Bilingualism Across the Lifespan: Factors Moderating Language Proficiency 9783110341249, 9783110340440, 9783110610468

Citation preview

Edited by Elena Nicoladis and Simona Montanari Bilingualism Across the Lifespan

14296-00-FM_4thPgs.indd 1

5/16/16 10:13 AM

Language and the Human Lifespan Series

Bilingualism Across the Lifespan: Factors Moderating Language Proficiency Edited by Elena Nicoladis and Simona Montanari

Bilingualism Across the Lifespan Factors Moderating Language Proficiency

Edited by Elena Nicoladis and Simona Montanari

American Psychological Association • Washington, DC

14296-00-FM_4thPgs.indd 3

5/16/16 10:13 AM

Copyright © 2016 by the American Psychological Association and Walter de Gruyter GmbH. All rights reserved. Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, including, but not limited to, the process of scanning and digitization, or stored in a database or retrieval system, without the prior written permission of the publishers. Published by American Psychological Association 750 First Street, NE Washington, DC 20002-4242 www.apa.org

Walter de Gruyter GmbH Genthiner Straße 13 10785 Berlin / Germany www.degruyter.com

To order in the United States and Canada: APA Order Department P.O. Box 92984 Washington, DC 20090-2984 Tel: (800) 374-2721; Direct: (202) 336-5510 Fax: (202) 336-5502; TDD/TTY: (202) 336-6123 Online: www.apa.org/pubs/books/ E-mail: [email protected]

To order in Europe: HGV Hanseatische Gesellschaft für Verlagsservice mbH Holzwiesenstr. 2 72127 Kusterdingen / Germany Tel.: +49 (0)7071 9353 – 55 Fax.: +49 (0)7071 9353 – 93 Online: www.degruyter.com E-mail: [email protected]

Other customers, including those in the United Kingdom, may order from either publisher. Typeset in DG Meta Serif Science by Circle Graphics, Inc., Columbia, MD Printer (U.S. & Canada): Edwards Brothers, Inc., Ann Arbor, MI Printer (Europe): CPI books GmbH, Leck, Germany Cover Designer: Walter de Gruyter GmbH, Berlin, Germany The opinions and statements published are the responsibility of the authors, and such opinions and statements do not necessarily represent the policies of the American Psychological Association or Walter de Gruyter GmbH. Library of Congress Cataloging-in-Publication Data Names: Nicoladis, Elena editor. | Montanari, Simona (Linguist), editor. Title: Bilingualism across the lifespan : factors moderating language proficiency / Edited by Elena Nicoladis and Simona Montanari. Description: Washington, DC ; American Psychological Association : Berlin ; Walter de Gruyter, [2016] | Series: Language and the Human Lifespan | Includes bibliographical references and index. Identifiers: LCCN 2016000486 | ISBN 9781433822834 | ISBN 1433822830 Subjects: LCSH: Language acquisition—Age factors. | Second language acquisition. | Bilingualism. | Cognitive learning. | Psycholinguistics. Classification: LCC P118.65 .B5 2016 | DDC 401/.93—dc23 LC record available at http://lccn.loc.gov/2016000486 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency has been published under the following ISBNs: APA: 978-1-4338-2283-4 DG: 978-3-11-034044-0 British Library Cataloguing-in-Publication Data A CIP record is available from the British Library. Bibliographic Information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the internet at http://dnb.dnb.de. Printed in the United States of America and Germany First Edition http://dx.doi.org/10.1037/14939-000

14296-00-FM_4thPgs.indd 4

5/16/16 10:13 AM

Contents Contributors 

 vii

Simona Montanari and Elena Nicoladis Introduction   3

1

Fred Genesee Shifting Perspectives on Bilingualism 

 9

I Early Bilingualism 2

Suzanne Quay and Simona Montanari Early Bilingualism: From Differentiation to the Impact of Family Language Practices   23

3

Christopher T. Fennell, Angeline Sin-Mei Tsui, and Tamara M. Hudon Speech Perception in Simultaneously Bilingual Infants   43

4

Barbara T. Conboy and Simona Montanari Early Lexical Development in Bilingual Infants and Toddlers 

5

W. Quin Yow, Ferninda Patrycia, and Suzanne Flynn Code-Switching in Childhood   81

II Factors Affecting Bilingualism Across the Lifespan 6

7

8

Sharon Unsworth Quantity and Quality of Language Input in Bilingual Language Development   103 Virginia C. Mueller Gathercole Factors Moderating Proficiency in Bilingual Speakers 

 123

Silvina Montrul Age of Onset of Bilingualism Effects and Availability of Input in First Language Attrition   141

 63

vi 

9

10

 Contents

David Birdsong and Jan Vanhove Age of Second-Language Acquisition: Critical Periods and Social Concerns   163 Jeff MacSwan Code-Switching in Adulthood 

 183

III Academic Achievement and Literacy in Bilinguals 11

12

Kathryn Lindholm-Leary Bilingualism and Academic Achievement in Children in Dual Language Programs   203 Judith F. Kroll, Jason Gullifer, and Megan Zirnstein Literacy in Adulthood: Reading in Two Languages 

 225

IV Cognitive Effects of Bilingualism 13

Ágnes Melinda Kovács Cognitive Effects of Bilingualism in Infancy 

14

Elena Nicoladis Bilingual Speakers’ Cognitive Development in Childhood 

15

Max R. Freeman, Anthony Shook, and Viorica Marian Cognitive and Emotional Effects of Bilingualism in Adulthood 

16

 269

 285

Hilary D. Duncan and Natalie A. Phillips The Contribution of Bilingualism to Cognitive Reserve in Healthy Aging and Dementia   305

