. Gubernick, D. J., & Nelson, R. J. (1989). Prolactin and paternal behavior in the biparental California mouse, Peromyscus californicus. Hormones and Behavior, 23(2), 203–210. Gubernick, D. J., & Nordby, J. C. (1993). Mechanisms of sexual fidelity in the monogamous California mouse, Peromyscus californicus. Behavioral Ecology and Sociobiology, 32(3), 211–219. Güroğlu, B., Haselager, G. J., van Lieshout, C. F., Takashima, A., Rijpkema, M., & Fernández, G. (2008). Why are friends special? Implementing a social interaction simulation task to probe the neural correlates of friendship. Neuroimage, 39(2), 903–910. Hazan, C., & Shaver, P. R. (1994). Attachment as an organizational framework for research on close relationships. Psychological Inquiry, 5(1), 1–22. Heinrichs, M., von Dawans, B., & Domes, G. (2009). Oxytocin, vasopressin, and human social behavior. Frontiers in Neuroendo­ crinology, 30(4), 548–557. Hoffman, K. A., Mendoza, S. P., Hennessy, M. B., & Mason, W. A. (1995). Responses of infant titi monkeys, Callicebus moloch, to removal of one or both parents: Evidence for

Witcz ak, Simmons, and Bales

347

paternal attachment. Developmental Psychobiology, 28(7), 399–407. Hubrecht, R. C. (1985). Home-range size and use and territorial behavior in the common marmoset, Callithrix jacchus, at the Tapacura Field Station, Recife, Brazil. International Journal of Primatology, 6(5), 533–550. Insel, T. R. (1997). A neurobiological basis of social attachment. American Journal of Psychiatry, 154, 726–735. Insel, T. R., & Hulihan, T. J. (1995). A gender-specific mechanism for pair bonding: Oxytocin and partner preference formation in monogamous voles. Behavioral Neuroscience, 109(4), 782–789. Insel, T.  R., Winslow, J.  T., Wang, Z., & Young, L.  J. (1998). Oxytocin, vasopressin, and the neuroendocrine basis of pair bond formation. In H.  H.  Zingg, C.  W.  Bourque, & D. G. Bichet (Eds.), Vasopressin and oxytocin (pp. 215–224). Boston, MA: Springer. Insel, T.  R., & Young, L.  J. (2001). The neurobiology of attachment. Nature Reviews Neuroscience, 2(2), 129–136. Israel, S., Lerer, E., Shalev, I., Uzefovsky, F., Riebold, M., Laiba, E., . . . Ebstein, R. P. (2009). The oxytocin receptor (OXTR) contributes to prosocial fund allocations in the dictator game and the social value orientations task. PloS One, 4(5), e5535. Kendrick, K. M. (2000). Oxytocin, motherhood and bonding. Experimental Physiology, 85, 111S–124S. Kendrick, K.  M., Keverne, E.  B., & Baldwin, B.  A. (1987). Intracerebroventricular oxytocin stimulates maternal behaviour in the sheep. Neuroendocrinology, 46(1), 56–61. Kessler, M.  S., Bosch, O.  J., Bunck, M., Landgraf, R., & Neumann, I. D. (2011). Maternal care differs in mice bred for high vs. low trait anxiety: Impact of brain vasopressin and cross-fostering. Social Neuroscience, 6(2), 156–168. Kholkute, S. D. (1984). Plasma estradiol and progesterone levels during post partum ovulation and early pregnancy in the com­ mon marmoset Callithrix jacchus. Primates, 25(4), 538–543. Kinnally, E. L., & Capitanio, J. P. (2015). Paternal early experiences influence infant development through non-social mechanisms in Rhesus macaques. Frontiers in Zoology, 12(s1), s14. Kleiman, D.  G. (1977). Monogamy in mammals. Quarterly Review of Biology, 52(1), 39–69. Koenig, A., & Rothe, H. (1994). Effects of familiarity on the behaviour towards intruders in captive common marmosets (Callithrix jacchus). Primates, 35(1), 89–93. Kuzawa, C. W., Gettler, L. T., Huang, Y. Y., & McDade, T. W. (2010). Mothers have lower testosterone than non-mothers: Evidence from the Philippines. Hormones and Behavior, 57(4), 441–447. Kuzawa, C. W., Gettler, L. T., Muller, M. N., McDade, T. W., & Feranil, A. B. (2009). Fatherhood, pairbonding and testosterone in the Philippines. Hormones and Behavior, 56(4), 429–435. Lim, M. M., & Young, L. J. (2006). Neuropeptidergic regulation of affiliative behavior and social bonding in animals. Hormones and Behavior, 50, 506–517. Lynn, S.  E., Hayward, L.  S., Benowitz-Fredericks, Z.  M., &  Wingfield, J.  C. (2002). Behavioural insensitivity to supplementary testosterone during the parental phase in the chestnut-collared longspur, Calcarius ornatus. Animal Behaviour, 63(4), 795–803. Madden, J.  R., & Clutton-Brock, T.  H. (2010). Experimental peripheral administration of oxytocin elevates a suite of cooperative behaviours in a wild social mammal. Proceedings of the Royal Society of London B: Biological Sciences, 278(1709), 1189–1194.

348

Social Bond Paradoxes

Maestripieri, D., & Georgiev, A. V. (2016). What cortisol can tell us about the costs of sociality and reproduction among freeranging rhesus macaque females on Cayo Santiago. American Journal of Primatology, 78(1), 92–105. Maestripieri, D., Klimczuk, A. C., Seneczko, M., Traficonte, D. M., & Wilson, M. C. (2013). Relationship status and relationship instability, but not dominance, predict individ­ ual differences in baseline cortisol levels. PloS One, 8(12), e84003. Magee, J. C., & Galinsky, A. D. (2008). 8 Social hierarchy: The self-reinforcing nature of power and status. Academy of Management Annals, 2(1), 351–398. Marshall, A. D. (2013). Posttraumatic stress disorder and partnerspecific social cognition: A pilot study of sex differences in the impact of arginine vasopressin. Biological Psychology, 93(2), 296–303. Maruska, K. P. (2015). Social transitions cause rapid behavioral and neuroendocrine changes. Integrative and Comparative Biology, 55(2), 294–306. Mason, W. A. (1966). Social organization of the South American monkey, Callicebus moloch: A preliminary report. Tulane Studies in Zoology, 13, 23–28. Mason, W.  A., & Sponholz, R.  R. (1963). Behavior of rhesus monkeys raised in isolation. Journal of Psychiatric Research, 1(4), 299–306. McEwen, B. S. (2004). Protection and damage from acute and chronic stress: Allostasis and allostatic overload and relevance to the pathophysiology of psychiatric disorders. Annals of the New York Academy of Sciences, 1032(1), 1–7. Mehta, P. H., & Josephs, R. A. (2010). Testosterone and cortisol jointly regulate dominance: Evidence for a dual-hormone hypothesis. Hormones and Behavior, 58(5), 898–906. Mendoza, S. P., & Mason, W. A. (1986a). Contrasting responses to intruders and to involuntary separation by monogamous and polygynous New World monkeys. Physiology & Behavior, 38(6), 795–801. Mendoza, S. P., & Mason, W. A. (1986b). Parental division of labour and differentiation of attachments in a monogamous primate (Callicebus moloch). Animal Behaviour, 34(5), 1336–1347. Muehlenbein, M.  P., & Watts, D.  P. (2010). The costs of dominance: Testosterone, cortisol and intestinal parasites in wild male chimpanzees. BioPsychoSocial Medicine, 4(1), 1. Nephew, B.  C., & Bridges, R.  S. (2008). Central actions of arginine vasopressin and a V1a receptor antagonist on maternal aggression, maternal behavior, and grooming in lactating rats. Pharmacology Biochemistry and Behavior, 91(1), 77–83. Nephew, B. C., Byrnes, E. M., & Bridges, R. S. (2010). Vasopressin mediates enhanced offspring protection in multiparous rats. Neuropharmacology, 58(1), 102–106. Nishitani, S., Ikematsu, K., Takamura, T., Honda, S., Yoshiura, K. I., & Shinohara, K. (2017). Genetic variants in oxytocin receptor and arginine-vasopressin receptor 1A are associated with the neural correlates of maternal and paternal affection towards their child. Hormones and Behavior, 87, 47–56. Numan, M. (2015). Neurobiology of social behavior: Toward an understating of the prosocial and antisocial brain. London, UK: Elsevier. Numan, M., & Insel, T. R. (2003). The neurobiology of parental behaviour. In G. F. Ball, J. Balthazart, & R. J. Nelson (Eds.), Hormones, brain, and behavior series. New York, NY: Springer. Nunes, S., Fite, J. E., & French, J. A. (2000). Variation in steroid hormones associated with infant care behaviour and experience

in male marmosets (Callithrix kuhlii). Animal Behaviour, 60(6), 857–865. Pavlidis, M., Sundvik, M., Chen, Y.  C., & Panula, P. (2011). Adaptive changes in zebrafish brain in dominant–subordinate behavioral context. Behavioural Brain Research, 225(2), 529–537. Pedersen, C.  A., Caldwell, J.  D., Walker, C., Ayers, G., & Mason, G.  A. (1994). Oxytocin activates the postpartum onset of rat maternal behavior in the ventral tegmental and medial preoptic areas. Behavioral Neuroscience, 108(6), 1163. Pfattheicher, S. (2017). Illuminating the dual-hormone hypothesis: About chronic dominance and the interaction of cortisol and testosterone. Aggressive Behavior, 43(1), 85–92. Prior, N. H., Yap, K. N., Adomat, H. H., Mainwaring, M. C., Fokidis, H. B., Guns, E. S., . . . Soma, K. K. (2016). Sex steroid profiles and pair-maintenance behavior of captive wild-caught zebra finches (Taeniopygia guttata). Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology, 202, 35–44. Proctor, C. M., Freeman, E. W., & Brown, J. L. (2010). Influence of dominance status on adrenal activity and ovarian cyclicity status in captive African elephants. Zoo Biology, 29(2), 168–178. Prudom, S. L., Broz, C. A., Schultz-Darken, N., Ferris, C. T., Snowdon, C., & Ziegler, T.  E. (2008). Exposure to infant scent lowers serum testosterone in father common marmosets (Callithrix jacchus). Biological Letters, 4, 603–605. Riedman, M. L. (1982). The evolution of alloparental care and adoption in mammals and birds. Quarterly Review of Biology, 57(4), 405–435. Rilling, J. K., & Young, L. J. (2014). The biology of mammalian parenting and its effect on offspring social development. Science, 345(6198), 771–776. Rodrigues, S.  M., Saslow, L.  R., Garcia, N., John, O.  P., & Keltner, D. (2009). Oxytocin receptor genetic variation relates to empathy and stress reactivity in humans. Proceedings of the National Academy of Sciences of the United States of America, 106(50), 21437–21441. Ross, H. E., Freeman, S. M., Spiegel, L. L., Ren, X., Terwilliger, E. F., & Young, L. J. (2009). Variation in oxytocin receptor density in the nucleus accumbens has differential effects on affiliative behaviors in monogamous and polygamous voles. Journal of Neuroscience, 29(5), 1312–1318. Ross, H.  E., & Young, L.  J. (2009). Oxytocin and the neural mechanisms regulating social cognition and affiliative behavior. Frontiers in Neuroendocrinology, 30(4), 534–547. Saito, A., & Nakamura, K. (2011). Oxytocin changes primate paternal tolerance to offspring in food transfer. Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology, 197(4), 329–337. Sakaguchi, K., Tanaka, M., Ohkubo, T., Doh-Ura, K., Fujikawa, T., Sudo, S., & Nakashima, K. (1996). Induction of brain prolactin receptor long-form mRNA expression and maternal behavior in pup-contacted male rats: Promotion by prolactin administration and suppression by female contact. Neuroen­ docrinology, 63(6), 559–568. Sapolsky, R.  M. (1990). Adrenocortical function, social rank, and personality among wild baboons. Biological Psychiatry, 28(10), 862–878. Sapolsky, R. M. (1992). Cortisol concentrations and the social significance of rank instability among wild baboons. Psychoneuroendocrinology, 17(6), 701–709. Sapolsky, R.  M. (2005). The influence of social hierarchy on primate health. Science, 308(5722), 648–652.

Sapolsky, R.  M., Alberts, S.  C., & Altmann, J. (1997). Hypercortisolism associated with social subordinance or social isolation among wild baboons. Archives of General Psychiatry, 54(12), 1137–1143. Schradin, C., Scantlebury, M., Pillay, N., & König, B. (2009). Testosterone levels in dominant sociable males are lower than in solitary roamers: Physiological differences between three male reproductive tactics in a sociably flexible mammal. American Naturalist, 173(3), 376–388. Shively, C. A., Laber-Laird, K., & Anton, R. F. (1997). Behavior and physiology of social stress and depression in female cynomolgus monkeys. Biological Psychiatry, 41(8), 871–882. Smith, A. S., Ågmo, A., Birnie, A. K., & French, J. A. (2010). Manipulation of the oxytocin system alters social behavior and attraction in pair bonding primates, Callithrix penicillata. Hormones and Behavior, 57(2), 255–262. Snowdon, C. T., Pieper, B. A., Boe, C. Y., Cronin, K. A., Kurian, A. V., & Ziegler, T. E. (2010). Variation in oxytocin is related to variation in affiliative behavior in monogamous, pairbonded tamarins. Hormones and Behavior, 58(4), 614–618. Snowdon, C. T., Ziegler, T. E., Schultz-Darken, N. J., & Ferris, C. F. (2006). Social odours, sexual arousal and pairbonding in primates. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 361(1476), 2079–2089. Stiver, K. A., & Alonzo, S. H. (2009). Parental and mating effort: Is there necessarily a trade-off? Ethology, 115(12), 1101–1126. Taborsky, M., & Limberger, D. (1981). Helpers in fish. Behavioral Ecology and Sociobiology, 8(2), 143–145. Taylor, S. E., Saphire-Bernstein, S., & Seeman, T. E. (2010). Are plasma oxytocin in women and plasma vasopressin in men biomarkers of distressed pair bond relationships? Psychological Science, 21(1), 3–7. Teichroeb, J. A., & Sicotte, P. (2008). Social correlates of fecal testosterone in male ursine colobus monkeys (Colobus vellerosus): The effect of male reproductive competition in aseasonal breeders. Hormones and Behavior, 54(3), 417–423. Tomonaga, M., Tanaka, M., Matsuzawa, T., Myowa-Yamakoshi, M., Kosugi, D., Mizuno, Y.,  .  .  .  Bard, K.  A. (2004). Development of social cognition in infant chimpanzees (Pan troglodytes): Face recognition, smiling, gaze, and the lack of triadic interactions. Japanese Psychological Research, 46(3), 227–235. Trainor, B.  C., & Marler, C.  A. (2001). Testosterone, paternal behavior, and aggression in the monogamous California mouse (Peromyscus californicus). Hormones and Behavior, 40(1), 32–42. Trainor, B. C., & Marler, C. A. (2002). Testosterone promotes paternal behaviour in a monogamous mammal via conversion to oestrogen. Proceedings of the Royal Society of London B: Biological Sciences, 269(1493), 823–829. Turan, B., Tackett, J. L., Lechtreck, M. T., & Browning, W. R. (2015). Coordination of the cortisol and testosterone responses: A dual axis approach to understanding the response to social status threats. Psychoneuroendocrinology, 62, 59–68. van Anders, S.  M. (2009). Androgens and diversity in adult human partnering. In P.  T.  Ellison & P.  B.  Gray (Eds.), Endocrinology of Social Relationships (pp. 340–363). Cambridge, MA: Harvard University Press. van Anders, S. M., Goldey, K. L., & Kuo, P. X. (2011). The steroid/ peptide theory of social bonds: Integrating testosterone and peptide responses for classifying social behavioral contexts. Psychoneuroendocrinology, 36(9), 1265–1275.

Witcz ak, Simmons, and Bales

349

van Anders, S. M., & Gray, P. B. (2007). Hormones and human partnering. Annual Review of Sex Research, 18(1), 60–93. van Anders, S. M., Hamilton, L. D., Schmidt, N., & Watson, N. V. (2007). Associations between testosterone secretion and sexual activity in women. Hormones and Behavior, 51(4), 477–482. van Anders, S. M., Tolman, R. M., & Volling, B. L. (2012). Baby cries and nurturance affect testosterone in men. Hormones and Behavior, 61(1), 31–36. van Anders, S. M., & Watson, N. V. (2006a). Relationship status and testosterone in North American heterosexual and nonheterosexual men and women: Cross-sectional and longitudinal data. Psychoneuroendocrinology, 31(6), 715–723. van Anders, S.  M., & Watson, N.  V. (2006b). Social neuroendocrinology. Human Nature, 17(2), 212–237. Van Roo, B.  L., Ketterson, E.  D., & Sharp, P.  J. (2003). Testosterone and prolactin in two songbirds that differ in paternal care: The blue-headed vireo and the red-eyed vireo. Hormones and Behavior, 44(5), 435–441. Veenema, A.  H. (2009). Early life stress, the development of aggression and neuroendocrine and neurobiological correlates: What can we learn from animal models? Frontiers in Neuroendocrinology, 30, 497–518. Veenema, A.  H., Blume, A., Niederle, D., Buwalda, B., & Neumann, I. D. (2006). Effects of early life stress on adult male aggression and hypothalamic vasopressin and serotonin. European Journal of Neuroscience, 24(6), 1711–1720. Virgin, C.  E., & Sapolsky, R.  M. (1997). Styles of male social behavior and their endocrine correlates among low-ranking baboons. American Journal of Primatology, 42(1), 25–39. Wang, Z., Ferris, C. F., & De Vries, G. J. (1994). Role of septal vasopressin innervation in paternal behavior in prairie voles (Microtus ochrogaster). Proceedings of the National Academy of Sciences of the United States of America, 91(1), 400–404. Weisman, O., Zagoory-Sharon, O., & Feldman, R. (2012). Oxytocin administration to parent enhances infant physiological and behavioral readiness for social engagement. Biological Psychiatry, 72(12), 982–989. Wingfield, J.  C. (2017). The challenge hypothesis: Where it began and relevance to humans. Hormones and Behavior, 92, 9–12. Wingfield, J. C., Hegner, R. E., Dufty, A. M., Jr., & Ball, G. F. (1990). The “challenge hypothesis”: Theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. American Naturalist, 136(6), 829–846. Wingfield, J. C., Lynn, S., & Soma, K. K. (2001). Avoiding the “costs” of testosterone: Ecological bases of hormone-behavior interactions. Brain, Behavior and Evolution, 57(5), 239–251.

350

Social Bond Paradoxes

Winslow, J. T., Hastings, N., Carter, C. S., Harbaugh, C. F., & Insel, T.  R. (1993). A role for central vasopressin in pair bonding in monogamous prairie voles. Nature, 365, 545–548. Wittenberger, J.  F., & Tilson, R.  L. (1980). The evolution of monogamy: Hypotheses and evidence. Annual Review of Ecology and Systematics, 11, 197–232. Wynne-Edwards, K.  E., & Timonin, M.  E. (2007). Paternal care in rodents: Weakening support for hormonal regulation of the transition to behavioral fatherhood in rodent animal models of biparental care. Hormones and Behavior, 52(1), 114–121. Young, K.  A., Liu, Y., & Wang, Z. (2008). The neurobiology of social attachment: A comparative approach to behavioral, neuroanatomical, and neurochemical studies. Comparative Biochemistry and Physiology Part C: Toxicology & Pharmacology, 148(4), 401–410. Young, L. J., Wang, Z., & Insel, T. R. (1998) Neuroendocrine bases of monogamy. Trends in Neurosciences, 21, 71–75. Ziegler, T.  E., Bridson, W.  E., Snowdon, C.  T., & Eman, S. (1987). Urinary gonadotropin and estrogen excretion during the postpartum estrus, conception, and pregnancy in the cotton-top tamarin (Saguinus oedipus oedipus). American Journal of Primatology, 12(2), 127–140. Ziegler, T.  E., Prudom, S.  L., Zahed, S.  R., Parlow, A.  F., & Wegner, F. (2009). Prolactin’s mediative role in male parenting in parentally experienced marmosets (Callithrix jacchus). Hormones and Behavior, 56(4), 436–443. Ziegler, T.  E., Scheffler, G., & Snowdon, C.  T. (1995). The relationship of cortisol levels to social environment and reproductive functioning in female cotton-top tamarins, Saguinus oedipus. Hormones and Behavior, 29(3), 407–424. Ziegler, T.  E., Wegner, F.  H., Carlson, A.  A., Lazaro-Perea, C., &  Snowdon, C.  T. (2000). Prolactin levels during the periparturitional period in the biparental cotton-top tamarin (Saguinus oedipus): Interactions with gender, androgen levels, and parenting. Hormones and Behavior, 38(2), 111–122. Zilioli, S., & Watson, N. V. (2012). The hidden dimensions of the competition effect: Basal cortisol and basal testosterone jointly predict changes in salivary testosterone after social victory in men. Psychoneuroendocrinology, 37(11), 1855–1865. Zilioli, S., & Watson, N.  V. (2013). Winning isn’t everything: Mood and testosterone regulate the cortisol response in competition. PLoS One, 8(1), e52582. Zimmermann-Peruzatto, J. M., Lazzari, V. M., de Moura, A. C., Almeida, S., & Giovenardi, M. (2015). Examining the role of vasopressin in the modulation of parental and sexual behaviors. Frontiers in Psychiatry, 6, 130.

CH A PT E R

20

Stress Hormones, Physiology, and Behavior

Justin K. Mogilski, Anna Wysocki, Simon D. Reeve, Virginia E. Mitchell, Jenna Lunge, and Lisa L. M. Welling

Abstract Stress, be it physical or psychological, can have a devastating long-term impact on an individual’s development, health, and well-being, and yet can be adaptive in the short term (e.g., promoting immediate survival, triggering the desire to remedy social conflict). The stress response system involves physiological processes in reaction to a real or perceived threat, which serve a variety of purposes. This chapter reviews pertinent topics and research within the social neuroendocrine study of stress, including acute versus chronic stress, and how stress influences social behavior and status. Where appropriate, it offers critiques of current theoretical models and includes suggestions for future directions within this research area. Key words:  stress, hormones, glucocorticoids, cortisol, health, behavior

Stress is any real or interpreted threat to the ­physiological or psychological integrity of an individual that results in physiological and/or behavioral responses (McEwen, 2000). It occurs in reaction to somatic challenges such as competition, environmental harshness, and disease, and is mediated by catecholamines (e.g., epinephrine and norepinephrine) and glucocorticoids (GCs; e.g., cortisol in humans and fish; corticosterone in amphibians, reptiles, and birds). GCs are a class of corticosteroids, which are steroid hormones that are released by the adrenal cortex as part of a physiological feedback system known as the hypothalamicpituitary-­adrenal (HPA) axis. The HPA axis regulates so-called stress hormones through positive and negative feedback systems initiated by neural input to the hypothalamus. Initial hypothalamic release of corticotropin-releasing hormone (CRH) to the anterior pituitary causes release of adrenocorticotropic hormone (ACTH) to the adrenal cortex. Next, GCs  produced by the adrenal cortex enter the bloodstream and promote a number of physiological changes (e.g., increased respiration, blood

o­ xygenation, glucose release), which divert energy from nonessential processes, such as reproduction and food digestion, to those that promote imme­ diate survival (e.g., muscle innervation, threat processing). Although the stress response can be adaptive in the short term, it can have maladaptive consequences for long-term functioning and health. In this chapter, we review several of the most well-researched topics within the social neuroendocrine study of stress, provide an overview and critical analysis of current theoretical models (where appropriate), and provide a summary of potential future directions within the current literature.

Acute Stress Response

The stress response is vital to an organism’s survival and has coevolved with a number of other physiological systems (e.g., dopaminergic reward systems, attachment and social buffering systems) to permit dynamic redistribution of somatic resources when the demands of internal or external events exceed immediately available resources (see, e.g., Lazarus & 351

Folkman,  1984). Acute stress can trigger survival-­ related and other adaptive behaviors, such as increases in attention and memory. For example, amygdala arousal from watching emotional films enhances episodic memory for these films compared to neutral controls (Cahill et al., 1996). Acute stress also reduces attention toward irrelevant information during a Stroop task (Booth & Sharma, 2009), suggesting that acute stress redirects attention toward temporally imperative information, but reduces more flexible attentional processing (Gagnon & Wagner, 2016). Similarly, other work has shown that acute stress impairs recall of complex informational arrays and hinders performance during delayed recall and memory tasks (Olver, Pinney, Maruff, & Norman, 2015). Together, this evidence suggests that acute stress functionally redirects cognitive effort toward immediate threats to enhance immediate survival at the cost of dampening long-term neural plasticity (Farmer, Park, Bullard, & Diamond, 2014). However, acute stress can also act as a prompt to compel an organism to act in several ways, depending on various personal and contextual factors. Research has largely focused on two specific threat responses: the fight-or-flight and the tend-and-befriend threat responses.

Fight-or-Flight Versus Tend-and-Befriend

Throughout evolutionary history, humans and other organisms have recurrently encountered situations that required split-second decision making (e.g., an approaching predator, an aggressive competitor). Stressful situations present an energy allocation problem: one may either divert resources to (1) confronting the threat (if there is a high probability of overpowering the threat) or (2) fleeing from it (when the threat is likely to win). This process, known as the fight-or-flight response, innervates the body to respond to immediate stress and is thus adaptive in situations of acute stress (Cannon, 1929). Upon recognition of a stressor, the sympathetic nervous system activates the fight-or-flight response (i.e., the sympathetic adrenal medullary [SAM] ­response) to begin redirecting energy. First, the dorsomedial amygdalar complex recognizes that a potential threat is near and sends neural impulses to the lateral and posterior hypothalamic regions (Roldan, AlvarezPelaez, & de Molina, 1974). The hypothalamus signals to the adrenergic neurons of the thoracic spinal cord to secrete norepinephrine, and to the chromaffin cells of the adrenal medullae to produce catecholamines in a process known as catecholaminogenesis (Everly & Lating,  2013). The adrenal medullary 352

catecholamines epinephrine and norepinephrine increase respiration, heart rate, and blood pressure, and improve blood flow to muscles (Everly & Lating, 2013). Likewise, they increase blood glucose levels and blood oxygenation and improve alertness, learning, and memory (Usdin, Kvetnansky, & Kopin, 1976). During this process, the adaptive fight-or-flight response halts biological functions that are unnecessary for immediate survival, including reproductive efforts, bladder muscle innervation, digestion, and blood flow to the skin and kidneys (Everly & Lating, 2013). Conversely, fight-or-flight increases heart rate to improve circulation, rate and depth of breathing to improve the exchange of gases, synthesis of glucose to provide energy, blood perfusion of muscles to increase strength and endurance, and blood clot reduction to minimize tissue damage (Nesse, Bhatnagar, & Ellis, 2016). Although it was originally believed that the fight-or-flight response decreased the function of the immune system to conserve energy (similar to its suppression of digestion and ovulation at times of stress), there is expanding evidence that immune responses actually increase in response to specific types of stressors (e.g., Dhabhar, 2002). Because some types of stress can lead to infection (e.g., an injury), immune system enhancement would be most beneficial for survival after specific types of acute stress such as an infection or injury (e.g., Dhabhar & McEwen, 2001). However, long-term exposure to this process may also intensify autoimmune and inflammatory diseases (Dhabhar, 2002), and thus, there likely exists an optimal balance that encourages the immuneenhancing or -suppressive responses depending on the situation. Indeed, during acute stressors like injury, glucocorticoids can enhance immune functioning, purportedly to prepare the immune system for potential infection (e.g., Dhabhar, Miller, Stein, McEwen, & Spencer, 1994). However, upon exposure to acute stressors such as a final examination (Glaser, Kiecolt-Glaser, Speicher, & Holliday, 1985) or life change (Cohen-Cole, Cogen, & Stevens, 1981), the immune system suppresses its response to conserve the energy needed to navigate the ­immediate stressor (McEwen et al., 1997). Overall, rather than acting specifically as an immune suppressant (e.g., Auphan, DiDonato, Rosette, Helmberg, & Karin, 1995), it appears glucocorticoids modulate their function to enhance or suppress the immune system response depending on the type of stress experienced (McEwen et al., 1997).

Stress Hormones, Physiology, and Behavior

Similar to fight-or-flight, the tend-and-befriend threat response is a hormone-driven behavioral ­reaction to stress that typically occurs more often in females (Taylor et al., 2000). Via the release of oxytocin (OT), this response manifests as an increased motivation to protect offspring (tend) and seek out social group members for mutual defense and aid (befriend). Testosterone and arginine v­asopressin (AVP) released during the fight-or-flight stress ­response have been found to exhibit the opposite effects of OT (see Taylor, 2006; Taylor et al., 2000, 2006). The original model developed by Taylor et al. (2000) describes tend-and-befriend as a female counterpart to the fight-or-flight response in males (Cannon, 1929). These authors highlight that, before 1995, research investigating the fight-or-flight ­response had been conducted predominantly on male participants, with females constituting a mere 17 percent of the participants, which may have led researchers to overlook the more female-typical tendand-befriend response. In response to threat, both males and females show increased activation of the autonomic ­nervous system, which causes the release of OT, AVP, and CRH. However, OT is released in greater quantities in females (Taylor et al., 2000). OT is believed by some to encourage affiliative behavior, including maternal nurturance and seeking social contact from peers (e.g., Insel, 1997; Carter, 1998; see also Grebe & Gangestad, this volume), and may actively alleviate the biological stress responses by, for ­example, decreasing heart rate, blood pressure, HPA activity (Light et al.,  2000), and cortisol level (Uvnas-Moberg, 1997, 1998). In addition, estrogen enhances the effects of OT, whereas androgens inhibit OT release (McCarthy, 1994), which may contribute to why women are more likely to respond to stress via tend-and-befriend than men. Certainly, women who report more deficiencies in their social relationships (i.e., reduced contact with various social support sources) and marital dissatisfaction have elevated OT levels (Taylor et al., 2006). Overall, women consistently show a stronger affiliative response to stress than men do (Tamres, Janicki, & Helgeson, 2002; Taylor et al., 2000), and therefore a social-support-seeking response to stress may be particularly adaptive for females. Both men and women rely on group living for successful ­defense against predators and outgroup members, but human females also face greater threats from in-group human males (e.g., rape, assault, abuse of offspring). Furthermore, women often take on more

parental responsibility for early offspring care, and pregnancy and nursing make women especially vulnerable to external threats (see Sear & Mace, 2008). Forming a network not only provides protection and help with raising offspring, but also serves to secure resources, such as housing and food. Given that a group is more likely than an individual to overcome or deter a threat, a social-support-seeking response is likely a protective mechanism for both the woman and her offspring. Correspondingly, evidence suggests that women, more than men, are geared for fostering and maintaining social relationships. For example, although the need for interpersonal connection seems to be a near-universal human trait, women tend to socialize more in new environments (Wheeler & Nezlek,  1977) and are more focused on maintaining belongingness than men (Baumeister & Leary, 1995). Even as children, girls have a stronger interest in maintaining meaningful and nurturing relationships, resulting in a higher number of relationships than male counterparts (Galambos,  2004; Nichols & Good,  1998). Also, women tend to score higher on measures of emotional intelligence and social skills than men (Bindu & Thomas,  2006; Petrides & Furnham,  2000), suggesting that women have specialized cognitive mechanisms for maintaining affiliation. Together, this research suggests that women compared to men are more likely to overtly prioritize social support responses to stress over fight-or-flight responses (e.g., Turton & Campbell, 2005) because evolutionary pressures favored an affiliative response. Although, as noted, OT is associated with seeking social support in response to stress, which can in turn alleviate stress, it is important to emphasize that the social environment itself can also be the source of stress. For example, cortisol has been shown to increase after social rejection in both men and women (Blackhart, Eckel, & Tice, 2007), although women show elevated cortisol responses to social rejection versus achievement failure manipulations compared to men (Stroud, Salovey, & Epel, 2002), suggesting they may be more physiologically reactive to social rejection than men. The relationship between OT and cortisol led Taylor (2006) to propose a potential model for a stress affiliation, which is ­dependent on the success or failure of gaining social support via the tend-and-befriend stress ­response (see Figure 20.1 for a simplified iteration of Taylor’s [2006] conceptual model). Although the buffering influence of OT on stress is discussed in more detail later (see Social Buffering), the model

Mogilski, Wysocki, Reeve, Mitchell, Lunge, and Welling

353

hormonal mechanism designed to promote surviving and thriving.

Stress

Chronic Stress

Increased Oxytocin Increased Affiliative Efforts Negative Social Outcome

Positive Social Outcome

Increased HPAaxis activity

Decreased HPAaxis activity

Increased Stress

Decreased Stress

Figure 20.1  A theoretical cascade model of an affiliative responses to stress, extrapolated from Taylor (2006). In this model, stress increases oxytocin (OT), which leads to increased affiliative efforts. When these efforts yield negative social outcomes, increased hypothalamic-pituitary-adrenal (HPA) axis activity leads to increased stress. When affiliative efforts yield positive social outcomes, HPA activity and, by extension, stress decrease.

proposes that stress triggers increased affiliative efforts via increased OT in line with the tend-and-befriend stress response system. The subsequent effect on stress is then dependent on the outcome of this effort; negative outcomes (e.g., failure to gain social support) will increase stress, whereas positive ­outcomes (e.g., success at gaining social support) will decrease stress. Taylor provides a selection of available evidence for each component of the model in support, although much of this evidence is derived from animal studies. However, recently a body of evidence in humans has begun to emerge, further supporting the theory (e.g., Cardoso, Ellenbogen, Serravalle, & Linnen, 2013; von Dawans, Fischbacher, Kirschbaum, Fehr, & Heinrichs, 2012). Hence, acute stress likely evolved to promote adaptive outcomes under difficult circumstances. Whereas the fight-or-flight response readies the body for action to escape or combat a threat, the tendand-befriend response encourages the individual to seek social support to combat threat and provide additional attention to vulnerable d ­ependents (e.g., children, kin), thus shielding their genetic (or social) assets from the threat. Either response highlights that, despite its unpleasant sensation, acute stress is typically an adaptive cognitive and 354

In contrast to acute stress, chronic stress is mal­ adaptive. One of the first studies on chronic stress (Selye, 1956) found that repeated shocks produced stomach ulcers and a lowered immune system ­response in rats. Prolonged exposure to GCs lowers immune response by suppressing cytokines, blocking cytokine receptors, disrupting lymphocyte development, and destroying lymphocytes (for review, see Segerstrom & Miller,  2004). As such, individuals who experience chronic stress are more likely to get the common cold and have more frequent cold sore flare-ups (Cohen, Tyrrell, & Smith,  1991,  1993). Additionally, chronic stress has been associated with an increased likelihood of getting a respiratory ­infection, the acceleration of autoimmune disorders, and increased recurrence of chronic allergies (Boyce et al., 1977, 1993; Monroe & Hadjiyannakis, 2002; Pereira & Penedo, 2005). Since Selye’s (1956) study, considerable research has investigated and confirmed the maladaptive effects of chronic stress (e.g., Coe & Lubach, 2003; Repetti, Taylor, & Seeman, 2002; Stowell, KiecoltGlaser, & Glaser, 2001; Taylor, Repetti, & Seeman, 1997). The stress response is a trade-off between long-term and short-term functioning. Processes essential for long-term survival (e.g., the immune system) are suspended to increase the likelihood of immediate survival. Because the stress response suppresses functions that are not vital for immediate survival, chronic stress can result in an increased susceptibility to infectious disease, psychological deficits, growth reduction, and reproductive issues (Ader, Felton, & Cohen, 1991; Glaser & KiecoltGlaser, 2014). Chronic stress continually mobilizes energy at the cost of energy storage, and the result is fatigue, muscle loss, and weakness (Bower et al., 2005, 2007). Additionally, the prolonged increase in heart rate weakens the heart muscles over time and increases plaque buildup (Booth-Kewley & Friedman, 1987; Rozanski, Blumenthal, & Kaplan, 1999). An increase in GCs is also associated with greater appetite and blockage of glucose reuptake, which can result in increased fat depositions, particularly in the midsection (Brindley & Rolland, 1989). As such, chronic stress is associated with an increased risk for cardiovascular disease, diabetes, and obesity (BoothKewley & Friedman, 1987). Growth functions are also suspended during the stress response by inhibiting the release of growth hormone (GH; Kosten,

Stress Hormones, Physiology, and Behavior

Jacobs, Mason, Wahby, & Atkins, 1984). When this occurs over a prolonged period in children, the result is psychosocial dwarfism, where a child typically grows to only one-half of the expected height for his or her age group (Green, Campbell, & David, 1984). These children have low endogenous GH and, depending on both their age and the duration of the chronic stress, can be unresponsive to exogenous GH supplementation (Albanese et al., 1994; Sapolsky, 1998). Chronic stress can also lead to a variety of reproductive issues (Rabin, Gold, Margioris, & Chrousos, 1988; Whirledge & Cidlowski, 2010). An increase in stress-related hormones (e.g., β-endorphins and CRH) inhibits the release of gonadotropin-releasing hormone (GnRH), resulting in reduced gonadotropins such as luteinizing hormone (LH) and follicle-stimulating hormone (FSH; Briski & Sylvester, 1991; Dubey & Plant, 1985). Both LH and FSH are involved in reproductive processes in males and females. In females, GCs inhibit the release of, and reduce sensitivity to, gonadotropins, which increases the likelihood of anovulatory cycles (i.e., menstrual cycles where an ovum is not released; Whirledge & Cidlowski, 2010). GCs also inhibit the secretion of progesterone, which is involved in the maturation of the uterine wall in preparation for egg implantation, decreasing the likelihood of successful implantation if ovulation occurs and in extreme cases halting menstruation completely. In males, GCs decrease testicular sensitivity to LH and lower testosterone, which can decrease sperm count and quality (Bambino & Hsueh,  1981; Saez, Morera, Haour, & Evain, 1977). Moreover, chronic stress activates the sympathetic nervous system and deactivates the parasympathetic ­nervous system, which increases the likelihood of erectile dysfunction and premature ejaculation (Agarwal, Nandipati, Sharma, Zippe, & Raina, 2006; Bancroft & Janssen, 2000). Chronic stress likewise has negative consequences on the brain and peripheral nervous system due to high levels of GCs causing cellular atrophy and impaired neurogenesis. Specifically, chronic stress has been linked to neural degeneration in areas of the brain such as the hippocampus, which is involved in learning and memory functions (McEwen & Sapolsky, 1995; Sapolsky, Krey, & McEwen, 1985). Several studies have supported the link between decreased hippocampal volume and decreased performance on spatial tasks in rats (Luine,  2002; Luine, Villegas, Martinez, & McEwen, 1994). In humans, this has best been studied in individuals with Cushing’s s­ yndrome, a disorder characterized

by the overproduction of GCs. Individuals with Cushing’s syndrome have lower hippocampal volume than average and perform worse on verbal recall tasks (Bourdeau et al., 2002; Starkman, Gebarski, Berent, & Schteingart, 1992). Furthermore, the HPA axis can be damaged by chronic stress, resulting in a stunted cortisol response to stressors (Glaser & Kiecolt-Glaser, 2014). This is particularly true when the stress occurs early in development; preand perinatal chronic stress has been linked to HPA axis malfunction in both rats and rhesus macaques (Clarke, 1993). Similarly, individuals who were sexually abused as children have been found to have damage to their HPA axis, which is thought to be linked to the chronic stress the child experienced as a result of the abuse (Heim, Newport, et al., 2000). In sum, there is a plethora of evidence supporting the link between consistent stressors in one’s environment and its negative impact on health. Although chronic stress has negative effects on health and bodily functions, the body has protective methods of preventing these effects (Kang, Coe, & McCarthy,  1996; Liu et al.,  2002). There are a number of characteristics that modulate the body’s resilience to the side effects of chronic stress (Coe & Lubach,  2003). One such characteristic is age, whereby children and the elderly are most susceptible to negative health outcomes from stress. An increased risk of disease when faced with a stressful environment was found in both elderly humans and monkeys (Bailey & Coe, 2002; Coe & Ershler, 2001; Kiecolt-Glaser, Marucha, Mercado, Malarkey, & Glaser, 1995; Uno, 1997). Infants also show a marked drop in immune function when removed from their mother in humans, rats, and squirrel monkeys (Ader & Friedman, 1965; Coe & Lubach, 2003). Another factor is the duration of the stressor; if a stressor is present for less than one month, there is no significant increase in the risk of illness. However, if a stressor is present for one month or more, then the risk of the individual developing an illness increases significantly (Cohen et al., 1998; Lepore, Miles, & Levy,  1997). This suggests that the body may be effective at combating the negative effects of chronic stress for a certain period of time only. Although there is compelling evidence that chronic stress is maladaptive, other specifics regarding how chronic stress impacts health are less clear. Some research suggests that cortisol is the main catalyst for the negative health outcomes associated with stress (Cohen, Kessler, & Gordon, 1995). There are two major theoretical models for the relationship between stress, cortisol, and health. The first

Mogilski, Wysocki, Reeve, Mitchell, Lunge, and Welling

355

model posits that chronic stress results in elevated cortisol (i.e., hypercortisolism) due to HPA axis hyperactivity. The increase in cortisol results in tissue damage and dysregulation of biological systems (Cohen et al., 1995; Schaeffer & Baum, 1984). The second model postulates that chronic stress results in decreased levels of cortisol (i.e., hypocortisolism) due to HPA axis habituation. Decreased cortisol results in fatigue and increased pain sensitivity and leaves the body more vulnerable to disease (Heim, Ehlert, & Hellhammer,  2000; Raison & Miller, 2003; Sternberg, Chrousos, Wilder, & Gold, 1992; Yehuda, 2000). For decades, it was accepted that chronic stress resulted in hypercortisolism, and a considerable number of studies have provided evidence to support this link (e.g., Arnetz et al., 1987; Baum, Gatchel, & Schaeffer,  1983; Schaeffer & Baum, 1984). For example, stressful tasks such as public speaking or mental arithmetic can increase cortisol levels (Kirschbaum, Pirke, & Hellhammer, 1993). However, recent studies by Yehuda and colleagues (Yehuda, 2000; Yehuda, Resnick, Schmeidler, Yang, & Pitman, 1998; Yehuda, Golier & Kaufman, 2005) have called into question whether there are exceptions to this relationship. Hypocortisolism has been most consistently found in individuals who have experienced a traumatic event and, subsequently, developed posttraumatic stress disorder (PTSD; see Yehuda, 1997). This relationship was first observed in Vietnam veterans and then replicated in Holocaust survivors, sexually abused women, and Bosnian prisoners of war (Bourne, Rose, & Mason,  1967, 1968; Dekaris et al., 1993; Yehuda, 2000; Yehuda et al., 1998, 2005). Hypocortisolism has also been found in individuals who have experienced chronic medical disorders, such as chronic pain, fibromyalgia, and asthma (Catley, Kaell, Kirschbaum, & Stone, 2000; Crofford et al.,  1995). Furthermore, hypocortisolism has been found in people without medical conditions who have other forms of chronic stress in their lives, for example, parents who had a child with a fatal illness (Friedman, Chodoff, Mason, & Hamburg,  1963) or individuals with a high amount of work stress (Caplan, Cobb, & French, 1979). What these two major findings suggest is that chronic stress has the ability to both increase and decrease cortisol levels in the body. Considering the strong evidence backing each of these models, the best explanation is an integration of the two (Gunnar & Vazquez,  2001; Heim, Ehlert, & Hellhammer, 2000; Miller, Chen, & Zhou, 2007; Raison & Miller, 2003). One factor that appears to have the greatest 356

impact on cortisol levels is time since the onset of the stressor, whereby time since onset is negatively related to cortisol level (Fries, Hesse, Hellhammer, & Hellhammer, 2005; Hellhammer & Wade, 1993; Miller, Cohen, & Ritchey,  2002). If the chronic stressor is no longer a part of the environment (e.g., a soldier coming home from a war zone), there is a greater likelihood of hypocortisolism. By comparison, someone who continues to experience the stressor (e.g., unemployment) is more likely to have hypercortisolism (Fries et al., 2005; Hellhammer & Wade, 1993; Miller et al.,  2007). This may explain why Yehuda and colleagues (1996, 1998, 2005) found hypocortisolism in those who were suffering from PTSD after experiencing a trauma. Thus, hypocortisolism seems to develop after a period of hyperactivity of the HPA axis, which would explain the perceived contradiction in previous research.

