Multilingual Text Comprehension: Cognitive, Developmental, and Educational Aspects (The Bilingual Mind and Brain Book Series) 3031433408, 9783031433405

Table of contents :
Preface
Acknowledgments
Contents
About the Author
Chapter 1: Issues and Questions in Text Comprehension
1.1 Text Comprehension and Text Production
1.2 Texts and Their Features
1.3 Text Genres
1.4 Text Processing
1.5 A Comprehensive View of Text Comprehension
1.6 Text Processing Modalities Listening and Reading
1.7 The Contribution of Working Memory
1.8 The Contribution of Long-Term Memory and Knowledge
1.9 The Contribution of Inferences
1.10 Comprehension Monitoring
1.11 Strategies to Improve Text Comprehension
1.12 Intelligence and Comprehension
1.13 Concluding Remarks
References
Chapter 2: Multiple Language Use and Text Comprehension
2.1 Multilingual Decoding and Encoding
2.1.1 Decoding
2.1.2 Encoding
2.2 Coactivation in Multilingual Text Processing
2.3 Comprehension of Content and Linguistic Surface Structure
2.4 Multilingual Tasks and Text Comprehension
2.5 Degree of Bilingualism and Its Consequences
2.5.1 Degree of Bilingualism and Automaticity of Picture Naming
2.5.2 Degree of Bilingualism and Comprehension
2.6 Cross- and Monolingual Tasks and Comprehension
2.7 Individual Differences and Text Comprehension
2.8 The Developmental Dynamics of Multilingualism
2.9 Concluding Remarks
References
Chapter 3: Assessment of Text Comprehension
3.1 Language-Dependent and -Independent Comprehension Assessment
3.1.1 Comprehension of Text and Graphics
3.1.2 Controlling for Decoding by Avoiding Verbal Input
3.2 Assessment of Text Readability
3.3 Comprehension in Two Modalities: Reading and Listening
3.3.1 Assessment of Reading Comprehension
3.3.2 Assessment of Listening Comprehension
3.4 Recall-Based Offline Measures
3.4.1 Recognition
3.4.2 Cued Recall of Text
3.4.3 Free Recall and Retell
3.4.4 Summary
3.4.5 Cloze Tests
3.5 Online Measures
3.5.1 Reading Fluency
3.5.2 Comprehension Monitoring and Control
3.5.3 Think-aloud Protocols and Probe Questions
3.5.4 Priming
3.6 Issues of Multilingual Assessment
3.7 Minimizing Effects of Decoding and Language
3.8 Age Effects in Multilingual Assessment
3.9 National and Cross-National Assessment
3.10 Concluding Remarks
References
Chapter 4: Multilingual Development of Text Comprehension
4.1 Multilingual Language Acquisition
4.1.1 Native Language in Discourse Development
4.1.2 Age of Multilingual Acquisition and Comprehension
4.1.3 Multilingual Vocabulary Growth and Comprehension
4.1.4 Acquisition of Multilingual Word Semantics Syntax and Comprehension
4.2 Development of Multilingual Narrative Skills
4.3 Effects of Degree of Bilingualism and Language of Text
4.4 Bilingual Children’s Cohesion and Coherence Construction in Texts
4.5 Concluding Remarks
References
Chapter 5: Higher Order Cognitive Skills in Text Comprehension
5.1 Foundations of Macronarrative Comprehension Skills
5.2 Constructing the Propositional Textbase
5.3 The Support of Working Memory
5.3.1 Suppression/Inhibition of Irrelevant Information
5.3.2 Updating Information in Text Processing
5.4 Long-term Memory as a Knowledge Base
5.4.1 World Knowledge
5.4.2 Language as a Knowledge Base
5.5 Development of Macronarrative Comprehension Skills
5.5.1 Understanding the Sequence of Narrative Events
5.5.2 Use of Conventional Text Structure
5.5.3 Situation Model Construction
5.5.4 Inference Making
5.6 Comprehension Monitoring
5.7 Concluding Remarks
References
Chapter 6: Text Comprehension Dynamics: Time Parameters
6.1 Time Constraints and Text Comprehension
6.2 Text Reading Fluency
6.3 Timing of Comprehension and Fluency of Recall
6.4 Consolidation of the Mental Text Representation
6.5 Separating Types of Pause
6.5.1 Language Conditions and Pauses
6.5.2 Crosslinguistic Text Processing and Pauses
6.6 Composite Processes of Comprehension and Pauses
6.7 Does Comprehension Continue During Recall?
6.8 The Function of Gist Pauses: A Process Analysis
6.9 Concluding Remarks
References
Chapter 7: A Theoretical Model of Multilingual Text Processing
7.1 Requirements of a Model of Multilingual Text Comprehension
7.2 Constructing a Multilingual Model of Text Comprehension
7.2.1 The Shallow Processing Hypothesis
7.2.2 Limited Automaticity in L2 Processing
7.2.3 Predictability of Text in L2
7.3 The Recall Component of Text Comprehension
7.4 Issues of Asymmetry
7.5 Retrieval of Mental Representation After Language Switching
7.6 A Model of Multilingual Text Comprehension
7.7 Developmental Perspective of the Model
7.8 Theoretical Afterthoughts
7.9 Concluding Remarks
References
Chapter 8: Educational Applications of Text Comprehension Research
8.1 Can Comprehension Be Trained?
8.2 Advantages and Disadvantages of Multilingualism
8.3 Modality-Specific Comprehension Training?
8.4 Effects of Instructions on Comprehension
8.5 Improving Comprehension by Strategy Use
8.5.1 Rehearsal Strategy
8.5.2 Text Structure Strategy
8.5.3 Instructions for Situation Model Construction
8.5.4 Questioning Strategy
8.5.5 Instructing Inference Making
8.5.6 Metacognitive Strategies: Monitoring Comprehension
8.5.7 Developmental Aspects of Monitoring
8.6 Improving Comprehension for L2 Learners
8.7 Text Comprehension as a Diagnostic Tool
8.8 Interventions for Diverse Groups of Children
8.9 Concluding Remarks
References
Chapter 9: Understanding Multilingual Children’s Text Comprehension
9.1 Discourse and Text Comprehension
9.2 Language Dominance and Text Comprehension
9.3 Cognitive Processes Contributing to Text Comprehension: A Framework
9.4 Assessment of Comprehension
9.5 Time Parameters of Text Retrieval
9.6 An Empirical Model of Multilingual Text Comprehension
9.7 Applications
9.8 Critical Issues
9.8.1 Coactivation
9.8.2 Online Assessment and Neuroscience
9.8.3 Semantic Networks and Multiple Language Use
9.8.4 Comprehension of Hypertexts
9.9 Deep Comprehension
References
Author Index
Subject Index

Citation preview

The Bilingual Mind and Brain Book Series

Ute Schönpflug

Multilingual Text Comprehension Cognitive, Developmental, and Educational Aspects

The Bilingual Mind and Brain Book Series Series Editors Roberto R Heredia, Department of Psychology and Communication Texas A&M International University Laredo, TX, USA Anna B. Cieslicka, Department of Psychology and Communication Texas A&M International University Laredo, TX, USA

Springer’s Bilingual Mind and Brain Book Series is intended to advance and contribute to our understanding of the bilingual/multilingual mind and brain, both as an academic discipline and as a maturing research field. The Bilingual Mind and Brain Book Series is interdisciplinary in its scope and examines the bilingual mind/ brain from such perspectives as psycholinguistics, cognitive psychology, cognitive science, and cognitive neuroscience, as well as applied linguistics and pedagogical approaches to second/foreign language learning. The Bilingual Mind and Brain Book Series is intended for the growing number of bilingual researchers and practitioners interested in understanding the behavioral aspects and neurobiology of bilingualism, as well as the dynamic character of the bilingual/multilingual/second language learner’s mind. This book series will be of interest to the growing number of advanced undergraduate and graduate students interested in The Psychology/ Psycholinguistics of Bilingualism, Bilingual Cognitive Psychology, Cognitive Science, and Cognitive Neuroscience. For this reason, all collective volumes and monographs will include a number of thought provoking questions, suggested student research projects, and suggested readings. In short, The Bilingual Mind and Brain Book Series provides a continuous update of the most current work in the behavioral and neuropsychological research in the fields of bilingualism/ multilingualism and second language acquisition. This book series contributes to the development and establishment of the Bilingual Cognitive Psychology and Bilingual Cognitive Neuroscience disciplines as subfields of Neuropsychology and Cognitive Neuroscience and fills a significant gap in the literature on the bilingual mind and brain. Have a book idea? Contact the Series Editors at bilingual.mind-­ [email protected].

Ute Schönpflug

Multilingual Text Comprehension Cognitive, Developmental, and Educational Aspects

Ute Schönpflug Department of Developmental Science Free University Berlin Berlin, Germany

ISSN 2945-5952     ISSN 2945-5960 (electronic) The Bilingual Mind and Brain Book Series ISBN 978-3-031-43340-5    ISBN 978-3-031-43341-2 (eBook) https://doi.org/10.1007/978-3-031-43341-2 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Paper in this product is recyclable.

Dedicated to Walter and Eileen Kintsch

Preface

Text comprehension is an important skill for personal fulfillment and for achieving academic and professional success. It is also a complex skill, involving cognitive abilities that interact at different text levels. The special perspective of this book builds on research on multilingual children’s comprehension of narrative and expository texts in the course of their cognitive development. Comparisons with corresponding findings of “monolingual” children raised and schooled in environments emphasizing L1 are intended to turn the readers’ attention to the issue of what impact languages may exert on text comprehension. The chapters focus on text comprehension beyond basic biologically grounded processes such as phonological decoding of oral or written input, as multilingual children—even when proficient in each of their languages—are challenged at the text level rather than at the word and grammar level. The author of this book attempts to find answers to the question of whether text processing in a second or more languages suffers when bi-/multilingual children perform in their other languages as compared to their native first language. The book addresses developmental periods from early childhood to adolescence and young adulthood and covers listening and reading comprehension. Research about reading comprehension dominates text comprehension literature, but the broad developmental span targeted in this book requires that listening comprehension will be in the focus as well. The nine chapters of this book present new insights into the cognitive processes enabling and regulating multilingual text comprehension, which bear the potential to be beneficial for knowledge of multilingualism in general and for research in the domains of cognitive development, discourse linguistics, and education. Texts in languages other than the comprehender’s native language can only be processed by multilingual individuals who have basic language proficiencies in each language. The question of the extent to which language, and especially multilingualism, plays a crucial role in the construction processes of gist formation represents the leading theme of this book. Chapter 1 outlines the scope of the book and offers access to the concepts and their meaning in text comprehension research. Chapter 2 reports findings that offer potential explanations for the complex relationship between knowing and using language(s) and understanding texts. Chapter 3 vii

viii

Preface

introduces common methods of assessing text comprehension in children with reference to the languages involved. General developmental issues addressing cognitive abilities and skills relevant for text comprehension are addressed in Chap. 4, while research findings from the domain of text comprehension in preschool- and school-aged children are discussed extensively in Chap. 5. Chapter 6 presents issues related to the dynamics of text comprehension, asking about the time parameters of composite processes of comprehension. Chapter 7 is devoted to theories and models of text comprehension given multilingual and developmental constraints. The book author suggests a model derived from her data that proposes inferences as mediators between cross- and monolingual text input/output conditions and text comprehension. The model introduces bilingual language proficiency as a moderator. The final Chaps. 8 and 9 point out applications developed for training and interventions to improve understanding in from kindergarten age on to secondary school and issues in need of further clarification. This book aims to develop a framework for fine-grained analyses of text comprehension skills that may foster our understanding of children’s text comprehension and comprehension problems. These issues have meanwhile been discussed in the literature to such an extent that a critical selection, structuring, and organization of the research areas are necessary. As linguistic issues have prevailed in the psycholinguistic research, the linguistic surface structure of the text has already received a great deal of attention. The psychological approach shifts the focus from the linguistic surface to the mental representation of the text. From the point at which the meanings of the linguistic structures are identified, psychological researchers provide appropriate expertise to explore comprehension processes leading to the mental representation of a text. The literature on children’s text comprehension has a long history in the educational sciences. Multilingual text comprehension research, by contrast, essentially only spans the last two decades. Earlier research on bilingualism and multilingualism favored word-level and sentence-level language comprehension processes. In view of current global mobility and migration, there is an urgent need to investigate higher-order cognitive processes like text comprehension and to examine how they are affected by a multilingual comprehension task performed by a multilingual child. The book author’s own research is selectively integrated into this overview of international findings on children’s text comprehension in the light of their multilingual competencies and multilingual texts. In addition to a summary of relevant literature, the author presents her own published and unpublished research results on L1-dominant, bi- and multilingual children’s and young adults’ text comprehension in their native language German, their second language English, and various other further languages. Whereas the L1-dominant sample consisted of three age groups (8-, 9-, and 10-year-olds), the bilingual sample comprised German fourth graders, aged 9 to 10 years, attending German-English bilingual schools, most of them from nursery school onward. The bilingual schools were public Europe Schools, a program dedicated to bilingual education from the first to sixth grade in Germany. The third group, comprising multilingual university students, were proficient in Polish (L1), German (L2), and English (L3).