V Conclusion Simona Montanari and Elena Nicoladis Concluding Remarks and Future Directions  Index 

 249

 331

About the Editors 

 353

 325

Contributors David Birdsong, PhD Department of French and Italian University of Texas at Austin Barbara T. Conboy, PhD Department of Communicative Disorders University of Redlands Redlands, CA Hilary D. Duncan, MA Department of Psychology Concordia University Montreal, Quebec Canada Christopher T. Fennell, PhD School of Psychology & Department of Linguistics University of Ottawa Ottawa, Ontario Canada

Tamara M. Hudon, PhD School of Psychology University of Ottawa Ottawa, Ontario Canada Ágnes Melinda Kovács, PhD Cognitive Development Center Central European University Budapest, Hungary Judith F. Kroll, PhD Department of Psychology The Pennsylvania State University University Park, PA Kathryn Lindholm-Leary, PhD Child and Adolescent Development Department San Jose State University San Jose, CA

Suzanne Flynn, PhD Department of Linguistics and Philosophy Massachusetts Institute of Technology Cambridge, MA

Jeff MacSwan, PhD Applied Linguistics and Language Education University of Maryland College Park, MD

Max R. Freeman, MS Roxelyn and Richard Pepper Department of   Communication Sciences and Disorders Northwestern University Evanston, IL

Viorica Marian, PhD Roxelyn and Richard Pepper Department of   Communication Sciences and Disorders Northwestern University Evanston, IL

Virginia C. Mueller Gathercole, PhD Linguistics Program Florida International University Miami, FL

Simona Montanari, PhD Department of Child and Family Studies California State University Los Angeles, CA

Fred Genesee, PhD Department of Psychology McGill University Montreal, Quebec Canada

Silvina Montrul, PhD Department of Spanish and   Portuguese/Linguistics University of Illinois Urbana, IL

Jason Gullifer, PhD Department of Psychology McGill University Montreal, Quebec Canada

Elena Nicoladis, PhD Department of Psychology University of Alberta Edmonton, Alberta Canada

viii 

 Contributors

Ferninda Patrycia, MS Singapore University of Technology and Design Singapore Natalie A. Phillips, PhD Department of Psychology Concordia University Montreal, Quebec Canada Suzanne Quay, PhD Department of Education and   Language Education International Christian University Tokyo, Japan Anthony Shook, PhD Roxelyn and Richard Pepper Department of   Communication Sciences and Disorders Northwestern University Evanston, IL Angeline Sin-Mei Tsui, MA School of Psychology University of Ottawa Ottawa, Ontario Canada

Sharon Unsworth, PhD Department of English Language and   Culture and Department of Linguistics Radboud University Nijmegen, the Netherlands Jan Vanhove, PhD Department of Multilingualism University of Fribourg Fribourg, Switzerland W. Quin Yow, PhD Singapore University of Technology and Design Singapore Megan Zirnstein, PhD Department of Psychology The Pennsylvania State University University Park, PA

Edited by Elena Nicoladis and Simona Montanari Bilingualism Across the Lifespan

14296-01-Introduction_4thPgs.indd 1

5/16/16 10:08 AM

Simona Montanari and Elena Nicoladis

Introduction

As we approach the third decade of the 21st century it is clear that the world’s citizens increasingly function in more than one language on an everyday basis, with consequences that go well beyond the realm of language. It is the purpose of this volume to present the latest research on bilingualism from this broader perspective, exploring the complex interaction between the acquisition, processing, and use of multiple languages throughout the lifespan and in different sociopolitical contexts and the effects of bilingualism on some nonlinguistic domains such as family functioning, literacy, cognition, and identity, to name a few. Primarily addressed to linguistics, psychology, and education students as well as to bilingualism scholars, the volume’s first goal is to underscore the variability of “the bilingual experience”—both in terms of language experiences and of language outcomes. At a superficial level, a bilingual speaker is defined as an individual who knows two languages, but the chapters in this book point to the inadequacy of this simple definition to capture the heterogeneity of bilingual speakers. Genesee (Chapter 1, this volume), in particular, expands on the myriad of ways bilingual speakers can differ from each other, from the pairs of languages spoken to the functions each language serves in bilingual speakers’ lives. The main implication is that because of their multidimensional nature, bilingual speakers must be studied in their own right, without reference to monolingual speakers as norms. Another crucial theme of the research presented in this volume is that language knowledge and use—be it mono- or multilingual—are deeply intertwined with sociocultural and sociopolitical issues, like identity, status, power, and ability to participate in the community. Hence, bilingualism can be understood only when examined in the context in which it unfolds. Code-switching, for example, is often seen as a sign of confusion or language incompetence in the United States (MacSwan, Chapter 10) but as an asset and evidence of language mastery in Singapore (Yow, Patrycia, & Flynn, Chapter 5). Similarly, the ability to speak Spanish and be bilingual in the United States is not valued in the same way as in countries that are supportive of bilingualism (Montrul, Chapter 8). Few studies have directly compared how different sociocultural settings affect bilingualism, so we will return to this issue in the volume’s Conclusion. The third goal of this volume is to challenge scholars of linguistics and psychol­ ogy to think about the generalizability of acquisition theories. A few decades ago, bilingualism was sometimes presented in language acquisition textbooks in the chapter on special populations. In one textbook used by one of the editors in graduate school, a discussion of bilingual first language acquisition was juxtaposed with learning to speak with cochlear implants. In recent years, researchers have begun http://dx.doi.org/10.1037/14939-001 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