Status Social Dominance and Dominance Hierarchies

Not only are stress hormones implicated in physiological reactions to stressful environmental cues, but also they appear to play an important role, along with the androgen testosterone, in the maintenance of social dominance and dominance hierarchies (Eisenegger, Haushofer, & Fehr, 2011; Mehta & Josephs, 2010). Dominance hierarchies form in social groups where individuals must compete to obtain resources (e.g., food, mating opportunities) and each individual’s ability to compete is different (Sapolsky, 2005). Dominance hierarchies can be described in terms of both their linearity (i.e., the number of binary dominance relationships established within a group and how permanent those relationships are) and their steepness (i.e., how successful an individual of one rank will be against another in a competitive encounter; DeVries, Stevens, & Vervaecke, 2006). In these scenarios, dominant individuals (i.e., those who are best able to acquire and monopolize resources) and subordinate animals (i.e., those who are not as capable) have distinctive physiological and psychological states. This, in part, is because one’s rank determines how much physical and psychological stress an individual typically encounters. How different these states are, however, depends on the type of hierarchy that is formed in each species and within each sex. Dominance hierarchies exist on a spectrum ranging from egalitarian, where no individual is assigned a rank and dominance is achieved with the support of subordinate animals, to despotic, where one individual is dominant and suppresses subordinate

Stress Hormones, Physiology, and Behavior

animals (van Schaik, 1989). In egalitarian hierarchies, subordination is not associated with increased levels of physiological stress, whereas rank can influence how much stress an individual is exposed to in ­despotic hierarchies (although this relationship depends on the stability of the hierarchy; Sapolsky, 2005). Despotic hierarchies can also be stable or ­unstable, depending on whether or not rank is inherited in that species (reviewed in Sapolsky, 2005). When rank is inherited, subordinate animals consistently experience the highest levels of stress. In groups where rank is fluid, whether or not dominants or subordinates experience the most stress depends on the stability of the hierarchy. In these types of h ­ ierarchies, high rank can be maintained through either physical aggression or psychological intimidation. In species where aggression is used to maintain rank, dominant animals at the top of the hierarchy bear a higher burden (e.g., increased ­parasite load) due to continually elevated levels of cortisol (Muehlenbein & Watts, 2010), putatively as a r­esponse to the d ­ emands of maintaining that dominance (e.g., via fighting, protecting resources, having to remain on high alert; Cavigelli & Caruso, 2015). Conversely, in groups where psychological intimidation is used by dominants to suppress subordinates, subordinates experience more stress, presumably because of their decreased access to necessary physical resources (Sapolsky, 2005). Males and females also tend to have different types of intrasexual dominance hierarchies. For males, maintaining social rank is associated with more frequent aggressive behaviors that may result in injury due to enhanced male weaponry (e.g., increased body size, specialized teeth or claws; Cavigelli & Caruso, 2015). In female hierarchies, on the other hand, rank is maintained by more complex, subtle aggression and affiliative behaviors (Sapolsky, 2005). Additionally, whereas males are more driven by access to mates and reproductive opportunities (Ellis, 1995), female dominance hierarchies determine access to quality food resources and protection (Sterck, Watts, & van Schaik, 1997). Acts of dominance and rankings in dominance hierarchies appear to be associated with the androgen testosterone, and testosterone may motivate individuals to perform behaviors that help them attain and maintain status within social groups (Eisenegger et al., 2011). Testosterone is produced by both sexes in primate species, and is positively correlated with aggression and rank in dominance hierarchies in male (Muller & Wrangham,  2004) and female (Beehner, Phillips-Conroy, & Whitten,

2005) primates. For example, aggressive behavior in male rhesus monkeys is positively associated with elevated levels of plasma testosterone (Rose, Holaday, & Bernstein, 1971), and higher ranking chimpanzees exhibit more aggressive behaviors and have higher levels of testosterone compared to their lower ranking counterparts (Muller & Wrangham, 2004; Muehlenbein, Watts, & Whitten,  2004). Testosterone level is also positively associated with aggressive behaviors (Archer, 2006) and self-perceived dominance (Welling, Moreau, Bird, Hansen, & Carré, 2016) in human males, although the relationship between dominant behavior and testosterone in humans is more mixed (e.g., Johnson, Burk, & Kirk, 2007; Mazur & Booth, 1998). It has recently been proposed that these discrepancies in findings between testosterone and dominant behaviors and rank in dominance hierarchies can be reconciled when also considering the role of cortisol in dominance interactions and cortisol’s relationship with testosterone (Mehta & Josephs, 2010). Cortisol is known to interact with testosterone in multiple physiological ways (reviewed in Viau, 2002). For example, cortisol is known to disrupt the hypothalamic-pituitary-gonadal (HPG) axis and reproductive function in both males and females (Handa, Burgess, Kerr, & O’Keefe, 1994). The HPG axis controls testosterone production, and increased levels of cortisol suppress the production of endogenous testosterone (Cumming, Quigley, & Yen,  1983). Accounting for these interactions between cortisol and testosterone, Mehta and Josephs (2010) proposed a dual-hormone hypothesis, wherein cortisol modulates the effect that testosterone has on behavior, such that higher levels of testosterone are associated with more dominant behaviors, but only in individuals who also have low levels of cortisol. They found that cortisol and testosterone coregulate leadership (i.e., dominance) behavior in men and women, and that testosterone and cortisol levels influenced men’s competitive behaviors. Importantly, the ratio of cortisol to testosterone was not a significant predictor of an individual’s dominant behavior: Testosterone was only associated with increased dominant behaviors in individuals low in cortisol. Indeed, either there was no relationship between testosterone and dominance behaviors in individuals with both high testosterone and cortisol or the relationship was actually reversed, such that participants with high testosterone and cortisol displayed less dominant behaviors. Overall, cortisol appears to strongly influence dominance and status in primates, including humans, but these

Mogilski, Wysocki, Reeve, Mitchell, Lunge, and Welling

357

effects may depend on interactions with other ­hormones, such as testosterone. Similarly, several hormones appear to provide a buffering effect that alleviates some of the harm of stress hormones.

Social Buffering Oxytocin and Arginine Vasopressin

OT and AVP share a long evolutionary history and mediate a number of socioemotional behaviors in both humans and nonhumans (Carter, Grippo, Pournajafi-­ Nazarloo, Ruscio, & Porges,  2008; Heinrichs & Domes,  2008; Meyer-Lindenberg, Domes, Kirsch, & Heinrichs,  2011). OT neural systems are believed to have originally emerged in mammals to promote affiliative, prosocial, and nurturing behaviors between mothers and infants (Donaldson & Young,  2008) and have putatively been co-opted to regulate basic aspects of sexuality (i.e., sexual arousal, motivation, and orgasm; Borrow & Cameron, 2012; Garrison et al., 2012) and other social bonds (Feldman, 2012; Feldman, Monakhov, Pratt, & Ebstein, 2016). Indeed, receptor distribution for mesotocin (i.e., the functional analogue of  OT in birds) is associated with flock size, and mesotocin administration increases flock formation, whereas mesotocin antagonists reduce this social behavior (Goodson, Schrock, Klatt, Kabelik, & Kingsbury,  2009). Likewise, AVP has been shown to mediate prosocial behaviors, aggression, and territoriality in several species, particularly in males (Caldwell, Lee, Macbeth, & Young, 2008; Donaldson & Young, 2008; Young & Wang, 2004). OT and AVP also appear to aid in social synchrony (ApterLevi, Zagoory-Sharon, & Feldman,  2014). For ­instance, Bowen and McGregor (2014) found that rats treated with OT and AVP increase defensive aggregation when exposed to an environmental stressor (i.e., cat fur), suggesting that these molecules coordinate social behaviors that assist in responding to environmental threats. OT and AVP receptors are distributed throughout various brain regions associated with stress and anxiety regulation (Landgraf & Neumann, 2004), and their activation has been shown to modulate experiences of stress in a­ daptive ways alongside the dopaminergic reward system (Ludwig & Leng, 2006). Some evidence suggests that OT attenuates HPA activity in rodents and nonhuman primates (Neumann & Landgraf, 2012; Parker, Buckmaster, Schatzberg, & Lyons, 2005) and is associated with lower plasma and salivary cortisol in response to environmental stressors (Cardoso, Ellenbogen, Orlando, Bacon, & Joober, 2013; Ditzen et al., 2009; Legros, 358

2001). Other studies have shown that peripheral and intranasally administered OT levels are positively associated with circulating cortisol, particularly when subjects expect a stressful experimental manipulation (Brown, Cardoso, & Ellenbogen, 2016). By comparison, AVP has anxiogenic effects (Heinrichs, von Dawans, & Domes, 2009). Rats who naturally or transgenically fail to produce AVP demonstrate lower anxiety (Bielsky, Hu, Szegda, Westphal, & Young,  2004; Zelena et al.,  2008). Moreover, elevated plasma AVP is present in several anxiety disorders (Surget & Belzung,  2008) and is associated with territoriality (Caldwell et al., 2008; Donaldson & Young, 2008; Young & Wang, 2004) and behavioral aggression in men after exposure to stress (Moons, Way, & Taylor, 2014). Together, this evidence suggests that OT and AVP play an important role in regulating physiological stress; however, the exact mechanism by which this function is accomplished has yet to be conclusively identified. Recent efforts at consolidating the disparate effects of OT and AVP on HPA activity have focused on their role in attenuating psychosocial stress (i.e., social buffering; Hostinar, Sullivan, & Gunnar,  2014). Social support and the presence of conspecifics appear to dampen the HPA axis response in both humans (e.g., Heinrichs, Baumgartner, Kirschbaum, & Ehlert, 2003; Rosal, King, Ma, & Reed, 2004; Taylor et al.,  2008) and nonhuman animals (e.g., Hennessy, 1984; Vogt, Coe, & Levine, 1981). For example, maternal contact appears to be formative during HPA axis development in infancy (Gunnar, Hostinar, Sanchez, Tottenham, & Sullivan, 2015). Indeed, maternally deprived infant rats show chronic HPA hyperresponding compared to control rats (Suchecki, Nelson, Van Oers, & Levine,  1995). Likewise, peers also reduce psychosocial stress, though this effect is moderated by individual and interpersonal features (e.g., gender, familiarity, species-typical social organization; Hennessy, Kaiser, & Sachser, 2009). This social buffering appears to be mediated, in part, by circulating levels of OT and AVP (Hostinar et al., 2014). Social activities enhance OT release (Carter, 1998) and, in humans, social support paired with OT administration dampens HPA axis activity in males (Heinrichs et al.,  2003). Likewise, Ditzen et al. (2007) found that support from a romantic partner lowers salivary cortisol, but only when paired with physical contact (i.e., a massage). In a sample of international migrants whose OT and AVP were m ­ easured shortly after arrival in  their host country and reassessed two and five months later, greater baseline levels of OT predicted

Stress Hormones, Physiology, and Behavior

increases in social relationship satisfaction and social support and decreases in loneliness over time. By comparison, greater social integration was associated with higher plasma AVP over time in the same study (Gouin, Pournajafi-Nazarloo, & Carter, 2015). Collectively, this evidence suggests that OT and AVP are functionally similar but will attenuate or enhance stress in a manner that depends on characteristics of an organism’s social and physical environment. The actions of OT and AVP in promoting and regulating social behavior and stress also appear to differ between sexes and depend on interactions with gonadal steroids. OT and its receptor are expressed in higher quantity in women (Carter, 2007), and OT gene promoters are stimulated by exposure to estrogen (Lee, Macbeth, Pagani, & Young, 2009; Mohr & Schmitz, 1991; Richard & Zingg, 1990), whereas AVP synthesis is stimulated by androgens (DeVries & Villalba, 1997). Plasma OT is associated with relationship dissatisfaction in women but not men, whereas plasma AVP is associated with relational distress in men but not women (Taylor, Saphire-Bernstein, & Seeman, 2010). These sex differences may underlie the previously noted divergent responses in males and females to social stressors, whereby women exhibit more OT-based affiliative responses to threat (i.e., tend-and-befriend) than do men (Taylor et al., 2000).

Opioids

Although commonly associated with analgesic effects (e.g., D’Amato, 1998; D’Amato & Castellano, 1989; Kieffer & Gavériaux-Ruff, 2002; Panksepp, 2004), opioids can also promote social attachments and buffer the experience of stress. The brain opioid systems are activated during play in rat pups (Panksepp & Bishop,  1981), social grooming in nonhuman primates (Panksepp,  2004), and positive social ­interactions in humans (Eisenberger,  2012; Hsu et  al.,  2013). Additionally, opioids are released in infants upon social (Blass & Fitzgerald,  1988; Panksepp & Bishop,  1981) and physical contact with mothers (Weller & Feldman,  2003) and are known to moderate infant distress vocalizations (DVs; e.g., Kalin, Shelton, & Barksdale,  1988). DVs are produced by infants separated from their normal social environment (e.g., their mother, littermates) and are seen in many vertebrate species, such as mice, rats, chickens, guinea pigs, kittens, puppies, monkeys, and humans (Panksepp, Herman, Conner, Bishop, & Scott,  1978). Furthermore, in rhesus macaques, the anterior cingulate cortex (which

has a high concentration of opiate receptors; e.g., Wise & Herkenham, 1982) has been found to play a primary role in the induction of separation calls (Robinson, 1967). In addition to endogenous opioids, exogenous opioids also alter DVs produced by infants. Low doses of morphine (an opioid receptor agonist; Eisenberger,  2012) reduce DVs from separated infant rats (Carden & Hofer,  1990b), guinea pigs (Herman & Panksepp,  1978), chicks (Panksepp, Vilberg, Bean, Coy, & Kastin, 1978), dogs (Panksepp et al., 1978), and primates (Kalin et al., 1988). By comparison, the administration of naloxone (an opioid antagonist; MacDonald & Leary,  2005) increases DVs in guinea pigs and chicks (Panksepp et al.,  1978) and reverses the mitigating effects of littermate presence and morphine administration on reduced DVs in infant rats (e.g., Carden & Hofer, 1990a, 1990b). Exogenous opioids also alter social affiliation in primates, guinea pigs, and rats (MacDonald & Leary, 2005). For example, opioid receptor agonists reduce social interactions with conspecifics, likely by mimicking the rewards of social interactions and thus removing the motivation of pursuing social interactions (Eisenberger, 2012). Conversely, opioid receptor antagonists increase social interaction attempts, likely by blocking the rewards of social interaction and thus motivating its pursuit (Eisenberger, 2012). Furthermore, the effects of both endogenous and exogenous opioids are regulated by μ-opioid receptors. Infant mice lacking the μ-opioid receptor gene, for example, do not experience pain relief from morphine (Eisenberger, 2012), and infant rats lacking the same gene do not exhibit DVs when separated from their mother and littermates (Kehoe & Blass, 1986; Moles, Kieffer, & D’Amato,  2004). It is apparent that social isolation reduces endogenous opioid levels, inducing social distress and DVs in many vertebrates. On the other hand, social interaction increases endogenous o­ pioids, which reduces social distress and simultaneously reinforces later social interactions (e.g.,  Nelson & Panksepp,  1998). Ultimately, the opioid system is essential in social behavior, as it both mediates stress response and encourages further social affiliations by rewarding prior social interactions and attachments.

Future Directions

Evidently, chronic stress is maladaptive for a variety of physical and psychological processes. However, the negative effects of chronic stress are mediated by factors such as age and duration of chronic stress,

Mogilski, Wysocki, Reeve, Mitchell, Lunge, and Welling

359

as well as other hormones like OT. Furthermore, future research should work to further integrate the hypo- and hypercortisolism models. At present, endocrinologists have a basic theoretical framework (Miller et al.,  2007), but it would be beneficial to  understand the catalysts for how the HPA axis switches from hyper- to hypoactivity. There is also a need for a better understanding of the role that cortisol has in dysregulating bodily functions, breaking down tissue, and increasing susceptibility to disease. Research should further investigate the relationship between stress hormones and dominance, particularly with respect to the interaction with testosterone (Mehta & Josephs,  2010), in both humans and nonhuman primates. Furthermore, possible interactions with other hormones, such as those involved in social buffering, could be explored. Similarly, the role of individual differences in mediating a person’s response to both long-term and short-term stress could lead to important clinical applications for the treatment of PTSD and other anxiety-related disorders. One such individual difference factor is sex, whereby women are more likely to suffer from an anxiety disorder during their lifetime compared to men (e.g., McLean, Asnaani, Litz, & Hofmann, 2011; see also Pigott et al., this volume). It is possible that such a sex difference is partially explained by differences in stress-related coping strategies, such as the increased tend-and-befriend response among women compared to men (Tamres et al.,  2002; Taylor et al., 2000) or, perhaps relatedly, women’s higher OT response to stress (e.g., Taylor et al., 2000). Indeed, recent research suggests that OT may serve to enhance the social salience of environmental cues (Shamay-Tsoory & Abu-Akel, 2016), which could explain women’s heightened response to social rejection (Stroud et al., 2002) and, by extension, increased social anxiety (Kessler et al., 2012). In general, more research is needed to further parse apart the various adaptive and maladaptive workings of the human stress response system.

References

Ader, R., Felton, D., & Cohen, N. (1991). Psychoneuroimmunology (2nd ed.). San Diego, CA: Academic Press. Ader, R., & Friedman, S.  B. (1965). Social factors affecting emotionality and resistance to disease in animals: V.  Early separation from the mother and response to a transplanted tumor in the rat. Psychosomatic Medicine, 27(2), 119–122. Agarwal, A., Nandipati, K. C., Sharma, R. K., Zippe, C. D., & Raina, R. (2006). Role of oxidative stress in the pathophysio­ logical mechanism of erectile dysfunction. Journal of Andrology, 27(3), 335–347. Albanese, A., Hamill, G., Jones, J., Skuse, D., Matthews, D. R., & Stanhope, R. (1994). Reversibility of physiological growth

360

hormone secretion in children with psychosocial dwarfism. Clinical Endocrinology, 40(5), 687–692. Apter-Levi, Y., Zagoory-Sharon, O., & Feldman, R. (2014). Oxytocin and vasopressin support distinct configurations of social synchrony. Brain Research, 1580, 124–132. Archer, J. (2006). Testosterone and human aggression: An evaluation of the challenge hypothesis. Neuroscience & Biobehavioral Reviews, 30(3), 319–345. Arnetz, B. B., Wasserman, J., Petrini, B., Brenner, S. O., Levi, L., Eneroth, P., . . . Theorell, T. (1987). Immune function in unemployed women. Psychosomatic Medicine, 49(1), 3–12. Auphan, N., DiDonato, J.  A., Rosette, C., Helmberg, A., & Karin, M. (1995). Immunosuppression by glucocorticoids: Inhibition of NF-kappaB activity through induction of IkappaB synthesis. Science, 270, 286–290. Bailey, M. T., & Coe, C. L. (2002). Endometriosis is associated with an altered profile of intestinal microflora in female rhesus monkeys. Human Reproduction, 17(7), 1704–1708. Bambino, T. H., & Hsueh, A. J. (1981). Direct inhibitory effect of glucocorticoids upon testicular luteinizing hormone receptor and steroidogenesis in vivo and in vitro. Endocrinology, 108(6), 2142–2148. Bancroft, J., & Janssen, E. (2000). The dual control model of male sexual response: A theoretical approach to centrally mediated erectile dysfunction. Neuroscience & Biobehavioral Reviews, 24(5), 571–579. Baum, A., Gatchel, R. J., & Schaeffer, M. A. (1983). Emotional, behavioral, and physiological effects of chronic stress at Three Mile Island. Journal of Consulting and Clinical Psychology, 51(4), 565–572. Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497–529. Beehner, J. C., Phillips-Conroy, J. E., & Whitten, P. L. (2005). Female testosterone, dominance rank, and aggression in an Ethiopian population of hybrid baboons. American Journal of Primatology, 67(1), 101–119. Bielsky, I. F., Hu, S. B., Szegda, K. L., Westphal, H., & Young, L.  J. (2004). Profound impairment in social recognition and reduction in anxiety-like behavior in vasopressin V1a receptor knockout mice. Neuropsychopharmacology, 29, 483–493. Bindu, P., & Thomas, I. (2006). Gender differences in emotional intelligence. Psychological Studies, 51(4), 261–268. Blackhart, G. C., Eckel, L. A., & Tice, D. M. (2007). Salivary cortisol in response to acute social rejection and acceptance by peers. Biological Psychology, 75(3), 267–276. Blass, E. M., & Fitzgerald, E. (1988). Milk-induced analgesia and comforting in 10-day-old rats: Opioid mediation. Pharmacology Biochemistry and Behavior, 29(1), 9–13. Booth, R., & Sharma, D. (2009). Stress reduces attention to irrelevant information: Evidence from the Stroop task. Motivation and Emotion, 33(4), 412–418. Booth-Kewley, S., & Friedman, H.  S. (1987). Psychological predictors of heart disease: A quantitative review. Psychological Bulletin, 101(3), 343–362. Borrow, A. P., & Cameron, N. M. (2012). The role of oxytocin in mating and pregnancy. Hormones and Behavior, 61(3), 266–276. Bourdeau, I., Bard, C., Noël, B., Leclerc, I., Cordeau, M.  P., Bélair, M., . . . Lacroix, A. (2002). Loss of brain volume in endogenous Cushing’s syndrome and its reversibility after correction of hypercortisolism. Journal of Clinical Endocrinology & Metabolism, 87(5), 1949–1954.

Stress Hormones, Physiology, and Behavior

Bourne, P.  G., Rose, R.  M., & Mason, J.  W. (1967). Urinary 17-OHCS levels: Data on seven helicopter ambulance medics in combat. Archives of General Psychiatry, 17(1), 104–110. Bourne, P. G., Rose, R. M., & Mason, J. W. (1968). 17-OHCS levels in combat: Special forces “A” team under threat of attack. Archives of General Psychiatry, 19(2), 135–140. Bowen, M. T., & McGregor, I. S. (2014). Oxytocin and vasopressin modulate the social response to threat: a preclinical study. International Journal of Neuropsychopharmacology, 17(10), 1621–1633. Bower, J. E., Ganz, P. A., Aziz, N., Olmstead, R., Irwin, M. R., & Cole, S. W. (2007). Inflammatory responses to psychological stress in fatigued breast cancer survivors: Relationship to glucocorticoids. Brain, Behavior, and Immunity, 21(3), 251–258. Bower, J. E., Ganz, P. A., Dickerson, S. S., Petersen, L., Aziz, N., & Fahey, J.  L. (2005). Diurnal cortisol rhythm and fatigue in breast cancer survivors. Psychoneuroendocrinology, 30(1), 92–100. Boyce, W. T., Chesterman, E. A., Martin, N., Folkman, S., Cohen, F., & Wara, D. (1993). Immunologic changes occurring at kindergarten entry predict respiratory illnesses after the Loma Prieta earthquake. Journal of Developmental & Behavioral Pediatrics, 14(5), 296–303. Boyce, W. T., Jensen, E. W., Cassel, J. C., Collier, A. M., Smith, A. H., & Ramey, C. T. (1977). Influence of life events and family routines on childhood respiratory tract illness. Pediatrics, 60(4), 609–615. Brindley, D.  N., & Rolland, Y. (1989). Possible connections between stress, diabetes, obesity, hypertension and altered lipoprotein metabolism that may result in atherosclerosis. Clinical Science, 77(5), 453–461. Briski, K. P., & Sylvester, P. W. (1991). Acute inhibition of pituitary LH release in the male rat by the glucocorticoid agonist decadron phosphate. Neuroendocrinology, 54(4), 313–320. Brown, C.  A., Cardoso, C., & Ellenbogen, M.  A. (2016). A meta-analytic review of the correlation between peripheral oxytocin and cortisol concentrations. Frontiers in Neuroendo­ crinology, 43, 19–27. Cahill, L., Haier, R. J., Fallon, J., Alkire, M. T., Tang, C., Keator, D., . . . & McGaugh, J.  L. (1996). Amygdala activity at encoding correlated with long-term, free recall of emotional information. Proceedings of the National Academy of Sciences of the United States of America, 93(15), 8016–8021. Caldwell, H.  K., Lee, H.  J., Macbeth, A.  H., & Young, W.  S. (2008). Vasopressin: Behavioral roles of an “original” neuro­ peptide. Progress in Neurobiology, 84, 1–24. Caplan, R. D., Cobb, S., & French Jr, J. R. (1979). White collar work load and cortisol: Disruption of a circadian rhythm by job stress?. Journal of Psychosomatic Research, 23(3), 181–192. Cannon, W. B. (1929). Bodily changes in pain, hunger, fear, and rage. New York, NY: Appleton-Century-Crofts. Carden, S.  E., & Hofer, M.  A. (1990a). Independence of benzodiazepine and opiate actions in the suppression of isolation distress in rat pups. Behavioral Neuroscience, 104, 160–166. Carden, S.  E., & Hofer, M.  A. (1990b). Socially mediated reduction of isolation distress in rat pups can be blocked by naltrexone but not by RO 15–1788. Behavioral Neuroscience, 104, 457–463. Cardoso, C., Ellenbogen, M. A., Orlando, M. A., Bacon, S. L., & Joober, R. (2013). Intranasal oxytocin attenuates the cortisol response to physical stress: A dose–response study. Psychoneuroendocrinology, 38, 399–407.

Cardoso, C., Ellenbogen, M. A., Serravalle, L., & Linnen, A. M. (2013). Stress-induced negative mood moderates the relation between oxytocin administration and trust: Evidence for the tend-and-befriend response to stress? Psychoneuroendo­ crinology, 38(11), 2800–2804. Carter, C.  S. (1998). Neuroendocrine perspectives on social attachment and love. Psychoneuroendocrinology, 23, 779–818. Carter, C. S. (2007). Sex differences in oxytocin and vasopressin: Implications for autism spectrum disorders? Behavioural Brain Research, 176, 170–186. Carter, C.  S., Grippo, A.  J., Pournajafi-Nazarloo, H., Ruscio, M. G., & Porges, S. W. (2008). Oxytocin, vasopressin and sociality. Progress in Brain Research, 170, 331–336. Catley, D., Kaell, A. T., Kirschbaum, C., & Stone, A. A. (2000). A naturalistic evaluation of cortisol secretion in persons with fibromyalgia and rheumatoid arthritis. Arthritis & Rheumatism, 13(1), 51–61. Cavigelli, S. A., & Caruso, M. J. (2015). Sex, social status and physiological stress in primates: The importance of social and glucocorticoid dynamics. Philosophical Transactions of the Royal Society B, 370(1669), 1–13. Clarke, A. S. (1993). Social rearing effects on HPA axis activity over early development and in response to stress in rhesus monkeys. Developmental Psychobiology, 26(8), 433–446. Coe, C. L., & Ershler, W. B. (2001). Intrinsic and environmental influences on immune senescence in the aged monkey. Physiology & Behavior, 73(3), 379–384. Coe, C. L., & Lubach, G. R. (2003). Critical periods of special health relevance for psychoneuroimmunology. Brain, Behavior, and Immunity, 17(1), 3–12. Cohen, S., Frank, E., Doyle, W. J., Skoner, D. P., Rabin, B. S., & Gwaltney, J. M., Jr. (1998). Types of stressors that increase susceptibility to the common cold in healthy adults. Health Psychology, 17(3), 214–223. Cohen, S., Kessler, R. C., & Gordon, L. U. (1995). Strategies for measuring stress in studies of psychiatric and physical disorders. In S.  Cohen, R.  C.  Kessler, & L.  U.  Gordon (Eds.), Measuring stress: A guide for health and social scientists (pp. 3–26). New York, NY: Oxford University Press. Cohen, S., Tyrrell, D. A., & Smith, A. P. (1991). Psychological stress and susceptibility to the common cold. New England Journal of Medicine, 325(9), 606–612. Cohen, S., Tyrrell, D. A., & Smith, A. P. (1993). Negative life events, perceived stress, negative affect, and susceptibility to the common cold. Journal of Personality and Social Psychology, 64(1), 131–140. Cohen-Cole, S., Cogen, R., & Stevens, A. (1981). Psychosocial, endocrine, and immune factors in acute necrotizing ulcerative gingivitis (“trenchmouth”). Psychosomatic Medicine, 43, 91–100. Crofford, L. J., Sano, H., Karalis, K., Webster, E. A., Friedman, T.  C., Chrousos, G.  P., & Wilder, R.  L. (1995). Local expression of corticotropin-releasing hormone in inflammatory arthritis. Annals of the New York Academy of Sciences, 771(1), 459–471. Cumming, D. C., Quigley, M. E., & Yen, S. S. C. (1983). Acute suppression of circulating testosterone levels by cortisol in men. Journal of Clinical Endocrinology & Metabolism, 57(3), 671–673. D’Amato, F.  R. (1998). Kin interaction enhances morphine analgesia in male mice. Behavioral Pharmacology, 9, 369–373. D’Amato, F. R., & Castellano, C. (1989). Behavioral effects of morphine in mice: Role of experimental housing. Pharmacology Biochemistry and Behavior, 34, 361–365.

Mogilski, Wysocki, Reeve, Mitchell, Lunge, and Welling

361

Dekaris, D., Sabioncello, A., Mažuran, R., Rabatić, S., SvobodaBeusan, I., Računica, N. L., & Tomašić, J. (1993). Multiple changes of immunologic parameters in prisoners of war: Assessments after release from a camp in Manjača, Bosnia. JAMA, 270(5), 595–599. DeVries, H.  A.  N., Stevens, J.  M., & Vervaecke, H. (2006). Measuring and testing the steepness of dominance hierarchies. Animal Behaviour, 71(3), 585–592. DeVries, G. J., & Villalba, C. (1997). Brain sexual dimorphism and sex differences in parental and other social behaviors. Annals of the New York Academy of Sciences, 807, 273–286. Dhabhar, F. S. (2002). Stress-induced augmentation of immune function: The role of stress hormones, leukocyte trafficking, and cytokines. Brain, Behavior, and Immunity, 16, 785–798. Dhabhar, F. S., & McEwen, B. S. (2001). Bidirectional effects of stress and glucocorticoid hormones on immune function: Possible explanations for paradoxical observations. In R. Ader, D. L. Felten, and N. Cohen (Eds.), Psychoneuro­ immunology (Vol. 1, pp. 301–338). New York, NY: Academic Press. Dhabhar, F.  S., Miller, A.  H., Stein, M., McEwen, B.  S., & Spencer, R. L. (1994). Diurnal and acute stress-induced changes in distribution of peripheral blood leukocyte subpopulations. Brain, Behavior, and Immunity, 8, 66–79. Ditzen, B., Neumann, I. D., Bodenmann, G., von Dawans, B., Turner, R. A., Ehlert, U., & Heinrichs, M. (2007). Effects of different kinds of couple interaction on cortisol and heart rate responses to stress in women. Psychoneuroendocrinology, 32(5), 565–574. Ditzen, B., Schaer, M., Gabriel, B., Bodenmann, G., Ehlert, U., & Heinrichs, M. (2009). Intranasal oxytocin increases positive communication and reduces cortisol levels during couple conflict. Biological Psychiatry, 65, 728–731. Donaldson, Z. R., & Young, L. J. (2008). Oxytocin, vasopressin, and the neurogenetics of sociality. Science, 322, 900–904. Dubey, A.  K., & Plant, T.  M. (1985). A suppression of gonadotropin secretion by cortisol in castrated male rhesus monkeys (Macaca mulatta) mediated by the interruption of hypothalamic gonadotropin-releasing hormone release. Biology of Reproduction, 33(2), 423–431. Eisenberger, N.  I. (2012). The pain of social disconnection: Examining the shared neural underpinnings of physical and social pain. Nature Reviews Neuroscience, 13, 421–434. Eisenegger, C., Haushofer, J., & Fehr, E. (2011). The role of testosterone in social interaction. Trends in Cognitive Sciences, 15(6), 263–271. Ellis, L. (1995). Dominance and reproductive success among nonhuman animals: A cross-species comparison. Ethology and Sociobiology, 16(4), 257–333. Everly, G.  S., Jr., & Lating, J.  M. (2013). The anatomy and physiology of the human stress response. In A clinical guide to the treatment of the human stress response (pp. 17–51). New York, NY: Springer. Farmer, G. E., Park, C. R., Bullard, L. A., & Diamond, D. M. (2014). Evolutionary, historical and mechanistic perspectives on how stress affects memory and hippocampal synaptic plasticity. In M. Popoli, D. Diamond, & G. Sanacora (Eds.), Synaptic stress and pathogenesis of neuropsychiatric disorders (pp. 167–182). New York, NY: Springer. Feldman, R. (2012). Oxytocin and social affiliation in humans. Hormones and Behavior, 61(3), 380–391. Feldman, R., Monakhov, M., Pratt, M., & Ebstein, R. P. (2016). Oxytocin pathway genes: evolutionary ancient system

362

impacting on human affiliation, sociality, and psychopathology. Biological Psychiatry, 79(3), 174–184. Friedman, S. B., Chodoff, P., Mason, J. W., & Hamburg, D. A. (1963). Behavioral observations on parents anticipating the death of a child. Pediatrics, 32(4), 610–625. Fries, E., Hesse, J., Hellhammer, J., & Hellhammer, D.  H. (2005). A new view on hypocortisolism. Psychoneuroendo­ crinology, 30(10), 1010–1016. Gagnon, S. A., & Wagner, A. D. (2016). Acute stress and episodic memory retrieval: Neurobiological mechanisms and behavioral consequences. Annals of the New York Academy of Sciences, 1369(1), 55–75. Galambos, N. L. (2004). Gender and gender role development in adolescence. Handbook of Adolescent Psychology, 2, 233–262. Garrison, J.  L., Macosko, E.  Z., Bernstein, S., Pokala, N., Albrecht, D.  R., & Bargmann, C.  I. (2012). Oxytocin/ vasopressin-related peptides have an ancient role in reproductive behavior. Science, 338(6106), 540–543. Glaser, R., & Kiecolt-Glaser, J. K. (Eds.). (2014). Handbook of human stress and immunity. San Diego, CA: Academic Press. Glaser, R., Kiecolt-Glaser, J.  K., Speicher, C.  E., & Holliday, J.  E. (1985). Stress, loneliness, and changes in herpes virus latency. Journal of Behavioral Medicine, 8, 249–260. Goodson, J.  L., Schrock, S.  E., Klatt, J.  D., Kabelik, D., & Kingsbury, M.  A. (2009). Mesotocin and nonapeptide receptors promote estrildid flocking behavior. Science, 325, 862–866. Gouin, J.  P., Pournajafi-Nazarloo, H., & Carter, C.  S. (2015). Changes in social functioning and circulating oxytocin and vasopressin following the migration to a new country. Physiology & Behavior, 139, 67–72. Green, W. H., Campbell, M., & David, R. (1984). Psychosocial dwarfism: A critical review of the evidence. Journal of the American Academy of Child Psychiatry, 23(1), 39–48. Gunnar, M.  R., Hostinar, C.  E., Sanchez, M.  M., Tottenham, N., & Sullivan, R. M. (2015). Parental buffering of fear and stress neurobiology: Reviewing parallels across rodent, monkey, and human models. Social Neuroscience, 10, 474–478. Gunnar, M. R., & Vazquez, D. M. (2001). Low cortisol and a flattening of expected daytime rhythm: Potential indices of risk in human development. Development and Psychopathology, 13(3), 515–538. Handa, R. J., Burgess, L. H., Kerr, J. E., & O’Keefe, J. A. (1994). Gonadal steroid hormone receptors and sex differences in the hypothalamo-pituitary-adrenal axis. Hormones and Behavior, 28(4), 464–476. Heim, C., Ehlert, U., & Hellhammer, D. H. (2000). The potential role of hypocortisolism in the pathophysiology of stress-related bodily disorders. Psychoneuroendocrinology, 25(1), 1–35. Heim, C., Newport, D. J., Heit, S., Graham, Y. P., Wilcox, M., Bonsall, R., . . . Nemeroff, C.  B. (2000). Pituitary-adrenal and autonomic responses to stress in women after sexual and physical abuse in childhood. JAMA, 284(5), 592–597. Heinrichs, M., Baumgartner, T., Kirschbaum, C., & Ehlert, U. (2003). Social support and oxytocin interact to suppress cortisol and subjective responses to psychological stress. Biological Psychiatry, 54, 1389–1398. Heinrichs, M., & Domes, G. (2008). Neuropeptides and social behaviour: Effects of oxytocin and vasopressin in humans. Progress in Brain Research, 170, 337–350. Heinrichs, M., von Dawans, B., & Domes, G. (2009). Oxytocin, vasopressin, and human social behavior. Frontiers in Neuroendocrinology, 30, 548–557.

Stress Hormones, Physiology, and Behavior

Hellhammer, D.  H., & Wade, S. (1993). Endocrine correlates of  stress vulnerability. Psychotherapy and Psychosomatics, 60(1), 8–17. Hennessy, M.  B. (1984). Presence of companion moderates arousal of monkeys with restricted social experience. Physiology & Behavior, 33, 693–698. Hennessy, M.  B., Kaiser, S., & Sachser, N. (2009). Social buffering of the stress response: Diversity, mechanisms, and functions. Frontiers in Neuroendocrinology, 30, ­­470–482. Herman, B. H., & Panksepp, J. (1978). Effects of morphine and naloxone on separation distress and approach attachment: Evidence for opiate mediation of social affect. Pharmacology, Biochemistry, and Behavior, 9, 213–220. Hostinar, C.  E., Sullivan, R.  M., & Gunnar, M.  R. (2014). Psychobiological mechanisms underlying the social buffering of the hypothalamic–pituitary–adrenocortical axis: A review of animal models and human studies across development. Psychological Bulletin, 140, 256–282. Hsu, D. T., Sanford, B. J., Meyers, K. K., Love, T. M., Hazlett, K. E., Wang, H., . . . & Zubieta, J. K. (2013). Response of the μ-opioid system to social rejection and acceptance. Molecular Psychiatry, 18(11), 1211–1217. Insel, T. R. (1997). A neurobiological basis of social attachment. American Journal of Psychiatry, 154, 726–735. Johnson, R.  T., Burk, J.  A., & Kirkpatrick, L.  A. (2007). Dominance and prestige as differential predictors of aggression and testosterone levels in men. Evolution and Human Behavior, 28(5), 345–351. Kalin, N. H., Shelton, S. E., & Barksdale, C. M. (1988). Opiate modulation of separation-induced distress in non-human primates. Brain Research, 440, 285–292. Kang, D. H., Coe, C. L., & McCarthy, D. O. (1996). Academic examinations significantly impact immune responses, but not lung function, in healthy and well-managed asthmatic adolescents. Brain, Behavior, and Immunity, 10(2), 164–181. Kehoe, P., & Blass, E. M. (1986). Opioid-mediation of separation distress in 10-day-old rats: Reversal of stress with maternal stimuli. Developmental Psychobiology, 19, 385–398. Kessler, R. C., Avenevoli, S., McLaughlin, K. A., Green, J. G., Lakoma, M.  D., Petukhova, M., . . . & Merikangas, K.  R. (2012). Lifetime co-morbidity of DSM-IV disorders in the US national comorbidity survey replication adolescent supplement (NCS-A). Psychological Medicine, 42(9), 1997–2010. Kiecolt-Glaser, J. K., Marucha, P. T., Mercado, A. M., Malarkey, W.  B., & Glaser, R. (1995). Slowing of wound healing by psychological stress. Lancet, 346(8984), 1194–1196. Kieffer, B.  L., & Gavériaux-Ruff, C. (2002). Exploring the opioid system by gene knockout. Progress in Neurobiology, 66, 285–306. Kirschbaum, C., Pirke, K.  M., & Hellhammer, D.  H. (1993). The “Trier Social Stress Test”—A tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology, 28(1–2), 76–81. Kosten, T.  R., Jacobs, S., Mason, J., Wahby, V., & Atkins, S. (1984). Psychological correlates of growth hormone response to stress. Psychosomatic Medicine, 46(1), 49–58. Landgraf, R., & Neumann, I.  D. (2004). Vasopressin and oxytocin release within the brain: A dynamic concept of multiple and variable modes of neuropeptide communication. Frontiers in Neuroendocrinology, 25, 150–176. Lazarus, R. S., & Folkman, S. (1984). Coping and adaptation. In W. D. Gentry (Ed.), The handbook of behavioral medicine (pp. 282–325). New York: Guilford.

Lee, H.  J., Macbeth, A.  H., Pagani, J.  H., & Young, W.  S. (2009). Oxytocin: The great facilitator of life. Progress in Neurobiology, 88, 127–151. Legros, J. J. (2001). Inhibitory effect of oxytocin on corticotrope function in humans: Are vasopressin and oxytocin ying–yang neurohormones? Psychoneuroendocrinology, 26, 649–655. Lepore, S. J., Miles, H. J., & Levy, J. S. (1997). Relation of chronic and episodic stressors to psychological distress, reactivity, and health problems. International Journal of Behavioral Medicine, 4(1), 39–59. Light, K.  C., Smith, T.  E., Johns, J.  M., Brownley, K.  A., Hofheimer, J. A., & Amico, J. A. (2000). Oxytocin respon­ sivity in mothers of infants: A preliminary study of relationships with blood pressure during laboratory stress and normal ambulatory activity. Health Psychology, 19, 560–567. Liu, L. Y., Coe, C. L., Swenson, C. A., Kelly, E. A., Kita, H., & Busse, W. W. (2002). School examinations enhance airway inflammation to antigen challenge. American Journal of Respiratory and Critical Care Medicine, 165(8), 1062–1067. Ludwig, M., & Leng, G. (2006). Dendritic peptide release and peptide-dependent behaviours. Nature Reviews Neuroscience, 7, 126–136. Luine, V. (2002). Sex differences in chronic stress effects on memory in rats. Stress, 5(3), 205–216. Luine, V., Villegas, M., Martinez, C., & McEwen, B. S. (1994). Repeated stress causes reversible impairments of spatial memory performance. Brain Research, 639(1), 167–170. MacDonald, G., & Leary, M.  R. (2005). Why does social exclusion hurt? The relationship between social and physical pain. Psychological Bulletin, 131, 202–223. Mazur, A., & Booth, A. (1998). Testosterone and dominance in men. Behavioral and Brain Sciences, 21(3), 353–363. McCarthy, M. M. (1994). Estrogen modulation of oxytocin and its relation to behavior. Advances in Experimental Medicine and Biology, 395, 235–245. McEwen, B.  S. (2000). Definition and concepts of stress. In G.  Fink (Ed.), Encyclopedia of stress (Vol. 3, pp. 508–509). San Diego, CA: Academic Press. McEwen, B.  S., Biron, C.  A., Brunson, K.  W., Bulloch, K., Chambers, W. H., Dhabhar, F. S., . . . & Weiss, J. M. (1997). The role of adrenocorticoids as modulators of immune function in health and disease: Neural, endocrine and immune interactions. Brain Research Reviews, 23, 79–133. McEwen, B. S., & Sapolsky, R. M. (1995). Stress and cognitive function. Current Opinion in Neurobiology, 5(2), 205–216. McLean, C. P, Asnaani, A., Litz, B. T., & Hofmann, S. G. (2011). Gender differences in anxiety disorders: Prevalence, course of illness, comorbidity and burden of illness. Journal of Psychiatric Research, 45(8), 1027–1035. Mehta, P. H., & Josephs, R. A. (2010). Testosterone and cortisol jointly regulate dominance: Evidence for a dual-hormone hypothesis. Hormones and Behavior, 58(5), 898–906. Meyer-Lindenberg, A., Domes, G., Kirsch, P., & Heinrichs, M. (2011). Oxytocin and vasopressin in the human brain: Social neuropeptides for translational medicine. Nature Reviews Neuroscience, 12, 524–538. Miller, G. E., Chen, E., & Zhou, E. S. (2007). If it goes up, must it come down? Chronic stress and the hypothalamicpituitary-adrenocortical axis in humans. Psychological Bulletin, 133(1), 25–45. Miller, G.  E., Cohen, S., & Ritchey, A.  K. (2002). Chronic psychological stress and the regulation of pro-inflammatory cytokines: a glucocorticoid-resistance model. Health psychology, 21(6), 531.

Mogilski, Wysocki, Reeve, Mitchell, Lunge, and Welling

363

Mohr, E., & Schmitz, E. (1991). Functional characterization of estrogen and glucocorticoid responsive elements in the rat oxytocin gene. Molecular Brain Research, 9, 293–298. Moles, A., Kieffer, B. L., & D’Amato, F. R. (2004). Deficit in attachment behavior in mice lacking the mu-opioid receptor gene. Science, 304, 1983–1986. Monroe, S.  M., & Hadjiyannakis, K. (2002). The social environment and depression: Focusing on severe life stress. In C. L. Hammen (Ed.), Handbook of depression (pp. 314–340). New York, NY: Guildford Press. Moons, W. G., Way, B. M., & Taylor, S. E. (2014). Oxytocin and vasopressin receptor polymorphisms interact with circulating neuropeptides to predict human emotional reactions to stress. Emotion, 14, 562–572. Muehlenbein, M.  P., & Watts, D.  P. (2010). The costs of dominance: Testosterone, cortisol and intestinal parasites in wild male chimpanzees. BioPsychoSocial Medicine, 4(1), 21. Muehlenbein, M.  P., Watts, D.  P., & Whitten, P.  L. (2004). Dominance rank and fecal testosterone levels in adult male chimpanzees (Pan troglodytes schweinfurthii) at Ngogo, Kibale National Park, Uganda. American Journal of Primatology, 64(1), 71–82. Muller, M.  N., & Wrangham, R.  W. (2004). Dominance, aggression and testosterone in wild chimpanzees: A test of  the “challenge hypothesis.” Animal Behaviour, 67(1), 113–123. Nelson, E. E., & Panksepp, J. (1998). Brain substrates of infant– mother attachment: Contributions of opioids, oxytocin, and  norepi­nephrine. Neuroscience & Biobehavioral Reviews, 22(3), 437–452. Nesse, R.  M., Bhatnagar, S., & Ellis, B. (2016). Evolutionary origins and functions of the stress response system. In Stress: Concepts, Cognition, Emotion, and behavior (pp. 95–101). Neumann, I.  D., & Landgraf, R. (2012). Balance of brain oxytocin and vasopressin: Implications for anxiety, depression, and social behaviors. Trends in Neurosciences, 35, 649–659. Nichols, S. L., & Good, T. L. (1998). Students’ perceptions of fairness in school settings: A gender analysis. Teachers College Record, 100(2), 369–401. Olver, J.  S., Pinney, M., Maruff, P., & Norman, T.  R. (2015). Impairments of spatial working memory and attention following acute psychosocial stress. Stress and Health, 31(2), 115–123. Panksepp, J. (2004). Affective neuroscience: The foundations of  human and animal emotions. London, UK: Oxford University Press. Panksepp, J., & Bishop, P. (1981). An autoradiographic map of (3H) diprenorphine binding in rat brain: Effects of social interaction. Brain Research Bulletin, 7, 405–410. Panksepp, J., Herman, B., Conner, R., Bishop, P., & Scott, J. P. (1978). The biology of social attachments: Opiates alleviate separation distress. Biological Psychiatry, 13, 607–618. Panksepp, J., Vilberg, T., Bean, N. J., Coy, D. H., & Kastin, A. J. (1978). Reduction of distress vocalization in chicks by opiatelike peptides. Brain Research Bulletin, 3, 663–667. Parker, K.  J., Buckmaster, C.  L., Schatzberg, A.  F., & Lyons, D. M. (2005). Intranasal oxytocin administration attenuates the ACTH stress response in monkeys. Psychoneuroendo­ crinology, 30, 924–929. Pereira, D. B., & Penedo, F. J. (2005). Psychoneuroimmunology and chronic viral infection: HIV infection. In K.  Vedhara and M. Irwin (Eds.), Human psychoneuroimmunology. Oxford, England, UK: Oxford University Press.