Preface

ix

This book reflects the need to clarify how proficiency in more than one language and language dominance can impact the higher-order cognitive processes of text comprehension under monolingual and crosslingual text input and output conditions. The author emphasizes developmental aspects of text processing, as educational applications should be based on the knowledge of how multilingual children’s cognitive processes change during the developmentally decisive years for text comprehension. Berlin, Germany  Ute Schönpflug Autumn 2023

Acknowledgments

Over the last two decades, a group of multilingual psychology graduates contributed to the research reported in this book. Given their multilingual background and their study situation, they were motivated to ask and find answers to questions concerning multilingual functioning when understanding texts. They broadened the scope of the research bringing in new aspects in the planning and the analysis stage of the project. Moreover, they conducted most of the experiments. The author thanks first of all Lenka Küpping-Faturikova and Felicitas Frische for their valuable ideas and work. Anna Vierling contributed her nearly balanced bilingual language proficiency to setting up the experimental materials and collecting part of the data. I owe many thanks to the numerous students who took part in the series of experiments assembled under the heading Monolingual and Multilingual Children’s Development of Text Comprehension and to their teachers who organized the experiments in their schools in the hope to be able to improve their students’ text comprehension performance.  With great patience and native language proficiency in English, Sarah Mannion de Hernandez smoothened the linguistic surface of the manuscript. As a Cambridge University graduate, she followed the book author’s discussion of the presented issues with critical text comprehension and thus helped to establish cohesion and coherence in the text of the book. The editing and production team chose the right format for what I had to say and I am grateful for their professional efforts. With respect I acknowledge the impact of the work of Walter Kintsch, who passed away this Spring, and Eileen Kintsch from the Institute of Cognitive Science at the University of Colorado: Their research and theorizing elicited my interest in text comprehension processes. This book is dedicated to them and to Wolfgang Schönpflug, Free University Berlin, whose day-to-day support maintained my engagement and perseverance. Thanks to Roberto Heredia and Anna Cieślicka’s engaged concern for bilingual research this book is published and may extend the aspects covered by their series on The Bilingual Mind and Brain to developmental and educational aspects. Published in this series, the book will hopefully reach an interested readership.

xi

Contents

1

 Issues and Questions in Text Comprehension ��������������������������������������    1 1.1 Text Comprehension and Text Production����������������������������������������    1 1.2 Texts and Their Features ������������������������������������������������������������������    3 1.3 Text Genres ��������������������������������������������������������������������������������������    4 1.4 Text Processing ��������������������������������������������������������������������������������    6 1.5 A Comprehensive View of Text Comprehension������������������������������    7 1.6 Text Processing Modalities Listening and Reading��������������������������   10 1.7 The Contribution of Working Memory ��������������������������������������������   16 1.8 The Contribution of Long-Term Memory and Knowledge��������������   17 1.9 The Contribution of Inferences ��������������������������������������������������������   17 1.10 Comprehension Monitoring��������������������������������������������������������������   19 1.11 Strategies to Improve Text Comprehension��������������������������������������   20 1.12 Intelligence and Comprehension������������������������������������������������������   21 1.13 Concluding Remarks������������������������������������������������������������������������   23 References��������������������������������������������������������������������������������������������������   23

2

 Multiple Language Use and Text Comprehension��������������������������������   29 2.1 Multilingual Decoding and Encoding����������������������������������������������   29 2.1.1 Decoding ������������������������������������������������������������������������������   29 2.1.2 Encoding ������������������������������������������������������������������������������   34 2.2 Coactivation in Multilingual Text Processing����������������������������������   35 2.3 Comprehension of Content and Linguistic Surface Structure����������   38 2.4 Multilingual Tasks and Text Comprehension������������������������������������   39 2.5 Degree of Bilingualism and Its Consequences ��������������������������������   41 2.5.1 Degree of Bilingualism and Automaticity of Picture Naming����������������������������������������������������������������������������������   41 2.5.2 Degree of Bilingualism and Comprehension������������������������   45 2.6 Cross- and Monolingual Tasks and Comprehension������������������������   47 2.7 Individual Differences and Text Comprehension������������������������������   49 2.8 The Developmental Dynamics of Multilingualism��������������������������   50 2.9 Concluding Remarks������������������������������������������������������������������������   51 References��������������������������������������������������������������������������������������������������   52 xiii

xiv

Contents

3

Assessment of Text Comprehension ������������������������������������������������������   57 3.1 Language-Dependent and -Independent Comprehension Assessment����������������������������������������������������������������������������������������   57 3.1.1 Comprehension of Text and Graphics����������������������������������   57 3.1.2 Controlling for Decoding by Avoiding Verbal Input������������   59 3.2 Assessment of Text Readability��������������������������������������������������������   60 3.3 Comprehension in Two Modalities: Reading and Listening������������   61 3.3.1 Assessment of Reading Comprehension������������������������������   61 3.3.2 Assessment of Listening Comprehension����������������������������   64 3.4 Recall-Based Offline Measures��������������������������������������������������������   65 3.4.1 Recognition ��������������������������������������������������������������������������   66 3.4.2 Cued Recall of Text��������������������������������������������������������������   67 3.4.3 Free Recall and Retell����������������������������������������������������������   68 3.4.4 Summary ������������������������������������������������������������������������������   71 3.4.5 Cloze Tests ��������������������������������������������������������������������������   72 3.5 Online Measures ������������������������������������������������������������������������������   73 3.5.1 Reading Fluency ������������������������������������������������������������������   74 3.5.2 Comprehension Monitoring and Control������������������������������   75 3.5.3 Think-aloud Protocols and Probe Questions������������������������   76 3.5.4 Priming����������������������������������������������������������������������������������   77 3.6 Issues of Multilingual Assessment����������������������������������������������������   79 3.7 Minimizing Effects of Decoding and Language ������������������������������   82 3.8 Age Effects in Multilingual Assessment������������������������������������������   84 3.9 National and Cross-National Assessment ����������������������������������������   85 3.10 Concluding Remarks������������������������������������������������������������������������   87 References��������������������������������������������������������������������������������������������������   88

4

 Multilingual Development of Text Comprehension������������������������������   95 4.1 Multilingual Language Acquisition��������������������������������������������������   95 4.1.1 Native Language in Discourse Development������������������������   97 4.1.2 Age of Multilingual Acquisition and Comprehension����������   98 4.1.3 Multilingual Vocabulary Growth and Comprehension ��������  101 4.1.4 Acquisition of Multilingual Word Semantics Syntax and Comprehension��������������������������������������������������  103 4.2 Development of Multilingual Narrative Skills����������������������������������  107 4.3 Effects of Degree of Bilingualism and Language of Text����������������  110 4.4 Bilingual Children’s Cohesion and Coherence Construction in Texts����������������������������������������������������������������������������������������������  113 4.5 Concluding Remarks������������������������������������������������������������������������  115 References��������������������������������������������������������������������������������������������������  116

5

 Higher Order Cognitive Skills in Text Comprehension�����������������������  121 5.1 Foundations of Macronarrative Comprehension Skills��������������������  121 5.2 Constructing the Propositional Textbase������������������������������������������  122 5.3 The Support of Working Memory����������������������������������������������������  125 5.3.1 Suppression/Inhibition of Irrelevant Information ����������������  126 5.3.2 Updating Information in Text Processing ����������������������������  128

Contents

xv

5.4 Long-term Memory as a Knowledge Base ��������������������������������������  130 5.4.1 World Knowledge ����������������������������������������������������������������  130 5.4.2 Language as a Knowledge Base�������������������������������������������  132 5.5 Development of Macronarrative Comprehension Skills������������������  133 5.5.1 Understanding the Sequence of Narrative Events����������������  135 5.5.2 Use of Conventional Text Structure��������������������������������������  136 5.5.3 Situation Model Construction ����������������������������������������������  139 5.5.4 Inference Making������������������������������������������������������������������  142 5.6 Comprehension Monitoring��������������������������������������������������������������  145 5.7 Concluding Remarks������������������������������������������������������������������������  148 References��������������������������������������������������������������������������������������������������  149 6

 Text Comprehension Dynamics: Time Parameters������������������������������  157 6.1 Time Constraints and Text Comprehension��������������������������������������  157 6.2 Text Reading Fluency ����������������������������������������������������������������������  159 6.3 Timing of Comprehension and Fluency of Recall����������������������������  159 6.4 Consolidation of the Mental Text Representation����������������������������  164 6.5 Separating Types of Pause����������������������������������������������������������������  167 6.5.1 Language Conditions and Pauses������������������������������������������  167 6.5.2 Crosslinguistic Text Processing and Pauses��������������������������  168 6.6 Composite Processes of Comprehension and Pauses ����������������������  170 6.7 Does Comprehension Continue During Recall? ������������������������������  172 6.8 The Function of Gist Pauses: A Process Analysis����������������������������  172 6.9 Concluding Remarks������������������������������������������������������������������������  175 References��������������������������������������������������������������������������������������������������  176

7

 Theoretical Model of Multilingual Text Processing��������������������������  179 A 7.1 Requirements of a Model of Multilingual Text Comprehension������  179 7.2 Constructing a Multilingual Model of Text Comprehension������������  181 7.2.1 The Shallow Processing Hypothesis������������������������������������  182 7.2.2 Limited Automaticity in L2 Processing��������������������������������  184 7.2.3 Predictability of Text in L2 ��������������������������������������������������  185 7.3 The Recall Component of Text Comprehension������������������������������  186 7.4 Issues of Asymmetry������������������������������������������������������������������������  188 7.5 Retrieval of Mental Representation After Language Switching ������  189 7.6 A Model of Multilingual Text Comprehension��������������������������������  191 7.7 Developmental Perspective of the Model ����������������������������������������  196 7.8 Theoretical Afterthoughts ����������������������������������������������������������������  199 7.9 Concluding Remarks������������������������������������������������������������������������  201 References��������������������������������������������������������������������������������������������������  202

8

 Educational Applications of Text Comprehension Research ��������������  205 8.1 Can Comprehension Be Trained? ����������������������������������������������������  205 8.2 Advantages and Disadvantages of Multilingualism��������������������������  206 8.3 Modality-Specific Comprehension Training������������������������������������  208 8.4 Effects of Instructions on Comprehension����������������������������������������  209

xvi

Contents

8.5 Improving Comprehension by Strategy Use������������������������������������  211 8.5.1 Rehearsal Strategy����������������������������������������������������������������  211 8.5.2 Text Structure Strategy����������������������������������������������������������  213 8.5.3 Instructions for Situation Model Construction ��������������������  216 8.5.4 Questioning Strategy������������������������������������������������������������  216 8.5.5 Instructing Inference Making������������������������������������������������  217 8.5.6 Metacognitive Strategies: Monitoring Comprehension��������  219 8.5.7 Developmental Aspects of Monitoring ��������������������������������  222 8.6 Improving Comprehension for L2 Learners ������������������������������������  223 8.7 Text Comprehension as a Diagnostic Tool����������������������������������������  225 8.8 Interventions for Diverse Groups of Children����������������������������������  229 8.9 Concluding Remarks������������������������������������������������������������������������  230 References��������������������������������������������������������������������������������������������������  230 9