4 

 Montanari and Nicoladis

to recognize that far from being a special population, bilingual and multi­lingual speakers make up a large proportion of the world’s population. To the extent that linguistic and psychological theories of language acquisition claim to be universal, they must account for the variability observed in bilingual speakers, in addition to monolingual acquisition. Consider, for example, usage-based theories of language acquisition that have had some success in recent years in explaining monolingual children’s language acquisition. Can these theories account for why bilingual children can sometimes attain similar outcomes to monolingual children of the same age, even though they use each of their languages less often than their monolingual peers (see Unsworth, Chapter 6, and Gathercole, Chapter 7)? How can the bilingual learning experience in all its variants be incorporated in such “universal” theories of language acquisition? It is with these three themes in mind—the heterogeneity of bilingual speakers, the importance of the social context to understand bilingualism, and the need to include bilingual development in general language acquisition theories—that this volume presents research on bilingualism across different stages of life, from infancy to old age. By framing the research within a lifespan perspective, we hope to show readers that age is yet another variable that affects bilingual outcomes. The volume focuses on four major areas of investigation in which many advances have been made in recent years: (a) early bilingualism, (b) factors affecting bilingualism across the lifespan, (c) academic achievement and literacy in bilinguals, and (d) cognitive effects of bilingualism. The contributions in these areas are not exhaustive of all bilingualism research but they are meant to provide the reader with a view of bilingualism that goes beyond traditional linguistic or psycholinguistic accounts and show the consequences of bilingualism on—and its interrelationship with—nonlinguistic domains such as literacy, cognition, and family functioning, to name a few. It is precisely in this framework that Quay and Montanari set their chapter (Chapter 2) in this book’s first section on early bilingualism. One important issue underlying research in this area and in all the chapters in this section is how children with early exposure to two languages learn to differentiate them. Bilingual children sometimes code-switch, or use both of their languages within an utterance or conversation. Does code-switching show that they cannot tell their languages apart? Quay and Montanari argue that the very phrasing of this issue implies that bilingual speakers are being compared to a monolingual norm. They propose to examine early bilingual development with a focus on what impacts it most—the social context, the language environment, and, in particular, the home and family language practices that allow children to become productive bilingual speakers. Fennell, Tsui, and Hudon (Chapter 3) then review studies on the processes by which bilingually exposed children learn to differentiate the rhythms, phonological patterns, and word forms of their two languages, how they come to perceive speech sounds, and how they categorize them as phonemic in each language. The chapter shows that, despite being exposed to more complex and variable input than their

Introduction 

 5

monolingual peers, bilingual infants meet most speech perception milestones—such as language discrimination, phonetic refinement of vowels and consonants, the acquisition of phonotactic rules, and the perception and use of phonological categories— following the same timeline as monolingual children. The authors argue that the perceptual gateway to bilinguals’ language separation might lie in their early ability to detect rhythmic differences between languages but also in their keen attention to talking faces and their sensitivity to the social cues that characterize speakers of each language. Not only must bilingual children differentiate the rhythms, sounds, and phono­ logical patterns of their two languages, but they also eventually must be able to develop vocabularies for each. Conboy and Montanari (Chapter 4) focus specifically on lexical processing and production in bilingual infants and toddlers. The review shows that bilingual infants develop two lexical systems by using the same segmentation, recog­ nition, comprehension, and processing mechanisms that monolingual infants use to build one system. However, young bilingual children might also develop unique abilities as they learn to manage two lexical systems, suggesting that differences from monolingual patterns may reflect appropriate responses to input factors and properties rather than to a delay or deviance induced by bilingualism. If bilingual children can differentiate their two languages, why then might they code-switch (i.e., alternate between two or more languages in the context of a single utterance or conversation)? Yow and colleagues (Chapter 5) consider data from Mandarin–English bilingual 5- and 6-year-old students in Singapore. The analyses reveal that the amount of code-switching by children is related to bilingual proficiency and that contexts that favor code-switching produce children with increased command of two languages. These results suggest that in a reality supportive of bilingualism, alternating languages in discourse is indicative of linguistic and sociolinguistic competence rather than lack of proficiency or language attrition. Part II delves more deeply into the factors that affect bilinguals’ knowledge and use of their two languages. Unsworth (Chapter 6) reviews the latest research that examines the role of input quantity and quality on bilingual children’s developing language skills. This review shows that more input generally leads to quicker rates of acquisition, especially in the areas of vocabulary and verbal morphology. However, input effects may be moderated by a variety of factors, including the variety and proficiency level of the input providers, the existence of older, school-age siblings, the regularity of language- and literacy-related activities in the home, and, most important, the extent to which children use the languages in question themselves. Gathercole (Chapter 7) extends Unsworth’s review by examining other factors that lead to and promote bilingual proficiency. After laying out the contributions of input in the home and the community and the role of the quality and sources of input, Gathercole shows that the complexity of linguistic forms in each language also affects bilingual outcomes, giving rise to profile effects. This means that language aspects that are complex take longer to emerge than aspects that are simpler, making