364

Petrides, K.  V., & Furnham, A. (2000). Gender differences in measured and self-estimated trait emotional intelligence. Sex Roles, 42, 449–461. Rabin, D., Gold, P.  W., Margioris, A.  N., & Chrousos, G.  P. (1988). Stress and reproduction: Physiologic and pathophysio­ logic interactions between the stress and reproductive axes. Mechanisms of Physical and Emotional Stress, 245, 377–387. Raison, C. L., & Miller, A. H. (2003). When not enough is too much: The role of insufficient glucocorticoid signaling in the pathophysiology of stress-related disorders. American Journal of Psychiatry, 160(9), 1554–1565. Repetti, R.  L., Taylor, S.  E., & Seeman, T.  E. (2002). Risky families: Family social environments and the mental and physical health of offspring. Psychological Bulletin, 128(2), 330–366. Richard, S., & Zingg, H. H. (1990). The human oxytocin gene promoter is regulated by estrogens. Journal of Biological Chemistry, 265, 6098–6103. Robinson, B. W. (1967). Vocalization evoked from forebrain in Macaca mulatta. Physiology and Behavior, 2, 345–354. Roldan, E., Alvarez-Pelaez, P., & de Molina, F. (1974). Electrographic study of the amygdaloid defense response. Physiology & Behavior, 13, 779–787. Rosal, M. C., King, J., Ma, Y., & Reed, G. W. (2004). Stress, social support, and cortisol: Inverse associations? Behavioral Medicine, 30, 11–22. Rose, R. M., Holaday, J. W., & Bernstein, I. S. (1971). Plasma testosterone, dominance rank and aggressive behaviour in male rhesus monkeys. Nature, 231, 366–368. Rozanski, A., Blumenthal, J. A., & Kaplan, J. (1999). Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation, 99(16), 2192–2217. Saez, J.  M., Morera, A.  M., Haour, F., & Evain, D. (1977). Effects of in vivo administration of dexamethasone, corti­ cotropin and human chorionic gonadotropin on steroido­ genesis and protein and DNA synthesis of testicular interstitial cells in prepuberal rats. Endocrinology, 101(4), 1256–1263. Sapolsky, R. M. (1998). The trouble with testosterone: And other essays on the biology of the human predicament. New York, NY: Simon and Schuster. Sapolsky, R.  M. (2005). The influence of social hierarchy on primate health. Science, 308(5722), 648–652. Sapolsky, R. M., Krey, L. C., & McEwen, B. S. (1985). Prolonged glucocorticoid exposure reduces hippocampal neuron number: Implications for aging. Journal of Neuroscience, 5(5), 1222–1227. Schaeffer, M. A., & Baum, A. (1984). Adrenal cortical response to stress at Three Mile Island. Psychosomatic Medicine, 46(3), 227–237. Sear, R., & Mace, R. (2008). Who keeps children alive? A review of the effects of kin on child survival. Evolution and Human Behavior, 29(1), 1–18. Segerstrom, S. C., & Miller, G. E. (2004). Psychological stress and the human immune system: A meta-analytic study of 30 years of inquiry. Psychological Bulletin, 130(4), 601–630. Selye, H. (1956). The stress of life. New York, NY: McGraw-Hill. Shamay-Tsoory, S. G., & Abu-Akel, A. (2016). The social salience hypothesis of oxytocin. Biological Psychiatry, 79(3), 194–202. Starkman, M.  N., Gebarski, S.  S., Berent, S., & Schteingart, D.  E. (1992). Hippocampal formation volume, memory dysfunction, and cortisol levels in patients with Cushing’s syndrome. Biological Psychiatry, 32(9), 756–765.

Stress Hormones, Physiology, and Behavior

Sterck, E.  H., Watts, D.  P., & van Schaik, C.  P. (1997). The evolution of female social relationships in nonhuman primates. Behavioral Ecology and Sociobiology, 41(5), 291–309. Sternberg, E. M., Chrousos, G. P., Wilder, R. L., & Gold, P. W. (1992). The stress response and the regulation of inflammatory disease. Annals of Internal Medicine, 117(10), 854–866. Stowell, J.  R., Kiecolt-Glaser, J.  K., & Glaser, R. (2001). Perceived stress and cellular immunity: When coping counts. Journal of Behavioral Medicine, 24(4), 323–339. Stroud, L. R., Salovey, P., & Epel, E. S. (2002). Sex differences in stress responses: Social rejection versus achievement stress. Biological Psychiatry, 52(4), 318–327. Suchecki, D., Nelson, D. Y., Van Oers, H., & Levine, S. (1995). Activation and inhibition of the hypothalamic-pituitaryadrenal axis of the neonatal rat: Effects of maternal deprivation. Psychoneuroendocrinology, 20, 169–182. Surget, A., & Belzung, C. (2008). Involvement of vasopressin in affective disorders. European Journal of Pharmacology, 583(2–3), 340–349. Tamres, L., Janicki, D., & Helgeson, V. S. (2002). Sex differences in coping behavior: A meta-analytic review. Personality and Social Psychology Review, 6, 2–30. Taylor, S. E. (2006). Tend and befriend: Biobehavioral bases of affiliation under stress. Current Directions in Psychological Science, 15, 273–277. Taylor, S. E., Burklund, L. J., Eisenberger, N. I., Lehman, B. J., Hilmert, C. J., & Lieberman, M. D. (2008). Neural bases of moderation of cortisol stress responses by psychosocial resources. Journal of Personality and Social psychology, 95, 197–211. Taylor, S.  E., Gonzaga, G., Klein, L.  C., Hu, P., Greendale, G.  A., & Seeman, S.  E. (2006). Relation of oxytocin to psychological and biological stress responses in older women. Psychosomatic Medicine, 68, 238–245. Taylor, S.  E., Klein, L.  C., Lewis, B.  P., Gruenewald, T.  L., Gurung, R. A. R., & Updegraff, J. A. (2000). Biobehavioral responses to stress in females: Tend-and-befriend, not fightor-flight. Psychological Review, 107, 411–442. Taylor, S.  E., Repetti, R.  L., & Seeman, T. (1997). Health psychology: What is an unhealthy environment and how does it get under the skin? Annual Review of Psychology, 48(1), 411–447. Taylor, S. E., Saphire-Bernstein, S., & Seeman, T. E. (2010). Are plasma oxytocin in women and plasma vasopressin in men biomarkers of distressed pair-bond relationships?. Psychological Science, 21(1), 3-7. Turton, S., & Campbell, C. (2005). Tend and befriend versus fight or flight: Gender differences in behavioral response to  stress among university students. Journal of Applied Biobehavioral Research, 10(4), 209–232. Uno, H. (1997). Age-related pathology and biosenescent markers in captive rhesus macaques. Age, 20(1), 1–13. Usdin, E., Kvetnansky, R., & Kopin, L. (1976). Stress and catecholamines. Oxford, UK: Pergamon Press. Uvnas-Moberg, K. (1997). Oxytocin linked antistress effects: The relaxation and growth response. Acta Psychologica Scandinavica, 640, 38–42.

Uvnas-Moberg, K. (1998). Oxytocin may mediate the benefits of positive social interaction and emotion. Psychoneuroendo­ crinology, 23, 819–835. von Dawans, B., Fischbacher, U., Kirschbaum, C., Fehr, E., & Heinrichs, M. (2012). The social dimension of stress reactivity: Acute stress increases prosocial behavior in humans. Psychological Science, 23(6), 651–660. van Schaik, C.  P. (1989). The ecology of social relationships amongst female primates. In V. Standen & R. Foley (Eds.), Comparative socioecology of mammals and man (pp. 195–218). Oxford, UK: Blackwell Scientific. Viau, V. (2002). Functional cross-talk between the hypothalamicpituitary-gonadal and-adrenal axes. Journal of Neuroendo­ crinology, 14(6), 506–513. Vogt, J.  L., Coe, C.  L., & Levine, S. (1981). Behavioral and adrenocorticoid responsiveness of squirrel monkeys to a live snake: Is flight necessarily stressful? Behavioral and Neural Biology, 32, 391–405. Weller, A., & Feldman, R. (2003). Emotion regulation and touch in infants: The role of cholecystokinin and opioids. Peptides, 24, 779–788. Welling, L. L. M., Moreau, B. J. P., Bird, B. M., Hansen, S., & Carré, J. M. (2016). Exogenous testosterone increases men’s perceptions of their own physical dominance. Psychoneuroendo­ crinology, 64, 136–142. Wheeler, L., & Nezlek, J. (1977). Sex differences in social participation. Journal of Personality and Social Psychology, 35(10), 742–754. Whirledge, S., & Cidlowski, J. A. (2010). Glucocorticoids, stress, and fertility. Minerva Endocrinologica, 35(2), 109–125. Wise, S. P., & Herkenham, M. (1982). Opiate receptor distribution in the cerebral cortex of the Rhesus monkey. Science, 218, 387–389. Yehuda, R. (1997). Sensitization of the hypothalamic-pituitaryadrenal axis in posttraumatic stress disorder. Annals of the New York Academy of Sciences, 821(1), 57–75. Yehuda, R. (2000). Biology of posttraumatic stress disorder. Journal of Clinical Psychiatry, 61, 14–21. Yehuda, R., Golier, J.  A., & Kaufman, S. (2005). Circadian rhythm of salivary cortisol in Holocaust survivors with and  without PTSD. American Journal of Psychiatry, 162, 998–1000. Yehuda, R., Resnick, H.  S., Schmeidler, J., Yang, R.  K., & Pitman, R. K. (1998). Predictors of cortisol and 3-methoxy4-hydroxyphenylglycol responses in the acute aftermath of rape. Biological Psychiatry, 43(11), 855–859. Yehuda, R., Teicher, M. H., Trestman, R. L., Levengood, R. A., & Siever, L. J. (1996). Cortisol regulation in posttraumatic stress disorder and major depression: a chronobiological analysis. Biological psychiatry, 40(2), 79–88. Young, L.  J., & Wang, Z. (2004). The neurobiology of pair bonding. Nature Neuroscience, 7, 1048–1054. Zelena, D., Domokos, Á., Barna, I., Mergl, Z., Haller, J., & Makara, G. B. (2008). Control of the hypothalamo-pituitaryadrenal axis in the neonatal period: adrenocorticotropin and corticosterone stress responses dissociate in vasopressindeficient brattleboro rats. Endocrinology, 149(5), 2576–2583.

Mogilski, Wysocki, Reeve, Mitchell, Lunge, and Welling

365

CH A PT E R

21

Hormones, Circadian Rhythms, and Mental Health

Yasmine-Marie Cissé, Jeremy C. Borniger, and Randy J. Nelson

Abstract Circadian rhythms permit adaptations to predictable temporal environmental changes. Daily 24-hour rhythms are controlled by molecular clockworks within the brain that are set by the daily light–dark cycle. Downstream endocrine signaling conveys temporal information throughout the body. Mood disorders often present with disruptions in circadian clock-controlled responses, such as sleep and cortisol secretion, whereas circadian rhythm disruptions via jet lag, night-shift work, or light at night increase disordered affective symptoms. Evidence suggests strong associations between circadian rhythms and mental health, but only recently have studies begun to discover the direct interactions between the circadian system and mood regulation. This chapter provides an overview of circadian rhythms and the circadian regulation of the endocrine system. It discusses how the circadian and endocrine systems interact to affect depressive, anxious, and addictive responses. Finally, it discusses the potential detrimental effects the widespread use of nighttime light has for mood and cognition. Keywords:  mental health, circadian rhythm, diurnal rhythm, mood, depression, anxiety

The earth’s revolution, axial tilt, and elliptical orbit around the sun have remained essentially constant for the last several billion years. During this time, life has evolved under precise day–night cycles of 24 hours. Organisms have been selected to develop endogenous time-keeping mechanisms (i.e., circadian systems that have self-sustaining rhythms of about 24 hours) that ensure survival and reproductive success in the face of predictable changes in the environment. In mammals, environmental light is used to ­entrain or synchronize the circadian system to produce precise 24-hour rhythms. Environmental light (i.e., zeitgeber/time-giver) acts through a unique population of intrinsically photosensitive retinal ganglion cells (ipRGCs; Hattar, Liao, Takao, Berson, & Yau, 2002). These cells relay light information along the retinohypothalamic tract (RHT) to the master circadian clock in the hypothalamus: the ­suprachiasmatic nuclei (SCN). The ipRGCs are

maximally activated by short-wavelength light (~460 to 484 nm, violet to blue) and are significantly less responsive to long-wavelength (i.e., red) light (Berson, Dunn, & Takao, 2002). Importantly, ipRGCs also project directly or indirectly to brain structures that control emotional behavior and memory (e.g., amygdala and hippocampus; LeGates et al., 2012; Vandewalle, Maquet, & Dijk, 2009), providing a direct link between light exposure and cognitive function. The SCN can also be entrained by nonphotic cues such as food restriction or physical exercise (Edgar & Dement, 1991; Mistlberger, 1994), but the focus for this chapter will be on light-mediated changes to the circadian system and downstream alterations to endocrine function and mental health. Most of the experimental work presented in this chapter is based on animal model research. Animal models can be useful to understand the endocrine mechanisms underlying affective disorders, but there 367

are limitations in their applicability to humans, especially in the context of disordered mood. ­ Importantly, any animal model for affective disorders must have face validity (i.e., how well the animal model resembles the human disorder), predictive validity (i.e., the expected responses to treatment that are effective in humans), and construct validity (i.e., the similarity of the underlying mechanisms of the disorder). Understanding the mechanisms of the human disorder and the development of therapeutic interventions requires construct validity. Because of the difficulties associated with assessing emotions and mood in nonverbal individuals, many studies of humans, rather than nonhuman animals, have been conducted to understand the causes and treatments of affective disorders. Studies that attempt to correlate human affective changes with endocrine events typically involve people reporting their moods, which are then correlated with measurements of hormone concentrations. The effects of endocrine manipulations or natural endocrine fluctuations are then assessed.

Review of Circadian Rhythms

The master clock has an endogenous rhythm of ~24 hours in the absence of all environmental cues (Welsh, Takahashi, & Kay, 2010), reflecting genetic programs that drive the clockwork independent of external stimuli. The SCN are composed of mainly inhibitory (GABAergic) cells that transmit photic

information to the rest of the brain and body through direct synaptic connections and neuroendocrine pathways. Many targets of SCN signaling are important nodes regulating arousal and emotional processing, including the locus coeruleus, ventral tegmental area, and multiple hypothalamic nuclei that control endocrine signaling along the hypothalamic-pituitaryadrenal (HPA) axis. At the molecular level, circadian rhythms in gene expression are driven by a primary autoregulatory transcriptional/translational feedback loop with a period of approximately 24 hours (Figure 21.1). In mammals, the core clock genes Clock and Bmal1 (Arntl) serve as the “positive” arms of the clock. Their protein products (CLOCK and BMAL1) heterodimerize (i.e., join two different products to become a single compound) and bind to the promoters of their own negative regulators, Per (1,2,3) and Cry (1,2), to enhance their transcription (i.e., the copying of DNA into RNA). Once transcribed and translated, PER and CRY proteins hetero- and homodimerize (i.e., join two of the same product to become a single compound), re-enter the nucleus, and inhibit their own transcription, creating the “negative” arm of the clock. Ancillary loops further modulate the period and phase of the rhythm. For example, RAR-related orphan receptor alpha (RORα) and reverse-ErbA alpha (REV-ERBα) nuclear receptors act as transcription factors (i.e., proteins ­involved in transcribing DNA into RNA) to influence and stabilize Bmal1 expression (Guillaumond, Dardente,

Clock Npas2 Bmal1 CLOCK/ NPAS2

RORE

BMAL1

PER CRY

E-box

Per 1, 2, 3

E-box

Cry 1, 2

E-box

RORα

E-box

REV-ERB

E-box

CCGs nucleus

REVERB

RORα

cytoplasm

Figure 21.1  Molecular components of the circadian clock. The CLOCK/NPAS2:BMAL1 heterodimer binds to the E-box element in the promoter region of Per, Cry, ROR, and Rev-ERB, as well as a host of clock-controlled genes (CCGs). Once translated, PER and CRY proteins heterodimerize in the cytoplasm, translocate to the nucleus, and act as negative regulators to inhibit their own transcription. Nuclear receptors provide additional modulation to the main transcriptional translation feedback loop. ROR-α and Rev-ERBs bind to the RORE element in the promoter region of the Bmal 1 gene and promote or repress gene expression.

368

Hormones, Circadian Rhy thms, and Mental Health

Giguère, & Cermakian, 2005). Because CLOCK and BMAL1 additionally control the transcription of hundreds of so-called clock-controlled genes (CCGs) involved in immunity, metabolism, endocrine, and neuronal responses (Bozek et al., 2009), rhythmic gene expression is essential for health. Dysregulation of this network via circadian disruption can have far-reaching effects throughout the organism (Laing et al., 2015). More generally, circadian rhythms in body temperature, blood pressure, glucocorticoid secretion, immune responses, motor activity, and cognitive performance are disrupted in patients with mental illness (Lanfumey, Mongeau, & Hamon,  2013; Turek, 2007), suggesting that this system is vital for mental health. Many animal models of circadian disruption (environmental and genetic) demonstrate a causal relationship between aberrant biological rhythms and development of addictive behavior and bipolar-, depressive-, and anxiety-like phenotypes (Spencer et al., 2013). To follow, we discuss the primary evidence linking circadian rhythms to mental health disorders. We further outline the potential detrimental effects of circadian disruption on mood and cognition. Because of the rapid and unprecedented rise and widespread use of electric lights, special emphasis is placed on aberrant light exposure as a circadian disruptor and important contributor to the rise of affective disorders.

Circadian Regulation of Hormones Glucocorticoids

The HPA axis regulates the host of adaptive physiological processes initiated in response to physical and psychological stressors. Descending input from the limbic system, in response to a stressor, stimulates corticotropin-releasing hormone (CRH) neurons of the paraventricular nucleus of the hypothalamus (PVN) to release CRH into the hypophyseal portal system. CRH acts on the pituitary and triggers the release of adrenocorticotropic hormone (ACTH) into the bloodstream. ACTH induces synthesis and release of glucocorticoids from the adrenal cortex. In addition to limbic input to the HPA axis, the SCN exerts an organizational force on HPA activity. CRH, ACTH, and glucocorticoids exhibit robust circadian rhythms peaking at the onset of the active phase (Kalsbeek et al., 2012). The SCN exerts both direct and indirect influence on the circadian rhythm of the HPA axis (Figure 21.2). The SCN inhibits CRH synthesis through indirect activation of GABAergic neurons in the dorsomedial hypothalamus (DMH) and the

sub-PVN. The SCN can affect glucocorticoid release independently of the HPA axis. Direct and indirect SCN innervation of the PVN regulates sympathetic and parasympathetic efferent nerve fibers, resulting in a circadian pattern of peak sympathetic tone early in the active phase and a trough during rest. The circadian rhythm in autonomic tone is responsible for setting the phase of rhythmicity in the ­periphery: immune organs, liver, and endocrine organs such as the adrenal cortex (Bartness, Song, & Demas,  2001; Elenkov, Wilder, Chrousos, & Vizi, 2000; Ishida et al., 2005; Kalsbeek, La Fleur, Van Heijningen, & Buijs, 2004). Cortisol appears to play a multifaceted role in major depressive disorder (MDD; Herbert, 2013). First, increased resistance to the feedback actions of glucocorticoids is often observed. Second, daily rhythms in cortisol are perturbed. Finally, resting cortisol concentrations and the postawakening cortisol surge are increased in people at risk for MDD (Herbert, 2013). The negative feedback features of the HPA axis appear to be impaired in depressed patients. Excessive cortisol production has been reported in nearly 50 percent of depressed patients examined (Carroll et al.,  2007). Cortisol concentrations are often at their highest three to four hours after sleep onset in depressed people, and decrease throughout the daylight hours (Carroll, 1980). Because cortisol is normally secreted in a pronounced circadian p ­ attern, with peak concentrations measured in the early morning, this disturbance of the diurnal rhythm of cortisol secretion suggests an abnormal disinhibition of the neural centers regulating the release of adrenocorticotropic hormone (ACTH), the tropic hormone from the anterior pituitary gland that stimulates adrenal output. Chronic dysregulation of the HPA axis and the resultant high cortisol concentrations are common findings in depression (see Carroll et al., 2007). In cases of hormonal changes in depressed patients, it is unclear whether depression causes changes in hormone production or whether changes in hormone production cause depression. When depression is secondary to some endocrine or immune dysfunction, then more definitive statements can be made about the direction of causality. For example, patients with Cushing syndrome have adrenals that produce excessive cortisol, and depression is often a symptom of this disorder; however, patients with Addison disease have adrenal glands that produce insufficient cortisol, and depression is a defining symptom of this disease as well (see Disruption of Circadian Rhythms and Mental Health section). Cissé, Borniger, and Nelson

369

(B) Hypothalamus

(A) Hippocampus

PVN SubPVN 3r d V

DMH

Pituitary

SCN

Cortisol

IML

Adrenal

6AM 12PM 6PM 12AM 6AM

Figure 21.2  Suprachiasmatic nucleus (SCN) regulation of the hypothalamic-pituitary-adrenal (HPA) axis/glucocorticoid secretion. The SCN directs glucocorticoid secretion through HPA-dependent and -independent pathways through projections to both neuroendocrine and preautonomic centers of the paraventricular nucleus (PVN), respectively (A). Neuroendocrine projections direct the production and release of corticotropin-releasing hormone (CRH) and initiation of the HPA axis, whereas autonomic fibers directly act on the adrenal medulla to induce glucocorticoid release. Within the hypothalamus, the SCN directly innervates both the dorsomedial hypothalamus (DMH) and the sub-PVN to differentially regulate subsets of cells in the PVN (B). SCN, suprachiasmatic nucleus; DMH, dorsomedial hypothalamus; PVN, paraventricular nucleus; 3rd V, Third ventricle; IML, intermediolateral column of the spinal cord.

All peripheral organs contain endogenous c­ircadian clocks. The phase setting of peripheral clocks is dependent on neural or humoral cues from the SCN, namely, rhythmic autonomic tone, glucocorticoids, and melatonin (Dibner, Schibler, & Albrecht, 2010). Endogenous rhythmic clock gene expression in the adrenal cortex drives circadian gating of ACTH sensitivity (Oster, Damerow, Hut, & Eichele, 2006; Oster, Damerow, Kiessling, et al., 2006). Clock gene expression in all tissues can provide temporal gating of glucocorticoid ­sensitivity (Balsalobre et al., 2000). Several clock genes have been implicated in rhythmic gating of glucocorticoid receptor (GR) expression, sensitivity, and function. The CLOCK:BMAL1 heterodimer acetylation of the GR prevents associations with glucocorticoid-responsive elements (GREs; Nader, Chrousos, & Kino, 2009). CRY proteins can either 370

directly bind to GRs or competitively bind to GREs and prevent GR transcriptional activity (Lamia et al., 2011). Disruption of clock gene expression can have down­stream effects on typical HPA function and signaling. Single clock gene mutants exhibit altered HPA function. Clock null mice exhibit decreased and nonrhythmic glucocorticoid secretion (Takita et al., 2013). Bmal1 mutant mice similarly decrease serum glucocorticoid concentrations, have a flattened daily rhythm, and have a blunted physiological and behavioral response to ACTH, forced swim, and repeated restraint challenge (Leliavski, Shostak, Husse, & Oster, 2014). Period and cryptochrome mutants also display flattened daily rhythms, but conversely increase serum glucocorticoid concentrations (Dallman et al., 1993; Destici et al., 2013; Lamia et al., 2011).

Hormones, Circadian Rhy thms, and Mental Health

PG

PVN

Lig

SCN

ht

IML

Eye SCG Figure 21.3  Suprachiasmatic nucleus (SCN) regulation of melatonin secretion. In the absence of light at night, GABAergic signals from the SCN are inhibited allowing for the activity of the preautonomic cells of the paraventricular nucleus (PVN). These PVN neurons project to the preganglionic cells of the intermediolateral column of the spinal cord (IML), which synapse onto the postganglionic cells of the superior cervical ganglion (SCG), driving melatonin synthesis in the pineal gland (PG).

Melatonin

Melatonin is the humoral signal of night. The pineal gland synthesizes and secretes the indoleamine hormone, melatonin, into blood and cerebrospinal fluid exclusively during the night (Reiter, 1993). In the absence of light, glutamatergic preautonomic parvocellular cells of the PVN activate preganglionic cells of the intermediolateral (IML) cell column (Figure 21.3). These preganglionic cells in turn activate postganglionic sympathetic cells of the superior cervical ganglion (SCG), which project directly to the pineal gland and drive melatonin synthesis and release (Moore, 1995). Light inhibits melatonin secretion (in mammals) by depolarizing GABAergic cells within the SCN (through the RHT), which inhibit the autonomic cells of the PVN (TeclemariamMesbah, Ter Horst, Postema, Wortel, & Buijs, 1999). In this way, light-induced inhibition of the pineal gland is relinquished during extended darkness, ensuring that melatonin is only secreted in the absence of photic stimulation to the SCN. Although the pineal gland exhibits rhythmic clock gene expression in vitro (Maronde & Stehle, 2007), these rhythms are not apparent in vivo in the absence of SCN input. Investigating pineal function in laboratory mice can be difficult due to the lack of melatonin production in most commercially available mouse strains. C3H mice retain appreciable melatonin synthesis and are generally used as a model

or genetically crossed onto melatonin-deficient strains to produce “melatonin proficient” mice. Mice harboring a mutation (exon 19 [Δ19]) in the gene Clock crossed to be melatonin proficient have functional melatonin rhythms (Kennaway, Owens, Voultsios, & Varcoe, 2006). The authors suggest a functional homolog of CLOCK, NPAS2, may be compensating for the nonfunctional CLOCK. Additional studies identify a distinct lack of NPAS2 specifically in the pineal gland and the SCN (Garcia et al., 2000), suggesting that NPAS2 may not play a role in the SCN at baseline. There are no data, however, on its role in the absence of CLOCK. In vitro, melatonin can reset the SCN and alter clock gene expression; melatonin decreases Per1 and Clock and increases Npas2 expression in striatal neurons (Imbesi et al., 2009). Considered together, these data suggest that melatonin feedback may play a role in maintaining central clock function. Melatonin acts as a messenger of the circadian system. It conveys time-of-day information to the rest of the body to synchronize central and peripheral circadian clocks, acting through multiple G-proteincoupled receptors (GPCRs) termed MT1 and MT2, as well as nuclear receptors (ROR/RZR). Melatonin is an ancient and pleiotropic hormone, and its receptors are located on cells throughout the neuroendocrine and immune systems (Weaver, Rivkees, & Reppert, 1989; Weil, Borniger, Cisse, Abi Salloum, Cissé, Borniger, and Nelson

371

& Nelson, 2015). In the absence of exogenous melatonin, melatonin-deficient mouse strains exhibit decreased amplitudes of peripheral clock gene expression in the pituitary and adrenal cortex (TorresFarfan, Seron-Ferre, Dinet, & Korf, 2006).

Circadian and Seasonal Rhythms

For many mammals, annual shifts in behavioral and physiological phenotypes increase fitness in seasonally shifting environmental conditions. Melatonin conveys time-of-year information; the duration of melatonin secretion, rather than amplitude, conveys night length as a proxy of duration of seasonal changes in day length. Although humans do not e­ xhibit many overt seasonal rhythms, some individuals are susceptible to seasonal changes in mood. Seasonal affective disorder (SAD) presents during the late autumn or winter and is characterized by depressed affect, lethargy, loss of libido, hypersomnia, excessive weight gain, carbohydrate cravings, anxiety, and inability to focus attention or concentrate (Rosenthal, Sack, Skwerer, Jacobsen, & Wehr, 1988). In the Northern Hemisphere, symptoms usually begin between October and December and go into remission during the spring. Individuals suffering from SAD in the Southern Hemisphere display symptoms six months out of phase with Northern Hemisphere residents (Terman,  1988). With the onset of summer, SAD patients regain their energy and become active and elated, often to the point of hypomania or mania. Three features atypical of depression—hyperphagia, carbohydrate cravings, ­ and hypersomnia—set SAD apart from nonseasonal d ­ epression. SAD is frequently diagnosed as “bipolar II” depression or “atypical bipolar disorder,” particularly if hypomania or mania is present (DSM-5, 2013). Exposure to at least 1,500 lux is necessary for the inhibition of human melatonin secretion (Lewy et al., 2001). This requirement may explain why standard indoor levels of artificial illumination are insufficient to relieve the symptoms of SAD; much brighter light must be used for effective treatment. Thus, people who develop SAD, who are mainly female and in their 20s through 40s, may have defects in the light transduction pathways. One study suggests that people suffering from SAD have a small genetic mutation in the melanopsin molecule; people diagnosed with SAD were 5.6 times more likely than people with no history of psychopathology to have a missense variant (rs2675703 [P10L]) of the melanopsin gene (Roecklein et al., 2009). Thus, genetic deficits in the nonvisual light input pathway from the eye to the central biological 372

clock may represent an important mediator of SAD and point to additional effective treatments (reviewed in Roecklein et al., 2013b). Indeed, people diagnosed with SAD showed an impaired postillumination pupil response compared with healthy controls (Roecklein et al.,  2013a). Individuals with variations in the melanopsin gene (OPN4) who may suffer from SAD may display differences in alertness, circadian entrainment, and melatonin secretion.

Disruption of Circadian Rhythms and Mental Health

Prevalent anxiety-mood disorders include major depression, specific or social phobias, posttraumatic stress disorder, generalized anxiety disorder, panic disorder, agoraphobia, bipolar disorder, and obsessive-compulsive disorder. Among these, the chances of an adult in the United States developing major depression or an anxiety-mood disorder over a lifetime vary between 9 and 30 percent, representing a sizeable portion of the population (Kessler, Petukhova, Sampson, Zaslavsky, & Wittchen, 2012). In some disorders, such as SAD, a clear link between the circadian system and affective behavior is evident, as SAD patients show altered sensitivity to light and timed light therapy is efficacious in the treatment of SAD (Golden et al., 2005; Thompson, Stinson, & Smith, 1990). Light therapy also modestly reduces symptoms of nonseasonal depression, suggesting some benefit to enhancing signals to the circadian system (Golden et al.,  2005; Martiny, Lunde, Unden, Dam, & Bech,  2005; Wirz-Justice et al., 2005). In this section we will discuss the influence of the c­ ircadian system in regulating mood and addictive behavior.

Depression

Patients diagnosed with MDD often display circadian rhythm disruptions in sleep, immune, and endocrine systems (Thase, 1999; Turek, 2007). As noted, many patients with MDD present with arrhythmic hypercortisolism that is resistant to feedback inhibition by dexamethasone (Beck-Friis et al., 1985; Keller et al., 2006). Rhythms in glucocorticoid set the phase of clocks in various peripheral tissues and provide a circadian pattern of glucocorticoid sensitivity (Leliavski, Dumbell, Ott, & Oster, 2015). Loss of rhythmicity and phase of glucocorticoid secretion can alter extra-SCN clock function and HPA responses. Indeed, patients with MDD experience disrupted sleep, body temperature, and serum cytokine rhythms, as well as poor HPA feedback inhibition (Alesci et al., 2005; Beck-Friis et al., 1985; Thase, 1999).

Hormones, Circadian Rhy thms, and Mental Health

Aberrant rhythms in melatonin production also underlie mood disorders such as bipolar disorder and depression, as patients have overall reduced melatonin concentrations or greatly disrupted rhythms in melatonin secretion (Claustrat, Chazot, Brun, Jordan, & Sassolas, 1984; Nurnberger et al., 2000; Srinivasan et al., 2006). Loss of a nighttime melatonin peak eliminates signals of time of day to the body, including mood centers. Indeed, knockout of the melatonin receptor MT1 increases depressive-like behavior in mice (Weil et al., 2006). Similarly, constant exposure to light or dim light at night decreases nocturnal melatonin secretion and increases depressive-like behavior (Bedrosian, Fonken, Walton, Haim, & Nelson, 2011; Bedrosian et al., 2013). Melatonin alters the adrenal cortex glucocorticoid response to ACTH (Campino et al., 2011; Torres-Farfan et al., 2003), whereby the loss of melatonin contributes to the dysregulation in HPA feedback from hypercortisolism. Antidepressant treatment returns melatonin and glucocorticoid rhythms to baseline in MDD patients (Golden et al., 1988; Linkowski et al., 1987). Agomelatine, an MT1/MT2 agonist and serotonin receptor 2C (5HT2C) antagonist, decreases symptom severity in patients with MDD (Olié et al., 2007). Together, these data suggest that a functional interplay between the melatonin and glucocorticoid rhythm may be necessary to regulate mood. Research in humans demonstrates marked changes in circadian function throughout the lifespan, coincident with the development of mental disorders. In two studies using high-quality postmortem brain tissue sampled at different times of the day, researchers were able to directly investigate clock gene expression in aged people and those with MDD (Chen et al., 2016; Li et al.,  2013). Because aging is associated with impaired sleep, cognition, and mood, understanding the contributions of circadian deregulation to the emergence of these problems is important for developing treatments. During the course of normal aging, the majority of genes become less rhythmic, but some compensatory transcripts increase their rhythmic expression (Chen et al.,  2016; Li et al., 2013). Furthermore, brains from people with MDD displayed dramatically reduced rhythmicity in many core clock components, suggesting that targeting the circadian system could be a viable strategy for the treatment of mental disorders.

Anxiety

Relationships among the circadian system and anxiety-like disorders were first established from basic research in mice with targeted disruption of core

molecular clock components. For instance, mice harboring the Δ19 mutation in the gene Clock display mania-like behavior, have reduced anxiety-like behavior, and are less fearful of aversive stimuli than wild-type (i.e., control) mice (Roybal et al., 2007). These changes seem to stem from disruptions in CLOCK-regulated cholecystokinin (CCK) expression in the ventral tegmental area (VTA; Arey et al., 2014). Reciprocally, mice lacking both Per1 and Per2 (but not either alone) have increased anxietylike behavior, increased social-defeat stress responses (a paradigm used to induce anxiety in rodents), and reduced Per1/2 gene expression within the nucleus accumbens (NAc; Spencer et al., 2013). Subsequent RNA interference-mediated inhibition of Per1/2 expression in the NAc of wild-type mice produced anxiety-like behavior, suggesting a causal role for these core clock components in the NAc for regulating anxiety. Lesions of the SCN produce antidepressant-like behavior in tests of behavioral despair, independent of alterations in anxiety (Tataroğlu, Aksoy, Yılmaz, & Canbeyli, 2004; Tuma, Strubbe, Mocaër, & Koolhaas, 2005), suggesting that clock gene actions outside the SCN are important for regulating anxiety-like behavior. Other studies have investigated how environmental disruption of circadian rhythms (i.e., with atypical light exposure) contributes to the development of anxiety-like behavior. In adult rats, chronic constant light induces depressive- and anxietylike behavioral responses (Tapia-Osorio, SalgadoDelgado, Angeles-Castellanos, & Escobar,  2013). However, results are inconsistent across species (Ashkenazy, Einat, & Kronfeld-Schor, 2009; Castro et al., 2005; Fonken et al., 2009) and depend on the developmental window during which circadian disruption occurs, as well as the type of light and its intensity (Bedrosian et al.,  2013; Borniger, McHenry, Abi Salloum, & Nelson,  2014; Cissé, Peng, & Nelson, 2016). Additional covariates are the extent to which light exposure alters sleep between different species of diurnal and nocturnal rodents (e.g., Borniger, Weil, Zhang, & Nelson, 2013; Stenvers et al.,  2016), as sleep disruption can be an independent contributor to the development of affective disorders (Jagannath, Peirson, & Foster, 2013; LeGates, Fernandez, & Hattar, 2014).

Addiction

Addiction, in common with other mental disorders, is associated with circadian disruption, though much of the disruption has been attributed to the types of drugs themselves, especially opiates and stimulants, Cissé, Borniger, and Nelson

373

even during withdrawal (Johanson, Roehrs, Schuh, & Warbasse, 1999; Schierenbeck, Riemann, Berger, & Hornyak,  2008; Watson, Bakos, Compton, & Gawin, 1992). Specifically, drugs and alcohol alter clock gene expression in mice and circadian rhythms in rodents and humans (Perreau-Lenz & Spanagel, 2008; Spanagel, Rosenwasser, Schumann & Sarkar, 2005). Addictive behaviors cycle in a circadian fashion and certain clock genes are necessary for drug responses in mice (Parekh & McClung,  2015). Mice with deletions of Clock increase cocaine reward and increase dopamine in the VTA (McClung et al., 2005; Ozburn, Larson, Self, & McClung,  2012; Roybal et al., 2007). Per genes have similarly been implicated in regulation of alcohol consumption and altered locomotor and reward response to cocaine (Abarca, Albrecht, & Spanagel,  2002; Spanagel, Pendyala, et al., 2005). The association between clock genes and cocaine addiction is weaker in humans (Malison, Kranzler, Yang, & Gelernter, 2006), but alcohol and heroin dependence are associated with polymorphisms in Bmal1 and Per3, respectively (Kovanen et al., 2010; Zou et al., 2007). The majority of the reward response–altering mutations mentioned previously are whole-body clock gene knockouts. The SCN indirectly innervates reward centers of the brain such as the ventral tegmental area (VTA; Luo & Aston,  2009). The VTA is a part of the mesolimbic dopamine system, commonly termed as the VTA-NAc (nucleus accumbens) circuit. The VTA is a major concentration of dopaminergic neurons that integrates external stimuli and, depending on the resulting dopamine release, designates whether a situation is rewarding or aversive (Russo & Nestler, 2013). Under normal conditions, this modulates the rewarding response to food and sex but is also directly acted upon by drugs of abuse. The NAc is a direct target of dopaminergic VTA projections. Together these brain regions project to and get reciprocal input from the amygdala, hippocampus, hypothalamus, and medial prefrontal cortex to mediate memory and response to rewarding and aversive stimuli. Global clock gene knockouts may alter reward responses due to lack of descending clock input to the reward system or altered clock gene expression in reward centers of the brain. Recent studies indicate that specific deletion of NPAS2 in the NAc decreases cocaine reward response and self-administration (Ozburn et al., 2015). Restoring Clock in the VTA of Clock Δ19 resolves the increase in cocaine reward (Roybal et al.,  2007). Indeed, specific deletion of 374

clock genes in the VTA and the NAc alters reward behavior, suggesting that clock function in individual regions of the reward pathway is important to mediating drug reward. Clock genes are not the sole messengers of the circadian system. Glucocorticoids modulate addictive behaviors via the reward system (Becker-Krail & McClung,  2016). Tyrosine hydroxylase (TH), the rate-limiting enzyme in dopamine production, contains a glucocorticoid response element, and glucocorticoids increased TH mRNA (Barrot et al., 2000; Fossom, Sterling, & Tank,  1992; Hagerty, Morgan, Elango, & Strong, 2008; Lewis, Harrington, & Chikaraishi, 1987). Deletion of GR in dopamine receptor 1a–expressing neurons decreases self-­ administration and VTA dopamine cell firing (Ambroggi et al., 2009). Moreover, melatonin modulates the reward system at the behavioral, dopamine content, release, and receptor sensitivity levels (Alexiuk & Vriend, 1993; Hamdi, 1998; Papp, Litwa, Łasoń-Tyburkiewicz, & Gruca,  2010; Zisapel, Egozi, & Laudon,  1982). MT1 exhibits circadian rhythms in expression in the central dopaminergic system (Uz et al., 2005), whereas MT2 is necessary for the reversal of morphine reward (Han, Xu, Yu, Shen, & Wei, 2008), and deletion of either melatonin receptor eliminates the methamphetamine reward response (Clough, Hutchinson, Hudson, & Dubocovich, 2014; Hutchinson, Hudson, & Dubocovich, 2012).

Conclusion

Life evolved on a rotating planet, and circadian rhythms in physiology and behavior developed as adaptations to predict day and night. The appropriate synchronization of the light–dark cycles with brain function is critical for appropriate endocrine function. Hormone secretion is synchronized by the circadian clocks to coincide with optimal receptor availability and overall physiological and behavioral functionality. The biological clock can be considered the conductor of the hormonal orchestra, keeping the endocrine system synchronized. Exposure to bright light during the day and dark environments at night has shaped how hormones and mood are regulated. Presumably, this arrangement harmonized with the lifestyles of our ancestors to optimize human functioning. The introduction of electric lights was a milestone in evolutionary history, allowing humans to control both inside and outside environments. The ability to illuminate the night led to safer, wealthier,

Hormones, Circadian Rhy thms, and Mental Health

and more productive individuals and societies. The study of circadian biology developed decades after the widespread adoption of electric lights. Only recently are we learning about the effects of artificial light at night on the brain and body. For example, in 2016, the American Medical Association issued a report cautioning against blue-rich LED streetlights, citing the detrimental effects on human health (American Medical Association,  2016). Based on several National Institutes of Health workshops, a growing scientific consensus suggests that light at night negatively affects affective state. Although LED lights are often cited as a big part of the problem, they can contribute to the solution of light pollution as well. LED lights tuned to longer wavelengths could be used at night to avoid disrupting circadian rhythms and thus avoid disrupting rhythms of endocrine function and dysregulated mood. By careful management of light exposure at night, circadian dysregulation and associated mood disorders may be diminished.

Acknowledgments

Preparation of this review was supported by NIH NRSA grant F31 ES026890 (YMC), a Pelotonia graduate student support grant (JCB), and NIH grants R01NS092388 and R21CA202745 (RJN).

References

Abarca, C., Albrecht, U., & Spanagel, R. (2002). Cocaine sensitization and reward are under the influence of circadian genes and rhythm. Proceedings of the National Academy of Sciences of the United States of America, 99(13), 9026–9030. http://doi.org/10.1073/pnas.142039099 Alesci, S., Martinez, P. E., Kelkar, S., Ilias, I., Ronsaville, D. S., Listwak, S.  J., . . . Gold, P.  W. (2005). Major depression is associated with significant diurnal elevations in plasma interleukin-6 levels, a shift of its circadian rhythm, and loss of physiological complexity in its secretion: Clinical implications. Journal of Clinical Endocrinology & Metabolism, 90(5), 2522–2530. http://doi.org/10.1210/jc.2004–1667 Alexiuk, N.  A.  M., & Vriend, J.  P. (1993). Melatonin reduces dopamine content in the neurointermediate lobe of male Syrian hamsters. Brain Research Bulletin, 32(4), 433–436. http://doi.org/10.1016/0361-9230(93)90211-S Ambroggi, F., Turiault, M., Milet, A., Deroche-Gamonet, V., Parnaudeau, S., Balado, E., . . . Tronche, F. (2009). Stress and addiction: Glucocorticoid receptor in dopaminoceptive neurons facilitates cocaine seeking. Nature Neuroscience, 12(3), 247–249. http://doi.org/10.1038/nn.2282 American Medical Association. (2016). Human and environmental effects of light emitting diode (LED) community lighting. Retrieved from https://www.ama-assn.org/about-us/councilscience-public-health-csaph-reports American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: Author.