 Understanding Multilingual Children’s Text Comprehension������������  237 9.1 Discourse and Text Comprehension�������������������������������������������������  237 9.2 Language Dominance and Text Comprehension������������������������������  238 9.3 Cognitive Processes Contributing to Text Comprehension: A Framework������������������������������������������������������������������������������������  240 9.4 Assessment of Comprehension ��������������������������������������������������������  242 9.5 Time Parameters of Text Retrieval����������������������������������������������������  243 9.6 An Empirical Model of Multilingual Text Comprehension��������������  244 9.7 Applications��������������������������������������������������������������������������������������  245 9.8 Critical Issues������������������������������������������������������������������������������������  246 9.8.1 Coactivation��������������������������������������������������������������������������  246 9.8.2 Online Assessment and Neuroscience����������������������������������  247 9.8.3 Semantic Networks and Multiple Language Use������������������  248 9.8.4 Comprehension of Hypertexts����������������������������������������������  249 9.9 Deep Comprehension������������������������������������������������������������������������  250 References��������������������������������������������������������������������������������������������������  251

Author Index����������������������������������������������������������������������������������������������������  253 Subject Index����������������������������������������������������������������������������������������������������  259

About the Author

Ute Schönpflug  PhD., Prof. adj. emer., received her education at Northwestern University and in Germany. She taught developmental/educational psychology at Martin-Luther University (Halle/Germany) and Free University Berlin. She was associate editor of the Journal of Cross-Cultural Psychology for ten years and published part of her research in a volume on Cultural Transmission. She is author and co-author of several psychology textbooks. As an adjunct member of ICC, Boulder, CO, USA, her research interests focused on text processing in multilingual children. She has served the UGC/RGC in Hong Kong as member and reviewer of the Linguistics/Psychology Panel for the last years.

xvii

Chapter 1

Issues and Questions in Text Comprehension

1.1 Text Comprehension and Text Production The human mind is constantly learning and interpreting new information, both from narrative texts (commonly referred to as stories) and from informational texts (also termed descriptive text or expository text). The word “text” is derived from the Latin verb “texere” meaning to weave. Most of our conversations, media, and early educational resources occur in the form of narrative discourse. Narratives typically encompass stories that start with an introduction containing characters and their goals, actions, conflicts, and events that lead to the accomplishment of these goals, to the resolution of conflicts. However, deviations from this conventional scheme may be found within literature and when drawing cross-cultural comparisons. Often, we experience narratives as oral presentations (e.g., verbal stories) or wordless picture sequences (e.g., picture books), audio-visual films (e.g., movies), and written text. Studies suggest that for all narrative forms, similar underlying processes are necessary for comprehension (Trabasso & Nickels, 1992; Graesser et al., 2013; Kintsch, 1998). Expository texts present information in more varied canonical text structures that will be discussed throughout the following chapters. Text comprehension and text production are considered as separate modes of text processing. However, conceptualizations of discourse indicate that production is intimately related to comprehension and the associated cognitive skills. For instance, the assessment of text comprehension may imply some form of text production, such as verbal recall of a text, retelling or summarization. This kind of discourse production provides a window into adequate comprehension performance. But there are also some indicators of comprehension that do not draw on text recall or retell protocols, such as recognition tests that simply ask, whether a word or a sentence had been in the input text. Text production is more than merely a form of recall protocol. Texts can be produced without previous listening or reading a text. The information to be © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 U. Schönpflug, Multilingual Text Comprehension, The Bilingual Mind and Brain Book Series, https://doi.org/10.1007/978-3-031-43341-2_1

1

2

1  Issues and Questions in Text Comprehension

communicated by producing a text starts in the producer’s own mind and is then transferred to mental functional units operating during speech planning, the purpose of which is to choose the appropriate linguistic forms for the message. Only then is the message articulated in oral or written form. Levelt’s (1993) speech production model for words may be helpful for conceptualizing the text production process, but given the great complexities of any specific text, it is certainly too simple. Research on text production often focuses on children’s or adults’ narrative skills, text length, text information details, and text cohesion and coherence, while research on text comprehension seeks to identify the mechanisms underlying the construction of narrative mental representations. Several authors (e.g., Pickering & Garrod, 2013a, b) argue against the traditional separation of comprehension and production, instead they model language production and comprehension as part of interactive processes between producer and comprehender. But how can production and comprehension both be involved in isolated instances of speaking or listening? Of course, production processes must be used when individuals produce language, and comprehension processes must be used when they comprehend language. However, it is possible that production processes are used during comprehension and that comprehension processes are used during production (Pickering & Garrod, 2013b). Furthermore, production processes are also involved in, for example, silent naming, when no utterance is observable. Silent naming thus involves some production processes (e.g., those associated with pre-utterance processes) but not those associated with articulation itself. Interlocutors often overlap when speaking, and therefore produce and comprehend at the same time, and they frequently complete each other’s utterances. Written and oral recall or reproduction of a text starts with the retrieval of a mental representation of the text constructed during text input. In the case of ideal memory functioning, the complete mental representation is retrieved and included in the recall. Recall can then be seen as occurring at the end of an imaginary bipolar scale that represents a verbatim repetition of the original text at one pole and free recall of content without linguistic overlap at the other. More typically, recall takes the form of a reproduction, in which minor omissions, elaborations and integrative inferences appear. If the product of retrieval shows great gaps in content (compared to the original text) or a lack of global coherences and/or local cohesion, constructive processes conveying meaning will remedy the deficiently retrieved mental representation, culminating in production of a new text. The distinguishing feature between verbatim recall, reproduction and production is the amount of retrieved content of the original text that appears in the retrieved product. Text researchers commonly posit that comprehension and production represent separate modules. Cain and Oakhill (2007) argued for a separation of comprehension and production by citing neuropathological findings of a functional dissociation between comprehension and production performance. More recent findings from the field of neuroscience, using functional magnetic resonance imaging (fMRI) methodology, indicate that when a person is telling a narrative to another person listening to this narrative, there is an extensive overlap in both persons’ brain regions that are activated across both hemispheres. Moreover, a direct comparison of the

1.2  Texts and Their Features

3

neural activity time courses during production or comprehension of the same narrative reveals areas in which neural activity is coupled in the speaker’s and the listener’s brains. This noteworthy finding suggests that comprehension and production processes may be more similar and more aligned regarding neural activity than previously assumed (Silbert et al., 2014). In an earlier overview of the research, Gernsbacher and Kaschak (2003) claimed that comprehension has to precede reproduction in the case of any text processing. The research findings offered until recently in the text comprehension literature leave one related question unanswered: Will the construction processes taking place during comprehension at the text input stage continue during the following reproduction or recall? The following literature survey, and the book author’s own research, seeks for answers to this question.

1.2 Texts and Their Features Children growing up in literate or in illiterate societies are confronted with the challenge of processing text to gain information about the world around them. Presenting a text either orally or in written format, and comprehending it, form part of the daily discourse of persons of all age groups, genders, social status and languages. To satisfy the human need for greater contextual meaning of information, texts encompass a more or less integrated sequence of verbal statements. The author of a text constructs the integration of the contextual meaning and communicates it in written or oral format to other persons. According to our cultural norms, the statements in a text should be as integrated as possible to create a cohesive and coherent text, that is, the connection of the statements among each other and to the main topic or idea of the text must be clear. The author’s language skills and specific expertise or general world knowledge determine how well the statements are integrated, and how explicitly the intended message of a text is expressed. The receivers of a text have to cognitively process the communication by listening or reading and by constructing a mental representation of the text. To gain insight into the intended contextual meaning, it is essential that the listener or reader of a text function at a similar cognitive level as the author in terms of language skills, background knowledge, and general cognitive skills such as inference making. Most scholars dealing with literature, science and arts suggest a rather narrow definition of the term “text” as a written form of discourse. This definition implies that the receiver of the text must be able to read. However, literacy is only a desired state of mankind, and not a global accomplishment. Literacy is acquired in childhood within literate cultures, but does that mean that preliterate children are not confronted with texts? Are they not challenged by searching for contextual meaning? Certainly, they hear texts read or told to them by their caregivers. Moreover, illiterate adults may listen to texts in the form of speeches, oral lectures, explanations and narratives. Hence, for the developmental perspective adopted for this book, it is recommendable to extend the definition of text to an oral production, in

4

1  Issues and Questions in Text Comprehension

which case text processing, and especially comprehension, take place by listening to the orally presented text. This book seeks to uncover whether the two modalities of comprehension, reading and listening, are subject to different methods of processing. According to the developmental perspective, comprehension processes during listening have to be followed up into adulthood, because the processes related to reading (defined as the skill of decoding verbal material) will interact with comprehension processes later in development. The “literate” definition of text fixates on the written format. By contrast, when repeatedly communicated and not learned verbatim by heart, the repetition of oral narratives will most likely not be identical. Are these variations the same text? In the strictest since, with regard to all text features they are not. Additions of short particles like “even” may change the meaning of a message: The new meaning of the statement encompasses an aspect of unexpectedness, surprise, unusualness, or extremeness. Systematic investigations of how to construct a cohesive and coherent text have yielded noteworthy findings. Coherence is defined as meaning gained by integration of context in texts (contextual meaning), tightening together the idea units of a text to a unifying main idea, while cohesion refers to the linguistic means that can be used to create contextual meaning. The conjunction “because” is the most prominent example of such linguistic means, relating cause and effect/consequence to each other and supplying information on why something appears, exists or happens. If no such explicit linguistic information is given, listeners or readers have to accomplish a cohesive mental representation by making inferences. Mostly, they will have to draw on their background knowledge to establish relations between statements. Experiences stored in long-term memory support the discovery of temporal, logical and referentially meaningful relations. However, the text also must allow for continuity in contextual meaning. Contradictions and breaks or other discontinuities hamper the construction of a coherent mental representation.

1.3 Text Genres The criteria used to identify different types of text involve the communicative situation, the contents, the author, and the recipient. Examples of communicative situations are dialogues or questionings. Types of contents in texts that serve as criteria for their identification include descriptive, narrative, informative/instructive/exposi tory texts, as well as argumentative texts. Typical content sequences (text structure or text grammar) contribute to these distinctions. A descriptive text reports experiences without great efforts invested in integrating construction. A narrative is characterized by a temporal sequence and a conventional text structure; one (commonly used in our culture is introduction, conflict/action, and resolution). Expository texts are also labelled as informative or instructive texts provide hierarchically structured information, whereas an argumentative text takes an explicitly analytical approach.

1.3  Text Genres

5

A form of an oral descriptive text might be, for instance, an oral report about a visit to the zoo. Examples of more conventional forms of narrative texts are novels or anecdotes, while expository texts may take the form of, e.g., science texts, manuals, recipes, etc. Psycholinguistic research on text types and text processing is commonly dedicated to narratives and expository texts. Coherent narratives, with their conventional text structure that organizes the text, rely on a similarity between the author and the recipient in terms of knowledge about the conventional structure. This common knowledge is essential for good comprehension. For instance, readings of the Mayan creation myths would be difficult to follow for people schooled in Western narrative schemata, as the myths lack coherence, and even simple cause/ effect structures may not be clear. In analyses developed in the field of literary science, with respect to the sequential structure of different forms of narrative (e.g., novel, short story), the narrative schema applies to the content elements. In addition to this content level, the discourse level is concerned with the linguistic surface form of a text (e.g., linguistic style, vocabulary, grammar). In psychological text analyses, the surface level is secondary, while psycholinguistic research and theorizing considers both levels. Cognitive psycholinguistics is interested in conventional text structure or text grammar reflecting the general organization. Psycholinguistics has developed its own perspectives on text processing of predominantly two distinct types of text: narratives or stories and informative science texts. Most psycholinguistic researchers agree on three text levels: basic linguistic surface structure, propositional textbase and main idea or situation model. Text structure or grammar is considered part of the highest text level, as it organizes ideas inherent in the text and the relationships among the ideas. Narrative and expository texts share these text levels. However, both text genres do not share text structure. Conventionally structured narrative texts commonly have the structure of introduction of the setting and the main figures, followed by action or conflict episodes and finally a resolution. In contrast, expository texts are known to be harder to understand than narrative texts (Mar et al., 2021), as more varied distinctive features have been used in the literature to describe types of expository text structures. Meyer et al. (1980) grouped expository text structures into six main types: compare-and-­ contrast (comparison), problem-and-solution (problem solving), cause-and-effect (causation), sequence (chronology), enumeration (collection, list), and description (categorization, generalization). Other researchers (Williams et al., 2009) have suggested two additional types of expository text structures: position-and-reason (persuasion/claim/support) and pro-and-con (argumentation). Each text structure type represents a distinct text organization and purpose. For example, compare-and-­ contrast text structures focus on the similarities or differences between ideas, things, or events, and problem-and-solution (problem solving) text structures focus on describing an unresolved issue and offering antidotes or solutions. Cause-and-effect text structures are used to describe how one event impacts another event, and sequence text structures are used to follow up how something changes over time. One prominent feature of expository texts is informativity. A text is informative if it provides new information to the recipient. The recipient does not always expect

6

1  Issues and Questions in Text Comprehension

the information, which may increase the intensity of processing, except for when the new information can build on the existing knowledge. Discontinuity due to missing information and discrepant (conflicting or contradictory) information hamper the processing of new information. In this case, recipients are unable to integrate the new information into their knowledge. Additional features that contribute to the informativity of a text concern the provision of facts, belief systems, (attitudes), and contradictions (negations).