6 

 Montanari and Nicoladis

bilingual learners perform similarly to monolingual speakers on some structures but distinctly on others. Gathercole further shows that, together with home language, socioeconomic status is one of the best predictors of performance on both receptive vocabulary and grammar measures, especially at later ages, pointing to the crucial role of socioeconomic and sociocultural issues in bilingualism. Unsworth and Gathercole discuss outcomes related to bilingualism when there is continuous exposure to both languages. Montrul (Chapter 8) reviews key studies on bilingual skills in populations for whom first language input is interrupted: heritage learners, the children of adult immigrants who receive reduced input to the heritage language, and international adoptees. The findings presented in this chapter show that for a language to develop, stabilize and not regress, a critical mass of input and use is required during an extended period of time, possibly until the period of late elementary schooling and later language development. These results put into question the notion that language acquisition is largely complete by age three or four, as traditionally put forth by monolingual acquisition studies, and calls for language acquisition theories that move beyond static accounts of bilingual development and take into account the full constellation of social and psychological factors that can affect—and the language skills that can result from—bilingualism across the lifespan. Next, Birdsong and Vanhove (Chapter 9) focus on one of the most influential factors that have been traditionally believed to determine bilingual attainment: age of onset of second-language acquisition. The relationship between age and ultimate level of attainment in a second language has been often thought of in terms of an “earlier is better” generalization, with young learners displaying fluent speech, effortless language processing, and native accent while late learners diverge from monolingual natives on accent, grammatical and lexical knowledge, and processing speed. This chapter reviews research on age effects on second language acquisition, in particular, whether such acquisition is truly constrained by a “critical period” after which nativelike attainment in the second language is not possible. Birdsong and Vanhove also review sociopsychological factors besides age—in particular, learners’ socialization patterns and their choices of integration with or segregation from specific sociolinguistic milieus—that might crucially condition the outcomes of second-language learning. The final chapter in this section (MacSwan, Chapter 10) focuses on code-switching in adulthood. Traditionally perceived, in monolingual cultures, as indicative of a language disability and seen as a coping strategy for incomplete language knowledge, code-switching is reanalyzed by MacSwan in a review that shows that bilingual speakers who code-switch are not only sensitive to the grammatical rules of their linguistic systems, but they use their two languages creatively to fulfill a variety of pragmatic, stylistic, and social needs. The third section of the volume addresses academic achievement in school-age bilingual children and literacy in bilingual readers. Lindholm-Leary (Chapter 11) reviews research on bilingualism and academic achievement among U.S. students

Introduction 

 7

enrolled in dual language programs, in which instruction occurs in two languages with the goal of promoting bilingualism, biliteracy, and grade-level achievement. The review shows that students in dual language programs become bilingual speakers and achieve at or above grade level in English and the partner language, and they perform at comparable or higher levels compared to peers who are enrolled in English mainstream programs. However, challenges exist to develop high levels of bilingualism within dual language programs beyond primary school because the U.S. educational context does not value bilingualism and school accountability and academic success are associated with demonstrating language proficiency and achievement in English but not in an added language. Turning to a psycholinguistic perspective of bilingual literacy, Kroll, Gullifer, and Zirnstein (Chapter 12) show that the bilingual speaker’s two languages are open to cross-language activation that influences reading in each language. This is the case not only for bilingual speakers who are highly proficient in two languages but also for those acquiring second language literacy beyond early childhood, and for bilingual readers of languages with different scripts or of only one language. Such crosslanguage activation requires bilingual speakers to control the activation of the language not in use, conferring a set of cognitive benefits in tasks that tap into executive function and require ignoring irrelevant information, resolving conflict, and switching between conditions. The main implication of these findings is that the model of the monolingual reader reading in a single, native language is no longer taken to characterize successful literacy. Rather, a new model of literacy should accommodate the presence of two or more languages, the interaction that occurs cross-linguistically, and its consequences on cognition and processing. If the experience of selecting languages for a context or a task affects cognitive processing, then such effects might well be observed across the lifespan. Indeed, in the next section on the cognitive effects of bilingualism, the authors review evidence showing bilingual advantages in some aspects of cognition over monolinguals across the lifespan. Kóvacs (Chapter 13) argues that early exposure to complex input leads not only to the development of enhanced cognitive abilities but also to complex social understandings in bilingual infants. Nicoladis (Chapter 14) shows that some of these effects may continue into early childhood: some studies have shown that bilingual children perform better than monolingual children on some cognitive tasks tapping flexibility, theory of mind, and selective attention. Freeman, Shook, and Marian (Chapter 15) demonstrate that managing multiple languages not only impacts cognitive control but also emotion. The review shows for instance that bilingual speakers process emotions differently across their first and second language and depending on context and cultural affiliation, suggesting interesting connections between identity, culture, language, and emotion processing. Finally, Duncan and Phillips (Chapter 16) examine how speaking more than one language may mitigate the cognitive decline often seen in aging and possibly delay the onset of dementia. Overall, these studies suggest that bilingualism may produce cognitive benefits across the lifespan and help

8 

 Montanari and Nicoladis

mitigate age-related cognitive decline. Yet, whether the hypothesized benefits come from bilingualism per se or from the many variables that characterize it remains a topic for future research. It is precisely with this issue in mind that we return, in the Conclusion, to the themes that emerge throughout the chapters but have received little direct research attention: the variability that characterizes bilingual speakers’ language experiences and out­comes, the often-unacknowledged relevance of sociopolitical and socio­cultural factors for the outcomes of bilingualism across the lifespan, and the need for future research to be more geographically diverse, socially inclusive, and more readily transmitted to the public at large.