Arey, R. N., Enwright, J. F., Spencer, S. M., Falcon, E., Ozburn, A. R., Ghose, S., . . . McClung, C. A. (2014). An important role for cholecystokinin, a CLOCK target gene, in the development and treatment of manic-like behaviors. Molecular Psychiatry, 19(3), 342–350. http://doi.org/10.1038/mp.2013.12 Ashkenazy, T., Einat, H., & Kronfeld-Schor, N. (2009). We are in the dark here: Induction of depression- and anxiety-like behaviours in the diurnal fat sand rat, by short daylight or melatonin injections. International Journal of Neuro­ psychopharmacology, 12(1), 83–93. Balsalobre, A., Brown, S.  A., Marcacci, L., Tronche, F., Kellendonk, C., Reichardt, H.  M., . . . Schibler, U. (2000). Resetting of circadian time in peripheral tissues by glucocorticoid signaling. Science, 289(5488), 2344–2347. Barrot, M., Marinelli, M., Abrous, D.  N., Rougé-Pont, F., Le Moal, M., & Piazza, P. V. (2000). The dopaminergic hyperresponsiveness of the shell of the nucleus accumbens is hormone-dependent. European Journal of Neuroscience, 12(3), 973–979. http://doi.org/10.1046/j.1460–9568.2000.00996.x Bartness, T.  J., Song, C.  K., & Demas, G.  E. (2001). SCN efferents to peripheral tissues: Implications for biological rhythms. Journal of Biological Rhythms, 16(3), 196–204. Becker-Krail, D., & McClung, C. (2016). Implications of circadian rhythm and stress in addiction vulnerability. F1000Research, 5, 59. http://doi.org/10.12688/f1000research.7608.1 Beck-Friis, J., Ljunggren, J.-G., Thorén, M., Rosen, D., von, Kjellman, B. F., & Wetterberg, L. (1985). Melatonin, cortisol and ACTH in patients with major depressive disorder and healthy humans with special reference to the outcome of the dexamethasone suppression test. Psychoneuroendocrinology, 10(2), 173–186. http://doi.org/10.1016/0306-4530(85)90055-1 Bedrosian, T.  A., Fonken, L.  K., Walton, J.  C., Haim, A., & Nelson, R. J. (2011). Dim light at night provokes depressionlike behaviors and reduces CA1 dendritic spine density in female hamsters. Psychoneuroendocrinology, 36(7), 1062–1069. http:// doi.org/10.1016/j.psyneuen.2011.01.004 Bedrosian, T. A., Vaughn, C. A., Galan, A., Daye, G., Weil, Z. M., & Nelson, R.  J. (2013). Nocturnal light exposure impairs affective responses in a wavelength-dependent manner. Journal of Neuroscience, 33(32), 13081–13087. Berson, D.  M., Dunn, F.  A., & Takao, M. (2002). Photo­ transduction by retinal ganglion cells that set the circadian clock. Science, 295(5557), 1070–1073. Borniger, J. C., McHenry, Z. D., Abi Salloum, B. A., & Nelson, R.  J. (2014). Exposure to dim light at night during early development increases adult anxiety-like responses. Physiology & Behavior, 133, 99–106. http://doi.org/10.1016/j.physbeh. 2014.05.012 Borniger, J. C., Weil, Z. M., Zhang, N., & Nelson, R. J. (2013). Dim light at night does not disrupt timing or quality of sleep in mice. Chronobiology International, 30(8), 1016–1023. http://doi.org/10.3109/07420528.2013.803196 Bozek, K., Relógio, A., Kielbasa, S. M., Heine, M., Dame, C., Kramer, A., & Herzel, H. (2009). Regulation of clockcontrolled genes in mammals. PLoS One, 4(3), e4882. http:// doi.org/10.1371/journal.pone.0004882 Campino, C., Valenzuela, F.  J., Torres-Farfan, C., Reynolds, H. E., Abarzua-Catalan, L., Arteaga, E., . . . Seron-Ferre, M. (2011). Melatonin exerts direct inhibitory actions on ACTH responses in the human adrenal gland. Hormone and Metabolic Research, 43(5), 337–342. http://doi.org/10.1055/s-00311271693

Cissé, Borniger, and Nelson

375

Carroll, B.  J. (1980). Dexamethasone suppression test in depression. Lancet, 2(8206), 1249. Carroll, B.  J., Cassidy, F., Naftolowitz, D., Tatham, N.  E., Wilson, W.  H., Iranmanesh, A., . . . Veldhuis, J.  D. (2007). Pathophysiology of hypercortisolism in depression. Acta Psychiatrica Scandinavica, 115(s433), 90–103. http://doi. org/10.1111/j.1600–0447.2007.00967.x Castro, J. P. M. V., Frussa-Filho, R., Fukushiro, D. F., Chinen, C. C., Abílio, V. C., & Silva, R. H. (2005). Effects of longterm continuous exposure to light on memory and anxiety in mice. Physiology & Behavior, 86(1), 218–223. http://doi. org/10.1016/j.physbeh.2005.07.009 Chen, C.-Y., Logan, R. W., Ma, T., Lewis, D. A., Tseng, G. C., Sibille, E., & McClung, C.  A. (2016). Effects of aging on circadian patterns of gene expression in the human prefrontal cortex. Proceedings of the National Academy of Sciences of the  United States of America, 113(1), 206–211. http://doi. org/10.1073/pnas.1508249112 Cissé, Y. M., Peng, J., & Nelson, R. J. (2016). Dim light at night prior to adolescence increases adult anxiety-like behaviors. Chronobiology International, 33(10), 1473–1480. http://doi. org/10.1080/07420528.2016.1221418 Claustrat, B., Chazot, G., Brun, J., Jordan, D., & Sassolas, G. (1984). A chronobiological study of melatonin and cortisol secretion in depressed subjects: Plasma melatonin, a biochemical marker in major depression. Biological Psychiatry, 19(8), 1215–1228. Clough, S. J., Hutchinson, A. J., Hudson, R. L., & Dubocovich, M. L. (2014). Genetic deletion of the MT1 or MT2 melatonin receptors abrogates methamphetamine-induced reward in C3H/HeN mice. Physiology & Behavior, 132, 79–86. http:// doi.org/10.1016/j.physbeh.2014.04.049 Dallman, M.  F., Strack, A.  M., Akana, S.  F., Bradbury, M.  J., Hanson, E. S., Scribner, K. A., & Smith, M. (1993). Feast and famine: Critical role of glucocorticoids with insulin in daily energy flow. Frontiers in Neuroendocrinology, 14(4), 303–347. http://doi.org/10.1006/frne.1993.1010 Destici, E., Jacobs, E. H., Tamanini, F., Loos, M., van der Horst, G.  T.  J., Oklejewicz, M., . . . Ndesendo, V. (2013). Altered phase-relationship between peripheral oscillators and envi­ ronmental time in Cry1 or Cry2 deficient mouse models for early and late chronotypes. PLoS One, 8(12), e83602. http:// doi.org/10.1371/journal.pone.0083602 Dibner, C., Schibler, U., & Albrecht, U. (2010). The mammalian circadian timing system: Organization and coordination of central and peripheral clocks. Annual Review of Physiology, 72(1), 517–549. http://doi.org/10.1146/annurev-physiol021909-135821 Edgar, D.  M., & Dement, W.  C. (1991). Regularly scheduled voluntary exercise synchronizes the mouse circadian clock. American Journal of Physiology, 261(4 Pt. 2), R928–R933. Elenkov, I.  J., Wilder, R.  L., Chrousos, G.  P., & Vizi, E.  S. (2000). The sympathetic nerve—An integrative interface between two supersystems: The brain and the immune system. Pharmacological Reviews, 52(4), 595–638. Fonken, L. K., Finy, M. S., Walton, J. C., Weil, Z. M., Workman, J. L., Ross, J., & Nelson, R. J. (2009). Influence of light at night on murine anxiety- and depressive-like responses. Behavioural Brain Research, 205(2), 349–354. http://doi. org/10.1016/j.bbr.2009.07.001 Fossom, L. H., Sterling, C. R., & Tank, A. W. (1992). Regulation of tyrosine hydroxylase gene transcription rate and

376

tyrosine  hydroxylase mRNA stability by cyclic AMP and glucocorticoid. Molecular Pharmacology, 42(5), 898–908. Garcia, J. A., Zhang, D., Estill, S. J., Michnoff, C., Rutter, J., Reick, M., . . . McKnight, S.  L. (2000). Impaired cued and contextual memory in NPAS2-deficient mice. Science, 288(5474), 2226–2230. Golden, R. N., Gaynes, B. N., Ekstrom, R. D., Hamer, R. M., Jacobsen, F. M., Suppes, T., . . . Nemeroff, C. B. (2005). The efficacy of light therapy in the treatment of mood disorders: A review and meta-analysis of the evidence. American Journal of Psychiatry, 162(4), 656–662. http://doi.org/10.1176/appi. ajp.162.4.656 Golden, R.  N., Markey, S.  P., Risby, E.  D., Rudorfer, M.  V., Cowdry, R.  W., & Potter, W.  Z. (1988). Antidepressants reduce whole-body norepinephrine turnover while enhancing 6-hydroxymelatonin output. Archives of General Psychiatry, 45(2), 150–154. Guillaumond, F., Dardente, H., Giguère, V., & Cermakian, N. (2005). Differential control of Bmal1 circadian transcription by REV-ERB and ROR nuclear receptors. Journal of Biological Rhythms, 20(5), 391–403. http://doi.org/10.1177/ 0748730405277232 Hagerty, T., Morgan, W. W., Elango, N., & Strong, R. (2008). Identification of a glucocorticoid-responsive element in the promoter region of the mouse tyrosine hydroxylase gene. Journal of Neurochemistry, 76(3), 825–834. http://doi.org/ 10.1046/j.1471–4159.2001.00072.x Hamdi, A. (1998). Melatonin administration increases the affinity of D2 dopamine receptors in the rat striatum. Life Sciences, 63(23), 2115–2120. http://doi.org/10.1016/S00243205(99)80008-3 Han, J., Xu, Y., Yu, C.-X., Shen, J., & Wei, Y.-M. (2008). Melatonin reverses the expression of morphine-induced con­ditioned place preference through its receptors within central nervous system in mice. European Journal of Pharmacology, 594(1), 125–131. http://doi.org/10.1016/j.ejphar. 2008.07.049 Hattar, S., Liao, H.-W., Takao, M., Berson, D.  M., & Yau, K.-W. (2002). Melanopsin-containing retinal ganglion cells: Architecture, projections, and intrinsic photosensitivity. Science, 295(5557), 1065–1070. Herbert, J. (2013). Cortisol and depression: Three questions for psychiatry. Psychological Medicine, 43(3), 449–469. http:// doi.org/10.1017/S0033291712000955 Hutchinson, A. J., Hudson, R. L., & Dubocovich, M. L. (2012). Genetic deletion of MT1 and MT2 melatonin receptors differentially abrogates the development and expression of methamphetamine-induced locomotor sensitization during the day and the night in C3H/HeN mice. Journal of Pineal Research, 53(4), 399–409. http://doi.org/10.1111/j.1600079X.2012.01010.x Imbesi, M., Arslan, A. D., Yildiz, S., Sharma, R., Gavin, D., Tun, N.,  .  .  .  Uz, T. (2009). The melatonin receptor MT1 is required for the differential regulatory actions of melatonin on neuronal “clock” gene expression in striatal neurons in vitro. Journal of Pineal Research, 46(1), 87–94. http://doi. org/10.1111/j.1600-079X.2008.00634.x Ishida, A., Mutoh, T., Ueyama, T., Bando, H., Masubuchi, S., Nakahara, D., . . . Okamura, H. (2005). Light activates the adrenal gland: Timing of gene expression and glucocorticoid release. Cell Metabolism, 2(5), 297–307. http://doi.org/ 10.1016/j.cmet.2005.09.009

Hormones, Circadian Rhy thms, and Mental Health

Jagannath, A., Peirson, S. N., & Foster, R. G. (2013). Sleep and circadian rhythm disruption in neuropsychiatric illness. Current Opinion in Neurobiology, 23(5), 888–894. http:// doi.org/10.1016/j.conb.2013.03.008 Johanson, C.-E., Roehrs, T., Schuh, K., & Warbasse, L. (1999). The effects of cocaine on mood and sleep in cocaine-dependent males. Experimental and Clinical Psychopharmacology, 7(4), 338–346. http://doi.org/10.1037/1064-1297.7.4.338 Kalsbeek, A., La Fleur, S., Van Heijningen, C., & Buijs, R. M. (2004). Suprachiasmatic GABAergic inputs to the paraven­ tricular nucleus control plasma glucose concentrations in the rat via sympathetic innervation of the liver. Journal of Neuroscience, 24(35), 7604–7613 Kalsbeek, A., van der Spek, R., Lei, J., Endert, E., Buijs, R. M., & Fliers, E. (2012). Circadian rhythms in the hypothalamo– pituitary–adrenal (HPA) axis. Molecular and Cellular Endocrinology, 349(1), 20–29. http://doi.org/10.1016/j.mce. 2011.06.042 Keller, J., Flores, B., Gomez, R. G., Solvason, H. B., Kenna, H., Williams, G.  H., & Schatzberg, A.  F. (2006). Cortisol circadian rhythm alterations in psychotic major depression. Biological Psychiatry, 60(3), 275–281. http://doi.org/10.1016/ j.biopsych.2005.10.014 Kennaway, D.  J., Owens, J.  A., Voultsios, A., & Varcoe, T.  J. (2006). Functional central rhythmicity and light entrainment, but not liver and muscle rhythmicity, are Clock independent. American Journal of Physiology, 291(4), R1172–R1180. Kessler, R. C., Petukhova, M., Sampson, N. A., Zaslavsky, A. M., & Wittchen, H.-U. (2012). Twelve-month and lifetime prevalence and lifetime morbid risk of anxiety and mood disorders in the United States. International Journal of Methods in Psychiatric Research, 21(3), 169–184. http://doi. org/10.1002/mpr.1359 Kovanen, L., Saarikoski, S. T., Haukka, J., Pirkola, S., Aromaa, A., Lönnqvist, J., & Partonen, T. (2010). Circadian clock gene polymorphisms in alcohol use disorders and alcohol consumption. Alcohol and Alcoholism, 45(4), 303–311. Laing, E.  E., Johnston, J.  D., Möller-Levet, C.  S., Bucca, G., Smith, C. P., Dijk, D.-J., & Archer, S. N. (2015). Exploiting human and mouse transcriptomic data: Identification of circadian genes and pathways influencing health. BioEssays, 37(5), 544–556. http://doi.org/10.1002/bies.201400193 Lamia, K. A., Papp, S. J., Yu, R. T., Barish, G. D., Uhlenhaut, N.  H., Jonker, J.  W., . . . Evans, R.  M. (2011). Crypto­ chromes mediate rhythmic repression of the glucocorticoid receptor. Nature, 480(7378), 552–556. http://doi.org/ 10.1038/nature10700 Lanfumey, L., Mongeau, R., & Hamon, M. (2013). Biological rhythms and melatonin in mood disorders and their treatments. Pharmacology & Therapeutics, 138(2), 176–184. http://doi.org/10.1016/j.pharmthera.2013.01.005 LeGates, T. A., Altimus, C. M., Wang, H., Lee, H.-K., Yang, S., Zhao, H., . . . Hattar, S. (2012). Aberrant light directly impairs mood and learning through melanopsin-expressing neurons. Nature, 491(7425), 594–598. http://doi.org/10.1038/ nature11673 LeGates, T. A., Fernandez, D. C., & Hattar, S. (2014). Light as a central modulator of circadian rhythms, sleep and affect. Nature Reviews Neuroscience, 15(7), 443–454. http://doi. org/10.1038/nrn3743 Leliavski, A., Dumbell, R., Ott, V., & Oster, H. (2015). Adrenal clocks and the role of adrenal hormones in the regulation of

circadian physiology. Journal of Biological Rhythms, 30(1), 20–34. http://doi.org/10.1177/0748730414553971 Leliavski, A., Shostak, A., Husse, J., & Oster, H. (2014). Impaired glucocorticoid production and response to stress in Arntl-deficient male mice. Endocrinology, 155(1), 133–142. http://doi.org/10.1210/en.2013–1531 Lewis, E. J., Harrington, C. A., & Chikaraishi, D. M. (1987). Transcriptional regulation of the tyrosine hydroxylase gene by glucocorticoid and cyclic AMP. Proceedings of the National Academy of Sciences of the United States of America, 84(11), 3550–3554. Lewy, A. J., Bauer, V. K., Hasler, B. P., Kendall, A. R., Pires, M. L., & Sack, R.  L. (2001). Capturing the circadian rhythms of free-running blind people with 0.5 mg melatonin. Brain Research, 918(1–2), 96–100. Li, J. Z., Bunney, B. G., Meng, F., Hagenauer, M. H., Walsh, D. M., Vawter, M. P., . . . Bunney, W. E. (2013). Circadian patterns of gene expression in the human brain and disruption in major depressive disorder. Proceedings of the National Academy of Sciences of the United States of America, 110(24), 9950–9955. http://doi.org/10.1073/pnas.1305814110 Linkowski, P., Mendlewicz, J., Kerkhofs, M., Leclercq, R., Golstein, J., Brasseur, M., . . . Van Cauter, E. (1987). 24-Hour profiles of adrenocorticotropin, cortisol, and growth hormone in major depressive illness: Effect of antidepressant treatment. Journal of Clinical Endocrinology & Metabolism, 65(1), 141–152. http://doi.org/10.1210/jcem-65-1-141 Luo, A., & Aston-Jones, G. (2009). Circuit projection from suprachiasmatic nucleus to ventral tegmental area: A novel circadian output pathway. European Journal of Neuroscience, 29(4), 748–760. Malison, R.  T., Kranzler, H.  R., Yang, B.-Z., & Gelernter, J. (2006). Human clock, PER1 and PER2 polymorphisms: Lack of association with cocaine dependence susceptibility and cocaine-induced paranoia. Psychiatric Genetics, 16(6), 245–249. http://doi.org/10.1097/01.ypg.0000242198.59020.ca Maronde, E., & Stehle, J.  H. (2007). The mammalian pineal gland: Known facts, unknown facets. Trends in Endocrinology & Metabolism, 18(4), 142–149. http://doi.org/10.1016/j.tem. 2007.03.001 Martiny, K., Lunde, M., Unden, M., Dam, H., & Bech, P. (2005). Adjunctive bright light in non-seasonal major depression: Results from patient-reported symptom and well-being scales. Acta Psychiatrica Scandinavica, 111(6), 453–459. http://doi. org/10.1111/j.1600–0447.2005.00532.x McClung, C.  A., Sidiropoulou, K., Vitaterna, M., Takahashi, J. S., White, F. J., Cooper, D. C., & Nestler, E. J. (2005). Regulation of dopaminergic transmission and cocaine reward by the Clock gene. Proceedings of the National Academy of Sciences of the United States of America, 102(26), 9377–9381. http://doi.org/10.1073/pnas.0503584102 Mistlberger, R. E. (1994). Circadian food-anticipatory activity: Formal models and physiological mechanisms. Neuroscience & Biobehavioral Reviews, 18(2), 171–195. http://doi.org/10.1016/ 0149-7634(94)90023-X Moore, R.  Y. (1995). Neural control of the pineal gland. Behavioural Brain Research, 73(1–2), 125–130. http://doi. org/10.1016/0166-4328(96)00083-6 Nader, N., Chrousos, G. P., & Kino, T. (2009). Circadian rhythm transcription factor CLOCK regulates the transcriptional activity of the glucocorticoid receptor by acetylating its hinge region lysine cluster: Potential physiological implications.

Cissé, Borniger, and Nelson

377

FASEB Journal, 23(5), 1572–1583. http://doi.org/10.1096/ fj.08–117697 Nurnberger, J. I., Adkins, S., Lahiri, D. K., Mayeda, A., Hu, K., Lewy, A., . . . Davis-Singh, D. (2000). Melatonin suppression by light in euthymic bipolar and unipolar patients. Archives of General Psychiatry, 57(6), 572–579. http://doi.org/10.1001/ ARCHPSYC.57.6.572 Olié, J.  P., Kasper, S., Anderson, I.  M., Armstrong, S.  M., McNulty, O.  M., Guardiola-Lemaitre, B., . . . GuardiolaLemaitre, B. (2007). Efficacy of agomelatine, a MT1/MT2 receptor agonist with 5-HT2C antagonistic properties, in major depressive disorder. International Journal of Neuro­ psycho­pharmacology, 10(5), 661–673. http://doi.org/10.1017/ S1461145707007766 Oster, H., Damerow, S., Hut, R.  A., & Eichele, G. (2006a). Transcriptional profiling in the adrenal gland reveals circadian regulation of hormone biosynthesis genes and nucleosome assembly genes. Journal of Biological Rhythms, 21(5), 350–361. http://doi.org/10.1177/0748730406293053 Oster, H., Damerow, S., Kiessling, S., Jakubcakova, V., Abraham, D., Tian, J., . . . Eichele, G. (2006b). The circadian rhythm of glucocorticoids is regulated by a gating mechanism residing in the adrenal cortical clock. Cell Metabolism, 4(2), 163–173. http://doi.org/10.1016/j.cmet.2006.07.002 Ozburn, A. R., Falcon, E., Twaddle, A., Nugent, A. L., Gillman, A.  G., Spencer, S.  M., . . . McClung, C.  A. (2015). Direct regulation of diurnal Drd3 expression and cocaine reward by NPAS2. Biological Psychiatry, 77(5), 425–433. http://doi. org/10.1016/j.biopsych.2014.07.030 Ozburn, A. R., Larson, E. B., Self, D. W., & McClung, C. A. (2012). Cocaine self-administration behaviors in ClockΔ19 mice. Psychopharmacology, 223(2), 169–177. http://doi. org/10.1007/s00213-012-2704-2 Papp, M., Litwa, E., Łasoń-Tyburkiewicz, M., & Gruca, P. (2010). Effects of melatonin in a place preference condi­ tioning depend on the time of administration. Pharmacological Reports, 62(6), 1023–1029. http://doi.org/10.1016/S17341140(10)70364-9 Parekh, P. K., & McClung, C. A. (2015). Circadian mechanisms underlying reward-related neurophysiology and synaptic plasticity. Frontiers in Psychiatry, 6, 187. http://doi.org/10.3389/ fpsyt.2015.00187 Perreau-Lenz, S., & Spanagel, R. (2008). The effects of drugs of abuse on clock genes. Drug News & Perspectives, 21(4), 211. http://doi.org/10.1358/dnp.2008.21.4.1213350 Reiter, R. J. (1993). The melatonin rhythm: Both a clock and a calendar. Experientia, 49(8), 654–664. http://doi.org/10.1007/ BF01923947 Roecklein, K. A., Rohan, K. J., Duncan, W. C., Rollag, M. D., Rosenthal, N.  E., Lipsky, R.  H., & Provencio, I. (2009). A missense variant (P10L) of the melanopsin (OPN4) gene in seasonal affective disorder. Journal of Affective Disorders, 114(1–3), 279–285. http://doi.org/10.1016/j.jad.2008.08.005 Roecklein, K., Wong, P., Ernecoff, N., Miller, M., Donofry, S., Kamarck, M., . . . Franzen, P. (2013a). The post illumination pupil response is reduced in seasonal affective disorder. Psychiatry Research, 210(1), 150–158. http://doi.org/10.1016/j. psychres.2013.05.023 Roecklein, K. A., Wong, P. M., Miller, M. A., Donofry, S. D., Kamarck, M.  L., & Brainard, G.  C. (2013b). Melanopsin, photosensitive ganglion cells, and seasonal affective disorder. Neuroscience & Biobehavioral Reviews, 37(3), 229–239. http://doi.org/10.1016/j.neubiorev.2012.12.009

378

Rosenthal, N. E., Sack, D. A., Skwerer, R. G., Jacobsen, F. M., & Wehr, T.  A. (1988). Phototherapy for seasonal affective disorder. Journal of Biological Rhythms, 3(2), 101–120. http:// doi.org/10.1177/074873048800300202 Roybal, K., Theobold, D., Graham, A., DiNieri, J.  A., Russo, S.  J., Krishnan, V., . . . McClung, C.  A. (2007). Mania-like behavior induced by disruption of CLOCK. Proceedings of  the National Academy of Sciences of the United States of America, 104(15), 6406–6411. http://doi.org/10.1073/pnas. 0609625104 Russo, S., & Nestler E. (2013). The brain reward circuitry in mood disorders. Nature Reviews Neuroscience, 14(9), 609–625. Schierenbeck, T., Riemann, D., Berger, M., & Hornyak, M. (2008). Effect of illicit recreational drugs upon sleep: Cocaine, ecstasy and marijuana. Sleep Medicine Reviews, 12(5), 381–389. http://doi.org/10.1016/j.smrv.2007.12.004 Spanagel, R., Pendyala, G., Abarca, C., Zghoul, T., SanchisSegura, C., Magnone, M.  C., . . . Albrecht, U. (2005). The clock gene Per2 influences the glutamatergic system and modulates alcohol consumption. Nature Medicine, 11(1), 35–42. http://doi.org/10.1038/nm1163 Spanagel, R., Rosenwasser, A.  M., Schumann, G., & Sarkar, D. K. (2005). Alcohol consumption and the body’s biological clock. Alcoholism: Clinical & Experimental Research, 29(8), 1550–1557. http://doi.org/10.1097/01.alc.0000175074. 70807.fd Spencer, S., Falcon, E., Kumar, J., Krishnan, V., Mukherjee, S., Birnbaum, S. G., & McClung, C. A. (2013). Circadian genes Period 1 and Period 2 in the nucleus accumbens regulate anxiety-related behavior. European Journal of Neuroscience, 37(2), 242–250. http://doi.org/10.1111/ejn.12010 Srinivasan, V., Smits, M., Spence, W., Lowe, A. D., Kayumov, L., Pandi-Perumal, S. R., . . . Cardinali, D. P. (2006). Melatonin in mood disorders. World Journal of Biological Psychiatry, 7(3), 138–151. http://doi.org/10.1080/15622970600571822 Stenvers, D.  J., van Dorp, R., Foppen, E., Mendoza, J., Opperhuizen, A.-L., Fliers, E., . . . Deboer, T. (2016). Dim light at night disturbs the daily sleep-wake cycle in the rat. Scientific Reports, 6, 35662. http://doi.org/10.1038/srep35662 Takita, E., Yokota, S., Tahara, Y., Hirao, A., Aoki, N., Nakamura, Y.,  .  .  .  Shibata, S. (2013). Biological clock dysfunction exacerbates contact hypersensitivity in mice. British Journal of Dermatology, 168(1), 39–46. http://doi.org/10.1111/ j.1365–2133.2012.11176.x Tapia-Osorio, A., Salgado-Delgado, R., Angeles-Castellanos, M., & Escobar, C. (2013). Disruption of circadian rhythms due to chronic constant light leads to depressive and anxiety-like behaviors in the rat. Behavioural Brain Research, 252, 1–9. http://doi.org/10.1016/j.bbr.2013.05.028 Tataroğlu, Ö., Aksoy, A., Yılmaz, A., & Canbeyli, R. (2004). Effect of lesioning the suprachiasmatic nuclei on behavioral despair in rats. Brain Research, 1001(1), 118–124. http://doi. org/10.1016/j.brainres.2003.11.063 Teclemariam-Mesbah, R., Ter Horst, G. J., Postema, F., Wortel, J., & Buijs, R. M. (1999). Anatomical demonstration of the suprachiasmatic nucleus-pineal pathway. Journal of Com­ parative Neurology, 406(2), 171–182. http://doi.org/10.1002/ (SICI)1096-9861(19990405)406:23.0. CO;2-U Terman, M. (1988). On the question of mechanism in phototherapy for seasonal affective disorder: Considerations of clinical efficacy and epidemiology. Journal of Biological Rhythms, 3(2), 155–172. http://doi.org/10.1177/074873048800300205

Hormones, Circadian Rhy thms, and Mental Health

Thase, M.  E. (1999). How should efficacy be evaluated in randomized clinical trials of treatments for depression? Journal of Clinical Psychiatry, 60(Suppl. 4), 23–31; discussion 32. Thompson, C., Stinson, D., & Smith, A. (1990). Seasonal affective disorder and season-dependent abnormalities of melatonin suppression by light. Lancet, 336(8717), 703–706. http:// doi.org/10.1016/0140-6736(90)92202-S Torres-Farfan, C., Richter, H. G., Rojas-García, P., Vergara, M., Forcelledo, M.  L., Valladares, L.  E., . . . Serón-Ferré, M. (2003). MT1 melatonin receptor in the primate adrenal gland: Inhibition of adrenocorticotropin-stimulated cortisol production by melatonin. Journal of Clinical Endocrinology & Metabolism, 88(1), 450–458. http://doi.org/10.1210/jc.2002– 021048 Torres-Farfan, C., Seron-Ferre, M., Dinet, V., & Korf, H.-W. (2006). Immunocytochemical demonstration of day/night changes of clock gene protein levels in the murine adrenal gland: Differences between melatonin-proficient (C3H) and melatonin-deficient (C57BL) mice. Journal of Pineal Research, 40(1), 64–70. http://doi.org/10.1111/j.1600-079X.2005.00279.x Tuma, J., Strubbe, J. H., Mocaër, E., & Koolhaas, J. M. (2005). Anxiolytic-like action of the antidepressant agomelatine (S 20098) after a social defeat requires the integrity of the SCN. European Neuropsychopharmacology, 15(5), 545–555. http://doi.org/10.1016/j.euroneuro.2005.02.004 Turek, F. W. (2007). From circadian rhythms to clock genes in depression. International Clinical Psychopharmacology, 22 (Suppl. 2), S1–S8. http://doi.org/10.1097/01.yic.0000277956. 93777.6a Uz, T., Arslan, A. D., Kurtuncu, M., Imbesi, M., Akhisaroglu, M., Dwivedi, Y.,  .  .  .  Manev, H. (2005). The regional and  cellular expression profile of the melatonin receptor MT1 in the central dopaminergic system. Molecular Brain Research, 136(1), 45–53. http://doi.org/10.1016/j.molbrainres. 2005.01.002 Vandewalle, G., Maquet, P., & Dijk, D.-J. (2009). Light as a modulator of cognitive brain function. Trends in Cognitive

Sciences, 13(10), 429–438. http://doi.org/10.1016/j.tics.2009. 07.004 Watson, R., Bakos, L., Compton, P., & Gawin, F. (1992). Cocaine use and withdrawal: The effect on sleep and mood. American Journal of Drug and Alcohol Abuse, 18(1), 21–28. http://doi.org/10.3109/00952999209001608 Weaver, D., Rivkees, S., & Reppert, S. (1989). Localization and characterization of melatonin receptors in rodent brain by in  vitro autoradiography. Journal of Neuroscience, 9(7), 2581–2590. Weil, Z. M., Borniger, J. C., Cisse, Y. M., Abi Salloum, B. A., & Nelson, R. J. (2015). Neuroendocrine control of photoperiodic changes in immune function. Frontiers in Neuroendocrinology, 37, 108–118. http://doi.org/10.1016/j.yfrne.2014.10.001 Weil, Z. M., Hotchkiss, A. K., Gatien, M. L., Pieke-Dahl, S., & Nelson, R. J. (2006). Melatonin receptor (MT1) knockout mice display depression-like behaviors and deficits in sensori­ motor gating. Brain Research Bulletin, 68(6), 425–429. http://doi.org/10.1016/j.brainresbull.2005.09.016 Welsh, D.  K., Takahashi, J.  S., & Kay, S.  A. (2010). Supra­ chiasmatic nucleus: Cell autonomy and network properties. Annual Review of Physiology, 72, 551–577. http://doi.org/ 10.1146/annurev-physiol-021909-135919 Wirz-Justice, A., Benedetti, F., Berger, M., Lam, R.  W., Martiny, K., Terman, M., & Wu, J.  C. (2005). Chrono­ therapeutics (light and wake therapy) in affective disorders. Psychological Medicine, 35(7), 939–944. http://doi.org/ 10.1017/S003329170500437X Zisapel, N., Egozi, Y., & Laudon, M. (1982). Inhibition of dopamine release by melatonin: Regional distribution in the rat brain. Brain Research, 246(1), 161–163. http://doi.org/ 10.1016/0006-8993(82)90157-3 Zou, Y., Liao, G., Liu, Y., Wang, Y., Yang, Z., Lin, Y., . . . Wang, Z. (2007). Association of the 54-nucleotide repeat polymorphism of hPer3 with heroin dependence in Han Chinese population. Genes, Brain and Behavior, 7(1), 26–30. http://doi.org/10.1111/ j.1601-183X.2007.00314.x

Cissé, Borniger, and Nelson

379

CH A PT E R

22

Hormones and Major Depressive Disorder

Mark A. Ellenbogen, Virginia Tsekova, and Lisa Serravalle

Abstract Hormonal influences figure prominently in the development of major depressive disorder (MD). The chapter addresses the hypothalamic-pituitary-adrenal (HPA) axis and gonadal (i.e., estrogen and testosterone) hormones that are most relevant to MD. There is substantial evidence of HPA dysfunction in persons with MD, including adrenal hypersensitivity leading to elevated cortisol levels and deficient negative feedback control of the axis. These abnormalities may represent a marker of vulnerability for MD, as they are observed in high-risk populations prior to the development of the disorder. Gonadal hormones are related to specific presentations of MD. Estrogen sensitivity appears to underlie a “reproductive” form of MD in women, as seen during the menstrual cycle, postpartum period, and perimenopausal transition. Low testosterone, as occurs during normal aging, is associated with an increased risk for MD in men. These hormonal changes may be important in defining subtypes of MD that might be treated with targeted interventions. Keywords:  hormones, depression, estrogen, testosterone, major depressive disorder

Approximately 16 percent of adults in North America suffer from major depressive disorder (MD) and an additional 10 percent from minor forms of depression over the course of their lifetime (Kessler et al., 2003). Similar lifetime prevalence rates (14.6 percent) have been reported in other high-income countries around the world, with lower rates (11 percent) in lowto middle-income countries (Bromet et al., 2011). Additionally, recurrence rates of depression are as high as 80 percent, and 60 percent of persons with MD report significant role impairment (Kessler et al., 2003; Kessler & Wang, 2009). The Global Burden of Disease study of the World Health Organization reported that MD in the year 2000 was the most burdensome disease in the Americas, ahead of ischemic heart disease, perinatal conditions, violence, road traffic accidents, and cerebrovascular disease (Üstün, Ayuso-Mateos, Chatterji, Mathers, & Murray, 2004). In Europe, MD was the third most burdensome disease in disability-adjusted life-years, falling just

below cerebrovascular disease but well ahead of ­alcohol use disorders, Alzheimer and other dementias, and road traffic accidents. The burden of MD around the world is based on five factors: high lifetime prevalence, high rate of recurrence, prolonged role impairment, frequent co-occurrence with other chronic diseases, and early age of onset (15 to 30 years; Bromet et al., 2011; Kessler, 2012; Van der Kooy et al., 2007). Depression in youth constitutes a major social-developmental problem because it co-occurs with social maladjustment, academic problems, family dysfunction, and interpersonal difficulties that transfer from one generation to the next (Joiner & Timmons,  2009; Lewinsohn, Rohde, Seeley, Klein, & Gotlib, 2003; Ostiguy, Ellenbogen, & Hodgins, 2012). Thus, MD is a recurrent, debilitating, and, to some extent, chronic disorder, despite the many advances in treatment (Segal, Pearson, & Thase, 2003). These facts highlight the importance of studying the etiology and developmental course of 381

this disorder, and the need for more efficacious treatments and prevention strategies. In this chapter, we review the relationship between MD and different hormonal systems, a key facet of the etiology of MD. Specifically, we show how hormonal systems are dysregulated in MD and propose that subtle changes in hormone functioning, such as rising levels of the glucocorticoid cortisol in youth, may serve as a marker of risk for MD. The latter point is particularly important as it demonstrates the possibility that changes in hormonal systems may represent an etiological risk factor for developing MD, rather than just a correlate or endocrine “symptom” of having MD. This chapter will focus on MD and will not review the literature on other affective disorders such as bipolar disorder or persistent depressive disorder (dysthymia), despite the fact that there are similar endocrine changes associated with these disorders (Watson, Gallagher, Ritchie, Ferrier, & Young, 2004). Although there are many endocrine factors that might be associated with MD, the chapter addresses those most relevant to MD, notably adrenal (i.e., cortisol) and gonadal (i.e., estrogen and testosterone) hormones. For reviews of other hormones, such as those from adipose tissue (leptin) and the gastrointestinal tract (­ghrelin), we refer the reader to Carvalho and colleagues (2014) and Wittekind and Kluge (2015). MD is defined as a syndrome characterized by depressed mood and/or anhedonia (loss of interest or pleasure in previously rewarding activities) that persists for at least two weeks, but typically longer, most of the day and nearly every day. Along with the two cardinal features of depression, other symptoms include weight loss or weight gain, insomnia or ­hypersomnia, psychomotor agitation or retardation, fatigue or loss of energy, diminished concentration or indecisiveness, feelings of worthlessness or excessive or inappropriate guilt, and recurring thoughts of death or suicidal behavior. The diagnosis of MD requires that persons display at least five of the aforementioned nine symptoms, with one of them sad mood or anhedonia, and that they report significant distress or impairment in social, occupational, or other important areas of functioning. Although the MD diagnosis has proven to be useful clinically, there are problems with the use of a categorical diagnostic system such as the Diagnostic and Statistical Manual for Mental Disorders, 5th edition, in studying the etiology of MD (Lupien et al., 2017), including high rates of comorbid presentations (Kessler & Wang,  2009) among patients with MD and the 382

use of an arbitrary number of criteria for defining the diagnosis (five out of nine symptoms rather than four, for example). There is accumulating evidence in prospective studies that “subclinical” depression is associated with significant negative outcomes at later follow-up assessments (Fergusson, Horwood, Ridder, & Beautrais, 2005; Klein et al., 2013), which argues against a categorical approach to MD. Because most of the studies reviewed in this chapter use the MD diagnosis, conclusions put forth in this chapter should be considered with these limitations in mind. MD is perhaps the prototype of a heterogeneous condition arising from a complex etiology. There are many variations of symptom presentation, from the melancholic subtype characterized by severe ­anhedonia, weight loss, and a lack of reactivity to pleasurable events to the atypical subtype characterized by reactivity to pleasurable events, weight gain, and a sensitivity to interpersonal rejection. There may be important psychobiological differences between these presentations (Harkness & Monroe, 2006). Heritability estimates of 40 to 60 percent have been reported in large twin studies of MD, with substantial variance attributed to nonshared unique environmental factors (influences that may be different between twins, such as stressful life events, peer groups, etc.; Kendler & Prescott, 1999; McGuffin et al., 2003). Early family factors, including maltreatment, maternal depression, and a parenting style characterized by unresponsiveness and low warmth, are linked to the development of MD (Goodman & Brand,  2009; Harkness & Lumley, 2008; McLeod, Weisz, & Wood,  2007). Transactional developmental theories that incorporate vulnerability factors (genetic risk, neurobiological factors), early environmental adversity, and proximal risk factors such interpersonal stress and cognitive factors (i.e., rumination, depressogenic attributions, etc.) are perhaps the most important and relevant models of the etiology of MD (Goodman & Brand, 2009; Hammen, 2015). Endocrine factors are central in these models in that they are sensitive to developmental and environmental factors and have an important impact on neural circuits associated with depressive affect and anhedonia (Heim, Newport, Mletzko, Miller, & Nemeroff, 2008; Lupien, McEwen, Gunnar, & Heim, 2009). In the next section, we address the hypothalamicpituitary-adrenal (HPA) axis, because this system appears to be dysfunctional during episodes of MD and may represent a key biological marker of risk for developing an affective disorder.

Hormones and Major Depressive Disorder

the hypophyseal transport system that guides the neuropeptide to the anterior portion of the pituitary gland, situated at the base of the brain. Adreno­ corticotropic hormone (ACTH) is then synthesized and released into circulation, where it stimulates the adrenal cortex, just above the kidneys, to produce and then secrete glucocorticoids, particularly cortisol in humans. Once in circulation, cortisol targets different organs to increase the availability of glucose through gluconeogenesis, promotes lipid and carbohydrate metabolism, and activates anti-inflammatory and other immunosuppressive effects. All these changes increase the availability of energy stores in circulation to promote short-term survival in threatening circumstances. Glucocorticoid actions occur through the stimulation of intracellular mineralocorticoid and glucocorticoid receptors (GRs), distributed widely through the brain with high concentrations in the hippocampus, amygdala, and hypothalamus. Once activated, these receptors in the cytosol of the cell migrate to the nucleus and enact changes in gene expression. In this way, the stimulation of mineralocorticoid and glucocorticoid receptors have short-term (i.e., gluconeogenesis) and long-term effects (i.e., changes

Hypothalamic-Pituitary-Adrenal Axis and Major Depressive Disorder The Hypothalamic-Pituitary-Adrenal Axis

Along with the fast-acting sympathetic nervous system, the HPA axis is a principal actor in orchestrating the mammalian stress response to prolonged challenges. The HPA axis is regulated via a complex interplay of serotonergic, noradrenergic, and suprachiasmatic nucleus circadian input, as well as cortical and limbic brain regions that detect and appraise threat and contextual factors (Chrousos,  1998; De Raedt & Koster, 2010; Holsboer, 1995; van de Werken et al., 2014). The sequence of events underlying the HPA stress response is as follows (see Figure 22.1). Real or perceived stress in response to a difficult challenge, typically those perceived to be beyond one’s ability to cope, activates a number of neural circuits including the prefrontal cortex, hippocampus, amygdala, septum, and hypothalamus. The neural response to an ongoing threat in the amygdala activates the HPA axis through the synthesis of corticotropin-releasing hormone (CRH) and arginine vasopressin in the paraventricular nuclei of the hypothalamus, which leads to the release of CRH into Amygdala + –

Hippocampus –

CRH –

PFC

Hypothalamus + CRH Vasopressin

–

Monoamines GABA

Pituitary –

ACTH BBB Adrenal cortex

Cortisol Figure 22.1  The hypothalamic-pituitary-adrenal (HPA) axis is depicted here, with the different factors that influence its functioning. In the context of a stressful or threatening situation, the amygdala both produces corticotropin-releasing hormone (CRH) in the central nucleus, which will stimulate other stress- and anxiety-related brain circuits, and activates the paraventricular neurons of hypothalamus to produce CRH and arginine vasopressin. These hormones are then released into the hypophyseal portal system, which delivers them to the anterior pituitary, at the base of the brain. Activation of the anterior pituitary causes the production and release of adrenocorticotropic hormone (ACTH), which then crosses the blood–brain barrier (BBB), enters the general circulation, and stimulates cells in the adrenal cortex to produce and release the glucocorticoid hormone cortisol. Because cortisol in the periphery crosses the BBB and enters the brain, levels of the hormone are regulated by negative feedback via activation of glucocorticoid receptors at various brain areas, most notably the hippocampus. The HPA system is also regulated by different neurotransmitter systems including monoamines and GABA, as well as the prefrontal cortex, which is involved in emotion regulation.

Ellenbogen, Tsekova, and Serravalle

383

in gene expression) that can alter how the system responds to subsequent stressors (McEwen,  2004; Sapolsky, Romero, & Munck, 2000). It is for this reason that there is interest in the effects of earlylife experiences on the development of the HPA system and its effects on physical and mental health (Heim et al., 2008; Lupien et al., 2009; McEwen & Gianaros, 2010). Because prolonged exposure to glucocorticoids can be neurotoxic and elicit negative effects such as the suppression of the immune system, there are restraining forces that decrease HPA activation following acute stress. These effects occur through the stimulation of GR at multiple levels of the axis, including the pituitary, hypothalamus, hippocampus, and frontal cortex, all of which inhibit the HPA response to stress. In addition to the HPA glucocorticoid response to stress, there are cognitive, behavioral, and autonomic aspects to the stress response (Nederhof, Marceau, Shirtcliff, Hastings, & Oldehinkel, 2015; van den Bos, Harteveld, & Stoop, 2009). These coordinated responses occur through CRH and its projections to a wide range of brain regions other than the HPA axis. Activation of CRH in the paraventricular hypothalamus stimulates neural projection to the thalamus, amygdala, and other parts of the limbic system, which serves to coordinate the autonomic and behavioral responses to threat. There is accumulating evidence that these circuits are fundamental in understanding the expression of different forms of psychopathology, but particularly MD (Kormos & Gaszner, 2013). In the following sections, we highlight key facets of HPA dysregulation in persons with MD and among those who are at risk for MD.

Studies of the Hypothalamic-PituitaryAdrenal Axis in Persons With Major Depressive Disorder

Converging evidence indicates that the HPA axis is severely compromised in persons with MD, with a cascade of related deficits and adaptations (Chrousos, 1998; Holsboer,  1995; Lopez-Duran, Kovacs, & George,  2009; Nemeroff & Vale,  2005). The primary dysfunction is believed to be increased CRH activity, as has been shown in studies of cerebrospinal fluid of depressed and suicidal patients (Nemeroff, 1996; Nemeroff, Owens, Bissette, Andorn, & Stanley, 1988; Pitts, Samuelson, Meller, & Bissette,  1995; Waters et al., 2015). In response to high CRH levels, the pituitary gland appears to become hyposensitive to CRH (Holsboer, Spengler, & Heuser,  1992). Adding to this dysregulation, the adrenal cortex 384

a­ ppears to be hypersensitive to ACTH, as approximately 50 percent of depressed patients, relative to nondepressed controls, show elevated basal levels of plasma or urinary cortisol (Halbreich, Asnis, Shindledecker, Zumoff, & Nathan, 1985; Nemeroff & Vale, 2005; Young, 2004). There may be important differences in HPA dysfunction across different presentations of MD. Cortisol hypersecretion, for example, is greater in MD with psychotic features than nonpsychotic depression (Duval et al., 2006; Keller et al.,  2006). Consistent with the evidence of cortisol hypersecretion, persons with MD have a larger adrenal gland than nondepressed volunteers, based on structural imaging studies (Kessing, Willer, & Knorr, 2011). In a meta-analysis of 354 studies and 18,374 participants (Stetler & Miller,  2011), cortisol levels were higher in persons with MD than nondepressed controls, and the group difference was estimated to be a medium effect size (d = 0.60). There is conflicting evidence, however, in the literature. A meta-analysis revealed that, when considering all published studies, CRH levels were not consistently elevated in depressed persons relative to nondepressed controls (Stetler & Miller, 2011). Elevated CRH in depression may be limited to certain types of studies. For example, CRH levels were elevated in MD when the study did not exclude persons on antidepressant treatment (Stetler & Miller, 2011). Perhaps CRH overdrive is characteristic of  severe MD where discontinuing antidepressant medication is not possible. CRH was also higher in depressed patients relative to controls for studies using blood rather than cerebrospinal fluid (Stetler & Miller, 2011). It is not clear why these differences are present, but they point to the need for further research on CRH in MD. Inconsistent findings have also been observed in the measurement of cortisol in persons with MD. Although many of the studies of basal cortisol have used inpatient samples and measures of the hormone in plasma, HPA dysregulation may be different in outpatients with MD, who make up the large majority of patients, and in studies measuring cortisol in saliva. To start, group differences in basal cortisol levels are greater in hospitalized depressed patients relative to outpatients with MD, and this does not appear to be due to differences in depression severity (Stetler & Miller, 2011). Studies of the cortisol response following awakening (CAR), assessed in saliva rather than blood, have found ­evidence of both elevated (Bhagwagar, Hafizi, & Cowen, 2005; Pruessner, Hellhammer, Pruessner, & Lupien, 2003; Ulrike, Reinhold, & Dirk, 2013;

Hormones and Major Depressive Disorder

Vreeburg et al.,  2009) and attenuated levels (Ellenbogen & Ostiguy, 2017; Huber, Issa, Schik, & Wolf, 2006; Knight, Avery, Janssen, & Powell, 2010; Stetler & Miller, 2005) of the hormone in depressed samples relative to age-matched controls. Other studies have found no differences in the cortisol response following awakening between depressed and control samples (Hellgren, Åkerud, Skalkidou, & Sundström-Poromaa, 2013; Vammen et al., 2014). As alluded to earlier, there is heterogeneity in depressed populations and some of the d ­ iscrepancies in the literature may be due to clinical differences in  the populations studied. For example, the Netherlands Study of Anxiety and Depression assessed cortisol levels in saliva at awakening and 30, 45, and 60 minutes later in 233 patients with MD and 543 controls (Lamers et al., 2013). The cortisol rise following awakening was larger in MD patients with melancholic features than controls, but there were no differences between the control group and MD patients with atypical features. These data were consistent with other studies that have shown that HPA hyperactivity is associated with melancholic features in depression (Stetler & Miller, 2011; Wong et al., 2000).