1.4 Text Processing Authors of texts intend to communicate a message, produce statements accordingly and transform their intended message into a text format according to their own or prescribed standards. The recipient of a text processes the statements to understand the message. Comprehension is the goal in processing the received message. Text comprehension refers to the acquisition of knowledge and the integration of the information given in a text into a representation of the meaning of the message that goes beyond words and sentences as units being processed. The comprehender will acquire information by understanding the text but will also use previously stored knowledge and activate inference making such that text-level interpretations result. Text processing concerns all levels of a text: The linguistic surface structure must be decoded, which means perceiving, recognizing, and transposing the written graphemes of letters (in languages with alphabetic scripts) into mental representations of letters and their combinations, making up words with meaning. In the case of processing an orally presented text, phonemes must be heard, recognized, and must be combined to fixed groups of phonemes, i.e., the words, that must be identified for meaning. Syntactic structures are decoded as well. For example, morphemes, the smallest units of a language conveying meaning, may indicate, e.g., past tense of events when the regular past tense morpheme -ed is attached to regularly conjugated verbs in the English language. Processing the second level, the textbase, requires the construction of mental representations of content units from the decoded linguistic surface of text. These mental text units are commonly referred to as idea units or propositions. The same propositions may be derived from different surface formats: e.g., “The earthquake tore down the roof” or “The roof was damaged by the earthquake” may yield the proposition “Earthquake destroy roof.” Texts differ in the number of propositions that may be derived from the text surface. The construction of propositions constitute the textbase may reveal repeated or redundant idea units. For the highest text level, the propositions must be integrated into one or a few higher-order units, the main idea or the situation model. The predominant mechanism of the integration process is inference making. Integration is a process with an open end. One assumption—albeit a questionable one—is that integration ends when readers or listeners are satisfied with the situation model, they have

1.5  A Comprehensive View of Text Comprehension

7

constructed. This issue touches one research issue presented by the author of this book, in conjunction with the question of what mechanisms are involved in the integration process. Readers of a text actively decode the author’s message in the search for meaning or information; hence, they are not passive recipients who decode and store the text in order to retrieve it when necessary. Rather, readers actively interact with the text: They decide when to start processing and when to stop, when to skim a text or when to process it more deeply, the parts to which they allocate their attention and the parts to neglect, which parts they repeatedly process, and whether they start at the end instead of the beginning. The choice of one or more of these options depends on the task demands, the features of the text, and on individual differences of the reader. Listeners have fewer options, but they also actively process what they hear. In a discourse situation involving a speaker and a listener, both parties can decide on different options. Only when the speaker is not present, such as in most media, or when the speaker’s role prescribes a “rigid” presentation without interaction, are the listener’s conditions for text processing less favorable. Listeners to “rigid” presentations cannot ask the speaker to repeat parts or the whole text while listening, and they must adapt to the speed of the speaker’s speech. They cannot, independently of the speaker, skip parts to allow skimming. But listeners can independently decide on the parts of the presentation to which they allocate their intensified attention. Coherence is an essential feature to call a sequence of sentences a text. Coherence is the result of an integration process involving the propositions of the textbase. Coherence may be limited, involving just two or three sentences or only a paragraph. Ideally, coherence encompasses the whole text and requires proceeding to the next higher level of comprehension and the integration of all propositions into a main idea or situation model. If the integration is limited to adjacent sentences the term cohesion is frequently used. Shallow processing involves only the decoding and construction of the textbase, whereas deep processing has the aim of constructing a comprehensive situation model and is more effective for the retrieval and recall of a text.

1.5 A Comprehensive View of Text Comprehension Several component abilities and skills enable text comprehension (Perfetti & Adlof, 2012), which need to be identified and analyzed in terms of their interaction in order to gain insights into the development of text comprehension from early childhood on (Kendeou et al., 2009; Oakhill & Cain, 2012). Cognitive processes, motivation/ interest and self-regulation interact to contribute to comprehension. Theoretical perspectives offer models of text comprehension. In their book “Models of understanding texts,” Britton and Graesser (1996) compile five approaches to organize these perspectives. The authors understand these approaches (to which they refer as metaphors) as shared among all the models offered.

8

1  Issues and Questions in Text Comprehension

Comprehension may be viewed as: 1. composition of a mental representation on different levels; 2. construction of a coherent mental representation; 3. a dynamic complex system; 4. an organizing process in working memory; 5. a generator of inferences. Based on these five metaphors, Britton and Graesser derived an agenda for investigating reading comprehension and a definition of text understanding: “Text understanding is the dynamic process of constructing coherent representations and inferences at multiple levels of text and context, within the bottleneck of a limited capacity working memory” (p. 350). A first pioneering approach refers to an influential theoretical model first developed by Kintsch and van Dijk (1978; Kintsch, 1998), who proposed the seminal idea that texts should be analyzed on the three levels mentioned above: linguistic surface structure, propositional textbase and global conceptual situation model. In subsequent years, Kintsch elaborated the three-level text processing model into the construction-integration theory of text comprehension. This theory suggests that after decoding of online incoming information during reading, propositions are constructed within the attention span of working memory. Within this processing cycle, the constructed propositions are kept active for an immediate integration of the activated propositions to build an integrated meaning unit that represents local cohesion. The result of this integration may be transferred to the next processing cycle, thus initiating new integrations with current preliminary situation models until the final situation model is achieved at the end of the input. A related theoretical conception following similar premises is the structure-­ building framework by Gernsbacher (1990). She posits that incoming verbal sentence or text information is decoded and progressively integrated and structured until a main idea or situation model is constructed. Forthcoming elaborations of multilevel text comprehension theories will have to consider the full range of constraints set up by each level to reach the next level. For instance, one issue that has not yet been fully clarified is the question whether a situation model can be constructed without having processed the complete textbase. Gernsbacher herself gives a positive answer to that question. She claims that processing of critical concepts may suffice at times. Consideration of the linguistic surface structure as the one basic level of the mental representation of text is usually neglected within the cognitive theories of comprehension. However, reading research must emphasize decoding processes and pay attention to the way in which linguistic surface structures are mentally represented and influence the building of the propositional textbase and the situation model. With respect to multilingual text processing, the linguistic surface level is of greater importance then for monolingual native language processing, as second or more languages are commonly not processed at the same degree of automaticity as the first language.

1.5  A Comprehensive View of Text Comprehension

9

The second approach looks at comprehension as a process of construction, progressing toward a coherent mental representation. First, propositions are constructed from the linguistic surface structure. Coherence among the propositions is the essential feature of a text representation, as opposed to a list of sentences. The construction of text coherence during input may be achieved by processing coherence indicators in the text, such as overlapping arguments expressed by the same subject in two propositions or by causal conjunctions like “because” or “as” (e.g., Sanders et al., 2007). This second approach uses linguistic analyses during text processing and emphasizes explicit text information (e.g., Das & Taboada, 2018). Das and Taboada discovered a wide variety of signals other than the common discourse markers, such as reference, lexical, semantic, syntactic, and graphical features. Hence, the signalling of coherence relations is much more elaborated than previously thought. The third approach views comprehension as a complex dynamic system of connected propositions. Propositions serve as nodes in working memory. Working memory, with its limited capacity, regulates the activation of propositional nodes during text processing and the activation of nodes in a semantic network in long-­ term memory (e.g., van den Broek et al., 1996). And again, the activation of nodes from long-term memory supports the processing of information in working memory. The level of activation of nodes in both functional units, working memory and long-term memory, is determined by the progress of the comprehension process. Activation values reach an “end state” when text comprehension has reached a satisfactory state (according to the standards of the comprehender). Whether the activation pattern reflects “correct” understanding must be validated by within-text and external non-text criteria (Baker, 1985). A fourth approach focuses on the functions of working memory. Early on, Baddeley (1992) distinguished a passive storage capacity and an active working component. In both functions, upper capacity limits appear to constrain the efficiency of working memory (Just et al., 1996). The critical issue is the manner of processing sentence by sentence in working memory. Turner et al. (1996) suggested that straightforward statements without elaborations are kept in working memory, whereas Fletcher et  al. (1996) claimed that overlapping propositions, goals and intentions of the protagonists are preferred in working memory for further processing. The longer a proposition is kept in working memory, the more central it is for the final comprehension product. Gernsbacher et al. (2004) argued that difficulties in understanding may be caused by unimportant information that was not discarded from working memory. A tracking process results in a mental representation that is built on the most relevant textual information, with subsequent information being either relevant or irrelevant to the pre-existing foundation. Therefore, by gaining benefits from information easy to integrate, and minimizing the costs incurred by less relevant information, a skilled reader is able to filter through narrative details with relatively little effort. In this book more recent research findings will be reported on the role of working memory in text comprehension. The fifth approach emphasizes the role of inferences in the process of comprehension. Inferences are indispensable for creating local cohesion and global

10

1  Issues and Questions in Text Comprehension

coherence of text and are automatically generated during or after text input. But less automatic inferences are also generated, elicited by questions or by incompatible arguments or information in a text or when elaborations are required (e.g., Kendeou, 2015). A sixth approach takes the developmental perspective by stating that investigations of the components of comprehension do not reveal what developmental state these cognitive composite processes have reached at what age, making it difficult to determine what performance can be expected of children of different age levels (e.g., Verhoeven & van Leeuwe, 2008). Whatever the results of this approach may be, they support the design of efficient scaffolding and intervention strategies for developing comprehenders. A seventh approach focuses on the role of multiple languages in the development of text comprehension. This frame sets a rigid constraint on the scope of the research being presented in this book. Integrated into the literature report is the author’s own research on bi- and multilingual children’s comprehension of bilingual texts. These seven approaches may demonstrate that researchers and theorists emphasize different aspects of the complex field of text processing depending on children’s limited resources. This first chapter starts with outlining fundamental issues and the concepts and terms necessary to understand the important research results that illuminate the functions of the composite processes of text processing from all seven approaches. However, within the scope of this book, such an overview is only feasible if the presentation is further constrained to text comprehension and neglects text production. The focus needs to remain on multilingual text comprehension complementing the wealth of findings in “monolingual” text processing in the native language. The majority of these “monolingual” children are L1-dominant as the Western school system promotes L2 learning from early on. Multilingual text comprehension should elicit sufficient interest in the light of the contemporary multilingual situation in schools and other educational organizations and their need to support children’s learning.

1.6 Text Processing Modalities Listening and Reading Conceptualizations of how text comprehension functions commonly posit that the processes involved are shared by both modalities, the auditory and the visual. But even the most basic process, decoding of either written or oral texts, already deals with different language aspects depending on the modality of text presentation. Listening comprehension and reading comprehension have a different developmental history. The development of listening comprehension begins prenatally, and in typically developing individuals ends with death. This lifelong functioning (sometimes with strongly reduced capacity) leads to a highly automatized functioning of the composite processes of comprehension. Reading is acquired later in childhood, at around 5–6 years of age in most cultures. As such, less automatic functioning of the composite processes of reading comprehension are to be expected.