Fred Genesee

1 Shifting Perspectives on Bilingualism To say that the study of bilingualism is broad would be an understatement. To begin, the population of interest itself is broad, ranging from neonates (and even prenatal babies), to young children, adolescents, and adults. At the same time, it can include studies of simultaneous bilingual speakers—those who acquire more than one language from birth and those who are second-language learners—who acquire additional languages after a first language has begun. Variation among bilingual speakers goes beyond issues related to when they acquire their two languages, a point I will return to later. The study of bilingualism can also arguably include special populations—such as individuals with cochlear implants and internationally adopted children insofar as these individuals acquire language after birth and, thus, contribute to our understanding of age effects on language acquisition—a key issue in understanding bilingualism (see Chapter 9, this volume). To complicate matters further, the term bilingualism is often also used to encompass not just individuals who acquire a second language, but individuals who learn and use more than two languages—multilinguals. It can also include individuals who acquire a signed language in addition to a spoken language. Studies of bilingualism, though normally associated with the acquisition of additional languages, also encompass studies of language loss as a second language replaces a first language (see Chapter 8). Indeed, transitions from one state of language competence to another, be it from monolingualism to bilingualism or vice versa, raise important and interesting issues about neural plasticity, linguistic representations and processing, and the role of input—all topics of relevance to bilingualism. Studies of bilingualism also include the acquisition and processing of written forms of two or more languages along with competence in oral modalities (see Chapters 11 and 12, for studies on reading). Of course, there is also variation associated with different stages of life and different ages when additional languages are acquired, which is the theme of this book. It is not just psycholinguists who are interested in bilingualism, but also neuroscientists, social psychologists, educators, linguists, political scientists, sociologists, anthropologists, speech and language pathologists, and others. This chapter will stick to the perspectives of psycholinguists and neuroscientists, as they are the focus of this book. Historically, whether the attention was on acquiring two languages, cognitive representation of each language, lexical access, or another topic, bilingual speakers have been compared with monolingual speakers on the assumption that monolingual speakers are the norm and that the way humans learn, represent, process, and use language is best understood by using monolingual speakers as the norm. Comparisons with monolingual speakers often took a simplistic theoretical view—either bilingualism is costly because learning two languages requires more work and more time than http://dx.doi.org/10.1037/14939-002 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

10 

 Fred Genesee

learning one language or, conversely, bilingualism is advantageous because learning two languages constitutes cognitive exercise that enhances cognitive resources. The deficit perspective tended to prevail. Moreover, comparisons with monolingual speakers were often based on the results of undifferentiated groups of bilingual speakers versus monolingual speakers, as if all bilingual speakers are the same. There has been a slow but inevitable shift in perspective in recent years as more and more evidence on bilingualism has accumulated. Increasingly, there is evidence that bilingual speakers are not two monolingual speakers in one; that bilingualism is not categorically better, harder, or more advantageous than monolingualism; and that not all bilinguals are the same. These emerging views are evident in research on bilingual speakers that covers the lifespan. In the following brief sections, I cannot provide a comprehensive view of research on bilingual speakers at different stages of development, but I selectively review studies that illustrate that a shift in perspective on bilingualism is occurring.

Early Stages of Life: Infants and Young Children Some of the earliest studies on bilingualism were conducted by Ronjat (1913), Leopold (1939–1949), and Burling (1959). These researchers examined the language development of their own children who were acquiring two languages—a not uncommon practice in early research because of the obvious convenience of recording and analyzing one’s own children’s earliest productions in two languages. There was a lull in research on bilingual speakers until the 1970s. The re-emergence of interest was due, in part at least, to the provocative and useful paper by Volterra and Taeschner (1978). In this landmark publication, Volterra and Taeschner laid out a theory of simultaneous bilingual acquisition that became a focus of research attention for many years. Specifically, they proposed that simultaneous bilingual speakers go through an initial stage when lexical and morphosyntactic development is not initially differentiated by language—what has come to be called the unitary language system hypothesis (Genesee, 1989). According to this hypothesis, it was only during the third year of life when separation or differentiation of the two languages occurred and, thus, when real bilingualism was evident. Leopold had made a similar claim much earlier (Leopold, 1939–1949). This issue continues to stimulate interest, as evident in a number of chapters in this volume (see Chapters 2–4). This was a radical conceptualization of bilingual acquisition insofar as it proposed that the neurocognitive system that supports language acquisition is essentially monolingual in nature and, by implication, that learning two languages simultaneously challenges language development resulting in confusion, slowed development, and possibly even impairment. Research that has examined young bilingual learners has refuted the unitary language hypothesis on a number of fronts. Chief among these lines of research are studies that have examined bilingual children’s language usage. Young bilingual



Shifting Perspectives on Bilingualism 

 11

children sometimes mix sounds, words, and phrases from their two languages in the same utterance; this was used as evidence for lack of differentiation of their underlying linguistic systems (Vihman, 1985; Volterra & Taeschner, 1978). Numerous studies since Volterra and Taeschner (1978) have shown that although bilingual children do indeed mix, they nevertheless use their two languages differentially and appropriately with interlocutors, including their parents (Genesee, Nicoladis, & Paradis, 1995) and even with adults with whom they have had no prior interaction (Genesee, Boivin, & Nicoladis, 1996). Moreover, young simultaneous bilingual children can make appropriate repairs to breakdowns in communication with strangers that result from the children’s use of a language that their interlocutor does not speak—by switching languages (Comeau, Genesee, & Mendelson, 2007). They are able to do this in the absence of explicit feedback that their choice of language has caused the breakdown, indicating that they can manage the additional constraints of communicating in two languages readily. That young bilingual children can use their two languages in these ways is not compatible with the unitary language system hypothesis. Even more direct evidence against the unitary language system hypothesis comes from careful examination of bilingual children code-mixing, the primary evidence used to support the hypothesis initially. If young children who are learning two languages simultaneously go through a stage when they treat both languages as part of one system, then one would expect them to code-mix extensively and, moreover, to violate the grammatical constraints of both languages because their grammatical systems are undifferentiated. In an early study on this issue, Genesee, Nicoladis, and Paradis (1995) found that French–English bilingual children (ages 1 year, 10 months– 2 years, 2 months) in Montreal mixed within utterances less than 3% of the time, on average, far less often than one would expect if they were unable to differentiate between French and English. In an independent sample of young French–English children in Montreal, Sauve and Genesee (2000) similarly found that code-mixing within utterances occurred less than 4% of the time and, as well, there were virtually no grammatical errors when code-mixing did occur (see also Paradis, Nicoladis, & Genesee, 2000). Researchers have also reported that the constraints that operate on children’s bilingual code-mixing are essentially the same as those that have been reported in adults (Paradis et al., 2000). There does not appear to be a stage in bilingual first-language acquisition when grammatical constraints do not operate, albeit the nature of the constraints may change as children’s grammars change. These findings are interesting for two reasons. First, they indicate that bilingual children acquire the grammatical constraints of each language; otherwise, how could one explain that they comply with the constraints of each most of the time. Second, and even more interesting, they indicate that bilingual children are able to activate and access both language systems at the same time to ensure that their mixed utterances conform to the constraints of both languages. The same findings have been reported in studies of children learning other language pairs (e.g., French and German [Meisel, 1989]; Inuktitut and English [Allen, Genesee, Fish, & Crago, 2002]).