Studies of the Cortisol Response to Psychosocial Stress in Adults With Major Depressive Disorder

A number of studies have assessed the cortisol response to stress in persons with MD (Burke, Davis, Otte, & Mohr, 2005; Ciufolini, Dazzan, Kempton, Pariante, & Mondelli, 2014) and MD in remission (Morris, Rao, Wang, & Garber, 2014). The findings have been mixed. Two meta-analyses of studies assessing the cortisol response to psychosocial stress found no evidence of increased stress reactivity across nine studies (Burke et al., 2005; Ciufolini et al., 2014), but one of the meta-analyses found evidence of increased cortisol levels during the recovery phase following the stress exposure (Burke et al., 2005). The effect size for group differences during the recovery phase (d = 1.39) was approximately five times higher than the effect size for the acute stress phase (d = 0.27), although it should be noted that the larger effect size was based on only four studies and was not replicated by Ciufolini et al (2014). The putative deficit in poststress recovery observed in persons with MD may reflect a deficit in the negative feedback control of the axis (see later). Moreover, this finding parallels a recent study showing poor habituation to repeated stressors in MD (Morris & Rao, 2014). In this study,

adolescents and young adults with MD underwent the Trier Social Stress Test (TSST; Kirschbaum, Pirke, & Hellhammer, 1993), a well-known psychosocial stress induction, on two occasions approximately six months apart. Although controls showed the expected attenuated cortisol response on the second TSST relative to the first (i.e., habituation), persons with MD showed no such effect. Thus, both poor habituation to stress and weakened negative feedback control of the axis may underlie more general self-regulation problems in depression, including deficits in emotion regulation and cognitive control, which influence HPA functioning (Ellenbogen, Schwartzman, Stewart, & Walker, 2006; Joormann & Tanovic, 2015; LeMoult & Joormann, 2014). A recent and larger meta-analysis found a small attenuation of the cortisol response to psychosocial stress in the MD group compared to controls (Zorn et al., 2017). More important, a robust sex difference among persons with MD was found: Depressed women exhibited attenuated cortisol reactivity to psychosocial stress, whereas depressed men displayed increased reactivity (Zorn et al., 2017). Interestingly, there were few robust differences between remitted and currently depressed patients, consistent with the view that some HPA dysfunctions may persist beyond the acute depressive state. In sum, there are mixed findings in studies examining the HPA response to stress in persons with MD. Tentatively, HPA dysregulation of the stress response consists of a weakened ability to mount a strong cortisol response to stress, particularly in women with MD, and possible deficits in regulating and normalizing cortisol levels poststress. In the next section, we focus on laboratory neuroendocrine challenge studies aimed at highlighting HPA dysfunctions and deficits in the negative feedback control of the axis.

Neuroendocrine Challenge Studies in Adults With Major Depressive Disorder

Along with CRH hyperactivity, the GR-mediated shutdown of the axis has been identified as a key deficit associated with MD (Holsboer & Ising, 2010). A number of neuroendocrine challenges have been developed to identify deficient negative feedback control of the HPA axis. Dexamethasone (DEX), a potent and long-lasting synthetic glucocorticoid, stimulates glucocorticoid and mineralocorticoid ­receptors and elicits a negative feedback suppression of the HPA axis, resulting in robust decreases in levels of ACTH and cortisol for approximately 24 hours. In early studies, these hormones remained

Ellenbogen, Tsekova, and Serravalle

385

elevated in patients with MD relative to controls following DEX administration, which was termed DEX nonsuppression and considered to be a clinical biomarker of MD, particularly the melancholic subtype (Bowie, Beaini, & Bowie, 1987; Holsboer, 1995; Rush et al., 1996). Unfortunately, it was soon realized that the reliability of the DEX test to identify MD was rather poor; identifying only 20 to 50 percent of patients, and it was abandoned as a possible diagnostic marker of the disorder (Nierenberg & Feinstein,  1988). However, DEX nonsuppression may be more robust in patients with psychotic MD than those with no psychotic features (Nelson & Davis,  1997), and it may have predictive value in identifying a poor prognosis following treatment (Ribeiro, Tandon, Grunhaus, & Greden, 1993). The DEX test was adapted into the DEX/CRH test (Holsboer, von Bardeleben, Wiedemann, Muller, & Stalla, 1987), which proved to be a more sensitive and reliable means of identifying HPA abnormalities and deficient negative feedback control of the HPA axis. Patients are given DEX in the evening, which reduces levels of ACTH and cortisol over the next 24 hours. On the next day, patients are challenged with intravenous CRH, activating the HPA axis in the context of DEX suppression. Following this procedure, cortisol levels remain high in 60 to 80 percent of untreated patients with MD (Heuser, Yassouridis, & Holsboer, 1994; Kunugi et al., 2006; Zobel et al., 2001), which has been confirmed in a recent meta-analysis (Mokhtari, Arfken, & Boutros, 2013). The DEX/CRH test distinguishes remitted patients from controls (Schmider et al.,  1995), predicts treatment response (Kunugi et al.,  2006; Paslakis, Heuser, Schweiger, & Deuschle,  2010), and prospectively predicts relapse (Zobel et al., 2001; Zobel, Yassouridis, Frieboes, & Holsboer, 1999). Thus, it holds promise as a putative biological marker of MD and treatment outcome. Cortisol and ACTH nonsuppression following the test, however, are not specific to MD, as it has been observed in other psychiatric disorders (Lammers et al., 1995) and during stressful circumstances (Wirtz et al.,  2010). Also, the mechanism underlying cortisol nonsuppression in patients in MD is not known. Although nonsuppression may be due to deficient GR-mediated negative feedback control of the HPA axis, it may also be related to the amplification of CRH release at the level of the pituitary by arginine vasopressin or changes in the sensitivity of the adrenal cortex to ACTH (Modell, Yassouridis, Huber, & Holsboer, 1997; Spijker & van Rossum, 2012). 386

Hypothalamic-Pituitary-Adrenal Functioning Prior to the Development of Major Depressive Disorder

Despite the evidence that the HPA axis is severely compromised in MD, it has been uncertain whether HPA abnormalities precede the onset of the disorder. The distinction is important because HPA abnormalities may represent part of the symptom expression of the disorder but have nothing to do with its etiology. Two strategies have been used to address the question of whether premorbid HPA abnormalities precede the development of MD. First, birth cohort and other large community-based longitudinal studies track large samples of children over time and determine predictors of negative health outcomes. Unfortunately, only a few such studies have assessed HPA function. The Tracking Adolescents’ Individual Lives Survey (TRAILS) followed a large cohort of children in the Netherlands from early adolescence to adulthood and assessed daytime cortisol levels in the natural environment. The findings from the TRAILS study are mixed. Although cross-sectional positive associations were found between measures of cortisol in the morning, including the CAR, and depressive symptoms at age 11 years (Dietrich et al., 2013), the CAR at 13 years of age did not predict MD at age 16 (Nederhof, van Oort, et al., 2015). Another 18-month longitudinal study of young adolescents found a ­positive relationship between cortisol reactivity to stressors in the environment and the development of depressive symptoms one year later (Susman, Dorn, Inoff-Germain, Nottelmann, & Chrousos, 1997), which is c­ onsistent with the TRAILS’s findings of the cohort in early adolescence. Perhaps the relationship between high cortisol levels and depression is specific to early adolescence, but there are too few studies to put forth strong conclusions. The second approach to studying premorbid HPA abnormalities prior to the development of MD is to identify select samples of youth who do not have MD but are at high risk for the development of the disorder. Different strategies have been used to identify target populations, including the recruitment of youth (1) exposed to multiple adversities (e.g., poverty, family conflict; Goodyer, Croudace, Dudbridge, Ban, & Herbert, 2010), (2) who have high trait neuroticism (Adam et al., 2014), or (3) who have a parent with a mental disorder. Studying children whose parents have an affective disorder, either MD or bipolar disorder, is a strategy commonly used because these children are at high risk for developing an ­affective disorder and other mental disorders, with

Hormones and Major Depressive Disorder

particularly high rates of MD (Nijjar, Ellenbogen, & Hodgins, 2014; Rasic, Hajek, Alda, & Uher, 2014). Offspring of parents with an affective disorder are approximately two and a half times more likely to be diagnosed with any mental disorder and three to four times more likely to be diagnosed with an affective disorder, particularly before the age of 20 years, than offspring of parents with no mental disorder (Lapalme, Hodgins, & LaRoche,  1997; Rasic et al., 2014).

Cortisol Levels in High-Risk Youth

With some exceptions (Ising, Lauer, Holsboer, & Modell, 2005; Ronsaville et al., 2006), there is evidence that subtle HPA abnormalities exist in populations at high risk for MD, and that these abnormalities may represent a marker of vulnerability for the disorder (Ellenbogen, Hodgins, Walker, Adam, & Couture, 2006; Goodyer, Herbert, Tamplin, & Altham, 2000; Lundy et al., 1999; Mannie, Harmer, & Cowen, 2007; Modell et al., 1998). Elevated levels of daytime cortisol measured in the natural environment may represent one such marker of vulnerability. The offspring of parents with bipolar disorder, for example, had higher daytime cortisol levels assessed in adolescence (Ellenbogen, Hodgins, et al., 2006), at 18 years of age (Ellenbogen, Santo, Linnen, Walker, & Hodgins, 2010), and at 20 years of age (Ostiguy, Ellenbogen, Walker, Walker, & Hodgins, 2011) compared to an age-matched control group having parents with no mental disorders. Elevated cortisol levels show trait-like stability and appear to persist over time, with evidence of stable high cortisol levels in high-risk youth sampled across 14 consecutive days (Ellenbogen et al., 2010; Owens et al., 2014). The higher cortisol levels among the offspring of a parent with bipolar disorder were not related to the small number of participants diagnosed with a mental disorder, nor were they associated with self- or parent reports of clinical symptoms, age, self-reported compliance with the saliva sampling protocol, time of awakening, smoking, food consumption, exercise, or oral contraceptive use. Similar findings have been reported in infant, adolescent, and young adult offspring of parents with MD (Halligan, Herbert, Goodyer, & Murray,  2004; Lundy et al.,  1999; Mannie et al., 2007).

Exogenous Challenge Studies in High-Risk Youth

Other aspects of HPA functioning have also been studied in persons at high risk of developing MD.

Nondepressed young adults having a parent with MD secreted higher levels of cortisol in response to the DEX/CRH challenge than healthy controls, but less than those who had MD (Holsboer, Lauer, Schreiber, & Krieg, 1995). As described previously, high cortisol in response to the DEX/CRH test indicates, in part, deficient negative feedback control of the HPA axis. Interestingly, the increased response to the DEX/CRH test persisted in a four-year ­follow-up of a subsample of at-risk participants who underwent the challenge a second time (Modell et al., 1998), which suggests that the response to the DEX/CRH challenge might be a stable vulnerability marker for MD. In contrast, a study of the response to CRH alone in offspring of parents with MD or bipolar disorder failed to uncover any evidence of increased sensitivity to CRH (Ronsaville et al., 2006). Thus, the increased sensitivity in young adults having a parent with MD to the DEX/CRH challenge (Holsboer et al., 1995) is likely due to the nonsuppression of cortisol release following pretreatment with DEX, rather than other CRH-related effects.

Studies of the Hypothalamic-PituitaryAdrenal Response to Stress in High-Risk Youth

For studies of stress reactivity, the findings are mixed. A number of studies indicate that infants and children exposed to postnatal maternal depression show increased cortisol reactivity to various normative challenges (i.e., immunization; Brennan et al., 2008; Dougherty, Klein, Rose, & Laptook, 2011; Dougherty, Tolep, Smith, & Rose, 2013; Essex, Klein, Eunsuk, & Kalin, 2002; Lundy et al., 1999; C. S. Waters et al., 2013). In a study of 7- and 8-year-olds, children with both internalizing problems and a depressed mother exhibited an increased cortisol response to a mild laboratory stressor and elevated baseline cortisol levels measured in the laboratory relative to the offspring of nondepressed parents (Ashman et al., 2002). Young adult offspring of parents with MD demonstrated an increased cortisol response to the TSST relative to a control group having parents with no mental disorder (Barry et al., 2015). The group differences in cortisol reactivity were independent of group differences in the offspring’s current levels of depression, anxiety, or stressful life events. However, other studies have not replicated these findings. In a study of 257 infants, no evidence of increased cortisol reactivity to stress was observed, but offspring of mothers with MD showed poor acclimatization to a novel environment (participating in a laboratory

Ellenbogen, Tsekova, and Serravalle

387

e­ xperiment) compared to offspring of parents with no mental disorder (Waters et al., 2013). Other studies of high-risk adolescents and young adults, in contrast, have found no evidence of a heightened response to stress (Bouma, Riese, Ormel, Verhulst, & Oldehinkel, 2011; Ellenbogen, Hodgins, et al., 2006). Although the adolescent offspring of parents with bipolar disorder did not show increased stress reactivity to the laboratory-based TSST relative to youth having parents with no mental disorders (Ellenbogen, Hodgins, et al.,  2006), they did show increased sensitivity to stressors occurring in the natural environment (Ostiguy et al., 2011). Offspring of parents with ­bipolar disorder who experienced high chronic stress displayed a larger CAR than the offspring of parents with bipolar disorder reporting low chronic stress. In addition, the offspring of parents with bipolar disorder who reported experiencing severe interpersonal episodic stress exhibited higher levels of daytime cortisol than the offspring of parents with bipolar disorder reporting low levels of interpersonal episodic stress. These relationships were substantially more robust in the offspring of parents with bipolar disorder than control offspring, even after controlling for psychopathology in the offspring. In sum, there are likely age-related and task-related factors that obfuscate the study of stress reactivity in highrisk youth (Hankin, Badanes, Abela, & Watamura, 2010). Despite the mixed findings, there is some evidence of altered HPA reactivity to stress in the offspring of parents with an affective disorder, and it may be important to study how high-risk youth respond to naturalistic stress in their environment, rather than artificial laboratory stressors.

Prospective Studies Linking Cortisol Levels to the Development of Depression Symptoms

Although the identification of group differences in HPA functioning between high-risk and control youth is consistent with the view that persons at risk for affective disorders exhibit subtle premorbid HPA abnormalities, it does not demonstrate that these markers are associated with the development of MD. For example, HPA abnormalities may simply highlight the fact that environmental adversities commonly aggregate in families having parents with MD (Goodman & Brand,  2009), and environmental ­adversity is known to alter the HPA system, either as sensitization (Heim et al., 2008) or blunted reactivity (Lovallo, Farag, Sorocco, Cohoon, & Vincent, 2012). Fortunately, a number of studies, but not all  (Carnegie et al.,  2014; Keenan et al.,  2013), have shown that elevated cortisol levels predict the 388

prospective development of MD (Adam et al., 2010; Colich, Kircanski, Foland-Ross, & Gotlib, 2015; Ellenbogen, Hodgins, Linnen, & Ostiguy,  2011; Goodyer et al., 2000, 2010; Harris et al., 2000) and depressive symptoms (Halligan, Herbert, Goodyer, & Murray, 2007; Susman et al., 1997). Odds ratios from these studies indicate that having elevated cortisol levels in the natural environment increase rates of depression by a factor of 1.6 to 7.1, although one study demonstrated that this relationship declines substantially at 2.5 years or more from the assessment of cortisol levels (Vrshek-Schallhorn et al., 2013). Given that the study by Carnegie and colleagues (2014) was the largest study to date (N = 668) and found no evidence of a prospective relationship between measures of cortisol and the development of depression at age 15 years, further research in this area is warranted, with an emphasis on moderators of the relationship. Pubertal status may represent one such moderator, as interactions between gonadal hormones and the HPA system may fundamentally change relations between cortisol levels and depressive symptoms (Colich et al., 2015; Hankin et al., 2010). It is also possible that the positive relationship between cortisol levels in youth and the later development of MD is limited to studies of high-risk populations and/or persons with a vulnerability to develop the disorder, which would be consistent with studies of the offspring of parents with an ­affective disorder (Ellenbogen et al., 2011; Halligan et al.,  2007) and youth with high neuroticism (Adam et al., 2010) or multiple risk factors (Goodyer et al., 2000). The sample studied by Carnegie et al (2014), in contrast, was a birth cohort. Studies of at-risk adolescents that used genotyping support this hypothesis (Goodyer, Bacon, Ban, Croudace, & Herbert, 2009; Goodyer et al., 2010). High cortisol levels in the morning predicted the development of MD one year later, but the strongest effects were found in adolescents having either the short allele of the ­serotonin transporter–linked promoter region polymorphism (5-HTTLPR) or the Val66Met variant of the brain-derived neurotropic factor gene polymorphism. The short allele 5-HTTLPR genotype is particularly relevant, as it is widely viewed as a marker of enhanced sensitivity to stress (Gibb, Beevers, & McGeary,  2013; Gotlib, Joormann, Minor, & Hallmayer,  2008). Adolescents having both the risky 5-HTTLPR genotype and elevated cortisol levels in the morning over four days of sampling were 7.6 times more likely to develop MD in the following year relative to adolescents with neither risk factor (Goodyer et al., 2010). By comparison,

Hormones and Major Depressive Disorder

adolescents having high cortisol levels, irrespective of their genotype, were 4.6 times more likely to develop MD than adolescents without elevated cortisol levels. In addition to genetic risk, other types of vulnerability, such as subclinical depressive symptoms (Fergusson et al., 2005), may be equally important in understanding the relation between cortisol levels and risk for MD (Ashman, Dawson, Panagiotides, Yamada, & Wilkinson,  2002). In a large study of adolescents, the combination of depressive symptoms and high morning cortisol levels in boys was associated with the development of MD over the next year, but not high cortisol levels alone (Owens et al., 2014). In adolescent boys, but not girls, both risk factors increased the risk for MD by a factor of over 14. In sum, elevated cortisol levels in adolescence may represent a biomarker of risk for the affective disorders, particularly in vulnerable populations.

Gonadal Hormones and Major Depressive Disorder

The gonadal hormones estrogen and testosterone have been extensively studied in persons with MD and may represent another key neuroendocrine system associated with risk for MD in vulnerable populations.

Estrogen

Estrogen is involved in the regulation of the female reproductive system and the development of secondary female sex characteristics. The three main forms of estrogen are estrone, estradiol, and estriol, with estradiol being the principal gonadal sex hormone. Estrogen synthesis in women is regulated primarily by the hypothalamic-pituitary-gonadal (HPG) axis. Gonadotropin-releasing hormone secreted by the hypothalamus stimulates the pituitary gland to release follicle-stimulating hormone. In turn, follicle-stimulating hormone stimulates the ovaries to produce estrogen. The liver, adrenal glands, breasts, and fat cells also secrete small amounts of estrogen (Nelson & Bulun, 2001). In addition to its role in the reproductive cycle, estrogen may play a role in the development of MD in women (Newhouse & Albert, 2015).

Depressive Symptoms in Women and Periods of Estrogen Fluctuation

Although rates of depression in boys and girls are comparable before puberty, girls become twice as likely to develop depressive symptoms or MD compared to their male counterparts during and after

puberty (Hankin & Abramson,  1999). Women remain at a higher risk for MD relative to men until menopause (Payne, 2003; Weissman et al., 1993). Estrogen fluctuations associated with different periods in the lifespan of women may play a role in the emergence and maintenance of the sex differences in the prevalence of MD. In addition to monthly variations of estrogen levels related to the menstrual cycle, the main periods associated with estrogen fluctuation in women are pregnancy, postpartum, and perimenopausal periods.

Late Luteal Phase of the Menstrual Cycle

The greatest monthly hormonal fluctuations occur at the end of the luteal phase of the menstrual cycle, during which levels of plasma estrogen fall sharply, and rise up subsequently during menstruation. During this period some women experience premenstrual dysphoric disorder (PMDD), a condition characterized by depressed mood, anxiety, irritability, changes in appetite and sleep, and different physical symptoms (e.g., breast pain, headaches; Payne, 2003). The prevalence of PMDD in the general population is 1.3 to 9 percent (Gehlert, Song, Chang, & Hartlage, 2009; Halbreich, Borenstein, Pearlstein, & Kahn, 2003), and there is a high rate of comorbidity (30 percent) with other affective disorders (Soares & Zitek, 2008). Known risk factors associated with PMDD include a history of MD and posttraumatic stress disorder, smoking, and fewer years of schooling (Soares & Zitek, 2008). Regarding the role of estrogen in the development of late luteal depressive symptoms, there is no consistent evidence that levels of estrogen differ between normal participants and those with PMDD (Rubinow, Schmidt, & Craft, 2013). Although estrogen levels are not directly associated with PMDD, fluctuations in estrogen levels associated with the menstrual cycle may trigger PMDD in women who are at risk for the development of an affective disorder by virtue of having had past episodes of MD or vulnerabilities in neurobiological systems associated with MD such as the serotonergic system or hypothyroidism (Rubinow et al., 2013). In a classic study manipulating ovarian hormones (Schmidt, Nieman, Danaceau, Adams, & Rubinow, 1998), women with PMDD treated with a GnRH agonist, which suppresses ovarian function, exhibited a decrease in PMDD symptoms relative to women on placebo. Symptoms reoccurred when the women with PMDD, but not controls, were given add-back estrogen or  progesterone. These findings have since been ­replicated in subsequent studies and confirmed via

Ellenbogen, Tsekova, and Serravalle

389

meta-analysis (Wyatt, Dimmock, Ismail, Jones, & O’Brien, 2004). There is evidence that it is the change in ovarian hormones that precipitates PMDD symptoms, and not simply the prolonged exposure to elevated ovarian hormone levels that occur following the add-back estradiol or progesterone treatment. Schmidt et al. (2017) first suppressed ovarian function before adding back combined estradiol and progesterone in women with PMDD for three months and found that PMDD symptoms increased in response to the add-back ovarian hormone treatment during the first menstrual cycle, but not during the second or third cycles when plasma levels of estradiol and progesterone were elevated but stable. Thus, changes in ovarian hormones such as estrogen may represent an important trigger of PMDD symptoms and stabilizing ovarian hormone levels may reduce PMDD symptoms, which suggests that women with PMDD may be more sensitive to normal fluctuations in hormone levels. This hypothesis was further tested by studying the effectiveness of combined oral contraceptives, containing estradiol and other hormones (e.g., drospirenone or levonorgestrel), as a means of reducing PMDD symptoms. Oral contraceptives dampen fluctuations in ovarian hormones. Unfortu­nately, the results of these studies have been inconsistent. Typical use of oral contraceptives, with 21 days of active hormones and 7 days hormone-free, has been ineffective in treating women with PMDD relative to placebo (Freeman et al., 2001; Graham & Sherwin, 1992). However, oral contraceptives with short hormonefree periods elicited greater symptom reduction in women with PMDD than placebo in some studies (Freeman et al., 2012; Pearlstein, Bachmann, Zacur, & Yonkers, 2005; Yonkers et al., 2005), but not in others (Eisenlohr-Moul, Girdler, Johnson, Schmidt, & Rubinow, 2017). Thus, the use of oral contraceptives to stabilize fluctuations in ovarian hormones has produced mixed findings with respect to reducing PMDD symptoms.

Pregnancy and Postpartum

Pregnancy and the postpartum period are also characterized by robust changes in ovarian and other hormones. Levels of estrogen gradually rise during pregnancy, reaching levels that are 200 to 300 times higher than nonpregnant women at week 20 of gestation (Brummelte & Galea, 2016). Estrogen levels, as well as progesterone, remain high until parturition, and then they precipitously decline to levels lower than any time during pregnancy. Antenatal depression, with a prevalence of 12 percent, is similar to 390

the 12-month prevalence rates of MD in the adult female population of childbearing age, which is in the range of 13 to 16 percent (Bennett, Einarson, Taddio, Koren, & Einarson,  2004; Brummelte & Galea, 2016; Kessler, McGonagle, Swartz, Blazer, & Nelson, 1993). The postpartum period, however, is well known for its perceived heightened risk for the development of MD, although there is some ­controversy over the veracity of this claim in part because of the wide range of prevalence estimates of postpartum depression (O’Hara & McCabe, 2013). Rates of postpartum depression vary from 10 to 30 percent depending on the criteria used for diagnosis in the general population (Brummelte & Galea, 2016; Vesga-Lopez et al., 2008). Of importance, rates of antenatal and postpartum depression are elevated in women who have a history of MD. In a prospective study, 43 percent of women with a history of MD experienced a relapse during pregnancy (Cohen, Altshuler, et al., 2006). Having a history of MD prior to pregnancy and having depressive or  anxious symptoms during pregnancy are the strongest predictors of the development of a postpartum depressive episode (O’Hara & McCabe, 2013; Robertson, Grace, Wallington, & Stewart, 2004). Other risk factors associated with postpartum depression are neuroticism, low self-esteem, stressful life event, marital problems, poor social support, low socioeconomic status, being single, unwanted pregnancy, obstetrical problems, and difficult infant temperament (Gelaye, Rondon, Araya, & Williams, 2016; O’Hara & McCabe, 2013; Robertson et al., 2004), highlighting the important role psychosocial factors play in the etiology of this disorder. There is little empirical evidence that women with postpartum depression vary in their levels of estrogen or progesterone from euthymic women in the postpartum period (Bloch, Daly, & Rubinow,  2003; Mehta et al., 2014; Workman, Barha, & Galea, 2012). As described for PMDD, it has been hypothesized that women at risk for postpartum depression are sensitive to changes in ovarian hormone levels, and women in pregnancy are exposed to a sudden decrease in levels of estrogen and progesterone following childbirth. Bloch and colleagues (2000) found that women with a history of postpartum depression were more likely to experience depressive symptoms following an induced hypogonadal state, elicited by the administration of a GnRH agonist, compared to never-depressed women. Furthermore, the women with a history of postpartum depression were more likely to experience depressive symptoms following add-back estradiol or progesterone treatment,

Hormones and Major Depressive Disorder

suggesting that it is the change in ovarian steroid hormones that is critical for the development of depressive symptoms in vulnerable women, not the withdrawal of these hormones. In another study (Frokjaer et al., 2015), the prolonged administration of a GnRH agonist was used as a model of the estrogen fluctuations occurring from pregnancy to postpartum because this procedure elicits an initial increase followed by a robust decrease in estradiol, but not progesterone, due to the desensitization of GnRH receptors. Relative to placebo, the prolonged administration of a GnRH agonist increased clinician-rated depressive symptoms, and this increase in symptoms was significantly and positively associated with the magnitude of change in estradiol and change in the level of the serotonin transporter in neocortex, as measured by positron emission tomography with a radiotracer. These findings demonstrate that an experimentally induced fluctuation in ovarian steroid hormones can trigger subclinical depressive symptoms in healthy volunteers. Moreover, the study raises the possibility that the serotonergic system is implicated in the relation between estrogen and depression, as high ­serotonin transporter indicates lower serotonergic functioning, a known risk factor for MD (Benkelfat, Ellenbogen, Dean, Palmour, & Young,  1994). One issue that should be raised with studies of GnRH agonists is that it also acts on GnRH receptors in brain areas outside the HPG axis and on synaptogenesis in the hippocampus, both of which may be related to depression (PrangeKiel et al., 2008; Skinner et al., 2009). Finally, a genome-wide association study compared gene ­expression during the first and third trimesters of pregnancy in women who were euthymic during pregnancy but developed postpartum depression versus women who were euthymic throughout the study, all of whom had a history of MD or bipolar disorder (Mehta et al.,  2014). One hundred and ­sixteen gene transcripts were differentially expressed between the women with postpartum depression and the euthymic women in the third trimester, and this gene expression profile was able to classify women in their respective categories with 88 percent accuracy. The observed gene expression included an overrepresentation of transcripts linked to estrogen signaling among the 116 transcripts predicting postpartum depression. Gene expression changes for the transcripts associated with estrogen signaling from the first to the third trimester and from the third trimester to the postpartum period were larger in the group with postpartum depression than the euthymic women, despite the fact that the

levels of estradiol did not differ between groups. Thus, women at risk for MD who develop postpartum depression show evidence of a heightened sensitivity to estrogen signaling relative to high-risk women who are euthymic during the postpartum period, both when comparing group differences and when conducting within-subject changes in gene expression across the pregnancy and postpartum periods.

Menopausal Transition

During the menopausal transition, there are periods of low and high estrogen levels caused by great variability in follicle-stimulating hormone concentrations. Rates of MD and depressive symptoms increase substantially during this period (Bromberger et al., 2007,  2011; Freeman et al.,  2004), by a factor of 1.3 to 1.8 (Gordon et al., 2015). However, one study reported that the menopausal transition, estimated via vasomotor symptoms (i.e., hot flashes and night sweats), was no longer a significant predictor of first lifetime episodes of MD in a sample of premenopausal and perimenopausal women when accounting for other factors, such as whether the women reported a highly stressful life event in the past year (Bromberger et al., 2009). Thus, despite strong evidence linking the menopausal transition to increased risk for MD, there is still debate on the nature of this relationship. As described previously, the fluctuations of ovarian steroid hormones, rather than the stable low levels that occur during the postmenopausal period, play a role in the development of MD in women during the menopausal transition. There is  little ­ evidence of stable differences in estradiol between perimenopausal women with and without MD (Barrett-Connor, von Muhlen, Laughlin, & Kripke, 1999). In contrast, longer exposure to fluctuating levels of estradiol caused by a prolonged menopausal transition may be a risk factor for perimenopausal depression (Avis, Brambilla, McKinlay, & Vass, 1994; Freeman, Sammel, Boorman, & Zhang,  2014; Schmidt, Murphy, Haq, Danaceau, & St. Clair, 2002). Because fluctuations in estradiol correlate with vasomotor and other menopausal symptoms (Freeman et al., 2007), the reported association b­ etween these symptoms and depression found in some (Cohen, Soares, Vitonis, Otto, & Harlow, 2006; Freeman et al., 2004) but not all studies (Bromberger et al., 2009) provides further evidence of a link between ovarian hormone fluctuations and risk for depression. In terms of direct evidence linking depression and fluctuating ovarian hormones levels, the findings have been mixed (Bromberger et al., 2011; Freeman, 2010; Freeman,

Ellenbogen, Tsekova, and Serravalle

391

Sammel, Lin, & Nelson, 2006; Woods et al., 2008). Inconsistent findings might be because many studies have not collected enough samples, or have collected them at infrequent intervals, to precisely quantify change in ovarian hormone levels (Gordon et al., 2015). In the one study that supports the link between depression and fluctuating estradiol levels (Freeman et al., 2006), sampling occurred at 10 time points across eight years, with two samples one month apart at each time point. More comprehensive sampling protocols using advanced statistical modeling are needed to effectively measure withinsubject changes in ovarian hormones.

Possible Mechanisms Underlying Link Between Estrogen and Depression

Across the three reproductive periods described previously, there is no consistent evidence for abnormalities in ovarian steroid hormone levels in women suffering from MD compared to controls (Payne, 2003). Oftentimes, women who experience perimenopausal depression have also experienced PMDD and postpartum depression, suggesting that there is a subgroup of women who are highly susceptible to intense changes in estrogen levels (Gordon et al., 2015). Some researchers have postulated that these episodes of depression constitute a specific “reproductive” subtype of MD and may require treatment that is different from those for nonreproductive depression (Payne, Palmer, & Joffe,  2009; Soares & Zitek,  2008). It has been proposed that reproductive MD may reflect an inability to successfully adapt to changing levels of ovarian steroid hormones (Payne et al., 2009). The precise biological substrate of the putative sensitivity to fluctuating ovarian steroid hormones is still largely unknown, although there is evidence that it may be related to the estrogen receptor and/or estrogen signaling pathways in areas known to be associated with mood and depression, such as the prefrontal cortex, amygdala, and hippocampus (Borrow & Cameron, 2014; Mehta et al., 2014). There is also evidence that reproductive subtypes of MD may be closely related to sensitivity in the serotonergic system and its interactions with sex steroid hormones (Borrow & Cameron, 2014; Payne et al., 2009). Estrogen can increase the activity of serotonergic (5-HT) neurons and alter the expression of specific serotonin receptors involved in depression (Moses et al., 2000; Robichaud & Debonnel, 2005). For example, Moses and colleagues (2000) found an increase in 5-HT2A receptor density following estrogen administration compared to baseline. Inactivation of the estrogen receptor β 392

system in mice through gene knockout similarly increased the expression of 5-HT2A receptors, but not other serotonergic receptors. Because 5-HT2A receptors are increased in persons with MD (Shelton, Sanders-Bush, Manier, & Lewis, 2009), these studies provide evidence that dramatic changes in estrogen levels or estrogen receptor function can stimulate changes in the serotonergic system associated with MD. In sum, vulnerable women, particularly those with a history of MD, may have a heightened sensitivity to fluctuations in estrogen levels, which may lead to symptoms of depression either directly through estrogen effects in brain areas associated with MD or in interaction with the 5-HT system or other systems (e.g., HPA) implicated in MD. Although progress has been made in this area, research is needed to clarify the exact mechanisms that make certain women more susceptible than others to changes in estrogen levels, and to better delineate the pathway from estrogen to the expression of symptoms of MD.

Testosterone

Testosterone is an androgen secreted by the gonads in both males and females and is responsible for the development of male reproductive tissues such as the testes and prostrate, as well as male secondary sex characteristics. In men, testosterone synthesis is mostly regulated by the HPG axis. GnRH secreted by the hypothalamus stimulates the pituitary gland, which then releases follicle-stimulating hormone and luteinizing hormone to stimulate Leydig cells in the testes to produce testosterone. In women, about half of the testosterone is secreted by the adrenal glands and ovaries, with the other half produced from circulating precursor molecules (Burger, 2002). One important difference between testosterone and estrogen levels in humans is that testosterone declines gradually with age (Davidson et al., 1983), in contrast to the precipitous decline of estrogen in the postpartum period and following menopause (Soares & Zitek, 2008). However, testosterone changes are substantive during the process of aging. Between the ages of 40 and 70 years, testosterone levels can decrease by as much as 40 percent (Seidman, 2007). In addition to its role in the development of male reproductive functioning and other physiological processes, testosterone has also been implicated in MD.

Testosterone Levels and Depression in Men

A number of studies conducted in male participants indicate an increased risk for depression in men with low levels of testosterone (Almeida, Waterreus,

Hormones and Major Depressive Disorder

Spry, Flicker, & Martins, 2004; Christiansen, 2001; Giltay et al., 2017; McIntyre et al., 2006), although there are inconsistencies in this literature (Ebinger, Sievers, Ivan, Schneider, & Stalla,  2009; Johnson, Nachtigall, & Stern, 2013). For example, a prospective study of over 3,000 older men showed that those with lower baseline levels of testosterone were twice as likely to have developed MD nine years later relative to men with normal androgen levels, even after controlling for medical comorbidities, lifestyle factors, and age (Ford et al., 2016). Similarly, in a prospective study of 469 participants over 65 years of age, men with MD had lower plasma testosterone levels than nondepressed men, and low testosterone predicted a worse course of the disorder over the two-year follow-up (Giltay et al., 2017). Other studies of men with atypically low levels of testosterone have found similar results. Men with hypogonadism, a condition characterized by low levels of testosterone, exhibit a higher prevalence of major depressive disorder compared to men with normal androgen levels (McHenry, Carrier, Hull, & Kabbaj,  2014). Over half of men referred to a hospital clinic for borderline total testosterone levels also had MD or depressive symptoms based on a recorded diagnosis during a chart review, self-report questionnaire, or documented use of antidepressant medication (Westley, Amdur, & Irwig, 2015). Men with hypogonadism were approximately four times more likely to develop a depressive illness (MD, dysthymia, or depressive disorder not otherwise specified) during a two-year follow-up compared to men with normal androgen levels (Shores et al., 2004). In sum, despite some inconsistent findings, there is increasing evidence, particularly among larger prospective studies (Ford et al.,  2016), that men with low levels of testosterone are at higher risk to develop MD. In addition to naturalistic studies of testosterone levels, researchers have investigated the effects of ­administering exogenous testosterone to men with MD, either as a standalone treatment or as an adjunct to traditional pharmacotherapy. With respect to the former, two meta-analyses have indicated a  positive effect of testosterone administration on ­depression in hypogonadal, but not eugonadal, men compared to placebo (Amanatkar, Chibnall, Seo, Manepalli, & Grossberg, 2014; Zarrouf, Artz, Griffith, Sirbu, & Kommor, 2009). In addition, the route of administration should be taken into consideration, as the transdermal application of testosterone appears to be more effective at alleviating depression than intramuscular injections (Zarrouf et al.,  2009). The exogenous administration of

t­estosterone has also been used as an augmentation agent in conjunction with antidepressants in male patients with MD. A systematic review conducted by Kleeblatt and colleagues (2017) concluded that testosterone administration is a viable method to enhance antidepressant efficacy, particularly in hypogonadal and older men. However, it is important to note that there are risks associated with repeated testosterone administration, which include polycythemia, prostate cancer, and cardiovascular problems (Surampudi, Wang, & Swerdloff, 2012). Thus, the use of testosterone as a treatment for MD in men has an important caveat, which may prevent its widespread use in patients with MD.

Testosterone Levels and Depression in Women

The association between testosterone levels and ­depressive status in women remains inconclusive (McHenry et al., 2014). For example, a large study found lower salivary testosterone levels in women diagnosed with MD relative to controls (Giltay et al., 2012). In an even larger study of elderly women, serum testosterone was inversely related to depressive symptoms (Morsink et al., 2007). However, results from a large multisite longitudinal study indicated that perimenopausal women with higher baseline serum testosterone levels were at greater risk to display elevated depressive symptoms eight years later (Bromberger et al., 2010). Furthermore, only marginally lower testosterone levels were found in women with MD relative to nondepressed women in a recent two-year prospective study (Giltay et al., 2017). Other studies have reported no difference in serum testosterone levels between women with and without MD (Erdincler, Bugay, Ertan, & Eker, 2004; Matsuzaka et al., 2013), and studies of exogenous testosterone administration in women have been ­inconclusive (McHenry et al., 2014). One possible explanation of these contrasting findings is that the relative ratio of testosterone to estradiol may be more relevant in the context of depression in women than testosterone levels alone (Bromberger et al., 2010).

Possible Mechanisms Underlying the Link Between Testosterone and Depression

Testosterone’s putative relationship with MD is likely due to its influence on different neuropeptide and neurotransmitter systems and its role in neurogenesis in the hippocampus, both of which are known to be dysregulated in MD (Ebinger et al., 2009). One possible specific link between testosterone and

Ellenbogen, Tsekova, and Serravalle

393

depression involves the extensive relationship ­between androgen and monoamine systems, which have repeatedly been implicated in the neurobiology of ­depression (McHenry et al., 2014; Robichaud & Debonnel,  2005). In rodents, testosterone administration increases the release of dopamine in the mesolimbic dopaminergic system (Alderson & Baum, 1981; de Souza Silva, Mattern, Topic, Buddenberg, & Huston, 2009; but see Triemstra, Sato, & Wood, 2008). Difference across studies may be due to methodological factors, as one study found that intranasal administration of testosterone in rats was superior to the subcutaneous injections of oxytocin in stimulating dopamine release (de Souza Silva et  al., 2009). Results from animal studies have shown that testosterone can modify the expression of various serotonergic receptors and alter the functioning of  the serotonergic system (Fink, Sumner, Rosie, Wilson, & McQueen,  1999). For example, testosterone administration in rats increases the firing of serotonergic neurons located in the dorsal raphe nucleus (Robichaud & Debonnel, 2005) and 5-HT2A receptor binding site densities (Sumner & Fink, 1998). Further translational research on how testosterone interacts with monoamine systems implicated in depression is needed to test these ­hypotheses in human populations. As another possibility, the role of testosterone in depression may be explained in part by the interaction between the HPA and HPG axes (Viau, 2002). Specifically, stress inhibits the production of testosterone precursors (gonadotrophins), which in turn lowers testosterone levels (McHenry et al.,  2014). Testosterone has also been shown to decrease glucocorticoid levels in rats (Viau & Meaney, 1996). However, the mechanism by which testosterone dampens HPA axis functioning is yet to be determined. This effect is probably indirect, as few gonadal steroid receptors are found in cells secreting CRH in the paraventricular nucleus of the hypothalamus. Testosterone might dampen the functioning of the HPA axis by acting on regions other than the paraventricular nucleus, such as the medial preoptic area and the amygdala (McHenry et al.,  2014). Again, identifying how neuropeptides influence behavior in humans, whether via the activation or dampening of brain circuits, via interactions with other hormonal and neurotransmitter systems, or via neurogenesis, represents the next phase of this research.

Conclusion and Future Directions

The chapter examined how hormones are implicated in the development and expression of MD by 394

r­eviewing the literature on the HPA axis and the gonadal hormones estrogen and testosterone. Although these hormones are central in understanding the etiology of depression, other adrenal hormones such as dehydroepiandrosterone and dehydroepiandrosterone-sulfate (Mocking et al., ­ 2015), growth hormone (Birmaher & Heydl, 2001), thyroid hormone (MacQueen & Joffe, 2002), and hormones from other tissues (i.e., leptin and ghrelin; Carvalho et  al., 2014; Wittekind & Kluge,  2015) may also be implicated in MD. Despite this limitation, two important trends in the neuroendocrine literature on MD were highlighted. First, MD is characterized by HPA dysfunction at multiple levels of the system, and there is evidence that subtle changes in HPA functioning, such as elevated cortisol levels or an increased CAR, indicate a vulnerability for the disorder. Understanding the relationship between neuroendocrine function and the subsequent development of MD will require a developmental approach to the study of the HPA axis, with a focus on the effects of stress and adversity across the lifespan (Lupien et al., 2009). Despite the progress in this area of research, inconsistencies across studies continue to pervade the literature and many key questions remain unanswered. It is unclear, for example, whether deficits in mounting the CAR (Knight et al.,  2010) or an elevated CAR (Bhagwagar et al., 2005) are associated with risk for MD and other health problems. It is not known how different vulnerabilities for affective disorders lead to, or interact with, subtle changes in the HPA system in the development of MD. Vulnerabilities might be genetic (Goodyer et al., 2010) or related to familial risk (Ostiguy et al., 2011) or associated with exposure to adversity (Rao, Hammen, Ortiz, Chen, & Poland, 2008), and understanding these types of risk will be critical in designing early interventions. We have designed, for example, a prevention program aimed at reducing salivary cortisol levels and the development of internalizing symptoms in the offspring of parents with affective disorders, who are high risk for developing MD and other mental disorders. Preliminary evidence indicates that the 12-week program, entitled Reducing Unwanted Stress in the Home (RUSH), which aimed to teach adaptive coping and emotion regulation strategies to families, successfully reduced internalizing symptoms in children at a six-month follow-up in those families whose parent–child interactions improved at the end of treatment (Iacono et al., 2018). We will be testing the hypothesis that children showing the greatest intervention-related lowering

Hormones and Major Depressive Disorder

of cortisol levels in the natural environment will show the greatest long-term reductions in internalizing symptoms. These types of targeted prevention programs aimed at high-risk children or interventions in persons showing early HPA markers of risk might be useful strategies in decreasing the prevalence of MD. Of course, these proposals require continued study of HPA abnormalities in high-risk populations to further delineate what are the best markers of risk (e.g., CAR, daytime cortisol mean, diurnal cortisol change/slope, cortisol response to stress, exogenous challenges) and rigorous tests of efficacy. At present, there are too many conflicting findings and nonreplications to promote any one of the potential markers listed previously. Future research will need to assess whether cortisol concentration in hair, which allows for the assessment of cortisol levels across months, may represent a useful HPA biomarker for MD (Herane Vives et al., 2015). The second important theme of this chapter is that hormonal markers may be important in defining different types of MD that could theoretically be treated with targeted interventions. Considering that approximately 40 to 50 percent of patients with MD either drop out of treatment prematurely or do not achieve full clinical remission following acute phase treatment with pharmacotherapy (Gitlin, 2014; Rush et al., 2004) or psychological treatments such cognitive-behavioral therapy (DeRubeis et al., 2005; Hollon et al., 2005), a hormonal-based personalized medicine approach might improve treatment efficacy and relapse prevention (Herbert, 2013). For example, patients with psychotic MD exhibit robust abnormalities in the HPA system, including cortisol hypersecretion (Keller et al.,  2006; J.  C. Nelson & Davis, 1997). Mifepristone, a glucocorticoid receptor antagonist, administered for seven days prior to standard antidepressant treatment improves efficacy over placebo in MD patients with psychotic features (DeBattista et al.,  2006), at least in those patients who show high plasma levels of mifepristone (Blasey, McLain, & Belanoff,  2013; Block et al., 2017). Molecules targeting CRH were also viewed with the potential to improve the treatment of MD. Despite the promise of CRH-targeted therapies in preclinical data (Holsboer & Ising, 2010), the use of CRH1 receptor antagonists in the treatment of MD has not been successful in randomized controlled trials (Griebel & Holsboer, 2012; Spierling & Zorrilla,  2017), although past studies have not matched CRH antagonist treatments with patients having CRH hypersecretion (Waters et al., 2015). Using HPA parameters as a means of predicting

the response to antidepressant medication, however, has not been particularly successful. A recent metaanalysis found no strong evidence that pretreatment measures of HPA functioning, such as basal cortisol or CRH levels, robustly predict treatment response to a variety of antidepressant medications (Fischer, Macare, & Cleare,  2017). One promising avenue might involve matching genetic markers associated with HPA axis functioning with the response to treatment. For example, a recent large-scale treatment efficacy study found that MD patients homozygous for the G allele of the rs28365143 variant of the CRHbinding protein gene displayed better response to antidepressant treatment than patients having the A allele (O’Connell et al., 2017). In contrast to the HPA system, gonadal hormone abnormalities highlight specific forms of MD that are associated with different transitions or stages of the lifespan. There is growing evidence of a “reproductive” form of MD that is linked to difficulties successfully adapting to changing levels of ovarian steroid hormones, as seen during the menstrual cycle, postpartum period, and perimenopausal transition (Payne, 2003). Testosterone appears to be implicated in the development of MD in men but limited to the situation when testosterone levels are low, as occurs during normal aging (Ford et al., 2016). For these specific presentations of MD, gonadal hormone augmentation shows great promise in improving the treatment of MD. The transdermal application of testosterone, either as a standalone treatment or an adjunct to traditional pharmacotherapy, appears to be effective at alleviating depression in men, at least among older and hypogonadal men (Amanatkar et al., 2014). The situation with estrogen in the treatment of MD is, however, more complex. Although earlier studies reported robust antidepressant effects of menopausal hormone replacement therapy (estrogen plus a progestin; Zweifel & O’Brien, 1997), a review of recent studies raised questions about the quality of evidence and concluded that estrogen replacement therapy might only benefit perimenopausal women with MD, with little evidence of therapeutic benefits in postmenopausal women (Rubinow, Johnson, Schmidt, Girdler, & Gaynes, 2015). There continues to be much to learn about gonadal hormones and MD. Few studies have attempted to determine predictors of treatment ­ ­response with gonadal hormone interventions, with the exception of low testosterone in men. It is not clear what length of treatment should be used, and  whether gonadal supplementation should be viewed as adjunct or standalone treatment of MD

Ellenbogen, Tsekova, and Serravalle

395

in vulnerable populations. There are few studies examining the childhood antecedents of these ­ ­reproductive forms of MD and whether gonadal hormones might be implicated before the development of these disorders. In one study, elevations of gonadal hormones, either estradiol or testosterone, between 11 and 13 years of age predicted the later development of MD in girls, even while controlling for stage of pubertal development ­ (Angold, Costello, Erkanli, & Worthman, 1999). Whether this early marker of increased gonadal sensitivity and its link to MD relates to later forms of sensitivity to fluctuating or declining gonadal hormones is unknown. Future research will need to consider a lifespan perspective in studying hormones and risk for MD.