11

1.6  Text Processing Modalities Listening and Reading

Table 1.1  Text levels, aspects of mental representation, and basic composite processes of listening and reading comprehension Text level Low

Intermediate

High

Mental representation Linguistic surface:  Vocabulary  Syntax Textbase:  Micropropositions  Macropropositions Mental model:  Main idea/situation model

Composite processes Decoding Working memory Attention Local inference making Comprehension monitoring Global inferences Long-term memory Comprehension monitoring

Kim (2015, 2020) defined and tested the component processes of listening comprehension of text based on the three-level conception of text features originally put forward by Kintsch and van Dijk (1978). Successful text comprehension involves the construction of a coherent mental representation, referred to as main idea or situation model. According to the construction-integration model of text comprehension (Butcher & Kintsch, 2013; Kintsch & Kintsch, 2005), mental representation of text is composed of three text levels (see Table 1.1) already introduced: (a) the first and lowest level is the surface code or linguistic surface, which consists of words and phrases in the text itself; (b) the second level is the textbase representation, which consists of initial, elementary propositions as derived from sentences, clauses, or phrases in the text; and (c) the highest level is the situation model, which is a conceptualization in the format of a mental model of the situation inherent in the text. These levels of the mental text representation have been hypothesized to be hierarchical, insofar as the situation model is built upon the textbase representation, which is in turn based on the linguistic surface representation. Working memory is hypothesized as a foundational cognitive function that contributes to the textbase representation, but potentially also to the situation model. It involves simultaneous storage and active manipulation of information as early as preschool years (Gathercole et al., 2004) and even earlier. Working memory is also posited to support higher-level cognitive skills such as comprehension monitoring and inference making (e.g., Oakhill et al., 2005). Working memory is needed for the construction of initial propositions (i.e., the textbase representation) because it temporarily stores information from the text while processing the new incoming linguistic input and transferring it to the propositional textbase level. Furthermore, it is suggested that working memory plays an important role in constructing the situation model, as it is necessary to hold and manipulate information in order to integrate propositions and establish local cohesion, e.g., within a sentence and global coherence across sentences and text sections (Cain et al., 2004). Evidence suggests that working memory makes a unique contribution to both listening comprehension (Florit et  al., 2009) and reading comprehension for children in middle primary grades and upper grades as well as in adults (Chrysochoou et al., 2011).

12

1  Issues and Questions in Text Comprehension

The highest level of representation, the situation model, requires the construction of propositions as well as the integration of adjacent and remote propositions to build the textbase representation. The textbase representation contains initial, explicit, potentially coherent, propositions. Therefore, coherence among propositions has to be constructed, evaluated, and integrated to establish the coherent situation model. The processes of construction, evaluation and resolution of incoherence require high-level cognitive skills. Comprehension monitoring is hypothesized to be a higher-order cognitive skill required for constructing and evaluating a coherent situation model. Comprehension monitoring refers to the ability to reflect on and evaluate one’s own mental representation as far as the linguistic surface, the textbase and the situation model is concerned. Given that most studies use written materials and investigate reading comprehension, current understanding of how comprehension monitoring in the oral context is related to listening comprehension is limited. Several existing studies suggest that English-speaking children at the primary school level do not spontaneously detect or identify inconsistent information presented in oral texts (Kim & Phillips, 2014), and children with language impairment show poorer performance in oral comprehension monitoring than do typically developing children. Furthermore, in Kim and Phillips’ study oral language comprehension monitoring was found to be substantially related to standardized listening comprehension tasks for English-speaking children in kindergarten and the first grade (r = 0.43 to r = 0.55). Constructing a coherent situation model is likely to rely on high-order complex reasoning, because the situation model requires to make inferences across initial propositions to establish local cohesion within a sentence or between adjacent sentences and global coherence. Previous studies have suggested a unique contribution of inference skill to listening comprehension and reading comprehension (Kendeou et al., 2008; Tompkins et al., 2013). Additionally, holding a Theory of Mind, a skill to take another person’s perspective, might be important in terms of constructing the situation model beyond inference-making skills, because many narrative stories require the individual to understand beliefs, thoughts, and intentions of interlocutors, storytellers, and story characters (Kim, 2015, 2016). Therefore, not only inference making but also social cognitive aspects might contribute to establishing the situation model. These higher-level cognitive skills capture processes “beyond literal meaning of clauses and sentences” (Perfetti et  al., 2007, p.  230), and it is implied that they are built on fundamental, low-level skills. For instance, it has been assumed that word meaning retrieval is necessary for comprehension monitoring and inference making, and that syntactic knowledge helps children to detect and correct reading errors, thus facilitating comprehension monitoring (Oakhill et al., 2003). Moreover, it has been hypothesized that working memory is important for comprehension monitoring and that poor comprehenders have lower working memory capacity than do good comprehenders (Oakhill et al., 2005). Overall, the results of Kim’s (2015, 2016) studies partially support the hierarchical structure of language and cognitive processing—low-order language and cognitive skills predict high-order cognitive skills, which in turn predict discourse-level listening comprehension. These findings also confirmed previous studies which

1.6  Text Processing Modalities Listening and Reading

13

Fig. 1.1  Hierarchical structure and relations of foundational cognitive skills (working memory and attention) and language skills (vocabulary and syntax) and of higher-order cognitive skills (inferences and comprehension monitoring) with listening comprehension

demonstrated that semantic and syntactic language skills underpin comprehension monitoring (Oakhill et  al., 2003), and suggested that vocabulary and syntactic knowledge are indeed foundational language skills. However, vocabulary and syntactic knowledge were also directly related to listening comprehension. Therefore, vocabulary and syntactic knowledge are the foundational skills for constructing the propositions needed not only for higher-order cognitive skills including inference making and the construction of the situation model at the discourse level. The conceptualization of the hierarchical structure (see Fig.  1.1) was among other researchers suggested by Kim’s (2016) analysis of listening comprehension. Her analyses revealed that acquired syntactic skills did not affect listening comprehension directly, but vocabulary exerted a direct and indirect effect on listening comprehension indicated by the construction of a coherent situation model. In an indirect path, comprehension monitoring mediated the relationship of vocabulary with listening comprehension. In Kim’s (2016) analyses comprehension monitoring also mediated the relationship between working memory and listening comprehension. Moreover, the direct relationships of working memory with syntax and of vocabulary with listening comprehension are noteworthy. Working memory played a more central role than other factors she tested, e.g., attention and theory of mind, in that it directly related with listening comprehension and monitoring as well as vocabulary and syntax. In recent years, research on reading comprehension has grown immensely. Are the results found for listening comprehension comparable to those for reading comprehension? Most researchers would agree that they are and restrain their research to either the auditory or the visual modality. However, Wolf et al. (2019) questioned the comparability between reading and listening comprehension at least in childhood. In a study of second and third graders, they introduced both modalities, and

14

1  Issues and Questions in Text Comprehension

found modality-specific effects. Moreover, the study tested a broad range of performance: reading fluency, vocabulary, long-term and working memory, attention, and inhibition of interference (the latter will be discussed in more detail in later chapters of this book). The analyses revealed that reading comprehension explained 34% of the performance in listening comprehension, whereas performance in listening comprehension explained 40% of the performance in reading comprehension. Only reading fluency and vocabulary contributed significantly to this overlap. This finding allows the conclusion that vocabulary is crucial for comprehension in general. The other factors, working memory, long-term memory, attention, and inhibition to interference, exert modality-specific effects. The findings reported by Kim (2015, 2016) are consistent with the lexical quality hypothesis discussed by Perfetti and Stafura (2014), which states that an individual’s representational lexical quality, including semantic, syntactic, and morphological features of a word, is key to the integration of meaning in comprehension. Perfetti and Adlof (2012) presented an encompassing model of reading comprehension, in which reading comprehension has to decode the orthographic system of a language. However, besides the orthographic system processed in the visual modality by explicitly learned decoding of letters or characters, further processing in reading comprehension parallels that of listening comprehension. Understanding of text is accomplished by activating lower-level processes that are automatic and therefore without conscious control, and higher-level processes that enable comprehension by activating cognitive strategies such as monitoring and other cognitive processes including inference making, focusing on attention and inhibition of interference. The author of this book suggests a related general framework for listening comprehension to offer a tentative systematic overview of the processes that might be involved (see Fig.  1.2). The model is suitable also for younger children as listening comprehension skills instead of orthographic skills are considered, and it is extended to the functioning of multilingual persons and to multilingual contexts. The central process of coactivation of languages (see Chap. 2) coordinates the two or more language systems. Depending on the number of languages learned more language systems will be coactivated with varying intensity. Multiple interference will have to be suppressed or inhibited to accomplish tasks in one target language. The framework may serve as a guideline for the presentations of the following chapters. It is a descriptive framework and not an explanatory model of text comprehension. Attempts to develop explanatory models for text comprehension will be reported in Chaps. 6 and 7. The framework presents the basic processes of decoding of phonological units to the identification of words and the analyses of morphological units as prerequisites of higher-order cognitive processes of text comprehension. An additional preparatory process to comprehension is the processing of meaning and form that becomes relevant, e.g., in tonal languages like Chinese in which tonal patterns change the meaning of words that share all other features. Higher order comprehension processes include parsing, construction of propositions to build a textbase, and construction of a situation model with the support of inferences. Recall of the constructed mental representation of a text will be subject to influences

Fig. 1.2  A framework of processes contributing to listening comprehension

1.6  Text Processing Modalities Listening and Reading 15

16

1  Issues and Questions in Text Comprehension

of distance of recall time to input time and presumably to reorganization of the mental representation. Perfetti and Adlof (2012) describe a related model to Fig. 1.2 for reading comprehension in the native language.

1.7 The Contribution of Working Memory Listening and reading comprehension rely heavily on working memory, a functional unit in our memory system that is constrained to rather short-term storage. In contrast to short-term memory itself, though, working memory can manipulate the briefly stored content units. Working memory is theoretically conceived as consisting of several subfunctions, an important one being the central executive function. This central executive function regulates attention focusing, planning, inhibition, and the selection and temporary storage of relevant content units (Oberauer, 2009). Psychological theories and models of text comprehension are not conceivable without the central executive component of working memory. In later chapters, more information is provided about the specific task of working memory in multilingual text comprehension. Texts are processed sequentially in both modalities, the auditory and the visual. A limited sequence of incoming verbal material can be temporarily stored in working memory for approximately 20 s. During this time, the input must be integrated to form condensed units that are selectively transferred to the next input cycle or discarded, depending on whether the listener or reader is able to construct an integrated version of the input information. If the comprehender is able to integrate, a first local coherence is established. If integration is not accomplished within the time working memory can hold information, the new incoming information will either overwrite the previous information and new attempts will be made for integration, or the previous units will be transferred to long-term storage and from there it may be activated for the next integration cycle (Ericsson & Kintsch, 1995, see Chap. 5). Working memory development in childhood has often been attributed to the role of increasing knowledge. Cowan et  al. (2015) re-examined this role using short-­ term recognition memory of an item within an array of English letters or unfamiliar characters as a working memory task. In 7-year-old children, the advantage of known English letters over unfamiliar characters was smaller than it was for older children and adults. From third grade to adulthood, there was no further developmental change in letter knowledge, but there was still a large increase in working memory performance with regard to other working memory measures. These findings indicate that capacity, and not only knowledge or use of strategies, increases with age. Inhibition mechanisms attributed to the executive function play an important role in the activities of working memory. For example, a story heard for the first time may invoke (long-term) memories of a similar story previously heard. If these memories are activated for operations in working memory, they might be helpful during the integration process, but they may also hamper integration. Successful inhibition

1.9 The Contribution of Inferences

17

processes regulated by the executive control function will keep irrelevant information from disturbing the formation of a coherent mental representation of the text (e.g., Borella & de Ribaupierre, 2014).