12 

 Fred Genesee

Also contrary to the unitary language system hypothesis, is that children acquiring two languages demonstrate evidence of differentiated phonological and morphosyntactic systems from the earliest stages of language development. Maneva and Genesee (2002) found that the babbling of the French–English infant in their study differed depending on whether he was interacting with his French-speaking father or his English-speaking mother and, furthermore, his babbling in each case was similar to that of monolingual infants learning the same languages with respect to the mean length of babbled utterances, syllable load, and syllable type. Detailed examination of the developing grammatical systems of French–English bilingual children in Montreal revealed that, for the most part, they were the same as those of monolingual children (Paradis & Genesee, 1996). Zwanziger, Allen, and Genesee (2005) reported evidence of differentiated grammatical development in children acquiring Inuktitut and English which, in contrast to French and English, have radically different morphosyntactic properties. Even under conditions of specific language impairment, bilingual children with impairment exhibit similar patterns of grammatical development as monolingual children with impairment (Paradis, Crago, Genesee, & Rice, 2003). Returning to the unitary language system hypothesis, the uniquely human neuro­ cognitive capacity for learning language is prepared to learn not just one language but two (or more) languages with relative facility—provided the learning conditions are adequate. Additionally, as more and more evidence about simultaneous bilingual speakers emerges, it becomes increasingly clear that bilingual speakers are not simply two monolingual speakers in one (see also Chapter 2, this volume). Differences among bilingual children and between bilingual and monolingual children exist, but to date most differences appear to be related primarily to characteristics of the learning environment, including the quantity, nature, and consistency of the input that bilingual children receive, rather than the fact of learning two languages per se (see Grüter & Paradis, 2014, and Chapter 6, this volume, for more extensive discussions of the role of input).

School-Age Learners Distinguishing school-age learners from pre-school and adult learners is tenuous at best because the lower and upper ends of this stage of development flow from and into the adjacent stages. Nevertheless, it is an important stage of development to consider separately because of its connection with schooling. It is at the onset of schooling when many children begin to acquire a second language. For example, immigrant children who are raised in homes where a minority language is used must learn the majority language for success in school. As well, there is a growing number of children who speak a majority language at home and participate in second-language immersion programs in school (see Genesee, 2004, for a review). Language learning in school is of additional interest because the demands on language proficiency



Shifting Perspectives on Bilingualism 

 13

increase and academic success is mediated through language. It is also in school that children learn to read and write (see Chapter 11, this volume). Investigations of bilingualism in the context of schooling are not a recent phenomenon. In fact, some of the earliest research on bilingual speakers examined the effects of bilingualism on intelligence and academic achievement in school-age children. This research dates back to the 1920s in Wales (e.g., Saer, 1923) and later in Ireland (Macnamara, 1966). As was true for early research on bilingual acquisition, thinking about bilingualism in school contexts at that time was on the basis of the assumptions that the human capacity for language learning is relatively limited (see Cummins, 1979, for an early explication of this thinking) and, therefore, acquiring more than one language entails negative effects on both language and academic development. Indeed, these early researchers reported that children in bilingual programs in Wales and Ireland scored lower than children in monolingual native-language programs on a variety of intelligence and language tests. These findings had a significant influence on scientific thinking about bilingualism and also on public policy on how best to educate immigrant children. Policies in many communities were established that forbade and even punished immigrant students for using their native language in school, and laws were passed that disallowed education in languages other than the majority language. However, this way of thinking changed dramatically with the publication of the landmark study by Peal and Lambert (1962). Their study implemented a number of important methodological controls that confounded interpretations of early findings, and they broadened their investigation to go beyond standardized tests of intelligence to include more broadly defined cognitive measures, including cognitive flexibility, creativity, and attentional control. In contrast to earlier studies, Peal and Lambert’s comparisons between bilingual and monolingual children revealed cognitive advantages in favor of bilingual children. The precise extent and nature of the cognitive consequences of bilingualism has continued to attract attention and controversy among researchers and the public at large. Bialystok, Craik, and Luk (2008) argued that a bilingual advantage can be found in the performance of tasks that call for selective attention, especially if inhibitory control of attention is called for. Bialystok et al., and others (Tao, Marzecová, Taft, Asanowicz, & Wodniecka, 2011), have argued that these cognitive advantages arise because learning and using two languages calls for selective attention to minimize interference between languages and ensure their appropriate use; this, in turn, enhances the development of executive control processes in general, not only in linguistic domains. These advantages have been found in both childhood and adulthood (Bialystok, Craik, Klein, & Viswanathan, 2004) and are most evident in bilingual speakers with relatively advanced levels of proficiency in two languages and who use two languages actively on a regular basis (Bialystok, Peets, & Moreno, 2014). Although these findings are controversial (see Baum & Titone, 2014, for a review), they mark a shift in perspective insofar as they argue that bilingualism is not a drag on cognitive development, but rather may be a boon. In other words, not only can the