References

Adam, E. K., Doane, L. D., Zinbarg, R. E., Mineka, S., Craske, M.  G., & Griffith, J.  W. (2010). Prospective prediction of major depressive disorder from cortisol awakening responses in adolescence. Psychoneuroendocrinology, 35, 921–931. doi:10.1016/j.psyneuen.2009.12.007 Adam, E. K., Vrshek-Schallhorn, S., Kendall, A. D., Mineka, S., Zinbarg, R.  E., & Craske, M.  G. (2014). Prospective associations between the cortisol awakening response and first onsets of anxiety disorders over a six-year follow-up—2013 Curt Richter Award Winner. Psychoneuroendocrinology, 44, 47–59. doi:10.1016/j.psyneuen.2014.02.014 Alderson, L.  M., & Baum, M.  J. (1981). Differential effects of gonadal steroids on dopamine metabolism in mesolimbic and nigro-striatal pathways of male rat brain. Brain Research, 218, 189–206. Almeida, O. P., Waterreus, A., Spry, N., Flicker, L., & Martins, R. N. (2004). One year follow-up study of the association between chemical castration, sex hormones, beta-amyloid, memory and depression in men. Psychoneuroendocrinology, 29, 1071–1081. doi:10.1016/j.psyneuen.2003.11.002 Amanatkar, H. R., Chibnall, J. T., Seo, B. W., Manepalli, J. N., & Grossberg, G.  T. (2014). Impact of exogenous testos­ terone on mood: A systematic review and meta-analysis of randomized placebo-controlled trials. Annals of Clinical Psychiatry, 26, 19–32. Angold, A., Costello, E.  J., Erkanli, A., & Worthman, C.  M. (1999). Pubertal changes in hormone levels and depression in girls. Psychological Medicine, 29, 1043–1053. doi:10.1017/ s0033291799008946 Ashman, S.  B., Dawson, G., Panagiotides, H., Yamada, E., & Wilkinson, C. W. (2002). Stress hormone levels of children of depressed mothers. Development & Psychopathology, 14, 333–349. doi:10.1017/s0954579402002080 Avis, N. E., Brambilla, D., McKinlay, S. M., & Vass, K. (1994). A longitudinal analysis of the association between menopause and depression. Results from the Massachusetts Women’s Health Study. Annals of Epidemiology, 4, 214–220. doi:10.1016/ 0378-5122(91)90286-y Barrett-Connor, E., von Muhlen, D., Laughlin, G. A., & Kripke, A. (1999). Endogenous levels of dehydroepiandrosterone sulfate, but not other sex hormones, are associated with depressed

396

mood in older women: The Rancho Bernardo Study. Journal of the American Geriatrics Society, 47, 685–691. doi:10.1111/ j.1532–5415.1999.tb01590.x Barry, T. J., Murray, L., Fearon, R. M. P., Moutsiana, C., Cooper, P., Goodyer, I. M., . . . Halligan, S. L. (2015). Maternal postnatal depression predicts altered offspring biological stress reac­ tivity in adulthood. Psychoneuroendocrinology, 52, 251–260. doi:10.1016/j.psyneuen.2014.12.003 Benkelfat, C., Ellenbogen, M. A., Dean, P., Palmour, R. M., & Young, S.  N. (1994). Mood-lowering effect of tryptophan depletion: Enhanced susceptibility in young men at genetic risk for major affective disorders. Archives of General Psychiatry, 51, 687–697. doi:10.1001/archpsyc.1994.03950090019003 Bennett, H. A., Einarson, A., Taddio, A., Koren, G., & Einarson, T.  R. (2004). Prevalence of depression during pregnancy: Systematic review. Obstetrics & Gynecology, 103, 698–709. doi:10.1097/01.AOG.0000116689.75396.5f Bhagwagar, Z., Hafizi, S., & Cowen, P. J. (2005). Increased salivary cortisol after waking in depression. Psychopharmacology, 182, 54–57. doi:10.1007/s00213-005-0062-z Birmaher, B., & Heydl, P. (2001). Biological studies in depressed children and adolescents. International Journal of Neuropsycho­ pharmacology, 4, 149–157. doi:10.1017/s1461145701002358 Blasey, C., McLain, C., & Belanoff, J. (2013). Trough plasma concentrations of mifepristone correlate with psychotic symptom reductions: A review of three randomized clinical trials. Current Psychiatry Reviews, 9, 148–154. doi:10.2174/ 1573400511309020009 Bloch, M., Daly, R. C., & Rubinow, D. R. (2003). Endocrine factors in the etiology of postpartum depression. Compre­ hensive Psychiatry, 44, 234–246. doi:10.1016/S0010440X(03)00034-8 Bloch, M., Schmidt, P. J., Danaceau, M., Murphy, J., Nieman, L., & Rubinow, D.  R. (2000). Effects of gonadal steroids in women with a history of postpartum depression. American Journal of Psychiatry, 157, 924–930. doi:10.1176/appi. ajp.157.6.924 Block, T., Petrides, G., Kushner, H., Kalin, N., Belanoff, J., & Schatzberg, A. (2017). Mifepristone plasma level and glucocorticoid receptor antagonism associated with response in patients with psychotic depression. Journal of Clinical Psychopharmacology, 37, 505–511. doi:10.1097/ JCP.0000000000000744 Borrow, A. P., & Cameron, N. M. (2014). Estrogenic mediation of serotonergic and neurotrophic systems: Implications for female mood disorders. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 54, 13–25. doi:10.1016/j.pnpbp.2014. 05.009 Bouma, E.  M., Riese, H., Ormel, J., Verhulst, F.  C., & Oldehinkel, A.  J. (2011). Self-assessed parental depressive problems are associated with blunted cortisol responses to a social stress test in daughters. The TRAILS Study. Psychoneu­ roendocrinology, 36, 854–863. doi:10.1016/j.psyneuen.2010. 11.008 Bowie, P.  C.  W., Beaini, A.  Y., & Bowie, L.  J. (1987). The prognosis of primary depressive illness: Its relationship to the dexamethasone suppression test. British Journal of Psychiatry, 150, 787–790. Brennan, P.  A., Pargas, R., Walker, E.  F., Green, P., Newport, D. J., & Stowe, Z. (2008). Maternal depression and infant cortisol: Influences of timing, comorbidity and treatment. Journal of Child Psychology & Psychiatry & Allied Disciplines, 49, 1099–1107. doi:10.1111/j.1469–7610.2008.01914.x

Hormones and Major Depressive Disorder

Bromberger, J.  T., Kravitz, H.  M., Chang, Y.  F., Cyranowski, J. M., Brown, C., & Matthews, K. A. (2011). Major depression during and after the menopausal transition: Study of Women’s Health Across the Nation (SWAN). Psychological Medicine, 41, 1879–1888. doi:10.1017/S003329171100016X Bromberger, J.  T., Kravitz, H.  M., Matthews, K., Youk, A., Brown, C., & Feng, W. (2009). Predictors of first lifetime episodes of major depression in midlife women. Psychological Medicine, 39, 55–64. doi:10.1017/S0033291708003218 Bromberger, J. T., Matthews, K. A., Schott, L. L., Brockwell, S., Avis, N.  E., Kravitz, H.  M., . . . Randolph, J.  F., Jr. (2007). Depressive symptoms during the menopausal transition: The Study of Women’s Health Across the Nation (SWAN). Journal of Affective Disorders, 103, 267–272. doi:10.1016/ j.jad.2007.01.034 Bromberger, J. T., Schott, L. L., Kravitz, H. M., Sowers, M., Avis, N.  E., Gold, E.  B., . . . Matthews, K.  A. (2010). Longitudinal change in reproductive hormones and depressive symptoms across the menopausal transition: Results from the Study of Women’s Health Across the Nation (SWAN). Archives of General Psychiatry, 67, 598–607. doi:10.1001/archgenpsychiatry. 2010.55 Bromet, E., Andrade, L. H., Hwang, I., Sampson, N. A., Alonso, J., de Girolamo, G., . . . Kessler, R. C. (2011). Cross-national epidemiology of DSM-IV major depressive episode. BMC Medicine, 9, 90. doi:10.1186/1741-7015-9-90 Brummelte, S., & Galea, L. A. M. (2016). Postpartum depression: Etiology, treatment and consequences for maternal care. Hormones and Behavior, 77, 153–166. doi:10.1016/j.yhbeh. 2015.08.008 Burger, H. G. (2002). Androgen production in women. Fertility and Sterility, 77(Suppl. 4), 3–5. doi:10.1016/s0015-0282(02) 02985-0 Burke, H. M., Davis, M. C., Otte, C., & Mohr, D. C. (2005). Depression and cortisol responses to psychological stress: A  meta-analysis. Psychoneuroendocrinology, 30, 846–856. doi:10.1016/j.psyneuen.2005.02.010 Carnegie, R., Araya, R., Ben-Shlomo, Y., Glover, V., O’Connor, T.  G., O’Donnell, K.  J., . . . Lewis, G. (2014). Cortisol awakening response and subsequent depression: Prospective longitudinal study. British Journal of Psychiatry, 204, 137–143. doi:10.1192/bjp.bp.113.126250 Carvalho, A.  F., Rocha, D.  Q.  C., McIntyre, R.  S., Mesquita, L. M., Köhler, C. A., Hyphantis, T. N., . . . Berk, M. (2014). Adipokines as emerging depression biomarkers: A systematic review and meta-analysis. Journal of Psychiatric Research, 59, 28–37. doi:10.1016/j.jpsychires.2014.08.002 Christiansen, K. (2001). Behavioural effects of androgen in men and women. Journal of Endocrinology, 170, 39–48. doi:10.1677/joe.0.1700039 Chrousos, G.  P. (1998). Stressors, stress, and neuroendocrine integration of the adaptive response: The 1997 Hans Selye memorial lecture. Annals of the New York Academy of Sciences, 851, 311–335. doi:10.1111/j.1749–6632.1998.tb0900.x Ciufolini, S., Dazzan, P., Kempton, M.  J., Pariante, C., & Mondelli, V. (2014). HPA axis response to social stress is attenuated in schizophrenia but normal in depression: Evidence from a meta-analysis of existing studies. Neuroscience and Biobehavioral Reviews, 47, 359–368. doi:10.1016/j.neubiorev. 2014.09.004 Cohen, L. S., Altshuler, L. L., Harlow, B. L., Nonacs, R., Newport, D. J., Viguera, A. C., . . . Stowe, Z. N. (2006). Relapse of major depression during pregnancy in women who maintain or

discontinue antidepressant treatment. JAMA, 295, 499–507. doi:10.1001/jama.295.5.499 Cohen, L.  S., Soares, C.  N., Vitonis, A.  F., Otto, M.  W., & Harlow, B. L. (2006). Risk for new onset of depression during the menopausal transition: The Harvard study of moods and cycles. Archives of General Psychiatry, 63, 385–390. doi:10.1001/ archpsyc.63.4.385 Colich, N. L., Kircanski, K., Foland-Ross, L. C., & Gotlib, I. H. (2015). HPA-axis reactivity interacts with stage of pubertal development to predict the onset of depression. Psychoneu­ roendocrinology, 55, 94–101. doi:10.1016/j.psyneuen.2015. 02.004 Davidson, J. M., Chen, J. J., Crapo, L., Gray, G. D., Greenleaf, W. J., & Catania, J. A. (1983). Hormonal changes and sexual function in aging men. Journal of Clinical Endocrinology and Metabolism, 57, 71–77. doi:10.1210/jcem-57-1-71 De Raedt, R., & Koster, E.  H.  W. (2010). Understanding vulnerability for depression from a cognitive neuroscience perspective: A reappraisal of attentional factors and a new conceptual framework. Cognitive, Affective, & Behavioral Neuroscience, 10, 50–70. doi:10.3758/cabn.10.1.50 de Souza Silva, M. A., Mattern, C., Topic, B., Buddenberg, T. E., & Huston, J. P. (2009). Dopaminergic and serotonergic activity in neostriatum and nucleus accumbens enhanced by intranasal administration of testosterone. European Neuropsychopharma­ cology, 19, 53–63. doi:10.1016/j.euroneuro.2008.08.003 DeBattista, C., Belanoff, J., Glass, S., Khan, A., Horne, R. L., Blasey, C., . . . Alva, G. (2006). Mifepristone versus placebo in the treatment of psychosis in patients with psychotic major depression. Biological Psychiatry, 60, 1343–1349. doi:10.1016/ j.biopsych.2006.05.034 DeRubeis, R.  J., Hollon, S.  D., Amsterdam, J.  D., Shelton, R. C., Young, P. R., Salomon, R. M., . . . Gallop, R. (2005). Cognitive therapy vs medications in the treatment of moderate to severe depression. Archives of General Psychiatry, 62, 409–416. doi:10.1001/archpsyc.62.4.409 Dietrich, A., Ormel, J., Buitelaar, J. K., Verhulst, F. C., Hoekstra, P. J., & Hartman, C. A. (2013). Cortisol in the morning and dimensions of anxiety, depression, and aggression in children from a general population and clinic-referred cohort: An integrated analysis. The TRAILS study. Psychoneuroendocri­ nology, 38, 1281–1298. doi:10.1016/j.psyneuen.2012.11.013 Dougherty, L.  R., Klein, D.  N., Rose, S., & Laptook, R.  S. (2011). Hypothalamic-pituitary-adrenal axis reactivity in the preschool-age offspring of depressed parents: Moderation by early parenting. Psychological Science, 22, 650–658. doi:10.1177/ 0956797611404084 Dougherty, L. R., Tolep, M. R., Smith, V. C., & Rose, S. (2013). Early exposure to parental depression and parenting: Associations with young offspring’s stress physiology and oppositional behavior. Journal of Abnormal Child Psychology, 41, 1299–1310. doi:10.1007/s10802-013-9763-7 Duval, F., Mokrani, M.-C., Monreal-Ortiz, J.  A., Fattah, S., Champeval, C., Schulz, P., & Macher, J.-P. (2006). Cortisol hypersecretion in unipolar major depression with melancholic and psychotic features: Dopaminergic, noradrenergic and thyroid correlates. Psychoneuroendocrinology, 31, 876–888. doi:10.1016/j.psyneuen.2006.04.003 Ebinger, M., Sievers, C., Ivan, D., Schneider, H.  J., & Stalla, G.  K. (2009). Is there a neuroendocrinological rationale for testosterone as a therapeutic option in depression? Journal of Psychopharmacology, 23, 841–853. doi:10.1177/ 0269881108092337

Ellenbogen, Tsekova, and Serravalle

397

Eisenlohr-Moul, T.  A., Girdler, S.  S., Johnson, J.  L., Schmidt, P. J., & Rubinow, D. R. (2017). Treatment of premenstrual dysphoria with continuous versus intermittent dosing of oral contraceptives: Results of a three-arm randomized controlled trial. Depression and Anxiety, 34, 908–917. doi:10.1002/ da.22673 Ellenbogen, M.  A., Hodgins, S., Linnen, A.  M., & Ostiguy, C.  S. (2011). Elevated daytime cortisol levels: A biomarker of  subsequent major affective disorder? Journal of Affective Disorders, 132, 265–269. doi:10.1016/j.jad.2011.01.007 Ellenbogen, M.  A., Hodgins, S., Walker, C.  D., Adam, S., & Couture, S. (2006). Daytime cortisol and stress reactivity in the offspring of parents with bipolar disorder. Psychoneuroendo­ crinology, 31, 1164–1180. doi:10.1111/j.1399–5618.2009. 00770.x Ellenbogen, M.  A., & Ostiguy, C.  S. (2017). The relationship between emotional information processing, stress, and diurnal cortisol levels among clinically depressed young adults. Manuscript in preparation. Ellenbogen, M. A., Santo, J. B., Linnen, A. M., Walker, C. D., & Hodgins, S. (2010). High cortisol levels in the offspring of  parents with bipolar disorder during two weeks of daily sampling. Bipolar Disorders, 12, 77–86. doi:10.1111/j. 1399–5618.2009.00770.x Ellenbogen, M. A., Schwartzman, A. E., Stewart, J., & Walker, C. D. (2006). Automatic and effortful emotional information processing regulates different aspects of the stress response. Psychoneuroendocrinology, 31, 373–387. doi:10.1016/j.psyneuen. 2005.09.001 Erdincler, D., Bugay, G., Ertan, T., & Eker, E. (2004). Depression and sex hormones in elderly women. Archives of  Gerontology and Geriatrics, 39, 239–244. doi:10.1016/j. archger.2004.03.008 Essex, M. J., Klein, M. H., Eunsuk, C., & Kalin, N. H. (2002). Maternal stress beginning in infancy may sensitize children to later stress exposure: Effects on cortisol and behavior. Biological Psychiatry, 52, 776–784. doi:10.1016/S00063223(02)01553-6 Fergusson, D. M., Horwood, L. J., Ridder, E. M., & Beautrais, A.  L. (2005). Subthreshold depression in adolescence and mental health outcomes in adulthood. Archives of General Psychiatry, 62, 66–72. doi:10.1001/archpsyc.62.1.66 Fink, G., Sumner, B., Rosie, R., Wilson, H., & McQueen, J. (1999). Androgen actions on central serotonin neurotrans­ mission: Relevance for mood, mental state and memory. Behavioural Brain Research, 105, 53–68. doi:10.1016/s01664328(99)00082-0 Fischer, S., Macare, C., & Cleare, A. J. (2017). Hypothalamicpituitary-adrenal (HPA) axis functioning as predictor of antidepressant response-Meta-analysis. Neuroscience & Biobehavioral Reviews, 83, 200–211. doi:10.1016/j.neubiorev. 2017.10.012 Ford, A. H., Yeap, B. B., Flicker, L., Hankey, G. J., Chubb, S. A., Handelsman, D.  J., . . . Almeida, O.  P. (2016). Prospective longitudinal study of testosterone and incident depression in older men: The Health in Men Study. Psychoneuroendocrinology, 64, 57–65. doi:10.1016/j.psyneuen.2015.11.012 Freeman, E.  W. (2010). Associations of depression with the transition to menopause. Menopause, 17, 823–827. doi:10.1097/ gme.0b013e3181db9f8b Freeman, E.  W., Halbreich, U., Grubb, G.  S., Rapkin, A.  J., Skouby, S. O., Smith, L., . . . Constantine, G. D. (2012). An overview of four studies of a continuous oral contraceptive

398

(levonorgestrel 90 mcg/ethinyl estradiol 20 mcg) on premenstrual dysphoric disorder and premenstrual syndrome. Contraception, 85, 437–445. doi:10.1016/j.contraception. 2011.09.010 Freeman, E. W., Kroll, R., Rapkin, A., Pearlstein, T., Brown, C., Parsey, K., . . . Group, P. P. R. (2001). Evaluation of a unique oral contraceptive in the treatment of premenstrual dysphoric disorder. Journal of Women’s Health and Gender-Based Medicine, 10, 561–569. doi:10.1089/15246090152543148 Freeman, E. W., Sammel, M. D., Boorman, D. W., & Zhang, R. (2014). Longitudinal pattern of depressive symptoms around natural menopause. JAMA Psychiatry, 71, 36–43. doi:10.1001/ jamapsychiatry.2013.2819 Freeman, E. W., Sammel, M. D., Lin, H., Gracia, C. R., Pien, G.  W., Nelson, D.  B., & Sheng, L. (2007). Symptoms associated with menopausal transition and reproductive hormones in midlife women. Obstetrics & Gynecology, 110, 230–240. doi:10.1097/01.AOG.0000270153.59102.40 Freeman, E.  W., Sammel, M.  D., Lin, H., & Nelson, D.  B. (2006). Associations of hormones and menopausal status with depressed mood in women with no history of depression. Archives of General Psychiatry, 63, 375–382. doi:10.1001/ archpsyc.63.4.375 Freeman, E. W., Sammel, M. D., Liu, L., Gracia, C. R., Nelson, D. B., & Hollander, L. (2004). Hormones and menopausal status as predictors of depression in women in transition to  menopause. Archives of General Psychiatry, 61, 62–70. doi:10.1001/archpsyc.61.1.62 Frokjaer, V.  G., Pinborg, A., Holst, K.  K., Overgaard, A., Henningsson, S., Heede, M., . . . Knudsen, G.  M. (2015). Role of serotonin transporter changes in depressive responses to sex-steroid hormone manipulation: A positron emission tomography study. Biological Psychiatry, 78, 534–543. doi:10.1016/j.biopsych.2015.04.015 Gehlert, S., Song, I.  H., Chang, C.  H., & Hartlage, S.  A. (2009). The prevalence of premenstrual dysphoric disorder in a randomly selected group of urban and rural women. Psychological Medicine, 39, 129–136. doi:10.1017/ S003329170800322X Gelaye, B., Rondon, M. B., Araya, R., & Williams, M. A. (2016). Epidemiology of maternal depression, risk factors, and child outcomes in low-income and middle-income countries. Lancet Psychiatry, 3, 973–982. doi:10.1016/S2215-0366(16)30284-X Gibb, B. E., Beevers, C. G., & McGeary, J. E. (2013). Toward an integration of cognitive and genetic models of risk for depression. Cognition and Emotion, 27, 193–216. doi:10.1080/ 02699931.2012.712950 Giltay, E. J., Enter, D., Zitman, F. G., Penninx, B. W., van Pelt, J., Spinhoven, P., & Roelofs, K. (2012). Salivary testosterone: Associations with depression, anxiety disorders, and antide­ pressant use in a large cohort study. Journal of Psychosomatic Research, 72, 205–213. doi:10.1016/j.jpsychores.2011.11.014 Giltay, E. J., van der Mast, R. C., Lauwen, E., Heijboer, A. C., de  Waal, M.  W.  M., & Comijs, H.  C. (2017). Plasma testosterone and the course of major depressive disorder in older men and women. American Journal of Geriatric Psychiatry, 25, 425–437. doi:10.1016/j.jagp.2016.12.014 Gitlin, M.  J. (2014). Pharmacotherapy and other somatic treatments for depression. In I. H. Gotlib & C. L. Hammen (Eds.), Handbook of depression (3rd ed., pp. 634–659). New York, NY: Guilford Press. Goodman, S. H., & Brand, S. R. (2009). Depression and early adverse experiences. In I. H. Gotlib & C. L. Hammen (Eds.),

Hormones and Major Depressive Disorder

Handbook of depression (2nd ed., pp. 249–274). New York, NY: Guilford Press. Goodyer, I. M., Bacon, A., Ban, M., Croudace, T., & Herbert, J. (2009). Serotonin transporter genotype, morning cortisol and subsequent depression in adolescents. British Journal of Psychiatry, 195, 39–45. doi:10.1192/bjp.bp.108.054775 Goodyer, I.  M., Croudace, T., Dudbridge, F., Ban, M., & Herbert, J. (2010). Polymorphisms in BDNF (Val66Met) and 5-HTTLPR, morning cortisol and subsequent depression in at-risk adolescents. British Journal of Psychiatry, 197, 365–371. doi:10.1192/bjp.bp.110.077750 Goodyer, I.  M., Herbert, J., Tamplin, A., & Altham, P.  M. (2000). Recent life events, cortisol, dehydroepiandrosterone and the onset of major depression in high-risk adolescents. British Journal of Psychiatry, 177, 499–504. doi:10.1192/bjp. 177.6.499 Gordon, J. L., Girdler, S. S., Meltzer-Brody, S. E., Stika, C. S., Thurston, R.  C., Clark, C.  T., . . . Wisner, K.  L. (2015). Ovarian hormone fluctuation, neurosteroids, and HPA axis dysregulation in perimenopausal depression: A novel heuristic model. American Journal of Psychiatry, 172, 227–236. doi:10.1176/appi.ajp.2014.14070918 Gotlib, I. H., Joormann, J., Minor, K. L., & Hallmayer, J. (2008). HPA axis reactivity: A mechanism underlying the associ­ ations among 5-HTTLPR, stress, and depression. Biological Psychiatry, 63, 847–851. doi:10.1016/j.biopsych.2007.10.008 Graham, C.  A., & Sherwin, B.  B. (1992). A prospective treat­ ment study of premenstrual symptoms using a triphasic oral contraceptive. Journal of Psychosomatic Research, 36, 257–266. doi:10.1016/0022-3999(92)90090-o Griebel, G., & Holsboer, F. (2012). Neuropeptide receptor ligands as drugs for psychiatric diseases: The end of the beginning? Nature Reviews. Drug Discovery, 11, 462–478. doi:10.1038/nrd3702 Halbreich, U., Asnis, G. M., Shindledecker, R., Zumoff, B., & Nathan, R. S. (1985). Cortisol secretion in endogenous depres­ sion. I. Basal plasma levels. Archives of General Psychiatry, 42, 904–908. Halbreich, U., Borenstein, J., Pearlstein, T., & Kahn, L.  S. (2003). The prevalence, impairment, impact, and burden of pre­menstrual dysphoric disorder (PMS/PMDD). Psychoneu­ roendocrinology, 28(Suppl. 3), 1–23. doi:10.1016/S03064530(03)00098-2 Halligan, S. L., Herbert, J., Goodyer, I. A., & Murray, L. (2004). Exposure to postnatal depression predicts elevated cortisol in adolescent offspring. Biological Psychiatry, 55, 376–381. doi:10.1016/j.biopsych.2003.09.013 Halligan, S.  L., Herbert, J., Goodyer, I., & Murray, L. (2007). Disturbances in morning cortisol secretion in association with maternal postnatal depression predict subsequent depressive symptomatology in adolescents. Biological Psychiatry, 62, 40–46. doi:10.1016/j.biopsych.2006.09.011 Hammen, C.  L. (2015). Stress and depression: Old questions, new approaches. Current Opinion in Psychology, 4, 80–85. doi:10.1016/j.copsyc.2014.12.024 Hankin, B. L., & Abramson, L. Y. (1999). Development of gender differences in depression: Description and possible explana­ tions. Annals of Medicine, 31, 372–379. doi:10.3109/ 07853899908998794 Hankin, B. L., Badanes, L. S., Abela, J. R., & Watamura, S. E. (2010). Hypothalamic-pituitary-adrenal axis dysregulation in dysphoric children and adolescents: Cortisol reactivity to psychosocial stress from preschool through middle adolescence.

Biological Psychiatry, 68, 484–490. doi:10.1016/j.biopsych. 2010.04.004 Harkness, K. L., & Lumley, M. N. (2008). Child abuse and neglect and the development of depression in children and adolescents. In J. R. Z. Abela & B. L. Hankin (Eds.), Handbook of depression in children and adolescents (pp. 466–488). New York, NY: Guilford Press. Harkness, K. L., & Monroe, S. M. (2006). Severe melancholic depression is more vulnerable than non-melancholic depression to minor precipitating life events. Journal of Affective Disorders, 91, 257–263. doi:10.1016/j.jad.2005.12.009 Harris, T.  O., Borsanyi, S., Messari, S., Stanford, K., Cleary, S. E., Shiers, H. M., . . . Herbert, J. (2000). Morning cortisol as a risk factor for subsequent major depressive disorder in adult women. British Journal of Psychiatry, 177, 505–510. doi:10.1192/bjp.177.6.505 Heim, C., Newport, D.  J., Mletzko, T., Miller, A.  H., & Nemeroff, C. B. (2008). The link between childhood trauma and depression: Insights from HPA axis studies in humans. Psychoneuroendocrinology, 33, 693–710. doi:10.1016/j.psyneuen. 2008.03.008 Hellgren, C., Åkerud, H., Skalkidou, A., & Sundström-Poromaa, I. (2013). Cortisol awakening response in late pregnancy in women with previous or ongoing depression. Psychoneuroen­ docrinology, 38, 3150–3154. doi:10.1016/j.psyneuen.2013. 08.007 Herane Vives, A., De Angel, V., Papadopoulos, A., Strawbridge, R., Wise, T., Young, A. H., . . . Cleare, A. J. (2015). The relationship between cortisol, stress and psychiatric illness: New insights using hair analysis. Journal of Psychiatric Research, 70, 38–49. doi:10.1016/j.jpsychires.2015.08.007 Herbert, J. (2013). Cortisol and depression: Three questions for psychiatry. Psychological Medicine, 43, 449–469. doi:10.1017/ S0033291712000955 Heuser, I., Yassouridis, A., & Holsboer, F. (1994). The combined dexamethasone/CRH test: A refined laboratory test for psychiatric disorders. Journal of Psychiatric Research, 28, 341–356. doi:10.1016/0022-3956(94)90017-5 Hollon, S.  D., DeRubeis, R.  J., Shelton, R.  C., Amsterdam, J. D., Salomon, R. M., O’Reardon, J. P., . . . Gallop, R. (2005). Prevention of relapse following cognitive therapy vs medi­ cations in moderate to severe depression. Archives of General Psychiatry, 62, 417–422. doi:10.1001/archpsyc.62.4.417 Holsboer, F. (1995). Neuroendocrinology of mood disorders. In  F.  Holsboer, F.  E.  Bloom, & D.  J.  Kupfer (Eds.), Psychopharmacology: The fourth generation of progress (pp. 957–969). New York, NY: Raven Press. Holsboer, F., & Ising, M. (2010). Stress hormone regulation: Biological role and translation into therapy. Annual Review of Psychology, 61, 81–109. doi:10.1146/annurev.psych.093008. 100321 Holsboer, F., Lauer, C. J., Schreiber, W., & Krieg, J. C. (1995). Altered hypothalamic-pituitary-adrenocortical regulation in healthy subjects at high familial risk for affective disorders. Neuroendocrinology, 62, 340–347. doi:10.1159/000127023 Holsboer, F., Spengler, D., & Heuser, I. (1992). The role of corticotropin-releasing hormone in the pathogenesis of Cushing’s disease, anorexia nervosa, alcoholism, affective disorders and dementia. Progress in Brain Research, 93, 385–417. doi:10.1016/s0079-6123(08)64586-0 Holsboer, F., von Bardeleben, U., Wiedemann, K., Muller, O. A., & Stalla, G. K. (1987). Serial assessment of corticotropinreleasing hormone response after dexamethasone in depression.

Ellenbogen, Tsekova, and Serravalle

399

Implications for pathophysiology of DST nonsuppression. Biological Psychiatry, 22, 228–234. doi:10.1016/0006-3223(87) 90237-x Huber, T.  J., Issa, K., Schik, G., & Wolf, O.  T. (2006). The cortisol awakening response is blunted in psychotherapy inpatients suffering from depression. Psychoneuroendocrinology, 31, 900–904. doi:10.1016/j.psyneuen.2006.03.005 Iacono, V., Serravalle, L., Wilson, A. L., Orlando, M. A., Tsekova, V., & Ellenbogen, M.  A. (2018). Parent-child interaction quality predicts reduced internalizing symptoms among the offspring of parents with bipolar disorder undergoing a stressreduction program: A proof-of-concept study. Manuscript in preparation. Ising, M., Lauer, C. J., Holsboer, F., & Modell, S. (2005). The Munich vulnerability study on affective disorders: Premorbid neuroendocrine profile of affected high-risk probands. Journal of Psychiatric Research, 39, 21–28. doi:10.1016/j.jpsychires. 2004.04.009 Johnson, J. M., Nachtigall, L. B., & Stern, T. A. (2013). The effect of testosterone levels on mood in men: A review. Psychosomatics, 54, 509–514. doi:10.1016/j.psym.2013.06.018 Joiner, T. E., Jr., & Timmons, K. A. (2009). Depression in its interpersonal context. In I.  H.  Gotlib & C.  L.  Hammen (Eds.), Handbook of depression (2nd ed., pp. 322–339). New York, NY: Guilford Press. Joormann, J., & Tanovic, E. (2015). Cognitive vulnerability to depression: Examining cognitive control and emotion regu­ lation. Current Opinion in Psychology, 4, 86–92. doi:10.1016/ j.copsyc.2014.12.006 Keenan, K., Hipwell, A., Babinski, D., Bortner, J., Henneberger, A., Hinze, A., . . . Sapotichne, B. (2013). Examining the develop­ mental interface of cortisol and depression symptoms in young adolescent girls. Psychoneuroendocrinology, 38, 2291–2299. doi:10.1016/j.psyneuen.2013.04.017 Keller, J., Flores, B., Gomez, R. G., Solvason, H. B., Kenna, H., Williams, G. H., & Schatzberg, A. F. (2006). Cortisol circadian rhythm alterations in psychotic major depression. Biological Psychiatry, 60, 275–281. doi:10.1016/j.biopsych.2005.10.014 Kendler, K.  S., & Prescott, C.  A. (1999). A population-based twin study of lifetime major depression in men and women. Archives of General Psychiatry, 56, 39–44. doi:10.1001/archpsyc. 56.1.39 Kessing, L. V., Willer, I. S., & Knorr, U. (2011). Volume of the adrenal and pituitary glands in depression. Psychoneuroendocri­ nology, 36, 19–27. doi:10.1016/j.psyneuen.2010.05.007 Kessler, R. C. (2012). The costs of depression. Psychiatric Clinics of North America, 35, 1–14. doi:10.1016/j.psc.2011.11.005 Kessler, R.  C., Berglund, P., Demler, O., Korentz, D., Merikangas, K. R., Rush, A. J., . . . Wang, P. S. (2003). The epidemiology of major depressive disorder. Result from the National Comorbidity Replication (NCS-R). JAMA, 289, 3095–3105. doi:10.1001/jama.289.23.3095 Kessler, R. C., McGonagle, K. A., Swartz, M. S., Blazer, D. G., & Nelson, C. B. (1993). Sex and depression in the National Comorbidity Survey I: Lifetime prevalence, chronicity and recurrence. Journal of Affective Disorders, 29, 85–96. doi:10.1016/0165-0327(93)90026-g Kessler, R. C., & Wang, P. L. (2009). Epidemiology of depression. In I.  H.  Gotlib & C.  L.  Hammen (Eds.), Handbook of depression (2nd ed.). New York, NY: Guilford Press. Kirschbaum, C., Pirke, K.  M., & Hellhammer, D.  H. (1993). The “Trier social stress test”—A tool for investigating

400

psychobiological stress responses in a laboratory setting. Neuropsychobiology, 28, 76–81. doi:10.1159/000119004 Kleeblatt, J., Betzler, F., Kilarski, L. L., Bschor, T., & Kohler, S. (2017). Efficacy of off-label augmentation in unipolar depres­ sion: A systematic review of the evidence. European Neuropsy­ chopharmacology, 27, 423–441. doi:10.1016/j.euroneuro.2017. 03.003 Klein, D. N., Glenn, C. R., Kosty, D. B., Seeley, J. R., Rohde, P., & Lewinsohn, P. M. (2013). Predictors of first lifetime onset of major depressive disorder in young adulthood. Journal of Abnormal Psychology, 122, 1–6. doi:10.1037/a0029567 Knight, J. M., Avery, E. F., Janssen, I., & Powell, L. H. (2010). Cortisol and depressive symptoms in a population-based cohort of midlife women. Psychosomatic Medicine, 72, 855–861. doi:10.1097/PSY.0b013e3181f4ab87 Kormos, V., & Gaszner, B. (2013). Role of neuropeptides in anxiety, stress, and depression: From animals to humans. Neuropeptides, 47, 401–419. doi:10.1016/j.npep.2013.10.014 Kunugi, H., Ida, I., Owashi, T., Kimura, M., Inoue, Y., Nakagawa, S., . . . Mikuni, M. (2006). Assessment of the dexamethasone/CRH test as a state-dependent marker for hypothalamic-pituitary-adrenal (HPA) axis abnormalities in major depressive episode: A multicenter study. Neuropsycho­ pharmacology, 31, 212–220. doi:10.1038/sj.npp.1300868 Lamers, F., Vogelzangs, N., Merikangas, K.  R., de Jonge, P., Beekman, A. T. F., & Penninx, B. W. J. H. (2013). Evidence for a differential role of HPA-axis function, inflammation and metabolic syndrome in melancholic versus atypical depression. Molecular Psychiatry, 18, 692–699. doi:10.1038/mp.2012.144 Lammers, C.-H., Garcia-Borreguero, D., Schmider, J., Gotthardt, U., Dettling, M., Holsboer, F., & Heuser, I.  J.  E. (1995). Combined dexamethasone/corticotropin-releasing hormone test in patients with schizophrenia and in normal controls: II. Biological Psychiatry, 38, 803–807. doi:10.1016/00063223(95)00065-8 Lapalme, M., Hodgins, S., & LaRoche, C. (1997). Children of parents with bipolar disorder: A meta-analysis of risk for mental disorders. Canadian Journal of Psychiatry, 42, 623–631. doi:10.1177/070674379704200609 LeMoult, J., & Joormann, J. (2014). Depressive rumination alters cortisol decline in major depressive disorder. Biological Psychology, 100, 50–55. doi:10.1016/j.biopsycho.2014.05.001 Lewinsohn, P.  M., Rohde, P., Seeley, J.  R., Klein, D.  N., & Gotlib, I. H. (2003). Psychosocial functioning of young adults who have experienced and recovered from major depressive disorder during adolescence. Journal of Abnormal Psychology, 112, 353–363. doi:10.1037/0021-843x.112.3.353 Lopez-Duran, N.  L., Kovacs, M., & George, C.  J. (2009). Hypothalamic-pituitary-adrenal axis dysregulation in depressed children and adolescents: A meta-analysis. Psychoneuroendo­ crinology, 34, 1272–1283. doi:10.1016/j.psyneuen.2009.03.016 Lovallo, W. R., Farag, N. H., Sorocco, K. H., Cohoon, A. J., & Vincent, A. S. (2012). Lifetime adversity leads to blunted stress axis reactivity: Studies from the Oklahoma Family Health Patterns Project. Biological Psychiatry, 71, 344–349. doi:10.1016/ j.biopsych.2011.10.018 Lundy, B. L., Jones, N. A., Field, T., Nearing, G., Davalos, M., Pietro, P. A., . . . Kuhn, C. (1999). Prenatal depression effects on neonates. Infant Behavior & Development, 22, 119–129. doi:10.1016/s0163-6383(99)80009-5 Lupien, S.  J., McEwen, B.  S., Gunnar, M.  R., & Heim, C. (2009). Effects of stress throughout the lifespan on the brain,

Hormones and Major Depressive Disorder

behaviour and cognition. Nature Reviews Neuroscience, 10, 434–445. doi:10.1038/nrn2639 Lupien, S.  J., Sasseville, M., Francois, N., Giguere, C.  E., Boissonneault, J., Plusquellec, P., . . . Lesage, A. (2017). The DSM5/RDoC debate on the future of mental health research: Implication for studies on human stress and presentation of the signature bank. Stress, 20, 95–111. doi:10.1080/10253890. 2017.1286324 MacQueen, G.  M., & Joffe, R.  T. (2002). A review of gender differences in studies of thyroid function and major depression. Psychiatric Annals, 32, 477–482. doi:10.3928/0048-571320020801-07 Mannie, Z. N., Harmer, C. J., & Cowen, P. J. (2007). Increased waking salivary cortisol levels in young people at familial risk of depression. American Journal of Psychiatry, 164, 617–621. doi:10.1176/ajp.2007.164.4.617 Matsuzaka, H., Maeshima, H., Kida, S., Kurita, H., Shimano, T., Nakano, Y., . . . Arai, H. (2013). Gender differences in serum testosterone and cortisol in patients with major depressive disorder compared with controls. International Journal of Psychiatry in Medicine, 46, 203–221. doi:10.2190/PM.46.2.g McEwen, B. S. (2004). Protection and damage from acute and chronic stress: Allostasis and allostatic overload and relevance to the pathophysiology of psychiatric disorders. Annals of the New York Academy of Sciences, 1032, 1–7. doi:10.1196/ annals.1314.001 McEwen, B.  S., & Gianaros, P.  J. (2010). Central role of the brain in stress and adaptation: Links to socioeconomic status, health, and disease. Annals of the New York Academy of Sciences, 1186, 190–222. doi:10.1111/j.1749–6632.2009.05331.x McGuffin, P., Rijsdijk, F. V., Andrew, M., Sham, P., Katz, R., & Cardno, A. (2003). The heritability of bipolar affective disorder and the genetic relationship to unipolar depression. Archives of General Psychiatry, 60, 497–502. doi:10.1001/archpsyc.60.5.497 McHenry, J., Carrier, N., Hull, E., & Kabbaj, M. (2014). Sex differences in anxiety and depression: Role of testosterone. Frontiers in Neuroendocrinology, 35, 42–57. doi:10.1016/j. yfrne.2013.09.001 McIntyre, R.  S., Mancini, D., Eisfeld, B.  S., Soczynska, J.  K., Grupp, L., Konarski, J.  Z., & Kennedy, S.  H. (2006). Calculated bioavailable testosterone levels and depression in middle-aged men. Psychoneuroendocrinology, 31, 1029–1035. doi:10.1016/j.psyneuen.2006.06.005 McLeod, B. D., Weisz, J. R., & Wood, J. J. (2007). Examining the association between parenting and childhood depression: A meta-analysis. Clinical Psychology Review, 27, 986–1003. doi:10.1016/j.cpr.2007.03.001 Mehta, D., Newport, D. J., Frishman, G., Kraus, L., Rex-Haffner, M., Ritchie, J. C., . . . Binder, E. B. (2014). Early predictive biomarkers for postpartum depression point to a role for estrogen receptor signaling. Psychological Medicine, 44, 2309–2322. doi:10.1017/S0033291713003231 Mocking, R.  J.  T., Pellikaan, C.  M., Lok, A., Assies, J., Ruhé, H. G., Koeter, M. W., . . . Schene, A. H. (2015). DHEAS and cortisol/DHEAS-ratio in recurrent depression: State, or trait predicting 10-year recurrence? Psychoneuroendocrinology, 59, 91–101. doi:10.1016/j.psyneuen.2015.05.006 Modell, S., Lauer, C. J., Schreiber, W., Huber, J., Krieg, J.-C., & Holsboer, F. (1998). Hormonal response pattern in the combined DEX-CRH test is stable over time in subjects at high familial risk for affective disorders. Neuropsychophar­ macology, 18, 253–262. doi:10.1016/s0893-133x(97)00144-9

Modell, S., Yassouridis, A., Huber, J., & Holsboer, F. (1997). Corticosteroid receptor function is decreased in depressed patients. Neuroendocrinology, 65, 216–222. doi:10.1159/ 000127275 Mokhtari, M., Arfken, C., & Boutros, N. (2013). The DEX/CRH test for major depression: A potentially useful diagnostic test. Psychiatry Research, 208, 131–139. doi:10.1016/j.psychres. 2012.09.032 Morris, M.  C., & Rao, U. (2014). Cortisol response to psychosocial stress during a depressive episode and remission. Stress: The International Journal on the Biology of Stress, 17, 51–58. doi:10.3109/10253890.2013.857398 Morris, M. C., Rao, U., Wang, L., & Garber, J. (2014). Cortisol reactivity to experimentally manipulated psychosocial stress in young adults at varied risk for depression. Depression and Anxiety, 31, 44–52. doi:10.1002/da.22125 Morsink, L. F., Vogelzangs, N., Nicklas, B. J., Beekman, A. T., Satterfield, S., Rubin, S. M., . . . Health, A. B. C. S. (2007). Associations between sex steroid hormone levels and depressive symptoms in elderly men and women: Results from the Health ABC study. Psychoneuroendocrinology, 32, 874–883. doi:10.1016/j.psyneuen.2007.06.009 Moses, E.  L., Drevets, W.  C., Smith, G., Mathis, C.  A., Kalro, B.  N., Butters, M.  A., . . . Berga, S.  L. (2000). Effects of estradiol and progesterone administration on human serotonin 2A receptor binding: A PET study. Biological Psychiatry, 48, 854–860. doi:10.1016/s0006-3223(00)00967-7 Nederhof, E., Marceau, K., Shirtcliff, E.  A., Hastings, P.  D., & Oldehinkel, A. J. (2015). Autonomic and adrenocortical interactions predict mental health in late adolescence: The TRAILS study. Journal of Abnormal Child Psychology, 43, 847–861. doi:10.1007/s10802-014-9958-6 Nederhof, E., van Oort, F. V. A., Bouma, E. M. C., Laceulle, O. M., Oldehinkel, A. J., & Ormel, J. (2015). Predicting mental disorders from hypothalamic-pituitary-adrenal axis functioning: A 3-year follow-up in the TRAILS study. Psychological Medicine, 45, 2403–2412. doi:10.1017/S0033291715000392 Nelson, J. C., & Davis, J. M. (1997). DST studies in psychotic depression: A meta-analysis. American Journal of Psychiatry, 154, 1497–1503. doi:10.1176/ajp.154.11.1497 Nelson, L. R., & Bulun, S. E. (2001). Estrogen production and action. Journal of American Academy of Dermatology, 45, S116–S124. doi:10.1067/mjd.2001.117432 Nemeroff, C. B. (1996). The corticotropin-releasing factor (CRF) hypothesis of depression: New findings and new directions. Molecular Psychiatry, 1, 336–342. Nemeroff, C. B., Owens, M. J., Bissette, G., Andorn, A. C., & Stanley, M. (1988). Reduced corticotropin releasing factor binding sites in the frontal cortex of suicide victims. Archives of General Psychiatry, 45, 577–579. Nemeroff, C.  B., & Vale, W.  W. (2005). The neurobiology of depression: Inroads to treatment and new drug discovery. Journal of Clinical Psychiatry, 66(Suppl. 7), 5–13. Newhouse, P., & Albert, K. (2015). Estrogen, stress, and depression: A neurocognitive model. JAMA Psychiatry, 72, 727–729. doi:10.1001/jamapsychiatry.2015.0487 Nierenberg, A. A., & Feinstein, A. R. (1988). How to evaluate a  diagnostic marker test. Lessons from the rise and fall of dexamethasone suppression test. JAMA, 259, 1699–1702. doi:10.1001/jama.259.11.1699 Nijjar, R., Ellenbogen, M. A., & Hodgins, S. (2014). Personality, coping, risky behavior, and mental disorders in the offspring