1.8 The Contribution of Long-Term Memory and Knowledge Substantive knowledge stored in long-term memory plays a crucial role in text comprehension. Knowledge structures grow with experience and are therefore to a certain extent dependent on age. After all, more information can be gathered until the age of 20 years than until 10 years. With growing literacy, these natural experiences may be reflected in multiple text units. Best et al. (2008) report that world knowledge exerts a stronger effect on comprehension than does decoding ability. This effect is weaker, however, when one is attempting to understand expository texts, because these texts usually encompass a greater density of new information than do narrative texts. If children do not have sufficient related knowledge about the concepts and ideas described within a text, they will be unable to generate accurate inferences, thus limiting their comprehension. This link between knowledge and comprehension of expository texts has been supported by research both in adults and children in middle school and high school (e.g., Cook & O’Brien, 2014). In fact, several studies in children and adults have suggested that knowledge has a greater influence on the comprehension of expository texts than do reading skills (e.g., Kostons & van der Werf, 2015). In Chap. 7, a discussion of the theoretical models of text comprehension will elaborate on the role of knowledge in text comprehension in greater detail. Background knowledge stored in long-term memory serves as a constraint to text comprehension  (Kintsch, 1988). When text comprehension starts with decoding either orally or visually presented verbal material, the comprehension starts from the bottom up, working its way to the textbase and from there to the construction of the situation model. When background knowledge is retrieved either at the textbase or more extensively at the level of construction of the situation model, background knowledge will elicit inferences to link propositional units in order to create meaningful coherence in the form of a situation model, and thus constrain or elaborate the result of this process. Constraints or elaborations elicited by knowledge are indicators of top-down processing in comprehension. Bottom-up and top-down processing interact when individuals comprehend texts (Kintsch, 2005).

1.9 The Contribution of Inferences Inferences connect propositions when a text is not sufficiently explicit. Propositions consist of the basic elements Argument 1 – Predicate – Argument 2. The sentences “The boy lit the candle” and “Suddenly, the napkin burned” may be transferred to

18

1  Issues and Questions in Text Comprehension

the propositions “Boy (Argument 1) lit (Predicate) candle (Argument 2),” “Napkin (Argument 1) burned (Predicate) suddenly (Argument 2).” The propositions have three basic elements. But a proposition may also lack Argument 2 or Argument 1. An inference may bridge the two statements by adding a causal relationship between them: “Lighting the candle caused the burning of the napkin.” Further elaborative inferences might construct a situation model containing, e.g., “The boy’s unskilful handling of the burning match” or “the boy dropped the match because the fire had burned his fingers.” The situation model might then be “Unskilful boy drops the match with which he wanted to light a candle on a paper napkin that started to burn.” The role of inferences for integration is obvious. Mental propositions and their integration with the support of inferences affect performance in text comprehension. Propositions and inferences that integrate them allow the construction of local cohesion between adjacent text elements, as shown in the causal relation of the previous example, and of global coherence in the main idea or situation model, as seen in the elaborative construction process in the example. The literature reports substantive support for the mental reality of the propositions and the inferences (e.g., Kintsch, 1998). Hence, inferences serve to close information gaps in a text. They are mostly elicited automatically by the gaps, but they may also be applied in a controlled strategic way. Inferences may be directed “backward,” as in the example above, in which a causal relationship was assumed for a past event sequence. But they may also be directed toward future events involving expectations and anticipations (e.g., “After the accident the boy will be more cautious when lighting matches in the future”). After the construction of propositions, the next level of integration is represented in macropropositions. When several basic propositions (micropropositions) are integrated by inferences to build a macroproposition that tells the same content on a more condensed abstract level. This is a decisive second step to achieve local cohesion in comprehension. The construction of macropropositions condenses the content as is usually accomplished in summaries. In a story included in a previous experiment of the author (Schönpflug & Küpping-Faturikova, 2020), several incidences are summarized in a macroproposition: “The dog ran out of the door, then came in and ran out again” was integrated into “The dog ran excitedly in and out of the door.” An inference was made from the reported behavior to the inner state of the dog. The efficiency of inferences can be detected by priming effects. If text input is preceded by a concept (e.g., “play”) and followed by the two sentences above (“The boy lit the candle” and “Suddenly, the napkin burned”), a bridging inference between the two sentences has increased the probability of causally connecting the two sentences and attributing the cause to the boy’s playing around with the lit match. Inferences are especially noticeable when a text is retrieved for free recall or retelling. In a free recall protocol, inferences are usually made explicit, mostly either verbally or visually, and their features may be analyzed. A decisive issue picked up in this book refers to when inferences arise. Most text comprehension researchers claim that inferences emerge during text input, e.g.,

1.10  Comprehension Monitoring

19

Kintsch (1998). Stressing the usefulness of retell or summary implies, however, that additional inferences are expected to emerge during retrieval and recall. This unclarified issue inspired the author to execute research which will be reported among other approaches in this monograph. The emergence of global coherence by constructing a situation model, preferably of the whole text, requires an intensive process of inferencing. One text may lead to more than one situation model. Expository or informative texts tend to have more than one main idea, as they usually convey more than one central piece of information. Forthcoming chapters will delve more deeply into situation model construction.

1.10 Comprehension Monitoring Comprehension monitoring is a skill that has important educational implications. The degree to which children observe their own comprehension of heard or read texts has therefore been analyzed by many investigators. Monitoring as a form of metacognition can be defined as conscious and deliberate thoughts about an individual’s behavior, emotions, and other thoughts (Flavell, 1979). One research issue addresses the extent to which the self-regulated, inquiry-centered, and explanation-­ centered learning processes are consciously detected, monitored, and regulated, as opposed to being relatively unconscious, automatic procedures. Educators have long since argued that students should keep track of their understanding and should inquire appropriately if they detect a failure of understanding. However, more recently, empirical studies found that children are surprisingly unlikely to carry out such evaluative and regulatory activities (e.g., Martin et al., 2016). Even older students and those identified as better readers show considerable potential for improvement. Studies of comprehension monitoring employ the basic paradigm of introducing an error or a problem into a passage of prose and assessing whether the subject is able to detect it. The underlying rationale for this paradigm is that if subjects are keeping a careful check on their understanding, they should notice the error or problem. However, the fact that subjects typically are not informed in advance that a passage is problematic may lead to a serious underestimation of their comprehension monitoring activities, as people tend to make sense of any input, however confusing (see Baker, 1985). Good comprehension is, of course, only partly due to the efficiency of monitoring during text input. Incoming information is integrated and evaluated online. Indicators such as the “feeling of knowing” new information, or evaluations such as “this is difficult,” “that contradicts what I heard before” or “I need to know more about that” point to monitoring or control processes during comprehension (Martin et al., 2016). Most relevant for education is an aspect of metacomprehension called “judgment of learning” (Nguyen & McDaniel, 2015). Nguyen and McDaniel documented that comprehenders’ evaluations of not having well understood a passage in a text or the whole text led to increased efforts of learning, e.g., allocation of more study time. Hence, judgments of how well one has learned given material is a potential motivating factor for learning.

20

1  Issues and Questions in Text Comprehension

Calibration is a common aspect of monitoring and refers to the degree to which individuals can provide judgments about their accuracy that correspond to an objective measure of their actual performance, such as the comprehension score in a test. Individuals whose performance judgments closely correspond to their actual performance can be well calibrated. To obtain measures of absolute monitoring accuracy, confidence judgments are plotted against the actual proportion of correct answers to the task questions. An alternative is to study relative monitoring accuracy by measuring metacognitive discrimination, i.e., how good a person is at distinguishing between correct and incorrect answers (Roebers & Spiess, 2017). Over the course of development, it appears that with increasing age, monitoring incorrect answers becomes more demanding and more subject to improvements compared to the monitoring of correct answers (Lyons & Ghetti, 2011). Mirandola et al. (2018) report that regardless of age, children with better reading comprehension also show greater accuracy in their metacognitive judgments.

1.11 Strategies to Improve Text Comprehension Approximately three decades ago, the teaching of text comprehension strategies to enhance the understanding of texts intensified. Psychologists, linguists, and educational scientists still invest efforts in this regard to the present day. Numerous studies have demonstrated that adequate comprehension can be trained if children have sufficient cognitive resources (Gersten et al., 2001). In Chap. 8, a detailed discussion will introduce the reader to common comprehension strategies and their efficiency for typically developing children as well as children struggling to comprehend age-appropriate texts. At this point it may suffice to briefly introduce the concept of strategy and distinguish it from comprehension skills. Text comprehension is an important skill to reach personal fulfilment, e.g., when reading poetry and to achieve both academic and professional success. A skill refers to the individual capacity to successfully cope with a task, here a comprehension task. Skills have a broader scope of applications than do strategies and are activated by the demands of the task. A strategy is a learned specific way of coping with the challenges of a comprehension task and may be called upon deliberately. Comprehension skill is not a unitary cognitive skill, but rather comprises several skill composites such as constructing main ideas, detecting, and ordering sequences of actions or events, identifying cause and effect relations, drawing inferences or conclusions, monitoring, controlling, and evaluating comprehension and so forth. Comprehension strategies for texts include plans and actions like generating questions to stimulate construction of coherence, predicting outcomes, summarizing information, recalling, and retelling, activating prior knowledge, or identifying text structure. Strategies are critical for designing intervention programs for text comprehension. They are readily trainable in an educational context. Effective intervention programs therefore require comprehensive knowledge about text comprehension

1.12  Intelligence and Comprehension

21

and what and how strategies may promote text comprehension. Chapter 8 discusses the possibilities and weaknesses related to the training of text comprehension strategies.

1.12 Intelligence and Comprehension A selection of cognitive abilities is subsumed under the concept of intelligence, among them the comprehension of text. General intelligence is assessed by means of standardized IQ (intelligence quotient) tests, most of which encompass some of the following subtests: performance scores for general world knowledge (long-term memory), concept formation and definition, breadth of vocabulary (synonyms), logical inferences (analogy, syllogism), arithmetic tasks, construction of a meaningful story from a series of pictures, listening/reading a text and recalling it, working memory performance (short-term memory of recalling a series of numbers forward and backward or other manipulations) as well as structure-building abilities in a non-verbal domain (constructing a mosaic pattern), and puzzle-like tasks. A look at the effects of general intelligence assessed with a selection of the above listed standardized scales of performance scores in intelligence subtests shows that text comprehension benefits from average and above-average general intelligence. Research by Tiu et al. (2003) combined results on the effects of general IQ on the two modalities of text comprehension: listening and reading. Using multiple regression, the authors observed different sizes of a common impact score: The significant effect of general intelligence on listening comprehension was three times higher than the impact of general intelligence on reading comprehension. General intelligence exerts an influence on text comprehension in both modalities, but a much weaker one when comprehending a written text by reading it. In the light of an assumed comparability of listening and reading comprehension, this finding seems puzzling at first glance. However, the authors’ complete path analyses revealed that the relationship between general intelligence and reading comprehension is mediated by a lower-level decoding ability. This mediated relationship was found to be highly significant for a group of third- and fourth graders with reading deficits, but only weakly significant for age-typical readers. Hence, decoding ability separates struggling readers from typically developed ones, but is irrelevant for listening comprehension. In other words, general intelligence is relevant for processing in either modality when decoding skills are controlled for. A few other tests based on verbal material appear most relevant for text comprehension. Some component abilities of intelligence affect text comprehension more than others (e.g., Zampini et al., 2013). Zampini et al. found no significant relation between a non-verbal intelligence test such as Raven’s Progressive Matrices (Raven, 2000), which calls for logical inferences based on non-verbal material, and text comprehension or text production. The Wechsler Intelligence Test for Children (WISC-III; Wechsler, 1991) is based on a two-factor theory of intelligence: verbal intelligence and performance

22

1  Issues and Questions in Text Comprehension

(non-­verbal) intelligence. In studies with German primary school children (ages 8–10 years) by Schönpflug (2005, 2008), no significant effect of non-verbal intelligence on story comprehension was found. However, verbal intelligence showed a moderate effect, which was mainly due to a subtest assessing world knowledge and common sense. Comprehension was assessed by 10 questions targeting the main idea of the text, which the children had to answer after listening to a story. The author found a significant effect of general intelligence on the number of correctly recalled propositions immediately after listening, but not 1 week later. Specific intelligence deficits may reveal the function of components of general intelligence while processing text. According to Wong and Wong (1986), even mild deficits hamper the understanding of texts, because children with mild deficits use different strategies to comprehend texts than do typically developing children. In their study, Wong and Wong included fifth and seventh grade students with good and moderate school achievements and a group of students with mild learning deficits. The students studied short, well-structured, or unorganized texts with frequent or infrequent vocabulary. The students with learning deficits spent more time studying the text with difficult vocabulary, whereas the more capable students spent more time studying the unorganized text. Although some other studies confirm that students with and without learning deficits process narrative and expository texts in a similar way, they also report that these groups use different strategies to process a text. Students with learning deficits concentrate their efforts at the textbase level, whereas typically developing children concentrate on the macronarrative or discourse level. These two processing modes are known as shallow versus deep processing, an issue followed up in Chap. 7. Vauras et al. (1994) argued that children with weak comprehension of texts have fewer resources available to initiate deep processing such as meaningful elaboration. These children struggle with decoding and construction of micropropositions, leaving limited cognitive resources for higher-order integrative processing leading to a coherent mental text representation. Moreover, students with learning deficits show more problems in understanding expository texts than narrative ones. Padeliadu and Antoniou (2014) posit that difficulties emerge because students with learning deficits struggle with the structure of the mental representation they must construct. The structure of an expository text is harder to capture than that of narratives because the conceptual density is higher, and students are less familiar with the concepts and the academic language. These processes leave little resources for structure building and constructing main ideas. Botsas (2017) identified a lack of strategies used by poor comprehenders that support deep processing like metacognitive comprehension monitoring, regulative and elaborative strategies; instead, students with learning deficits prefer verbatim repetition of texts or stay close to the reproduction of the propositional textbase. The literature also includes studies that found no influence of intelligence on text comprehension: In their longitudinal study, Florit and Levorato (2012) highlighted the importance of passive vocabulary for listening comprehension in preschool-age children. General intelligence, and even verbal intelligence played no significant role, either directly or as a mediating variable.