14 

 Fred Genesee

human brain handle two languages but it benefits from bilingualism. These findings also paint a picture of bilingual speakers as different from monolingual speakers in ways that speak to neurocognitive plasticity and adaptability in response to monolingual versus bilingual environments. Our own research on young second-language learners who acquire a second language before 3 years of age indicates that they draw on different neurocognitive systems when processing a second language than do monolingual speakers (Pierce, Chen, Delcenserie, Genesee, & Klein, 2014). These differences in underlying neurocognitive engagement were found despite the fact that there were no significant differences between the monolingual speakers and secondlanguage learners on behavioral measures of language proficiency. Taken together, these findings highlight the neural plasticity of the brain to acquire language in different ways. Finally, these kinds of studies on bilingualism demonstrate the contribution that such studies have to theories of neurocognition in general. Arguably, theories of language learning as well as theories of neurocognitive development that are based exclusively on findings from monolingual speakers fail to tell the whole story because they do not take into account learning and development in the full range of contexts in which they typically occur.

Adulthood Finally, findings from studies of bilingual adults also speak to a view of bilingualism that goes beyond the deficit view. In this regard, we have known for some time that for bilingual speakers both languages are activated to different degrees even when only one language is being used (Blumenfeld & Marian, 2013). These findings speak to the capacity to access two languages simultaneously. That there may be a cost to switching between languages or when accessing one language rather than another is best viewed as a reflection of bilingual speakers having to manage an extended language repertoire in comparison to monolingual speakers—knowing more means having to sort through more. As well, extensive research on code switching in bilingual adults, dating back to the 1970s (Scotton & Ury, 1977), confirms findings on code switching in children—it is usually grammatically constrained (see Chapter 5, this volume) and, thus, cannot be seen as a sign of linguistic confusion or incompetence. It is now well recognized that bilingual adults engage in code switching for many communicative purposes—to express ethnic identity, to establish affinity with or distance from an interlocutor, to express status or subordination in relationship to an interlocutor, or to mark social context (e.g., Genesee & Bourhis, 1988; Scotton & Ury, 1977), among others. Findings from studies of children showing that bilingualism might be a boon to development and not a burden have been corroborated by recent research on adults. Our own research on successive and simultaneous bilingual adults found that they outperform monolingual adults on measures of both verbal and nonverbal working memory and that the simultaneous bilingual adults performed significantly better



Shifting Perspectives on Bilingualism 

 15

than the successive bilingual adults (Delcenserie & Genesee, 2014). Regression analyses revealed further that both age of acquisition of the second language and proficiency in that language were significant predictors of performance on the memory tasks. What is important here is that although superior performance on the memory tasks was found for all groups of bilingual speakers, it was the simultaneous bilingual speakers who were differentially favored, arguing that the effects of bilingualism on executive control functions are mediated by age of acquisition as well as proficiency (see also Bialystok et al., 2014, for evidence of proficiency effects). Research on aging adults suggests that the nonlinguistic effects of bilingualism may be broad in scope. More specifically, Bialystok and others report that the onset of dementia among older individuals who suffer from Alzheimer’s disease may be delayed by up to four years in bilingual speakers in comparison to monolingual speakers (Alladi et al., 2013; Bialystok, Craik, & Freedman, 2007). These authors argue that being bilingual confers additional cognitive reserves that stave off the pernicious effects of early stages of dementia. Chapter 16 of this volume indicates that the link between bilingualism and dementia may be much more complex. Nevertheless, the important point here is that, taken together, these studies present more varied and nuanced views of bilingualism than earlier views which saw it as categorically negative or positive. The focus in a number of the chapters on bilingual adults in the present volume (Chapters 7, 12, and 15) and of a growing number of others are exploring how to uncover the ways in which the bilingual experience differs in ways that matter for language acquisition, representation, and processing. These studies are moving away from a view of bilingual speakers as a single undifferentiated group to recognize that bilingualism and, therefore, bilingual speakers are multidimensional and, thus, that simple characterizations of bilingual adults are untenable. Whitford and Titone (2015) found significant differences in word and text reading among bilingual adults as a function of differences in their current exposure to each language. More specifically, they found that bilingual adults with more current exposure to their second language demonstrated facilitated reading in that language in comparison to bilingual adults with less current exposure—specifically, more exposure enhanced reading fluency and perceptual reading span. Of particular note, current second-language exposure also affected word processing in the first language—the more current second-language exposure, the more word processing in the first language was impeded. In short, there was a trade-off between first- and second-language reading performance as a result of current second-language exposure. Whitford and Titone (2014) also found interesting developmental differences between younger and older bilingual speakers. Although more current second-language exposure enhanced reading fluency in the second language of older bilingual speakers, as was found in young bilingual speakers, it did not have an effect on their perceptual reading span in the second language, nor did it alter their first-language reading, contrary to what was found in young bilingual speakers. The authors argue that because older bilingual speakers had had considerably more