Ellenbogen, Tsekova, and Serravalle

401

of parents with bipolar disorder: A comprehensive psychosocial assessment. Journal of Affective Disorders, 166, 315–323. doi:10.1016/j.jad.2014.04.047 O’Connell, C. P., Goldstein-Piekarski, A. N., Nemeroff, C. B., Schatzberg, A. F., Debattista, C., Carrillo-Roa, T., . . . Williams, L. M. (2017). Antidepressant outcomes predicted by genetic variation in corticotropin-releasing hormone binding protein. American Journal of Psychiatry, 175, 251–261. doi:10.1176/ appi.ajp.2017.17020172 O’Hara, M. W., & McCabe, J. E. (2013). Postpartum depression: Current status and future directions. Annual Review of Clinical Psychology, 9, 379–407. doi:10.1146/annurev-clinpsy-050212185612 Ostiguy, C.  S., Ellenbogen, M.  A., & Hodgins, S. (2012). Personality of parents with bipolar disorder and interpersonal functioning among their offspring: A prospective 10-year study. Development and Psychopathology, 24, 573–587. doi:10.1017/ S095457941200017X Ostiguy, C. S., Ellenbogen, M. A., Walker, C. D., Walker, E. F., & Hodgins, S. (2011). Sensitivity to stress among the offspring of parents with bipolar disorder: A study of daytime cortisol levels. Psychological Medicine, 41, 2447–2457. doi:10.1017/ S0033291711000523 Owens, M., Herbert, J., Jones, P. B., Sahakian, B. J., Wilkinson, P. O., Dunn, V. J., . . . Goodyer, I. M. (2014). Elevated morning cortisol is a stratified population level biomarker for major depression in boys only with high depressive symptoms. Proceedings of the National Academy of Sciences of the United States of America, 111, 3638–3643. doi:10.1073/pnas.1318786111 Paslakis, G., Heuser, I., Schweiger, U., & Deuschle, M. (2010). A single DEX/CRH test in male drug-free depressed patients is  associated with the clinical response to treatment with fluoxetine. Journal of Psychiatric Research, 44, 1154–1157. doi:10.1016/j.jpsychires.2010.04.010 Payne, J. L. (2003). The role of estrogen in mood disorders in women. International Review of Psychiatry, 15, 280–290. doi:10.1080/0954026031000136893 Payne, J. L., Palmer, J. T., & Joffe, H. (2009). A reproductive subtype of depression: Conceptualizing models and moving toward etiology. Harvard Review of Psychiatry, 17, 72–86. doi:10.1080/10673220902899706 Pearlstein, T.  B., Bachmann, G.  A., Zacur, H.  A., & Yonkers, K. A. (2005). Treatment of premenstrual dysphoric disorder with a new drospirenone-containing oral contraceptive formulation. Contraception, 72, 414–421. doi:10.1016/j. contraception.2005.08.021 Pitts, A.  F., Samuelson, S.  D., Meller, W.  H., & Bissette, G. (1995). Cerebrospinal fluid corticotropin-releasing hormone, vasopressin, and oxytocin concentrations in treated patients with major depression and controls. Biological Psychiatry, 38, 330–335. doi:10.1016/0006-3223(95)00229-a Prange-Kiel, J., Jarry, H., Schoen, M., Kohlmann, P., Lohse, C., Zhou, L., & Rune, G. M. (2008). Gonadotropin-releasing hormone regulates spine density via its regulatory role in hippocampal estrogen synthesis. Journal of Cell Biology, 180, 417–426. doi:10.1083/jcb.200707043 Pruessner, M., Hellhammer, D. H., Pruessner, J. C., & Lupien, S.  J. (2003). Self-reported depressive symptoms and stress levels in healthy young men: Associations with the cortisol response to awakening. Psychosomatic Medicine, 65, 92–99. doi:10.1097/01.psy.0000040950.22044.10 Rao, U., Hammen, C., Ortiz, L.  R., Chen, L.-A., & Poland, R. E. (2008). Effects of early and recent adverse experiences

402

on adrenal response to psychosocial stress in depressed adolescents. Biological Psychiatry, 64, 521–526. doi:10.1016/j. biopsych.2008.05.012 Rasic, D., Hajek, T., Alda, M., & Uher, R. (2014). Risk of mental illness in offspring of parents with schizophrenia, bipolar disorder, and major depressive disorder: A meta-analysis of family high-risk studies. Schizophrenia Bulletin, 40, 28–38. doi:10.1093/schbul/sbt114 Ribeiro, S. C., Tandon, R., Grunhaus, L., & Greden, J. F. (1993). The DST as a predictor of outcome in depression: A metaanalysis. American Journal of Psychiatry, 150, 1618–1629. doi:10.1176/ajp.150.11.1618 Robertson, E., Grace, S., Wallington, T., & Stewart, D.  E. (2004). Antenatal risk factors for postpartum depression: A synthesis of recent literature. General Hospital Psychiatry, 26, 289–295. doi:10.1016/j.genhosppsych.2004.02.006 Robichaud, M., & Debonnel, G. (2005). Oestrogen and testosterone modulate the firing activity of dorsal raphe nucleus serotonergic neurones in both male and female rats. Journal of Neuroendocrinology, 17, 179–185. doi:10.1111/ j.1365–2826.2005.01292.x Ronsaville, D.  S., Municchi, G., Laney, C., Cizza, G., Meyer, S.  E., Haim, A., . . . Martinez, P.  E. (2006). Maternal and environmental factors influence the HPA axis response to corticotropin releasing hormone infusion in offspring of mothers with or without mood disorders. Development and Psychopathology, 18, 173–194. doi:10.1017/s095457940606010x Rubinow, D.  R., Johnson, S.  L., Schmidt, P.  J., Girdler, S., & Gaynes, B. (2015). Efficacy of estradiol in perimenopausal depression: So much promise and so few answers. Depression and Anxiety, 32, 539–549. doi:10.1002/da.22391 Rubinow, D. R., Schmidt, P. J., & Craft, C. D. (2013). Gonadal steroids and mood disorders. In D. S. Charney, J. D. Buxbaum, P. Sklar, E. J. Nestler, D. S. Charney, J. D. Buxbaum, P. Sklar, & E.  J.  Nestler (Eds.), Neurobiology of mental illness (pp. 483–495). New York, NY: Oxford University Press. Rush, A. J., Giles, D. E., Schlesser, M. A., Orsulak, P. J., Parker, C. R., Jr., Weissenburger, J. E., . . . Vasavada, N. (1996). The dexamethasone suppression test in patients with mood disorders. Journal of Clinical Psychiatry, 57, 470–484. Rush, A. J., Trivedi, M., Carmody, T. J., Biggs, M. M., ShoresWilson, K., Ibrahim, H., & Crismon, M. L. (2004). One-year clinical outcomes of depressed public sector outpatients: A benchmark for subsequent studies. Biological Psychiatry, 56, 46–53. doi:10.1016/j.biopsych.2004.04.005 Sapolsky, R. M., Romero, L. M., & Munck, A. U. (2000). How do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocrine Reviews, 21, 55–89. doi:10.1210/edrv.21.1.0389 Schmider, J., Lammers, C.-H., Gotthardt, U., Dettling, M., Holsboer, F., & Heuser, I.  J.  E. (1995). Combined dexamethasone/corticotropin-releasing hormone test in acute and remitted manic patients, in acute depression, and in normal controls: I.  Biological Psychiatry, 38, 797–802. doi:10.1016/0006-3223(95)00064-X Schmidt, P.  J., Martinez, P.  E., Nieman, L.  K., Koziol, D.  E., Thompson, K. D., Schenkel, L., . . . Rubinow, D. R. (2017). Premenstrual dysphoric disorder symptoms following ovarian suppression: Triggered by change in ovarian steroid levels but not continuous stable levels. American Journal of Psychiatry, 174, 980–989. doi:10.1176/appi.ajp.2017.16101113 Schmidt, P. J., Murphy, J. H., Haq, N., Danaceau, M. A., & St. Clair, L.  S. (2002). Basal plasma hormone levels in

Hormones and Major Depressive Disorder

depressed perimenopausal women. Psychoneuroendocrinology, 27, 907–920. doi:10.1016/S0306-4530(02)00004-5 Schmidt, P. J., Nieman, L. K., Danaceau, M. A., Adams, L. F., & Rubinow, D. R. (1998). Differential behavioural effects of gonadal steroids in women with and in those without premenstrual syndrome. New England Journal of Medicine, 338, 209–216. doi:10.1056/NEJM199801223380401 Segal, Z. V., Pearson, J. L., & Thase, M. E. (2003). Challenges in preventing relapse in major depression. Journal of Affective Disorders, 77, 97–108. doi:10.1016/s0165-0327(02)00112-x Seidman, S.  N. (2007). Androgens and the aging male. Psychopharmacology Bulletin, 40, 205–218. Shelton, R. C., Sanders-Bush, E., Manier, D. H., & Lewis, D. A. (2009). Elevated 5-HT 2A receptors in postmortem prefrontal cortex in major depression is associated with reduced activity of protein kinase A. Neuroscience, 158, 1406–1415. doi:10.1016/ j.neuroscience.2008.11.036 Shores, M. M., Sloan, K. L., Matsumoto, A. M., Moceri, V. M., Felker, B., & Kivlahan, D.  R. (2004). Increased incidence of  diagnosed depressive illness in hypogonadal older men. Archives of General Psychiatry, 61, 162–167. doi:10.1001/ archpsyc.61.2.162 Skinner, D. C., Albertson, A. J., Navratil, A., Smith, A., Mignot, M., Talbott, H., & Scanlan-Blake, N. (2009). Effects of gonadotrophin-releasing hormone outside the hypothalamicpituitary-reproductive axis. Journal of Neuroendocrinology, 21, 282–292. doi:10.1111/j.1365–2826.2009.01842.x Soares, C.  N., & Zitek, B. (2008). Reproductive hormone sensitivity and risk for depression across the female life cycle: A continuum of vulnerability? Journal of Psychiatry & Neuroscience, 33, 331–343. Spierling, S. R., & Zorrilla, E. P. (2017). Don’t stress about CRF: Assessing the translational failures of CRF₁ antagonists. Psychopharmacology, 234, 1467–1481. doi:10.1007/s00213017-4556-2 Spijker, A.  T., & van Rossum, E.  F. (2012). Glucocorticoid sensitivity in mood disorders. Neuroendocrinology, 95, 179–186. doi:10.1159/000329846 Stetler, C., & Miller, G. E. (2005). Blunted cortisol response to awakening in mild to moderate depression: Regulatory influ­ ences of sleep patterns and social contacts. Journal of Abnormal Psychology, 114, 697–705. doi:10.1037/0021-843x.114.4.697 Stetler, C., & Miller, G. E. (2011). Depression and hypothalamicpituitary-adrenal activation: A quantitative summary of four decades of research. Psychosomatic Medicine, 73, 114–126. doi:10.1097/PSY.0b013e31820ad12b Sumner, B. E., & Fink, G. (1998). Testosterone as well as estrogen increases serotonin 2A receptor mRNA and binding site densities in the male rat brain. Molecular Brain Research, 59, 205–214. Surampudi, P. N., Wang, C., & Swerdloff, R. (2012). Hypogonadism in the aging male diagnosis, potential benefits, and risks of testosterone replacement therapy. International Journal of Endocrinology, 2012, 625434. doi:10.1155/2012/625434 Susman, E.  J., Dorn, L.  D., Inoff-Germain, G., Nottelmann, E.  D., & Chrousos, G.  P. (1997). Cortisol reactivity, distress behavior, and behavioral and psychological problems in young adolescents: A longitudinal perspective. Journal of Research on Adolescence, 7, 81–105. doi:10.1207/s15327795jra0701_5 Triemstra, J. L., Sato, S. M., & Wood, R. I. (2008). Testosterone and nucleus accumbens dopamine in the male Syrian hamster. Psychoneuroendocrinology, 33, 386–394. doi:10.1016/j. psyneuen.2007.12.006

Ulrike, S., Reinhold, L., & Dirk, H. (2013). Major depression in young girls is related to altered cortisol awakening response. European Child and Adolescent Psychiatry, 22, 379–384. doi:10.1007/s00787-012-0371-9 Üstün, T. B., Ayuso-Mateos, J. L., Chatterji, S., Mathers, C., & Murray, C. J. L. (2004). Global burden of depressive disorders in the year 2000. British Journal of Psychiatry, 184, 386–392. Vammen, M.  A., Mikkelsen, S., Hansen, Å. M., Grynderup, M.  B., Andersen, J.  H., Bonde, J.  P., . . . Thomsen, J.  F. (2014). Salivary cortisol and depression in public sector employees: Cross-sectional and short term follow-up findings. Psychoneuroendocrinology, 41, 63–74. doi:10.1016/j. psyneuen.2013.12.006 van de Werken, M., Booij, S. H., van der Zwan, J. E., Simons, M.  J., Gordijn, M.  C., & Beersma, D.  G. (2014). The biological clock modulates the human cortisol response in a  multiplicative fashion. Chronobiology International, 31, 572–580. doi:10.3109/07420528.2013.868472 van den Bos, R., Harteveld, M., & Stoop, H. (2009). Stress and decision-making in humans: Performance is related to cortisol reactivity, albeit differently in men and women. Psychoneuroendocrinology, 34, 1449–1458. doi:10.1016/j. psyneuen.2009.04.016 Van der Kooy, K., van Hout, H., Marwijk, H., Marten, H., Stehouwer, C., & Beekman, A. (2007). Depression and the risk for cardiovascular diseases: Systematic review and meta analysis. International Journal of Geriatric Psychiatry, 22, 613–626. doi:10.1002/gps.1723 Vesga-Lopez, O., Blanco, C., Keyes, K., Olfson, M., Grant, B. F., & Hasin, D. S. (2008). Psychiatric disorders in pregnant and postpartum women in the United States. Archives of General Psychiatry, 65, 805–815. doi:10.1001/archpsyc.65.7.805 Viau, V. (2002). Functional cross-talk between the hypothalamicpituitary-gonadal and -adrenal axes. Journal of Neuroendo­ crinology, 14, 506–513. doi:10.1046/j.1365–2826.2002. 00798.x Viau, V., & Meaney, M.  J. (1996). The inhibitory effect of testosterone on hypothalamic-pituitary-adrenal responses to stress is mediated by the medial preoptic area. Journal of Neuroscience, 16, 1866–1876. Vreeburg, S. A., Hoogendijk, W. J. G., Pelt, J. V., DeRijk, R. H., Verhagen, J.  C.  M., van Dyck, R., . . . Penninx, B.  W.  J.  H. (2009). Major depressive disorder and hypothalamic-pituitaryadrenal axis activity: Results from a large cohort study. Archives of General Psychiatry, 66, 617–626. doi:10.1001/ archgenpsychiatry.2009.50 Vrshek-Schallhorn, S., Doane, L. D., Mineka, S., Zinbarg, R. E., Craske, M. G., & Adam, E. K. (2013). The cortisol awakening response predicts major depression: Predictive stability over a 4-year follow-up and effect of depression history. Psychological Medicine, 43, 483–493. doi:10.1017/S0033291712001213 Waters, C.  S., van Goozen, S., Phillips, R., Swift, N., Hurst, S.-L., Mundy, L., . . . Hay, D. F. (2013). Infants at familial risk for depression show a distinct pattern of cortisol response to experimental challenge. Journal of Affective Disorders, 150, 955–960. doi:10.1016/j.jad.2013.04.054 Waters, R.  P., Rivalan, M., Bangasser, D.  A., Deussing, J.  M., Ising, M., Wood, S. K., . . . Summers, C. H. (2015). Evidence for the role of corticotropin-releasing factor in major depressive disorder. Neuroscience and Biobehavioral Reviews, 58, 63–78. doi:10.1016/j.neubiorev.2015.07.011 Watson, S., Gallagher, P., Ritchie, J. C., Ferrier, I. N., & Young, A. H. (2004). Hypothalamic-pituitary-adrenal axis function

Ellenbogen, Tsekova, and Serravalle

403

in patients with bipolar disorder. British Journal of Psychiatry, 184, 496–502. Weissman, M. M., Bland, R. C., Joyce, P. R., Newman, S., Wells, J. E., & Wittchen, H. U. (1993). Sex differences in rates of depression: Cross-national perspectives. Journal of Affective Disorder, 29, 77–84. Westley, C. J., Amdur, R. L., & Irwig, M. S. (2015). High rates of depression and depressive symptoms among men referred for borderline testosterone levels. Journal of Sexual Medicine, 12, 1753–1760. doi:10.1111/jsm.12937 Wirtz, P.  H., Siegrist, J., Schuhmacher, A., Hoefels, S., Maier, W., & Zobel, A.  W. (2010). Higher overcommitment to work is associated with higher plasma cortisol but not ACTH responses in the combined dexamethasone/CRH test in apparently healthy men and women. Psychoneuroendocrinology, 35, 536–543. doi:10.1016/j.psyneuen.2009.09.002 Wittekind, D.  A., & Kluge, M. (2015). Ghrelin in psychiatric disorders—A review. Psychoneuroendocrinology, 52, 176–194. doi:10.1016/j.psyneuen.2014.11.013 Wong, M. L., Kling, M. A., Munson, P. J., Listwak, S., Licinio, J., Prolo, P., . . . Gold, P. W. (2000). Pronounced and sustained central hypernoradrenergic function in major depression with melancholic features: Relation to hypercortisolism and corticotropin-releasing hormone. Proceedings of the National Academy of Sciences of the United States of America, 97, 325–330. doi:10.1073/pnas.97.1.325 Woods, N.  F., Smith-DiJulio, K., Percival, D.  B., Tao, E.  Y., Mariella, A., & Mitchell, S. (2008). Depressed mood during the menopausal transition and early postmeno­ pause: Observations from the Seattle Midlife Women’s Health Study. Menopause, 15, 223–232. doi:10.1097/gme. 0b013e3181450fc2 Workman, J. L., Barha, C. K., & Galea, L. A. (2012). Endocrine substrates of cognitive and affective changes during pregnancy and postpartum. Behavioral Neuroscience, 126, 54–72. doi:10.1037/a0025538

404

Wyatt, K. M., Dimmock, P. W., Ismail, K. M., Jones, P. W., & O’Brien, P. M. (2004). The effectiveness of GnRHa with and without “add-back” therapy in treating premenstrual syndrome: A meta analysis. BJOG: An International Journal of Obstetrics & Gynaecology, 111, 585–593. doi:10.1111/ j.1471–0528.2004.00135.x Yonkers, K. A., Brown, C., Pearlstein, T. B., Foegh, M., SampsonLanders, C., & Rapkin, A. (2005). Efficacy of a new lowdose oral contraceptive with drospirenone in premenstrual dysphoric disorder. Obstetrics & Gynecology, 106, 492–501. doi:10.1097/01.AOG.0000175834.77215.2e Young, A.  H. (2004). Cortisol in mood disorders. Stress, 7, 205–208. doi:10.1080/10253890500069189 Zarrouf, F. A., Artz, S., Griffith, J., Sirbu, C., & Kommor, M. (2009). Testosterone and depression: Systematic review and meta-analysis. Journal of Psychiatric Practice, 15, 289–305. doi:10.1097/01.pra.0000358315.88931.fc Zobel, A. W., Nickel, T., Sonntag, A., Uhr, M., Holsboer, F., & Ising, M. (2001). Cortisol response in the combined dexamethasone/CRH test as predictor of relapse in patients with remitted depression: A prospective study. Journal of  Psychiatric Research, 35, 83–94. doi:10.1016/S00223956(01)00013-9 Zobel, A.  W., Yassouridis, A., Frieboes, R.-M., & Holsboer, F. (1999). Prediction of medium-term outcome by cortisol response to the combined dexamethasone-CRH test in patients with remitted depression. The American Journal of Psychiatry, 156, 949–951. doi:10.1176/ajp.156.6.949 Zorn, J. V., Schür, R. R., Boks, M. P., Kahn, R. S., Joëls, M., & Vinkers, C. H. (2017). Cortisol stress reactivity across psychiatric disorders: A systematic review and meta-analysis. Psychoneuro­ endocrinology, 77, 25–36. doi:10.1016/j.psyneuen.2016.11.036 Zweifel, J. E., & O’Brien, W. H. (1997). A meta-analysis of the effect of hormone replacement therapy upon depressed mood. Psychoneuroendocrinology, 22, 189–212. doi:10.1016/ s0306-4530(96)00034-0

Hormones and Major Depressive Disorder

CH A PT E R

23

Sex Differences in Anxiety Disorders

Teresa A. Piggott, Alexandra N. Duran, Isha Jalnapurkar, Tyler Kimm, Stephanie Linscheid, and Melissa K. Allen

Abstract Women are more likely than men to meet lifetime criteria for an anxiety disorder. Moreover, anxiety is a risk factor for the development of other psychiatric conditions, including major depression. Numerous studies have identified evidence of sex differences in anxiety disorders, and there is considerable research concerning factors that may contribute to vulnerability for anxiety in females. In addition to psychosocial influences, biological components such as the female reproductive hormone cycle have also been implicated. Although psychotropic medication is more likely to be prescribed to women, there is little controlled data available concerning sex differences in the efficacy and/or tolerability of pharmacotherapy in anxiety disorders. This chapter provides an overview of the impact of gender in the epidemiology, phenomenology, course, and treatment response in generalized anxiety disorder (GAD), social anxiety disorder (SAD), posttraumatic stress disorder (PTSD), panic disorder (PD), and obsessive-compulsive disorder (OCD). Keywords:  anxiety disorders, sex differences, disorders in women, generalized anxiety disorder, social anxiety disorder, posttraumatic stress disorder, panic disorder, obsessive-compulsive disorder

Nearly one-fourth of adults in the United States will meet criteria for an anxiety disorder during their lifetime, with females having a greater risk than males (Kessler & Wang,  2008). In the National Institute of Mental Health (NIMH) Collaborative Psychiatric Epidemiology Studies (CPES) survey of over 20,000 U.S. adults, 33 percent of women met lifetime criteria for an anxiety disorder compared to 22 percent of men, suggesting that women were 1.7 times more likely to have an anxiety disorder than men (McLean, Asnaani, Litz, & Hofmann, 2011). An earlier population survey conducted in the United States, the National Comorbidity Survey (NCS), also found that women (30.5 percent) were much more likely than men (19.2 percent) to meet lifetime criteria for an anxiety disorder (Kessler et al., 1994). The NCS also reported that women were also more likely than men to meet criteria for panic disorder (PD), agoraphobia (AG), specific phobia

(SP), social anxiety disorder (SAD), and posttraumatic stress disorder (PTSD) during their lifetime. Anxiety disorders typically start in childhood, adolescence, or early adulthood until they reach a peak in middle age, and then tend to decrease in older age. More females than males meet criteria for an anxiety disorder at every age throughout the lifespan, although there is a narrowing in the prevalence rate between genders after the age of 65. The cumulative effects of anxiety-related mortality, difficulty differentiating between cognitive impairment and an anxiety disorder, and the impact of female reproductive hormone cycle cessation have been implicated in this finding (Krasucki, Howard, & Mann, 1998). Several studies have reported a similar age of anxiety disorder onset in females and males, in general, as well as for  specific anxiety disorder (Lewinsohn, Gotlib, Lewinsohn, Seeley, & Allen, 1998; McLean et al., 2011). However, others have d ­ etected a later onset of 405

PD, GAD, and OCD in females than in males (Clayton, Stewart, Fayyad, & Clary, 2006; Lochner et al.,  2004; Ruscio, Stein, Chiu, & Kessler,  2010; Yonkers, Bruce, Dyck, & Keller, 2003). The presence of an anxiety disorder may have important long-term consequences. Adolescents with an anxiety disorder are more likely to experience pregnancy and/or parenthood than those without an anxiety disorder (Kessler et al., 1997). Adults with anxiety disorders are reported to have increased functional impairment, reduced educational and occupational opportunities, and elevated morbidity and mortality rates in comparison to those without an anxiety disorder (Katerndahl & Realini, 1997; Kessler & Walters, 1998; Leon, Portera, & Weissman, 1995; Schneier, Johnson, Hornig, Liebowitz, & Weissman, 1992). An anxiety disorder diagnosis has been linked to increased emergency medical and mental health utilization rates (Wittchen, Zhao, Kessler, & Eaton, 1994). The NIMH CPES survey also indicated that the illness burden associated with anxiety disorders was greater in females than in males (McLean et al., 2011). Comorbid psychiatric disorders are common in anxiety disorders. The NCS reported that two-thirds of those meeting criteria for an anxiety disorder would also be expected to meet criteria for a mood disorder during their lifetime, particularly major depressive disorder (Kessler et al., 2007; Kessler & Walters, 1998; Regier et al., 1988). Women with an anxiety disorder are more likely to meet criteria for an additional lifetime anxiety disorder and are also more likely to be diagnosed with lifetime major depressive disorder (MDD) or bulimia nervosa (BN), whereas men with an anxiety disorder were more likely to be diagnosed with a lifetime substance use disorder, attention-deficit/hyperactivity disorder (ADHD), or intermittent explosive disorder in the CPS (McLean et al., 2011). Sex may also have an impact on treatment response in anxiety disorders. The hepatic P450 system is critical to the metabolism of the most commonly prescribed medications for anxiety disorders, including antidepressants and benzodiazepines. Given that sex-related phenotypic differences have been identified within the P450 isoenzyme system, differences in psychotropic plasma concentrations at the same dose of medication would be expected to occur; this variance may have a substantial impact on efficacy and tolerability when medications are prescribed in the treatment of anxiety disorders. In addition, the female hormone progesterone is associated with reduced gastric acid production, which 406

may have a substantial impact on the bioavailability and absorption of psychotropic medication. Despite these pharmacokinetic differences, there is limited data concerning sex differences in the efficacy and tolerability of pharmacotherapy in the treatment of anxiety disorders (Howell, Brawman-Mintzer, Monnier, & Yonkers, 2001; Ronfeld, Tremaine, & Wilner, 1997; Steiner et al., 2005). Sex differences reported in anxiety disorders are likely to arise from numerous factors, including genetic, neurodevelopmental, environmental, and neurobiological influences. There appears to be genetic heterogeneity within anxiety disorders such that the relative contribution of hereditary factors may vary between individual anxiety disorders (Kendler, Neale, Kessler, Heath, & Eaves, 1992b; Kendler et al., 1995; Kessler & Walters, 1998). Although more than 98 percent of the studies investigating brain structure and function pertinent to anxiety disorders such as fear conditioning and exposure have focused on male subjects, structural and functional differences have been reported between males and females. Moreover, the brain regions reported to have sex differences are those considered relevant to anxiety, including the prefrontal cortex, hippocampus, and extended amygdala complex (Lebron-Milad & Milad, 2012). Female reproductive function and related behaviors may also predispose women to develop anxiety disorders. Various attributes observed in females, including superior social cognition and nurturing capacity, may enhance childrearing behavior. Researchers have suggested that these same traits may convey more sensitivity to separation, rejection, and criticism, resulting in a greater risk for anxiety disorders (Altemus, Sarvaiya, & Epperson, 2014; Cyranowski, Frank, Young, & Shear, 2000; A.  H.  Taylor, 2000; Zahn-Waxler, Shirtcliff, & Marceau, 2008). Given their potent actions within the central nervous system (CNS), the female reproductive hormones, estrogen and progesterone, may also have a role in increasing vulnerability for developing anxiety disorders (Altshuler, Hendrick, & Cohen, 1998; Noshirvani, Kasvikis, Marks, Tsakiris, & Monteiro, 1991; Weiss, Baerg, Wisebord, & Temple, 1995; Yonkers & Ellison, 1996). In turn, estrogen and progesterone may provide critical CNS modulating effects that influence the presentation, course, and treatment response of anxiety disorders in women (McEwen & Parsons, 1982; M. V. Seeman, 1997; Shear, 1997; Stahl, 1997). The female reproductive cycle is characterized by periodic fluctuations in estrogen and progesterone

Sex Differences in Anxiet y Disorders

beginning at menarche, during the menstrual cycle, pregnancy, and the postpartum period, followed by the perimenopause and postmenopausal periods. These vacillations in reproductive hormones are greater in females than males. The female hormone fluctuations also precipitate changes within the hypothalamic-pituitary-adrenal (HPA) axis (Altemus et al., 2014). The effects of the broad changes in gonadal steroid- and glucocorticoid-responsive brain systems have been linked to the enhanced vulnerability for anxiety disorders in females, as well as to changes in anxiety symptom severity observed during puberty, pregnancy, lactation, and menopause (Altemus,  2006; Pigott,  1999). Indeed, Altemus (2006) has argued that the increased prevalence of anxiety disorders in females may be a recent development. That is, in evolutionary terms, the ability of gonadal hormones to suppress the HPA axis and the catecholamine stress response system during pregnancy and lactation may have protected females against the development of anxiety disorders. However, since present-day females spend less time pregnant or lactating between puberty and menopause, the defense strategy provided by gonadal hormones is diminished and pathological anxiety states have become more common. In contrast, the primary male reproductive hormone, testosterone, reduces stress responsiveness by suppressing HPA axis activity and as such reduces anxiety and enhances fear extinction in animal models (McHenry, Carrier, Hull, & Kabbaj, 2014). Although PTSD and OCD are no longer classified within anxiety disorders in the most recent edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V), there is limited research incorporating these changes. In contrast, there are substantial data available for anxiety disorders that included PTSD and OCD in combination with GAD, PD, and SAD. Therefore, this chapter will include OCD and PTSD in a review of the impact of sex on the epidemiology, phenomenology, course, and treatment response of anxiety disorders.

Generalized Anxiety Disorder Overview and Epidemiology

GAD has consistently been recognized as one of the most common anxiety disorders. Data from the NCS and NCS Replication (NCS-R) studies estimated the lifetime prevalence of GAD between 5 and 6 percent (Kessler, Chiu, Demler, & Walters, 2005; Kessler et al., 1994). There is a sex difference in lifetime prevalence rates, with women reported to have approximately two times greater risk than men

for GAD (Beesdo, Pine, Lieb, & Wittchen, 2010; Gum, King-Kallimanis, & Kohn, 2009; McLean et al., 2011). In primary care settings, GAD is the second most common psychiatric disorder after depression (Yonkers et al., 2003). In a 14-country World Health Organization (WHO)-sponsored study conducted in primary care settings, 8 percent of patients met criteria for GAD in the past month. Even in the absence of comorbid conditions, GAD is associated with deficits in social and role functioning, general health, and bodily pain. The patients with GAD also had more disability days and a greater number of physician visits than patients without GAD (Kroenke, Spitzer, Williams, Monahan, & Löwe, 2007; Maier et  al., 2000), and GAD is associated with overuse of health care resources, elevated rates of medically unexplained symptoms, elevated rates of disability and social impairment, increased psychotropic medication use, and an increased risk of suicide compared to controls (Wittchen, Zhao, Kessler, & Eaton, 1994; Wittchen et al., 2000). Comorbid psychiatric disorders occur in up to 90 percent of patients with GAD. In the NCS, mood disorders were the most common lifetime comorbid disorder among patients with GAD, with unipolar depression (67 percent) four times more likely to occur than bipolar disorder (17 percent; Wittchen et al., 1994). Yonkers et al. (2003) found that 83 percent of patients with GAD had an additional psychiatric diagnosis, with PD (41 percent) and MDD (37 percent) the most common comorbid diagnoses. In patients with GAD seeking treatment, SAD and PD were the most common comorbid psychiatric disorders (Keller,  2002). Despite the availability of efficacious treatments, most patients with GAD remain untreated. For example, in 127 patients meeting criteria for GAD, the average interval between the onset of GAD and the initiation of the first adequate medication trial was 81.6 months (Dell’osso, Camuri, Benatti, Buoli, & Altamura, 2013). Selective serotonin reuptake inhibitor (SSRI) antidepressants are considered first-line pharmacotherapy for GAD. However, serotonin–norepinephrine reuptake inhibitor (SNRI) antidepressants (duloxetine and venlafaxine) are also highly effective pharmacotherapy for GAD. If effective, antidepressant treatment for GAD should be continued for at least 12 months (Gelenberg, 2000). Although benzodiazepines are effective anxiolytic agents for shortterm use, they should not be given over the long term because of the danger of addiction. Buspirone was effective for GAD in some trials. Psychotherapies,

Piggot t, Duran, Jalnapurkar, Kimm, Linscheid, and Allen

407

particularly cognitive-behavioral therapy (CBT), have also demonstrated significant efficacy and similar treatment effect sizes as those associated with SSRI and SNRI antidepressants in GAD (Hunot et al., 2007; Linden et al., 2005). The range of response rates in GAD is 47 to 75 percent with CBT and 44 to 81 percent for pharmacotherapy (Bandelow et al., 2013). Although data are limited, combination (medication plus CBT) treatment in GAD has not demonstrated improved efficacy over pharmacotherapy or CBT alone (Bond, Wingrove, Curran, & Lader, 2002; Crits-Christoph et al., 2011).

Sex Differences in Generalized Anxiety Disorder

As previously noted, women are at least twice as likely as men to meet criteria for lifetime GAD (Beesdo et al.,  2010; Gum et al.,  2009; McLean et al., 2011). Women with GAD may endorse more somatic complaints, such as fatigue and muscle tension, than men with GAD (Vesga-López et al., 2008). This increase in somatic complaints may reflect the influence of social and sex-specific factors, as women may be more prone to internalizing disorders than men (McLean et al.,  2011). Since traits such as negative affect and neuroticism have been identified as risk factors in the development of anxiety in general, as well as GAD in particular, some authors have suggested that this may contribute to the increased prevalence of GAD in females (Clark, Watson, & Mineka,  1994). Psychosocial and/or cultural issues may also have an important influence on sex differences in GAD prevalence. In the Collaborative Study on Psychosocial Problems in General Healthcare (Gater et al.,  1998), rates of psychiatric illness were examined in 26,969 patients attending 15 different primary care centers across four continents. Although the odds ratio in prevalence rates was higher in women than in men for current depression (1.60) and agoraphobia or PD (1.63), there was substantial variance in the sex prevalence rates associated with GAD. In fact, three distinct groups of centers were identified with odds ratios of 0.46, 1.34, and 3.09. Most population surveys have failed to detect a sex difference in age of onset or in the clinical course or chronicity of GAD (Beesdo et al., 2010; Kessler et al.,  2005; McLean et al.,  2011; Vesga-López et al., 2008). In contrast, data from clinical samples indicate sex differences in the onset and course of GAD. The Harvard/Brown Anxiety Research Program (HARP) study found an earlier onset of GAD in females than males (Yonkers et al., 2003) 408

and also demonstrated that women with GAD were less likely to achieve remission than men with GAD during the seven-year study. Moreover, remission occurred later in females than males with GAD (Yonkers et al., 2003). A more recent study conducted on a sample of primary care patients also found that men were more likely than women to achieve a partial recovery from GAD (Rodriguez et al., 2006). Sex differences in the comorbid conditions present in GAD have also been reported. The 2001–2002 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) study was a cross-sectional survey of over 43,000 participants in the United States (Vesga-López et al., 2008). Men in comparison to women meeting criteria for GAD had higher rates of comorbid alcohol and drug use disorders, nicotine ­dependence, and antisocial personality disorder, but lower rates of comorbid mood disorders (except bipolar disorder) and anxiety disorders (except SAD) in the NESARC. The men also reported greater use of alcohol and drugs to relieve GAD symptoms, whereas women with GAD were more likely to report a family history of depression and greater levels of disability. Although few of the participants in the NESARC sought treatment for GAD, men were even less likely than women to pursue treatment (Vesga-López et al., 2008). Comorbid depression in GAD has been associated with increased functional impairment and greater risk of suicide. Since women with GAD are more likely to have comorbid depression, this may contribute to the finding that women have a more chronic course of GAD and greater symptom severity. Data from bivariate female twin pairs with GAD suggest that genetic factors accounted for approximately 30 percent of the risk of GAD development, with environmental factors explaining the remainder of the variance (Kendler, Neale, Kessler, Heath, & Eaves, 1992a; Kendler et al., 1995). In addition to genetic factors, the female reproductive cycle also appears to have an impact on the course of GAD. Mixed results have been reported concerning GAD symptoms and the menstrual cycle. An early prospective study failed to detect any difference in GAD symptom severity across the menstrual cycle (McLeod, Hoehn-Saric, Foster, & Hipsley, 1993), but in a more recent study over half (52 percent) of the women with GAD reported premenstrual worsening (Hsaio, Hsaio & Liu, 2004). Several studies suggest that the risk of GAD is increased during pregnancy, with prevalence rates during pregnancy (8.5 to 10.8 percent) exceeding those reported for nonpregnant women (Adewuya,

Sex Differences in Anxiet y Disorders

Ola, Aloba, & Mapayi, 2006; Buist, Gotman, & Yonkers, 2011; Misri, Abizadeh, Sanders, & Swift, 2015). GAD symptoms were more pronounced during the first and third trimesters in two studies that examined the course of GAD across pregnancy (Lee et al., 2007; Teixeira, Figueiredo, Conde, Pacheco, & Costa, 2009). In a large prospective study conducted in pregnant women with a history of MDD (n = 2,793), 9.5 percent met criteria for GAD during their pregnancy and GAD symptoms were most severe during the first trimester and improved across pregnancy. Several risk factors were associated with an increased risk for GAD during pregnancy, including a previous history of GAD, a lower education level and support, and a history of childhood abuse in the same report (Buist, Gotman, & Yonkers,  2011). Perinatal loss (stillbirth after 20 weeks of gestational age or infant death in the first month) is also associated with risk of GAD (Gold, Boggs, Muzik, & Sen, 2014). GAD prevalence appears more variable in the postpartum period (ranging from 4.4 to 10.8 percent; Misri et al., 2015; Navarro et al., 2008; Phillips, Sharpe, & Matthey, 2007; Wenzel, Haugen, Jackson, & Robinson, 2003). Women diagnosed with GAD 10 weeks after childbirth were more likely to report sexual fear, avoidance, and body image self-consciousness compared to postpartum controls (Blair, Glynn, Sandman, & Davis, 2011). Women with GAD may also be prone to developing depression in the postpartum period (e.g., Wenzel et al., 2003). In a study of perinatal women with GAD (Grigoriadis et al., 2011), 50 percent had comorbid MDD and comorbid anxiety disorders were common, including specific phobias (20 p ­ ercent), PD (10 percent), agoraphobia (9 percent), and OCD (4 percent). The presence of GAD in pregnancy was also associated with lower levels of fetal brainderived neurotrophic factor in one study, raising concerns about a negative impact on fetal neurodevelopment (Uguz et al., 2013), although further data are lacking. There was no evidence of sex differences in GAD treatment response in a double-blind, placebo-­ controlled trial of sertraline (Steiner et al., 2005), but females with GAD were less likely to respond than males in another much smaller SSRI trial (Simon et al., 2006). Unfortunately, there do not appear to be additional reports that have specifically addressed this issue. Results from a survey completed by more than 60,000 patients seen in general practice settings revealed that women in comparison to men were twice as likely to receive

a  first prescription and repeat prescriptions for ­benzodiazepines for anxiety (Van Der Waals, Mohrs, & Foets, 1993). Although benzodiazepines are not considered first-line pharmacotherapy for GAD or any of the other specific anxiety disorders, this suggests that women may be less likely than men to be prescribed first-line pharmacotherapy (antidepressant medication) for an anxiety disorder including GAD. The previous findings indicate that sex differences exist in the prevalence, clinical features, and comorbid conditions that may complicate GAD. The emerging picture is that GAD is more common in women than men and is also more likely to be chronic, complicated by comorbid psychiatric disorders, and associated with more functional impairment. Yet, women with GAD may also be less likely than men to receive first-line treatment with antidepressant medication. Pregnancy has been associated with an increased risk for GAD, and women with pre-existing GAD appear to be at risk for developing depression and additional anxiety disorders during the postpartum period. Little is known about sex differences in treatment response in GAD.

Social Anxiety Disorder Overview and Epidemiology

SAD is one of the most common psychiatric disorders, with lifetime prevalence estimates exceeding 13 percent (Boyd et al.,  1990; Kessler et al.,  1994; Schneier, Johnson, Hornig, Liebowitz, & Weissman, 1992). Females are more likely than males to develop SAD during their lifetime, with odds ratios between 1.2 and 1.5 (Kessler et al., 2012). Public speaking– related anxiety appears to be the most commonly feared situation in SAD. Other anxiety-provoking circumstances include small-group activities, meeting strangers, eating in public, cashing checks in public, and using public restrooms. The onset of SAD is usually in adolescence or early adulthood, often before the age of 18; onset after the age of 25 is uncommon. Younger age (18 to 29 years), lower socioeconomic status, lack of social support, and single marital status are linked to an increased risk for SAD (Boyd et al., 1990; Schneier et al., 1992). SAD follows a chronic and unremitting clinical course. The presence of a comorbid alcohol use disorder and an earlier age of onset have been linked to a more chronic clinical course in SAD (Yonkers et al., 2003). SAD has been associated with long-term functional consequences, including lower education attainment, increased workplace impairment, and an increased reliance

Piggot t, Duran, Jalnapurkar, Kimm, Linscheid, and Allen

409

on welfare, and these effects appear independent of the effects of depression (Abelson et al., 2004). In the DSM-IV, SAD was classified into a generalized and a nongeneralized subtype. Epidemiological data suggested that two-thirds of those meeting criteria for SAD were in the generalized rather than the nongeneralized subtype (Kessler & Walters, 1998; Wittchen, Stein, & Kessler, 1999). However, in the DSM-V, the SAD subtypes were replaced by a specifier for a “performance-only type” of SAD. This change was predicated on the assumption that the SAD subtypes likely represented symptom severity more than distinct groups defined by shared genetic, neurobiological, or other features. The performance-only specifier is limited to performance fears that are most impairing in professional settings or in roles that require regular public speaking, although they may also manifest in work, school, or academic settings. Individuals with the performanceonly type of SAD do not fear or avoid nonperformance social situations. Using the DSM-V criteria, the National Survey of Mental Health and Well-Being (NSMHWB) conducted on over 8,800 adults in Australia reported an overall lifetime prevalence rate of 8.4 percent for SAD, with 0.3 percent meeting criteria for the performance-only specifier (Crome et al., 2015). SAD was also characterized by higher prevalence rates in females, frequent (70 percent) comorbid conditions, and relatively low rates of treatment seeking (20 percent). Due to the limited research reported using DSM-V criteria, this review will include data derived from studies using DSM-IV criteria, including the generalized and nongeneralized subtypes of SAD. The estimated lifetime risk for comorbid psychiatric disorders in SAD is 60 to 80 percent in most reports (Kessler et al., 1994; Kessler & Walters, 1998; Merikangas & Angst, 1995; Schneier et al., 1992). Lifetime mood disorders are common in SAD, with NCS data suggesting an increased risk of MDD, dysthymia, and bipolar disorder; in addition, the course of mood disorders appears to be more severe and chronic when comorbid with SAD (Kessler, Stang, Wittchen, Stein, & Walters, 1999). Patients with SAD and comorbid disorders also have greater clinical severity and greater treatment utilization than those with SAD alone (Merikangas & Angst, 1995). Although an increased risk of suicide attempts has also been associated with SAD, this appears to be largely attributable to other comorbid conditions (Schneier et al., 1992). Neuroimaging studies have implicated abnormalities in serotonin and also in pathways regulating 410

social reward versus punishment in the development of SAD. In a recent study, subjects with SAD in comparison to controls had greater rates of serotonin synthesis in the amygdala, a key site for fear regulation (Frick et al.,  2016). Moreover, symptom improvement was associated with a subsequent reduction in amygdala serotonin synthesis rates. These results suggest that enhanced serotonergic tone in the amygdala may exert an anxiogenic influence and that effective treatment in SAD may be mediated by decreasing serotonin formation in fear management pathways. The complex interplay between social reward and punishment is thought to be mediated through neural pathways within the striatum. Another functional neuroimaging study revealed that subjects with SAD had reduced striatal activation for reward versus punishment trials compared to the control group (Cremers et al., 2015). The authors postulated that SAD may be characterized by an attenuated preference for the anticipation of social reward rather than punishment. SSRI and SNRI antidepressants are considered first-line pharmacotherapy in SAD (D.  J.  Stein, Ipser, & Balkom, 2004). The clinical effects of SSRI treatment for SAD typically require four to six weeks to have an impact; maximal benefit can require as long as 16 weeks. Although less well studied than the SSRIs, the SNRI venlafaxine extendedrelease appeared equally effective for SAD on the basis of a comparable effect size compared with various SSRIs in meta-analysis (Liebowitz, Gelenberg, & Munjack, 2005; Liebowitz, Mangano, Bradwejn, & Asnis,  2005). Although there is no consensus concerning the optimal length of treatment in SAD, most guidelines suggest at least six months of pharmacotherapy. CBT is also a well-established firstline therapy in SAD that may be a helpful adjunct in nonresponders to pharmacological treatments (Blanco, Bragdon, Schneier, & Liebowitz, 2013). Both group and individual psychotherapy are also established treatments for SAD. A recent meta-analysis of 36 randomized controlled trials revealed medium to large positive effects for cognitive-behavioral group therapies (CBGTs) in comparison to wait-list controlled trials in alleviating symptoms of SAD. No differences were detected in the direct comparisons of group or individual psychotherapy or pharmacotherapy for SAD (Barkowski et al., 2016).