References

23

1.13 Concluding Remarks The framework for processes leading to text comprehension and the psychological functions involved, highlight the many facets involved in text processing. Text comprehension is a complex process per se and any analysis involving language switches within one text or between text input and text recall by multilingual persons is still more complex. The majority of theories of text comprehension draw a distinction between the three text levels: linguistic surface structure, the textbase, and the main idea or situation model. The question of whether multilingual task conditions or the degree of multilingual proficiency affect all three levels will be examined throughout this book. Most researchers agree that besides two sensory systems, vision and hearing, working memory and long-term memory are activated when text processing is required. Both of these memory units enable inferences and strategies that contribute to the integration of text information, aiming at the construction of a main idea or situation model. However, the most pressing question to be resolved by researchers of multilingual text processing is how coactivation of languages while comprehending a text relates to the text levels and the psychological functional units represented in the framework. The selective review of issues pertaining to text processing provided in the first chapter presents a research area that reaches beyond the psychological, the cognitive and linguistic, and the educational perspective: Literature and neuroscience may gain and provide insights that could enlighten all those who work with texts. The contents of this book stretch the basic issues of text processing listed here to issues related to multilingual texts to be understood by multilingual individuals. However, this book will refrain from addressing issues related to multilingualism associated with different scripts, as its emphasis is on listening comprehension as well as reading comprehension. The author posits that it is first necessary to detect fundamental interdisciplinary research issues and findings in the domain of text processing in order to understand the problems and questions related to multilingual text processing. These issues are reserved for later chapters and will be discussed in detail in view of the many facets of the relationship between multilingualism and text comprehension.

References Baddeley, A. (1992). Working memory: The interface between meaning and cognition. Journal of Cognitive Neuroscience, 4(3), 281–288. https://doi.org/10.1162/jocn.1992.4.3.281 Baker, L. (1985). How do we know when we don’t understand? Standards for evaluating comprehension. In D.-L. Forrest-Pressley, E. McKinnon, & T. G. Waller (Eds.), Metacognition, cognition and human performance (pp. 155–205). Academic Press. Best, R.  M., Floyd, R.  G., & McNamara, D.  S. (2008). Differential competencies contributing to children’s comprehension of narrative and expository texts. Reading Psychology, 29(2), 137–164. https://doi.org/10.1080/02702710801963951

24

1  Issues and Questions in Text Comprehension

Borella, E., & de Ribaupierre, A. (2014). The role of working memory, inhibition, and processing speed in text comprehension in children. Learning and Individual Differences, 34, 86–92. https://doi.org/10.1016/j.lindif.2014.05.001 Botsas, G. (2017). Differences in strategy use in the reading comprehension of narrative and science texts among students with and without learning disabilities. Learning Disabilities: A Contemporary Journal, 15(1), 139–162. Special Issue: Creating Argument. Britton, B. K., & Graesser, A. C. (Eds.). (1996). Models of understanding text (1st ed.). Lawrence Erlbaum. Butcher, K. R., & Kintsch, W. (2013). Text comprehension and discourse processing. In I. B. Weiner, A. F. Healy, & R. W. Proctor (Eds.), Handbook of psychology (Vol. 4, pp. 578–604). Wiley. Cain, K., & Oakhill, J. V. (2007). Children comprehension problems in oral and written language. A cognitive perspective. Guilford Press. Cain, K., Oakhill, J.  V., & Bryant, P.  E. (2004). Children’s reading comprehension ability: Concurrent prediction by working memory, verbal ability, and component skills. Journal of Educational Psychology, 96(1), 31–42. https://doi.org/10.1037/0022-­0663.96.1.31 Chrysochoou, E., Bablekou, Z., & Tsigilis, N.  S. (2011). Working memory contributions to reading comprehension components in middle childhood children. The American Journal of Psychology, 124(3), 275–289. https://www.jstor.org/stable/10.5406/amerjpsyc.124.3.0275 Cook, A. E., & O’Brien, E. J. (2014). Knowledge activation, integration, and validation during narrative text comprehension. Discourse Processes, 51(1–2), 26–49. https://doi.org/10.108 0/0163853x.2013.855107 Cowan, N., Ricker, T. J., & Clark, K. M. (2015). Knowledge cannot explain the developmental growth of working memory. Developmental Science, 18(1), 132–145. https://doi.org/10.1111/ desc.12197 Das, D., & Taboada, M. (2018). Signaling of coherence relations in discourse, beyond discourse markers. Discourse Processes, 55(8), 743–770. https://doi.org/10.1080/0163853X. 2017.1379327 Ericsson, A., & Kintsch, W. (1995). Long-term working memory. Psychological Review, 102(2), 211–245. https://doi.org/10.1037/0033-­295x.102.2.211 Flavell, J.  H. (1979). Metacognition and cognitive monitoring. A new area of cognitive-­ developmental inquiry. American Psychologist, 34(10), 906–911. Special Issue: Psychology and Children: Current Research and Practice. https://doi.org/10.1037/0003-­066X.34.10.906 Fletcher, C. R., van den Broek, P., & Arthur, E. J. (1996). A model of narrative comprehension and recall. In B. K. Britton & A. C. Graesser (Eds.), Models of understanding text (1st ed., pp. 141–163). Lawrence Erlbaum. Florit, E., & Levorato, M. C. (2012). A longitudinal study on listening text comprehension and receptive vocabulary in preschoolers. Rivista di Psicolinguistica Applicata, 12(1–2), 65–79. Florit, E., Roch, M., Altoè, G., & Levorato, M. C. (2009). Listening comprehension in preschoolers: The role of memory. Journal of Developmental Psychology, 27(4), 935–951. https://doi. org/10.1348/026151008X397189 Gathercole, S. E., Pickering, S. J., Ambridge, B., & Wearing, H. (2004). The structure of working memory from 4 to 15 years of age. Developmental Psychology, 40(2), 177–190. https://doi. org/10.1037/0012-­1649.40.2.177 Gernsbacher, M. A. (1990). Language comprehension as structure building. Lawrence Erlbaum Associates. Gernsbacher, M. A., & Kaschak, M. P. (2003). Neuroimaging studies of language production and comprehension. Annual Review of Psychology, 54, 91–114. https://doi.org/10.1146/annurev. psych.54.101601.145128 Gernsbacher, M. A., Robertson, R. A. W., Palladino, P., & Werner, N. K. (2004). Managing mental representations during narrative comprehension. Discourse Processes, 37(2), 145–164. https:// doi.org/10.1207/s15326950dp3702_4 Gersten, R., Fuchs, L. S., Williams, J. P., & Baker, S. (2001). Teaching reading comprehension strategies to students with learning disabilities: A review of research. Review of Educational Research, 71(2), 279–320. https://doi.org/10.3102/00346543071002279

References

25

Graesser, A. C., Olde, B., & Klettke, B. (2013). How does the mind construct and represent stories? In M. C. Green, J. J. Strange, Brock, & C. Timothy (Eds.), Narrative impact: Social and cognitive foundations (pp. 229–262). Psychology Press. Just, M. A., Carpenter, P. A., & Keller, T. A. (1996). The capacity theory of comprehension: New frontiers of evidence and arguments. Psychological Review, 103(4), 773–780. https://doi.org/1 0.1037/0033-­295X.103.4.773 Kendeou, P. (2015). A general inference skill. In E. J. O’Brien, A. E. Cook, & R. F. Lorch (Eds.), Inferences during reading (pp. 160–181). Cambridge University Press. Kendeou, P., Bohn-Gettler, C., White, M.  J., & van den Broek, P. (2008). Children’s inference generation across different media. Journal of Research in Reading, 31(3), 259–272. https://doi. org/10.1111/j.1467-­9817.2008.00370.x Kendeou, P., van den Broek, P., White, M. J., & Lynch, J. S. (2009). Predicting reading comprehension in early elementary school: The independent contributions of oral language and decoding skills. Journal of Educational Psychology, 101(4), 765–778. https://doi.org/10.1037/a0015956 Kim, Y.-S. (2015). Language and cognitive predictors of task comprehension: Evidence from multivariate analysis. Child Development, 86(1), 128–144. https://doi.org/10.1111/cdev.12293 Kim, Y.-S. (2016). Direct and mediated effects of language and cognition skills on comprehension of oral narrative texts for children. Journal of Experimental Child Psychology, 141(1), 101–120. https://doi.org/10.1016/jecp.2015.08.003 Kim, Y.-S. (2020). Hierarchical and dynamic relations of language and cognitive skills to reading comprehension: Testing the Direct and Indirect Effects Model of Reading (DIER). Journal of Educational Psychology, 112(4), 667–684. https://doi.org/10.1037/edu0000407 Kim, Y.-S., & Phillips, B. (2014). Cognitive correlates of listening comprehension. Reading Research Quarterly, 49(3), 269–281. https://doi.org/10.1002/rrq.74 Kintsch, W. (1988). The role of knowledge in discourse comprehension: A construction-integration model. Psychological Review, 95(2), 163–182. https://doi.org/10.1037/0033-­295X.95.2.163 Kintsch, W. (1998). Comprehension: A paradigm for cognition. Cambridge University Press. Kintsch, W. (2005). An overview of top-down and bottom-up effects in comprehension: The CI perspective. Discourse Processes, 39(2&3), 125–128. https://doi.org/10.1207/ s15326950dp3902&3_2 Kintsch, W., & Kintsch, E. (2005). Comprehension. In S. G. Paris & S. A. Stahl (Eds.), Children’s reading comprehension and assessment (pp. 71–92). Lawrence Erlbaum Associates. Kintsch, W., & van Dijk, T. (1978). Toward a model of text comprehension and production. Psychological Review, 85(5), 363–394. https://doi.org/10.1037/0033-­295x.85.5.363 Kostons, D., & van der Werf, G. (2015). The effects of activating prior topic and metacognitive knowledge on text comprehension scores. British Journal of Educational Psychology, 85(3), 264–275. https://doi.org/10.1111/bjep.12069 Levelt, W. J. M. (1993). Speaking: From intention to articulation. MIT Press. Lyons, K. E., & Ghetti, S. (2011). The development of uncertainty monitoring in early childhood. Child Development, 82(6), 1778–1787. https://doi.org/10.1111/j.1467-­8624.2011.01649 Mar, R. A., Li, J., Nguyen, A. T. P., & Ta, C. P. (2021). Memory and comprehension of narrative versus expository texts: A meta-analysis. Psychonomic Bulletin & Review, 28(3), 732–749. https://doi.org/10.3758/s13423-­020-­01853-­1 Martin, N.  D., Nguyen, K., & McDaniel, M.  A. (2016). Structure building differences influence learning from educational text: Effects on encoding, retention, and metacognitive control. Contemporary Educational Psychology, 46(July), 52–60. https://doi.org/10.1016/j. cedpsych.2016.03.005 Meyer, B. J. F., Haring, M., Brandt, D. M., & Walker, C. M. (1980). Comprehension of stories and expository texts. Poetics, 9(1), 203–211. Mirandola, C., Ciriello, A., Gigli, M., & Cornoldi, C. (2018). Metacognitive monitoring of text comprehension: An investigation on postdictive judgments in typically developing children and children with reading comprehension difficulties. Frontiers in Psychology, 9, Article 2253. https://doi.org/10.3389/fpsyg.2018.02253