Sex Differences in Social Anxiety Disorder

As previously noted, there is a slightly elevated lifetime risk of SAD for women in comparison to men. Some sex differences in clinical phenomenology

Sex Differences in Anxiet y Disorders

have also been reported. In a community sample of young adults, women meeting criteria for SAD were more likely to endorse feared situations related to eating or drinking in public, writing while someone was watching, talking to others, and participating in social events than men with SAD (Wittchen et al., 1999; see also Turk et al., 1998). Only two feared situations (urinating in public bathrooms and returning goods to a store) were reported more by males than females (Turk et al., 1998). Men with SAD may also be more likely to seek treatment for SAD encountered in dating situations (Hart, Turk, Heimberg, & Liebowitz, 1999). Results from female twin studies suggest that SAD results from the combined effect of a slightly stronger genetic influence and nonspecific environmental experiences (Kendler et al., 1992b). Family studies also provide support for the importance of genetic factors in the development of generalized SAD (Kendler et al., 1992a, 1995; Stein & Chavira, 1998; Stein, Jang, & Livesley, 1999). Females with SAD may be more likely than males with SAD to have comorbid psychiatric disorders, especially mood disorder. However, shared genetic vulnerability may contribute to this association between SAD and mood disorders in women (Hettema, Neale, Myers, Prescott, & Kendler, 2006; Xu et al., 2012). Of the anxiety disorders, SAD may be most impacted by social influences. Females with SAD may be prone to develop comorbid internalizing disorders such as mood disorders, whereas men with SAD may be more likely to develop “more socially acceptable” externalizing disorders such as substance use disorders (Landrine, Bardwell, & Dean, 1988). Data from the NESARC survey conducted in the United States provide additional support for this idea. Women meeting criteria for SAD not only reported a greater number of social fears but also were more likely to have comorbid mood disorders and were more likely to seek treatment with medication than men with SAD. In contrast, men meeting criteria for SAD were more likely to report dating-related fears, have comorbid substance use disorders, and use alcohol and illicit drugs for symptom relief (Xu et al., 2012). Indeed, comorbid alcohol use disorders are common in SAD, and males are more likely than females to report greater use of alcohol and illicit drugs to relieve SAD symptoms (Buckner, Ledley, Heimberg, & Schmidt, 2008; Randall, Thomas, & Thevos, 2001). Although data from an epidemiologic catchment area health survey found that being male, being younger, feeling stigmatized, and being impulsive

were predictors of substance dependence, it also revealed that females meeting criteria for substance dependence were more likely to have comorbid SAD than males with substance dependence (Fleury, Grenier, Bamvita, Perreault, & Caron, 2014). However, another community survey of adolescents found that girls meeting criteria for SAD were less likely to use drugs (Wu et al., 2010). Childhood trauma has been linked to an increased risk of SAD. In the NCS, childhood sexual assaults by a relative and chronic exposure to verbal outbursts between parents were linked to the onset of SAD in females, but not males. In fact, there was no link between childhood adverse experiences and onset of SAD in males (Magee,  1999). Findings from the NCS-R survey among adolescent girls further suggest a unique interplay between early exposure to trauma, onset of menarche, and risk of SAD. That is, trauma during puberty in comparison to other developmental periods conferred a greater risk of an anxiety disorder diagnosis (primarily SAD) within two years after menarche. In contrast, trauma occurring in grade school conferred a greater risk for a depressive disorder diagnosis in the teenage girls. Early onset of menarche predicted a lifetime diagnosis of SAD, PTSD, and simple phobia, whereas late-onset menarche was associated with an increased rate of SAD. The authors suggested that menarche may amplify social sensitivity in females, making them vulnerable to the effect of trauma during puberty (Weingarden & Renshaw, 2012). No sex differences have been consistently reported in the clinical course or outcome of SAD. After 65 weeks of observation as part of the HARP study, clinical course and outcome were similar between the male and female patients with SAD (Reich, Goldenberg, Goisman, Vasile, & Keller, 1994). Remission rates were low in both sexes at an eight-year follow-up. Women were more likely to have concurrent agoraphobia, whereas comorbid substance use disorders were more common in men. Women with poor baseline functioning and a history of suicide attempts did have a more chronic course than men with the same characteristics in the HARP study (Yonkers et al., 2003). Most studies have reported a similar treatment response in men and women with SAD, including a pooled analysis of data from the three multicenter, placebo-controlled paroxetine trials (Stein, Stein, Pitts, Kumar & Hunter,  2002). Women with SAD report more distress related to family and social functioning than men (Randall et al., 2001), although men may be more likely to seek treatment (Weinstock, 1998).

Piggot t, Duran, Jalnapurkar, Kimm, Linscheid, and Allen

411

There may be some differences in brain structure between men and women with SAD. In a three-­ dimensional structural magnetic resonance imaging study, patients with SAD had reduced amygdala (13 percent) and hippocampal (8 percent) size in comparison to controls. Further analysis revealed that the reduction in amygdala size was only statistically significant in the men with SAD. Smaller right-sided hippocampal volumes in the patients with SAD were also related to greater severity of SAD (Irle et al., 2010). As summarized by Goodman, Chenausky, and Freeman (2014), prevalence rates for SAD reported during pregnancy have been variable, but most have ranged from 2 to 4 percent, including those conducted in the United States (3 percent, n = 453), Italy (3.8 percent, n = 1,066; 4.1 percent, n = 590), Brazil (4.6 percent, n = 239), Sweden (2.7 percent, n = 453), and France (3.2 percent, n = 309). Lower rates were reported in Sweden (0.4 percent, n = 1,556) and Malaysia (0.6 percent, n = 175), whereas the highest rate for SAD in pregnancy was found in Nigeria (I6.4 percent, n = 172). Three studies compared rates of SAD in pregnant versus nonpregnant controls. Adewuya and colleagues (2006) reported a  greater prevalence rate for SAD in pregnant (6.4 percent) than in nonpregnant Nigerian females (2.8 percent). In contrast, rates of SAD were similar in pregnant (3 percent) compared to nonpregnant females (2.8 percent) in studies conducted in the United States (Vesga-López et al., 2008) and Turkey (3.2 percent during pregnancy versus 2.8 percent in nonpregnant controls; Uguz et al., 2013). In a retrospective analysis of the course of SAD across pregnancy, most (59 percent) patients reported no change in SAD symptoms across pregnancy. However, in the subgroup that reported symptom change, pregnancy was associated with an improvement in SAD symptoms, whereas the postpartum period was associated with worsening resulting in a return to prepregnancy levels of severity (Van Veen, Jonker, Van Vliet, & Zitman, 2009). The presence of SAD during pregnancy may also be associated with an increased risk for postpartum depression (Coelho, Murray, Royal-Lawson, & Cooper,  2011; Mauri et al., 2010). These findings indicate that sex differences exist in the prevalence, clinical features, and comorbid conditions that may complicate SAD. Women are more likely than men to meet lifetime criteria for SAD, and they also have a greater number of social anxiety–related fears and greater overall symptom severity. Women with SAD may also have an increased 412

risk of comorbid agoraphobia, whereas men with SAD may be more likely to have coexisting substance use disorders. Consistent sex differences in the clinical course or treatment response in patients with SAD have not been reported. There is some evidence suggesting sex differences in brain structure in SAD. There are also data suggesting that childhood stress or trauma may increase risk for the later emergence of SAD, the timing of which may be particularly important in females, and that SAD symptoms may improve during pregnancy, but that the postpartum period is associated with a return to prepregnancy SAD symptom severity.

Posttraumatic Stress Disorder Overview and Epidemiology

Although PTSD is the most common psychopathology developed in response to a traumatic event (Galovski, Blain, Mott, Elwood, & Houle, 2012), the risk of PTSD after trauma may only be ~25 percent (Kessler et al.,  1994; Kessler, Sonnega, Bromet, Hughes, & Nelson,  1995). In the NCS, the lifetime prevalence rate for PTSD was estimated at approximately 8 percent, with a twofold greater lifetime prevalence rate for PTSD in women (10.4 percent) than men (5.0 percent; Kessler et al., 1995). PTSD is also associated with a greater risk for a family history of psychiatric illness, parental poverty, child abuse, parental separation or divorce before the age of 10, and greater job instability (Karno, Golding, Sorenson, & Burnam, 1988; Weissman, 1998). Although PTSD can emerge at any age, certain traumatic events are more likely to result in PTSD than others. Interpersonal violence is the most likely trauma type to be associated with PTSD (Kessler, 2000). In fact, Foa (1997) reported that 95 percent of rape victims and 75 percent of those sustaining nonsexual assaults developed PTSD within two weeks of the assault. Data from community samples suggest that the sudden, unexpected death of a loved one is also likely to precipitate PTSD (Breslau et al., 1998). The most common traumas associated with PTSD in women are sexual assault, sexual molestation, and childhood physical abuse, whereas combat exposure is the most common trauma associated with PTSD in men (Breslau, Chilcoat, Kessler, Peterson, & Lucia,  1999; Breslau et al., 1998; Kendler et al., 1995). Numerous risk factors have been implicated in the emergence of PTSD after trauma exposure. An epidemiological cohort study (Maes, Delmeire, Mylle, & Altamura,  2001) conducted on victims trapped in a ballroom fire found that being female,

Sex Differences in Anxiet y Disorders

greater amounts of previous trauma, a history of simple phobia, the extent of trauma exposure, hospitalization for trauma-related injuries, and the presence of burns increased the odds of PTSD. In contrast, a sense of control during the trauma and consumption of or intoxication with alcohol decreased the odds of PTSD. Three of the factors (the amount of previous trauma, a history of simple phobia, and loss of control) independently predicted a greater risk for PTSD. These results suggest that the emergence of PTSD after trauma exposure is likely conveyed by an interaction with numerous temporal factors including pre- (e.g., sex), peri(e.g., alcohol consumption), and post- (e.g., traumarelated injuries/hospitalization) trauma variables, although these results also suggest that certain peritrauma factors (e.g., sense of control, alcohol consumption and/or intoxication) may protect against PTSD (Maes et al., 2001). Comorbid psychiatric conditions are common in PTSD, with epidemiological studies suggesting a 70 to 80 percent lifetime risk. Mood and substance use disorders, in general, and MDD and GAD, in particular, are common in PTSD. Individuals with PTSD are more than four times as likely as those without PTSD to have substance use disorder (Leeman et al., 2017). In the NCS, both PTSD and trauma exposure were factors associated with elevated lifetime risks for alcohol abuse (28.1 percent) and alcohol dependence (6.5 percent; Kessler et al., 1999). The presence of certain comorbid conditions may also increase the risk of PTSD. Pre-existing depression appears to convey an increased risk for exposure to traumatic events, as well as the development of PTSD once trauma occurs (Breslau et al., 1998). PTSD has also been linked to elevated rates of suicide attempts (Davidson, Hughes, Blazer, & George, 1991). Although the pathophysiology of PTSD is likely multifactorial, dysregulation of the glutamatergic, monoamine neurotransmitter (noradrenergic and serotonergic), and neuroendocrine pathways have been implicated (Nutt, 2000). The development of PTSD may be facilitated by an atypical biological response in the immediate aftermath of severe trauma exposure that may then predispose the individual to develop sustained neurobiological abnormalities and a maladaptive psychological state (Yehuda, McFarlane, & Shalev,  1998). PTSD has also been associated with abnormalities in brain structure and function (Brunello et al., 2001). SSRI antidepressants constitute first-line pharmacotherapy for PTSD (Ballenger et al., 2004). Duration of

a therapeutic trial of an SSRI should be a minimum of six to eight weeks before concluding that the medication has failed. In addition, it is common practice to start at a low dose but eventually push the dose to the high end of the therapeutic range (to the extent that this is tolerated by the patient) before concluding that a therapeutic trial has failed. Common PTSD symptoms, such as anxiety, insomnia, and an exaggerated startle response, often improve during SSRI treatment. Moreover, SSRI treatment may ameliorate the intrusive traumarelated recollections, feelings of emotional numbing, and avoidance behaviors (Davidson & Connor, 1999). Although there are fewer studies assessing the efficacy of SNRI than SSRI antidepressants in PTSD, two randomized trials found venlafaxine extended-release (ER) to be more effective in reducing PTSD symptoms than placebo (Davidson et al., 2006).

Sex Differences in Posttraumatic Stress Disorder

Although women have higher lifetime PTSD rates than men, the sex differences in prevalence do not appear to be due to differential trauma exposure rates. In fact, some studies have reported that men had greater rates of lifetime trauma exposure than women (Galovski et al., 2012; Kessler et al., 1995). Certain types of trauma are 2.30 to 2.49 times more likely to be associated with PTSD in females than males (Frans, Rimmö, Åberg, & Fredrikson, 2005; Norris, Kaniasty, Conrad, Inman, & Murphy, 2002; Rosenman,  2002; Stein, Walker, & Forde,  2000). Females may also have greater exposure to certain traumas, such as rape and childhood sexual abuse, than males. Women may be more likely to experience avoidance and numbing symptoms in response to assault than men (Breslau et al.,  1999). Women victimized by sexual assault have a high risk of development of PTSD. For example, three months posttrauma, female rape victims were found to be twice as likely to have PTSD as women victimized by nonsexual crimes (48 percent vs. 25 percent; Foa, 1997). Another report found that women were at an increased risk for PTSD following nonsexual assault (e.g., mugging), but not after nonviolent trauma (Stein et al., 2000). Moreover, women with histories of childhood abuse were reported to have a level of functional impairment commensurate with that of women with recent abuse (McCauley et al., 1997). Girls may be particularly vulnerable to the negative effects of childhood sexual abuse (Walker, Carey, Mohr, Stein, & Seedat,  2004),

Piggot t, Duran, Jalnapurkar, Kimm, Linscheid, and Allen

413

whereas boys may be more vulnerable to adverse effects from childhood neglect (Teicher et al., 2004). The age at which trauma exposure occurs may represent a particularly critical issue. For example, females have higher PTSD rates than males after childhood trauma exposure before age 15 in comparison with exposure after age 15 (Breslau, Davis, Andreski, Peterson & Schultz 1997; 1998; Kessler et al., 1995). Women victimized by domestic violence are more likely to develop anxiety symptoms as well as PTSD, whereas male victims of domestic violence are at greater risk of developing substance use disorders (Mason & O’Rinn, 2014). An elevated risk of depression and an increased number of physical and psychological health problems have also been reported in women exposed to ongoing domestic violence. The severity of domestic violence or the presence of injuries sustained was not predictive of the development of psychiatric symptoms or PTSD (Sutherland, Bybee, & Sullivan,  1998). Although this may seem counterintuitive, it may reflect the importance of “perceived threat” in the formation of PTSD. That is, a victim’s perception of danger or possibility of death during an assault or exposure to trauma may be more important for developing PTSD as compared to more objective assessments of life-threatening events. These findings suggest that sexual assault, childhood abuse, and individual assessment of threat during the occurrence of trauma may be the strongest predictors of subsequent PTSD development in women. They also confirm that childhood abuse has particularly devastating complications that are likely to persist into adulthood in women. Certain factors appear to increase the risk of development of PTSD and other complications in women, including (1) exposure to sexually related trauma or aggression, (2) occurrence of abuse or severe trauma during childhood or before the age of 15, and (3) perception within the victim that the traumatic event is life threatening or that escape is unlikely (Breslau et al., 1998, 1999; Kessler et al., 1995). Fullerton and colleagues (2001) examined the relationship between various factors (i.e., previous trauma, PTSD, MDD, anxiety disorder besides PTSD, passenger injury, and peritraumatic dissociation) and occurrence of acute PTSD in women and men after a serious motor vehicle accident. Women did not differ from men in meeting the overall re-experiencing criterion for a diagnosis of PTSD, but women were nearly five times as likely to meet the overall avoidance/numbing criterion and 414

almost four times as likely to meet the overall arousal criterion for PTSD. The sex differences noted in acute PTSD were not associated with previous trauma, PTSD, peritraumatic dissociation, MDD, or anxiety disorder not including PTSD, or with passenger injury. However, dissociative symptoms at the time of the accident were associated with a higher risk for acute PTSD in women than in men. Sex differences in peritraumatic dissociation may help explain differences in risk for PTSD and for some PTSD symptoms in women and men (Fullerton et al., 2001). Coexisting somatoform pain disorder is more common in women than in men with PTSD (Elklit & Christiansen,  2009). The development of conduct disorder may also increase the risk for PTSD, particularly in girls, by exposing youth to situations in which they are traumatized (Reebye, Moretti, Wiebe, & Lessard, 2000). The increased prevalence of pre-existing anxiety or MDD in women has also failed to account for the sex difference in the prevalence of PTSD. In the NCS, there was substantial overlap in factors that predicted an increased risk for trauma exposure and development of PTSD. Once the overlapping risk factors were excluded, only history of affective disorder predicted PTSD in women, whereas history of anxiety disorder and parental mental disorder were associated with an increased risk of PTSD in men (Bromet, Sonnega, & Kessler, 1998). These findings highlight the importance of differentiating between variables that are more predictive of trauma exposure than PTSD. Combat exposure is the most common trauma associated with PTSD in men, and male military veterans are diagnosed with PTSD at a much greater rate than female military veterans. This is attributed to higher rates of combat exposure in male than female veterans. However, there are limited data concerning the effects of combat-related exposure on women. In a retrospective chart review of military veterans, veterans exposed to higher levels of combat-related stress were more likely to have increased PTSD symptomatology (Pereira, 2002). In the same study, male and female veterans exposed to similar levels of combat-related stress were equally likely to have PTSD symptoms. These results suggest that female veterans may be underdiagnosed with combatrelated PTSD. Elevated rates of chronic medical illness may be one of the many adverse consequences that arise from failing to identify and diagnose PTSD in women military veterans. In a report of female Vietnam War–era veterans, there was a

Sex Differences in Anxiet y Disorders

relationship detected between PTSD symptoms and reported health problems in the women with previous trauma exposure (Kimerling, Clum, & Wolfe, 2000); in addition, the hyperarousal PTSD symptom cluster was most strongly associated with increased health complaints. These results suggest that women may have a similar risk for PTSD after combat-related trauma exposure as men and that persistent PTSD symptoms, particularly hyperarousal, may result in elevated rates of chronic health problems in female military veterans. However, a combination of factors is likely involved in the expression of PTSD (Hansen & Elklit, 2011). Sex differences have been identified in the biological alterations associated with PTSD. An elevated norepinephrine-to-cortisol ratio has been reported in men with PTSD, whereas women with PTSD demonstrated elevated daily levels of urinary norepinephrine, epinephrine, dopamine, and cortisol (Lemieux & Coe, 1995). Cortisol appears to modulate emotionally influenced memory, and sex differences in the relationship between stress-induced cortisol levels and memory have been reported (Wolf et al., 2001). Low cortisol levels may also be associated with specific dimensions of PTSD symptomatology, such as emotional numbing (Hawk, Dougall, Ursano, & Baum, 2000). Sex differences in reactivity to psychological stress within the HPA axis have also been reported (Biondi & Picardi, 1999; Kirschbaum, Kudielka, Gaab, Schommer, & Hellhammer, 1999; Kudielka & Kirschbaum, 2005; Matthews, Gump, & Owens, 2001; T. E. Seeman, Singer, Wilkinson, & McEwen, 2001; Traustadottir, Bosch, & Matt, 2003). Although women between puberty and menopause show lower HPA axis responses than men of the same age, HPA response is higher and poststress cortisol levels approach those of men during the luteal phase of the menstrual cycle. The HPA axis also has strong, multilevel inhibi­ tory effects on the female reproductive hormones. Because HPA axis alterations are implicated in PTSD, the marked fluctuations in estrogen and progesterone levels that characterize the female reproductive cycle may have a significant impact on the course of PTSD. The relative hypercortisolism that occurs during the third trimester of pregnancy is speculated to cause a transient suppression of the adrenals during the postpartum period (Chrousos, Torpy, & Gold, 1998).This suggests that women with pre-existing PTSD may experience changes in symptomatology during pregnancy or the postpartum period. There are limited data available concerning the prevalence, onset, course, risk factors, or course of

PTSD during pregnancy or the postpartum period. Pregnant women with PTSD have a higher risk for ectopic pregnancy, spontaneous abortion, hyperemesis, preterm contractions, and excessive fetal growth in comparison to women without PTSD (Seng et al., 2010). There is also evidence that women who experience increased symptoms of psychosocial stress after exposure to a natural disaster (flood) may be at an increased risk of pregnancy loss and perinatal complications (Neuberg et al., 1998). As summarized in Goodman et al. (2014), prevalence rates reported for PTSD during pregnancy have ranged from 0.6 to 7.9 percent in cross-cultural studies. Two of these studies compared rates of PTSD in pregnant women with nonpregnant controls. Nigerian women in late pregnancy were more likely to meet criteria for a DSM-IV anxiety disorder (39 percent) than a comparison group (16 percent) of nonpregnant women, but rates of PTSD were low and not different between the pregnant (0.6 percent) and the nonpregnant (0 percent) population (Adewuya et al., 2006). Another study (Seng et al., 2010) was conducted in 1,500 women pregnant with their first child that were interviewed at their first prenatal care visit. Twice as many of the pregnant women (7.9 percent) met criteria for PTSD than in the comparison group (3.9 percent). In another community sample of pregnant women with a lifetime PTSD diagnosis, certain factors (child abuse history, demographic risk, and lifetime PTSD symptom count) were associated with an increased risk of PTSD during pregnancy, whereas gestational PTSD symptoms were best predicted by interim trauma and labor anxiety (Muzik et al., 2016). Women with the greatest increase in PTSD symptoms during pregnancy were most likely to suffer postpartum depression and they also reported the most impaired bonding with their infants at the sixweek postpartum visit. There are limited data concerning sex differences in PTSD treatment response. There was some suggestion of a sex difference in the sertraline multicenter PTSD trials in that women appeared to have a greater degree of PTSD symptom reduction than men (Davidson, Rothbaum, van der Kolk, Sikes, & Farfel,  2001). However, there was no evidence of sex differences or differential response of PTSD symptom clusters reported in the paroxetine multicenter, placebo-controlled PTSD trial (Marshall, Beebe, Oldham, & Zaninelli,  2001). There also was no sex differences identified in the venlafaxine ER multicenter, placebo-controlled trials conducted in PTSD.

Piggot t, Duran, Jalnapurkar, Kimm, Linscheid, and Allen

415

In a Veterans Affairs (VA) study examining prescribing patterns for psychotropic medication (antidepressants, antipsychotics, and hypnotics) over an 11-year period (1999 to 2009) in veterans diagnosed with PTSD (Bernardy et al., 2013), women were more likely than men to receive medication across all classes except prazosin, whereas men had higher prescribing frequency. The proportion of women receiving first-line pharmacotherapy treatments for PTSD (SSRI or SNRI antidepressant) increased from 56 percent in 1999 to 66 percent in 2009, which was a higher rate than that seen in the men with PTSD (49 percent with increase to 58  percent). Atypical antipsychotic prescriptions increased from 15 to 26 percent overall, but were more likely to be prescribed in women than men with PTSD. However, the most notable sex discrepancy was observed for benzodiazepines, where prescriptions decreased for men with PTSD (from 37 to 30 percent) but steadily increased for women with PTSD (from 33 to 38 percent) despite PTSD guidelines that recommend against the use of benzodiazepines in PTSD (Bernardy et al., 2013). These results suggest that women are twice as likely to meet lifetime criteria for PTSD. Numerous factors including the type and time of exposure to trauma, as well as pre-existing features and sex-related neurobiological factors, likely contribute to this finding. Comorbid conditions are common with PTSD, and some sex differences have been identified. There is also evidence that PTSD prevalence may increase during pregnancy and may be associated with adverse consequences including an increased risk for postpartum depression and worse outcomes. Sex differences have also been identified in the biological alterations associated with PTSD. Most pharmacotherapy studies conducted in PTSD have failed to detect sex differences. However, women may be more likely than men to be prescribed benzodiazepines, even though their use is not ­consistent with treatment guidelines for PTSD.

Panic Disorder Overview and Epidemiology

PD is a condition wherein people experience multiple, recurrent panic attacks (PAs), with at least some of these attacks occurring at unexpected times. The overall lifetime prevalence rate of PD varies by country and by study (see, e.g., American Psychiatric Association,  2013; Bandelow & Michaelis,  2015), but estimates are generally below 5 percent. To meet DSM-V criteria for a PA, it must have an acute onset, peak within minutes, fulfill a set number of 416

somatic and psychological criteria, and also be ­associated with avoidance behavior and/or anticipatory anxiety (American Psychiatric Association, 2013). Prior to the DSM-V, agoraphobia was classified as a subordinate condition within the diagnosis of PD (Wittchen et al., 2010). However, DSM-V categorized PD and AG as separate disorders, each with its own distinct diagnostic criteria. Individuals with AG (from Greek for “fear of the marketplace”) experience anxiety symptoms related to real or imagined exposure to public places. The phobic focus associated with AG centers on the inability to escape and/or the negative outcomes that may occur if assistance is unavailable (American Psychiatric Association, 2013). The presence of PAs alone appears to increase the odds of developing phobias, GAD, OCD, or PTSD, with the estimated risk ranging from 3- to 16-fold in the literature. PAs have also been associated with an increased risk for mood disorders, in general, and MDD, in particular, as well as personality disorders and substance use disorders (Craske et al.,  2010). PD is associated with a chronic course, high rates of comorbid disorders, and significant disability. Highly comorbid with MDD, the presence of MDD conveys an increased risk for substance use disorders, as well as for additional anxiety disorders in PD (Wittchen & Essau, 1993). PD also has the highest rate of medical utilization among anxiety disorders. More than half of those with PD have been reported to seek treatment; this rate is twice that associated with other anxiety disorders (American Psychiatric Association, 2013; Regier et al., 1993). In contrast, AG in the absence of PD may have relatively less disability in comparison to PD, and while anxiety disorders may be more common, comorbid mood or substance use disorders appear not to be more common in AG alone (Wittchen, Gloster, Beesdo-Baum, Fava, & Craske,  2010). However, PD complicated by AG was associated with more severe PD, an elevated suicide risk, and higher levels of anxiety sensitivity, neuroticism, and trait anxiety than PD without AG. Comorbid hypomania and SAD were also more likely to occur in PD with AG (Inoue, Kaiya, Hara, & Okazaki, 2016). SSRI antidepressants are considered first-line pharmacotherapy for PD. A systematic review and meta-analysis of 12 trials of acute treatment for PD found the SSRIs to be efficacious compared to placebo, with a medium effect size (Otto, Tuby, Gould, McLean, & Pollack,  2001). There appears to be comparable efficacy among the SSRI in PDs (Stein et al.,  2009). Venlafaxine ER has also been

Sex Differences in Anxiet y Disorders

found in randomized trials to be an efficacious treatment for PD (Cowley, Ha, & Roy-Byrne, 1997; Pollack et al., 2007). Sixteen randomized trials (11 for imipramine and 5 for clomipramine) have also found tricyclic antidepressants (TCAs) to be superior to placebo in PD. Although TCAs reduce the frequency of panic attacks, their effects on anticipatory anxiety and phobic avoidance may be more variable and less robust than that seen with SSRIs and venlafaxine (Bakker, van Balkom, & Spinhoven, 2002). Numerous randomized trials have also reported that benzodiazepines were efficacious for each of the three components of PD (attack frequency, anticipatory anxiety, and phobic avoidance; Stein et al.,  2009), with effect sizes reported as similar to those seen for SSRIs or tricyclic antidepressants in meta-analysis (Mitte, 2005; Wilkinson, Balestrieri, Ruggeri, & Bellantuono, 1991). However, benzodiazepines have the potential for abuse and they also lack efficacy in the comorbid mood disorders that often complicate PD, so most PD guidelines suggest using venlafaxine in patients who fail to respond to one or more trials of an SSRI. The duration of pharmacotherapy should be at least one year after symptom control has been attained.

Sex Differences in Panic Disorder

Women are 2.5 to 3 times more likely than men to meet criteria for panic disorder during their lifetime (3.4 percent vs. 0.9 percent; Eaton et al., 1989). The lifetime prevalence estimate for panic disorder was also 2.5 times greater in women (5 percent) than in men (2 percent) in the NCS study (Kessler et al., 1994). Women were reported to have higher prevalence rates for panic disorder in the CrossNational Epidemiology of Panic Disorder study (Weissman et al., 1997). Women are more likely to endorse more individual panic-related symptoms than men and also to display more avoidance symptoms (Pigott, 1999). Results from a National Comorbidity Survey in respondents meeting DSM-III-R criteria for PD or PAs revealed that female respondents were more likely than males to experience respirationrelated difficulties during panic attacks. Men with PD displayed greater general fatigue symptoms and reduced activity in comparison to control subjects, whereas women with PD scored higher for physical fatigue but were similar to controls in general fatigue symptoms and activity levels in one report (Kaiya et al., 2008). Women with PD also appear more susceptible to comorbid conditions then men with PD including AG, MDD, GAD, and somatoform disorders

(­reviewed in Pigott, 2003). Sex differences have also been reported in patients with PD with AG (PDA). Women were more likely than men to have PDA and have a greater prevalence of comorbidities, including hypomanic episodes, social phobia, and a higher risk of suicide (Inoue et al., 2016). Women with PDA were more likely to avoid buses, enter ­unfamiliar situations alone, and avoid agoraphobic situations by staying home than males with PDA, whereas the men with PDA were more likely to avoid staying at home alone in another study (Starcevic et al.,  2008). Females with PDA appear more impaired and also display more severe agoraphobic avoidance than PDA males. Interestingly, PDA males were more likely to worry about their physical health and also about the potential for serious somatic consequences from having panic attacks than PDA females (Latas & Starcevic, 2008). No sex differences were detected in the age of onset, illness duration, panic attack severity or frequency, or severity of anxiety, depression, or general psychiatric symptoms in PDA samples (Latas & Starcevic, 2008; Starcevic, Djordjevic, Latas, & Bogojevic, 1998). Sex differences in parenting dynamics have also been reported in PD samples. Males with PD reported greater rates of “overprotective mothers” as measured by the Parental Bonding Instrument than control subjects, whereas the females with PD reported greater rates of fathers that were “overprotective” and/or had reduced “warmth” in comparison to control subjects (Parker, Tupling, & Brown, 1979). This combination of parental “overprotection” and “authoritarianism” has been implicated as a risk factor for childhood trauma and perhaps increased PD development by some authors (Seganfredo et al., 2009). The preponderance of respiratory-related symptoms in females in comparison to males with PD appears to arise from sex differences in CO2 sensitivity, as well as in the threshold for panic attacks during hypoxic and hypercapnic states (Sheikh, Leskin, & Klein, 2002). That is, when exposed to a CO2 concentration consistent with a panic state, healthy women and men had similar autonomic responses, but the women reported higher subjective fear than the men, suggesting that anxiety sensitivity (AS) is higher in women than in men (Stewart, Taylor, & Baker, 1997). Not only is AS considered a cognitive risk factor for PD, but also people with anxiety disorders have greater AS than those without anxiety disorders. AS has also been shown to be highly predictive of the frequency and intensity of PAs (Plehn & Peterson, 2002).

Piggot t, Duran, Jalnapurkar, Kimm, Linscheid, and Allen

417

The female reproductive hormones may play a critical role in mediating the sex differences reported in PD. For example, females appear to have a greater susceptibility to anxiety-inducing challenges during the late luteal or premenstrual phase of the menstrual cycle (Donner & Lowry, 2013; Nillni, Rohan, & Zvolensky, 2012). A major metabolite of progesterone, the steroid allopregnanolone, is a potent positive allosteric modulator of GABAA receptors, a subset of receptors that make up the predominant inhibitory neurotransmitter system in the brain (Nillni et al., 2012). One site rich in GABAA receptors, the periaqueductal gray (PAG), has been implicated in panic symptomatology as a subcortical panic-generating brain region (Johnson, Molosh, Truitt, Fitz, & Shekhar, 2012). In animal models, direct GABAA antagonist injections into the PAG elicit panic symptoms (Graeff, Silveira, Nogueira, Audi, & Oliveira,  1993). This panic circuitry is more excitable via decreased GABAergic inhibition in rats during the late diestrous phase, which is a state of low progesterone (and, thereby, a relative decrease in allopregnanolone). In contrast, the highest rates of anxiolytic and antidepressant behaviors in female pubertal rats as measured by behavioral paradigms occur in the proestrous phase of the estrous cycle, when allopregnanolone levels are at their peak (Nillni et al., 2012). The increase in PD prevalence in females at puberty has been linked to the decrease in, or relative withdrawal of, allopregnanolone in the hippocampus during that time (Lovick, 2014). Synaptic pruning also occurs at that time, and this loss of inhibition, from the ventromedial prefrontal cortex to subcortical areas, may also contribute to less attenuation of panic-related circuits (Johnson, Federici, & Shekhar, 2014). A few studies conducted in women with PD have reported an association between menstrual cycle phase and changes in PD symptom severity. In a retrospective study, 79 percent of women with PD reported an increase in anxiety premenstrually, with 58 percent reporting increased frequency of panic attacks and 47 percent recalling more phobic avoidance (Cook et al.,  1990). Women with PD who endorsed premenstrual worsening of their PD symptoms were also 2.5 times more likely to be classified as suicidal when compared to women without premenstrual worsening of their PD symptoms in another report (Basoglu, Cetin, Semiz, Agargun, & Ebrinc, 2000). Premenstrual worsening of symptoms was reported by 51 percent of women with AG in another study (Breier, Charney, & Heninger, 1986). 418

Although pregnancy and the postpartum period are also times of reproductive hormone changes, the course of PD during pregnancy appears to be highly variable. In a systematic review of 12 studies examining anxiety disorders during pregnancy, Goodman and colleagues (2014) found that prevalence rates for new-onset PD during pregnancy varied from 0 to 53.8 percent. The course of PD during pregnancy was also examined in nine studies mostly retrospective in nature. Four (44 percent) of the studies demonstrated an improvement in PD symptoms, four (44 percent) revealed minimal change in PD, and one (12 percent) reported worsening PD symptoms. There was a single study in their review that showed decreased infant birth weight for mothers with PD compared to mothers with GAD, mothers with depression, or controls without psychiatric disorders (Goodman et al.,  2014). There is also a report that the risk of cleft lip with or without cleft palate may be increased in infants born to mothers with PD (Acs, Banhidy, Horvath-Puho, & Czeizel, 2006). In contrast to the variable course of PD during pregnancy, the postpartum period appears to be more consistently associated with an increased risk of onset of PD and also an increased likelihood of PD exacerbation. In the largest sample (n = 128) reported, there was a large rise (a 132-fold increase) in the incidence of PD occurring during the postpartum period and there was also an increase in overall panic symptomatology during the postpartum period in comparison to pregnancy and the nonpregnancy period (Bandelow et al.,  2006). There was also evidence that women with PDA were more symptomatic than women with PD without AG in the same report. Although women with PD are reported to have a threefold higher rate of relapse compared to men with PD (Donner & Lowry, 2013), there are limited data regarding sex differences in the treatment of PD. Clayton, Stewart, Fayyad, and Clary (2006) analyzed four double-blind, placebo-controlled outpatient industry-sponsored trials of sertraline for the treatment of PD. The women with PD had an increased response rate compared to the men with PD in terms of improvement from baseline panic attack frequency and also in decreased time spent worrying about future attacks. In a post hoc analysis of a placebo-controlled trial of gabapentin in PD (n = 103), the women with severe PD show a greater response than men (Pande et al., 2000). These results suggest that women are at least twice as likely as men to meet lifetime criteria for PD.

Sex Differences in Anxiet y Disorders

Numerous factors including environmental and sex-related neurobiological factors likely contribute to this finding. Comorbid conditions are common with PD and AG, and some sex differences have been identified. There is also evidence that the PD symptoms may change during the menstrual cycle and during pregnancy and that the postpartum period may be associated with both an increased risk of onset of PD and worsening in PD symptoms. Sex differences have also been identified in the biological alterations associated with PD. Most pharmacotherapy studies conducted in PD have failed to detect sex differences.

Obsessive-Compulsive Disorder Overview and Epidemiology

Prevalence rates for OCD range between 2 and 3 percent worldwide, and women are about 1.5 times more likely than men to meet criteria for OCD during their lifetime (Karno et al.,  1988; Weissman,  1998), although some work reports more similar prevalence rates for OCD between men and women (Bekker & van Mens-Verhulst, 2007; McLean & Anderson, 2009). The onset of OCD is reported to be early with two peaks of onset, the first occurring between 10 and 19 years followed by another between 20 and 29 years (Kolada, Bland, & Newman, 1994). Most studies conducted on OCD have identified the most commonly endorsed recurrent thoughts (i.e., obsessions) as those related to aggression, contamination, symmetry, saving/collecting, sexual impulses, or religious matters. Checking, cleaning, counting, ordering, and hoarding rituals (i.e., compulsions) are the most frequent compulsive behaviors (Antony, Purdon, Huta, & Swinson, 1998; Baer, 1994; Rasmussen & Eisen, 1990, 1992). The most common comorbid disorders in OCD are MDD (60 to 80 percent lifetime risk), an ­additional anxiety disorder diagnosis (30 to 50 percent), eating disorders (10 to 20 percent), and Tourette’s disorder (8 to 15 percent; Antony et al., 1998; Fahy, Osacar, & Marks,  1993; Leckman et al., 1994; McDougle, Barr, Goodman, & Price, 1999; Noshirvani et al., 1991; Pigott, L’Heureux, Dubbert, Bernstein, & Murphy, 1994; Rasmussen & Eisen,  1990; Rubenstein, Pigott, L’Heureux, Hill, & Murphy,  1992). A lifetime diagnosis of OCD is also associated with an increased likelihood of substance abuse, schizophrenia, body dysmorphic disorder (BDD), hypochondriasis, and personality disorders (Antony et al., 1998; Hollander et al., 1996; Karno et al., 1988; Kolada et al., 1994; Pigott et al., 1994; Rasmussen & Eisen, 1992, 1994; Rubenstein

et al., 1992). Most OCD patients (60 to 70 percent) report severe impairment in psychosocial and occupational functioning (Hollander & Benzaquen, 1996; Koran, Thienemann, & Davenport,  1996). An OCD diagnosis is also associated with elevated use of medical and mental health services (Hollander et  al.,  1996; Kennedy & Schwab,  1997). In the largest nationwide survey of consumers with OCD (Hollander et al., 1997, the average delay between the onset of OCD symptoms (mean age = 14.5) and the initiation of appropriate treatment interventions (mean age = 31.5) was 17.5 years. This is a particularly disturbing finding since 13 percent reported a history of suicide attempts and most of the respondents felt their OCD was directly responsible for impaired academic achievement (58 percent), lowered career aspirations (66 percent), and disrupted family relationships (73 percent). Serotonin (5-HT) dysregulation has long been considered the critical neurobiological abnormality mediating OCD symptom development (Zohar & Insel,  1987). Much of this assumption has been based on the consistent finding that antidepressants with potent serotonin-reuptake-inhibiting (SRI) effects have preferential efficacy in the treatment of OCD (Pigott, 1996; Pigott & Seay, 1999; Zohar & Insel, 1987). Moreover, antidepressants with nonselective serotonergic or primary noradrenergic actions lack efficacy in OCD. Challenge studies conducted with serotonergic probes have also revealed evidence of altered behavioral or neuroendocrine responses in patients with OCD compared with controls (Aouizerate et al., 2005). In contrast, pharmacological challenge paradigms conducted in patients with OCD have failed to detect evidence of altered noradrenergic function compared with control subjects. OCD has also been linked to hypersensitivity of postsynaptic serotonin receptors (Gross, Sasson, Chorpa, & Zohar,  1998). Genetic association studies conducted in OCD have focused on attractive candidate genes in the 5-HT pathways (Enoch, Greenberg, Murphy, & Goldman, 2001). Inconsistent results have also been reported concerning the transmission of alleles associated with polymorphisms in the 5-HT1D β-receptor gene (Di Bella, Cavallini, & Bellodi, 2002; Mundo et al., 2002) and the 5-HT2A receptor gene (Enoch et al., 2001). There has also been excitement about findings concerning the promoter region of the 5-HT transporter gene (5-HTTLPR) in OCD (Cavallini, Di Bella, Siliprandi, Malchiodi, & Bellodi, 2002). However, more recent research (Saiz et al., 2008) failed to replicate these findings,

Piggot t, Duran, Jalnapurkar, Kimm, Linscheid, and Allen

419

including a meta-analysis (Bloch et  al., 2008). Although the meta-analysis found no evidence of an association between genetic variation at the 5-HTTLPR locus and OCD, a stratified metaanalysis revealed a potential association between the l-allele and specific OCD subgroups in the same report (Bloch et al.,  2008). In the largest metaanalysis (S. Taylor, 2013) investigating the relationship between OCD and 230 polymorphisms from 113 genetic association studies, OCD was associated with serotonin-related polymorphisms (5-HTTLPR and HTR2A). Neuroimaging studies have been influential in shaping neurobiological models of OCD (Atmaca, 2013). In particular, the cortico-striato-thalamocortical circuit has been implicated in the mediation of OCD symptoms (Breiter & Rauch, 1996; Pauls, Abramovitch, Rauch, & Geller,  2014; Radua & Mataix-Cols,  2009). In fact, numerous functional neuroimaging studies have demonstrated increased metabolism in the frontal cortex, anterior cingulate, caudate nucleus, and thalamus during symptom provocation in OCD patient studies (Breiter & Rauch, 1996; Rauch et al., 1994). These key brain regions are likely relevant to the pathophysiology of OCD (Whiteside, Port, & Abramowitz, 2004). In addition, the amygdala is also implicated in OCD due to its important role in fear conditioning, and it is also an important target of the SRIs (Atmaca, 2013). From an evolutionary standpoint, the same key brain areas implicated in OCD are activated during the processing of danger and threat (Steimer, 2002). Thus, although OCD compulsions are excessive and often disabling for the individual, they are based on behaviors that may have increased ancestral survival and reproduction (see Gonda et al., 2008). The SRIs (SSRI antidepressants and the TCA clomipramine) remain the cornerstone of the pharmacological treatment for OCD. However, due to their favorable risk profile, preference should be given to the SSRIs (Baldwin et al., 2014; Bandelow, 2008; Koran & Simpson, 2013). SSRIs are more effective than placebo in achieving clinical response in reducing symptoms, with no difference between the individual SSRI drugs and response rates (Soomro et al., 2008), although most patients with OCD will continue to have residual symptoms even after an adequate trial of treatment.

Sex Differences in Obsessive-Compulsive Disorder

Several major epidemiological studies reported OCD to be slightly more prevalent among women 420

than men (e.g., Karno et al.,  1988; Kolada et  al., 1994), but data from clinical samples fail to detect sex differences in prevalence rates in adults OCD (Rasmussen & Eisen,  1992). Numerous studies have examined sex differences in the types of obsessions and compulsions. Most studies have found that males with OCD are more likely to endorse symptoms related to sexual, religious, or aggressive content (Cherian et al.,  2014; Labad et al.,  2008; Lensi et al.,  1996; Mathis et al.,  2011; Torresan et  al.,  2009), whereas women with OCD report more symptoms with contamination or cleaning content (Labad et al.,  2008; Mathis et al.,  2011). These findings have led some to conclude that the content and clinical expression of OCD symptoms may be influenced by sociocultural factors (Khanna & Mukherjee, 1992; Lensi et al., 1996). There are also substantial data indicating sex differences in the onset and clinical course associated with OCD. In a review of 63 studies concerning sex differences in OCD, Mathis et al. (2011) concluded that male patients were more likely than females to be single, have earlier symptom onset, have a chronic clinical course, and have greater social impairment. The mean age of onset for OCD in males (20 years) is earlier than in females (25 years), and males are also much more likely to develop OCD during childhood (Ruscio, Stein, Chiu, & Kessler, 2010). In fact, OCD onset before the age of 10 is associated with being male, tic disorder, a positive family history (Leonard et al., 1992), and symptom severity (McLean & Anderson,  2009). Comorbid tic disorder has an especially strong association with the male sex and early age of onset in OCD. Boys with OCD and adult men with childhood-onset OCD are more likely to have comorbid tics and attention-deficit disorders (Leckman et al., 2010; Ruscio et al., 2010). Tourette’s syndrome is also much more pronounced in males compared to females with OCD (Mathis et al., 2011). In a meta-analysis of 26 studies conducted with OCD patients, 31 percent had comorbid tic disorder, and this was associated with male sex and early age of onset. The OCD group with tics was more likely to endorse touching, twitching, repeating, rubbing, and symmetry rituals, as well as “just right” perceptions prior to their compulsions, whereas the OCD group without tics was more likely to endorse contamination obsessions, washing rituals, and anxiety preceding their compulsive behaviors (Fibbe, Cath, & van Balkom,  2011). Among pediatric patients, those with OCD onset before age 12 years have a higher frequency of comorbid tic disorder and

Sex Differences in Anxiet y Disorders

disruptive behavior disorders, whereby symptoms related to order/symmetry are more frequent in males and had the highest comorbidity with tics (Masi et al., 2010). Consistent with other reports, tic comorbidity is also associated with an earlier onset of OCD and a heavier comorbidity with ADHD and other disruptive behavior disorders. Comorbid ADHD in turn is associated with an earlier onset of OCD and a poorer response to treatment. In contrast, females are more likely to endorse OCD symptoms related to contamination or cleaning. Contamination/cleaning rituals are associated with the lowest clinical severity, but a greater rate of comorbid anxiety and depression (Masi et al., 2010). Differences in comorbidity patterns in OCD emerge at the age of 10 and are more pronounced at the age of 17 (de Mathis et al., 2008). There also appears to be differences in the comorbidity pattern between sexes and between prepubertal and adolescent-onset cases of OCD. In a study of children and adolescents diagnosed with OCD, most (83.6 percent) participants had at least one comorbid psychiatric disorder, and oppositional defiant disorder (ODD) and contamination/somatic obsessions were higher in boys than in girls (Tanidir et al., 2015). Further analysis revealed that disruptive behavior disorders were more frequent in those with tic disorder than in those without tics (Tanidir et al., 2015). These findings provide further support for the inclusion of tic-related OCD as a specifier in the DSM-V. In contrast to early-onset OCD, women are more likely to report later onset of OCD symptoms. Late-onset OCD is associated with being female, a history of chronic (>10 years) subclinical obsessivecompulsive symptoms, the co-occurrence of PTSD, and a history of recent pregnancy or pregnant partner (Frydman et al.,  2014). In an Indian sample, 37  percent reported juvenile-onset (