26

1  Issues and Questions in Text Comprehension

Nguyen, K., & McDaniel, M. A. (2015). The JOIs of text comprehension. Journal of Experimental Psychology: Applied, 21(4), 1–13. https://doi.org/10.1037/xap0000066 Oakhill, J. V., & Cain, K. (2012). The precursors of reading ability in young readers: Evidence from a four-year longitudinal study. Scientific Studies of Reading, 16(2), 91–121. https://doi. org/10.1080/10888438.2010.529219 Oakhill, J. V., Cain, K., & Bryant, P. E. (2003). The dissociation of word reading and text comprehension: Evidence from component skills. Language and Cognitive Processes, 18(4), 443–468. https://doi.org/10.1080/01690960344000008 Oakhill, J. V., Hartt, J., & Samols, D. (2005). Levels of comprehension monitoring and working memory in good and bad comprehenders. Reading and Writing, 18(7–9), 657–686. https://doi. org/10.1007/s11145-­005-­3355-­z Oberauer, K. (2009). Design for a working memory. In B. H. Ross (Ed.), Learning and motivation (pp. 45–100). Elsevier (Academic Press). Padeliadu, S., & Antoniou, F. (2014). The relationship between reading comprehension, decoding, and fluency in Greek: A cross-sectional study. Reading and Writing Quarterly, 30(1), 1–31. https://doi.org/10.1080/10573569.2013.758932 Perfetti, C. A., & Adlof, S. M. (2012). Reading comprehension: A conceptual framework from word meaning to text meaning. In J. Sabatini & E. Albro (Eds.), Assessing reading in the 21st century: Aligning and applying advances in the reading and measurement sciences (pp. 3–20). Rowman & Littlefield Education. Perfetti, C.  A., & Stafura, J. (2014). Word knowledge in a theory of reading comprehension. Scientific Studies of Reading, 18(1), 22–37. https://doi.org/10.1080/10888438.2013.827687 Perfetti, C. A., Landi, N., & Oakhill, J. V. (2007). The acquisition of reading comprehension skill. In M. J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 227–247). Wiley. Pickering, M.  J., & Garrod, S. (2013a). An integrated theory of language production and comprehension. Behavioral and Brain Sciences, 36(4), 329–347. https://doi.org/10.1017/ S0140525X12001495 Pickering, M. J., & Garrod, S. (2013b). How tightly are production and comprehension interwoven? Frontiers in Psychology, 4, Article 238. https://doi.org/10.3389/fpsyg.2013.00238 Raven, J. (2000). The Raven’s progressive matrices: Change and stability over culture and time. Cognitive Psychology, 41(1), 1–48. https://doi.org/10.1006/cogp.1999.0735 Roebers, C. M., & Spiess, M. (2017). The development of metacognitive monitoring and control in second graders: A short-term longitudinal study. Journal of Cognition and Development, 18(1), 110–128. https://doi.org/10.1080/15248372.2016.1157079 Sanders, T., Land, J., & Mulder, G. (2007). Linguistic markers of coherence improve text comprehension in functional contexts. Information Design Journal, 15(3), 219–235. Special Issue: Discourse, Cognition and Communication. https://doi.org/10.1075/idj.15.3.04san Schönpflug, U. (2005). Wörtliches und inhaltliches Erinnern gehörter Erzählungen bei Kindern [Children’s verbatim and content memory of heard stories]. In K.-P. Dahle & R. Volbert (Eds.), Entwicklungspsychologische Aspekte der Rechtspsychologie [Developmental aspects of law psychology] (pp. 272–278). Hogrefe. Schönpflug, U. (2008). Pauses in elementary school children’s verbatim and gist free recall of a story. Cognitive Development, 23(3), 385–394. https://doi.org/10.1016/j.cogdev.2008.05.002 Schönpflug, U., & Küpping-Faturikova, L. (2020). Bilingual children’s story comprehension across language boundaries: Does retelling improve comprehension? International Journal of Bilingual Education and Bilingualism, 23(6), 657–675. https://doi.org/10.1080/1367005 0.2018.1479727 Silbert, L. J., Honey, C. J., Simony, E., Poeppel, D., & Hasson, U. (2014). Coupled neural systems underlie production and comprehension of naturalistic narrative speech. Proceedings of the National Academy of Sciences of the United States of America PNAS, 111, Article 43, E4687– E4696. https://www.pnas.org/cgi/doi/10.1073/pnas.1323812111 Tiu, R.  D., Jr., Thompson, L.  A., & Lewis, B.  A. (2003). The role of IQ in a component model of reading. Journal of Learning Disabilities, 36(5), 424–436. https://doi. org/10.1080/02702710500468724

References

27

Tompkins, V., Guo, Y., & Justice, L. M. (2013). Inference generation, comprehension, and language skills in the preschool years. Reading and Writing, 26(3), 403–429. https://doi.org/10.1007/ s11145-­012-­9374-­7 Trabasso, T., & Nickels, M. (1992). The development of goal plans of action in the narration of a picture story. Discourse Processes, 15(3), 249–275. https://doi.org/10.1080/01638539209544812 Turner, A., Britton, B.  K., Andraessen, P., & McCutchen, D. (1996). A predication semantics model of text comprehension. In B. K. Britton & A. C. Graesser (Eds.), Models of understanding text (pp. 33–71). Lawrence Erlbaum. van den Broek, P., Risden, K., Fletcher, C. R., & Thurlow, R. (1996). A “landscape” view of reading: Fluctuation, patterns of activation and the construction of a stable memory representation. In B. K. Britton & A. C. Graesser (Eds.), Models of understanding text (1st ed., pp. 165–187). Lawrence Erlbaum. Vauras, M., Kinnunen, R., & Kuusela, L. (1994). Development of text-processing skills in high-, average-, and low-achieving primary school children. Journal of Reading Behavior, 26(4), 361–389. Verhoeven, L., & van Leeuwe, J. (2008). Prediction of the development of reading comprehension: A longitudinal study. Applied Cognitive Psychology, 22(3), 407–423. https://doi.org/10.1002/ acp.1414 Wechsler, D. (1991). The Wechsler Intelligence Scale for Children (WISC III) (3rd version). The Psychological Corporation. Williams, J. P., Stafford, K. B., Lauer, K. D., Hall, K. M., & Pollini, S. (2009). Embedding reading comprehension training in content area instruction. Journal of Educational Psychology, 101(1), 1–20. https://doi.org/10.1037/a0013152 Wolf, M. C., Muijselaar, M., Boonstra, A. M., & de Bree, E. H. (2019). The relationship between reading and listening comprehension: Shared and modality-specific components. Reading and Writing, 32(7), 1747–1767. https://doi.org/10.1007/s11145-­018-­9924-­8 Wong, B., & Wong, R. (1986). Study behavior as a function of metacognitive knowledge about critical task variables: An investigation of above average, average, and learning-disabled readers. Learning Disabilities Research, 1(2), 101–111. Zampini, L., Suttora, C., D’Odorico, L., & Zanchi, P. (2013). Sequential reasoning and listening text comprehension in preschool children. European Journal of Developmental Psychology, 10(5), 563–579. https://doi.org/10.1080/17405629.2013.766130

Chapter 2

Multiple Language Use and Text Comprehension

2.1 Multilingual Decoding and Encoding The first chapter introduced to several composite processes of text comprehension. This chapter will focus on the role of language, specifically multiple languages, in the functioning of some composite processes of text comprehension. The acquisition of additional languages to an individual’s native language, and the usage of these languages, may change the way in which mental representations of texts are constructed. Languages are first represented by the linguistic surface structure. Phonemes in cases of oral text input and grapheme units in cases of written input are partially shared by two or more languages when they belong to the same language family. The lexicon or vocabulary, the morphology, and the syntax are to a large degree language-specific. Thus, to uncover effects of multiple language use on text comprehension, one must start with the processing of the lowest level of text representation: multilingual decoding ability. When text processing is assessed by free recall, also encoding processes come into play. Two aspects of multilingualism will be discussed in the following sections: the influence of an individual’s multilingual proficiency and the influence of multilingual tasks on text comprehension. The overarching developmental perspective of this book narrows down the selection of findings to those that are essential for understanding multilingual children’s text comprehension.

2.1.1 Decoding Text processing begins with the discrimination of sounds (phonemes) constituting the words of a text. Processing of the phonemes builds up phonological awareness, an ability to recognize and distinguish the phoneme units of a language. At © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 U. Schönpflug, Multilingual Text Comprehension, The Bilingual Mind and Brain Book Series, https://doi.org/10.1007/978-3-031-43341-2_2

29

30

2  Multiple Language Use and Text Comprehension

preliterate age, phonological awareness is assessed by finding or recognizing rhymes to words, phoneme repetition at the beginning, middle or end of words, and syllable identification. Phonological awareness or ability is a prerequisite for decoding, which refers to the ability to recognize the smallest sound units and their combination into meaningful words of a language. After acquisition of literacy children know how to match phonemes to (written) graphical symbols in a script. The script can be alphabetic, with letters, or have other graphemes or characters. Hence, depending on the modality, auditory or written, decoding converts sounds into the phoneme repertoire of a language or converts graphic symbols into the grapheme system that contains, for example, the letters of an alphabet. Following the multilevel model of text comprehension presented in Table 1.1, comprehension is prepared while listening, for example, to a story, and single phonemes and phoneme groupings are recognized and distinguished, i.e., decoded, and associated with learned meaning. In cases of multilingual speakers phonemes with their specific features and sequential order, must be identified as belonging to words of a specific language. By decoding constituents of words and word order, the decoder recognizes their syntactic structure. Decoding of phoneme sequences presented in word units during listening is highly automatized in the overlearned native or first language heard from the prenatal stage onwards. The acquisition of the lexicon of a language (vocabulary) supports the decoding of phonemes or graphemes, because in the case of familiar frequent words, it is already possible to recognize the whole word even before all phonemes of the word have been decoded, as was observed by Schönpflug (2000) who employed a Word Fragment Completion Test to trilingual students (Polish-German-English). The point of earliest recognition or uniqueness point is assessed by the number of identified letters in percent of the total number of letters in the word necessary for the identification of the word. According to Schönpflug’s findings, this point of early recognition or uniqueness point is apparently later in a stronger L2 German compared to an additionally learned weaker L3 English. Moreover, the author found that the uniqueness points were higher for concrete opposed to abstract words (Pavio et al., 1968), with longer words opposed to shorter ones, and words with one translation equivalents (from German to English and vice versa, data by Schönpflug, 1997a) opposed to more than one translation equivalent. The mean uniqueness points in percent for these word categories are listed in Table 2.1. Table 2.1  Uniqueness points (percent of total number of letters in word needed) dependent on eight word categories (trilingual students with competence in L1 Polish, L2 German, and L3 English)

Word category L3 (English) L2 (German) Abstract Concrete Short Long 1 Translation equivalent >1 Translation equivalent

Uniqueness point (%) 37.9 44.1 40.1 41.0 40.1 41.8 42.5 39.0

2.1  Multilingual Decoding and Encoding

31

The word fragment completion test requires a letter-by-letter processing and the retrieval of words (nouns) that start with the processed sequence of letters. Table 2.1 shows the results of a new analysis of variance (ANOVA) of the data collected by Schönpflug (2000). The language makes a significant difference, as L2 German words were recognized later than L3 English ones (F(1,28)  =  21.41, p  1 translation equivalent (F(1,28) = 48.53, p