16 

 Fred Genesee

accumulated exposure to their second language, current exposure to that language had fewer and smaller effects. Pivneva, Palmer, and Titone (2012) recently showed that even individual differences in domain-general cognitive factor can influence language production in bilingual speakers. Of particular note, bilingual speakers with higher inhibitory capacity were more efficient in planning and producing specific features of spontaneous speech than bilingual speakers with less inhibitory capacity. Individual differences in inhibitory capacity have also been shown to modulate both within-language and betweenlanguage competition during word recognition in bilingual speakers; these effects are, in turn, modulated by task demands and language exposure and proficiency (Mercier, Pivneva, & Titone, 2014). These findings indicate that the link between bilingualism and inhibitory control is bidirectional—with evidence that bilingualism can enhance inhibitory control, as argued by Bialystok and Martin (2004), and that inhibitory control can also affect language processing in bilingual speakers. Interesting developmental, or lifespan, shifts emerge from a comparison of child and adult studies. Thus, the findings by Whitford and Titone (2012, 2015) provide direct evidence that word and text reading in younger and older bilingual speakers are similar in some respects and different in others. Studies of bilingual code switching indicate that a primary reason for switching by bilingual children is to fill lexical gaps (Genesee, 2002), whereas a primary reason for bilingual adult code switching is often social or communicative in nature. This makes sense insofar as bilingual children are still in the processing of acquiring two languages and need to draw on the resources of both languages to express themselves fully, whereas bilingual adults have likely reached their ultimate attainment and have recourse to code switching for other reasons. Studies of the nonlinguistic cognitive consequences of bilingualism reveal different effects in adults and children—the effects of bilingualism in bilingual children are reflected in enhanced inhibitory control in comparison with monolingual children, whereas the effects of bilingualism in bilingual adults are reflected in delayed onset of the pernicious effects of Alzheimer’s in comparison to monolingual adults. Arguably, the results for both bilingual children and adults in these studies reflect the influence of bilingualism on executive control functions more generally.

Conclusion As we acquire greater knowledge about bilingual speakers across the lifespan, we are compelled to eschew comparisons with monolingual speakers that depict one or the other as more favorable, more difficult, or more “normal.” Studies that examine bilingual acquisition and processing in depth reveal complexities and nuances that are lost when such unidimensional comparisons are made. The focus of current studies on variation among bilingual speakers further challenges simplistic comparisons between



Shifting Perspectives on Bilingualism 

 17

bilingual speakers and monolingual speakers that disregard diversity among bilingual speakers. The importance of discovering what is characteristic among bilingual speakers and what differs among bilingual speakers in response to external (e.g., input) and internal (e.g., domain-general cognitive) factors is inescapable. In short, we are compelled to recognize that learning more than one language, simultaneously or successively, is simply another way of learning language and merits attention on its own. References to “nativelike” that are inexorably linked to monolingual development as a characterization of bilingual performance are no longer tenable.

References Alladi, S., Bak, T. H., Duggirala, V., Surampudi, B., Shailaja, M., Shukla, A. K., . . . Kaul, S. (2013). Bilingualism delays age at onset of dementia, independent of education and immigration status. Neurology, 81, 1938–1944. http://dx.doi.org/10.1212/01.wnl.0000436620.33155.a4 Allen, S. E. M., Genesee, F. H., Fish, S. A., & Crago, M. B. (2002). Typological constraints on code mixing in Inuktitut–English bilingual adults. In M.-A. Mahieu & N. Tersis (Eds.), Variations on polysynthesis: The Eskaleut languages (pp. 273–306). Amsterdam, the Netherlands: Benjamins. Baum, S., & Titone, D. (2014). Moving toward a neuroplasticity view of bilingualism, executive control, and aging. Applied Psycholinguistics, 35, 857–894. http://dx.doi.org/10.1017/ S0142716414000174 Bialystok, E., Craik, F., & Luk, G. (2008). Cognitive control and lexical access in younger and older bilinguals. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34, 859–873. http://dx.doi.org/10.1037/0278-7393.34.4.859 Bialystok, E., Craik, F. I. M., & Freedman, M. (2007). Bilingualism as a protection against the onset of symptoms of dementia. Neuropsychologia, 45, 459–464. http://dx.doi.org/10.1016/ j.neuropsychologia.2006.10.009 Bialystok, E., Craik, F. I. M., Klein, R., & Viswanathan, M. (2004). Bilingualism, aging, and cognitive control: Evidence from the Simon task. Psychology and Aging, 19, 290–303. http://dx.doi.org/10.1037/0882-7974.19.2.290 Bialystok, E., & Martin, M. (2004). Attention and inhibition in bilingual children: Evidence from the dimensional change card sort task. Developmental Science, 7, 325–339. http://dx.doi.org/ 10.1111/j.1467-7687.2004.00351.x Bialystok, E., Peets, K. F., & Moreno, S. (2014). Producing bilinguals through immersion education: Development of metalinguistic awareness. Applied Psycholinguistics, 35, 177–191. http:// dx.doi.org/10.1017/S0142716412000288 Blumenfeld, H. K., & Marian, V. (2013). Parallel language activation and cognitive control during spoken word recognition in bilinguals. Journal of Cognitive Psychology, 25, 547–567. http:// dx.doi.org/10.1080/20445911.2013.812093 Burling, R. (1959). Language development of a Garo and English-speaking child. Word, 15, 45–68. Comeau, L., Genesee, F., & Mendelson, M. (2007). Bilingual children’s repairs of breakdowns in communication. Journal of Child Language, 34, 159–174. http://dx.doi.org/10.1017/ S0305000906007690

18 

 Fred Genesee

Cummins, J. (1979). Linguistic interdependence and the educational development of bilingual children. Review of Educational Research, 49, 222–251. http://dx.doi.org/10.3102/00346543049002222 Delcenserie, A., & Genesee, F. (2014). Language and memory abilities of internationally adopted children from China: Evidence for early age effects. Journal of Child Language, 41, 1195–1223. http://dx.doi.org/10.1017/S030500091300041X Genesee, F. (1989). Early bilingual development: One language or two? Journal of Child Language, 16, 161–179. http://dx.doi.org/10.1017/S0305000900013490 Genesee, F. (2002). Portrait of the bilingual child. In V. Cook (Ed.), Portraits of the Second Language User (pp. 170–196). Clevedon, England: Multilingual Matters. Genesee, F. (2004). What do we know about bilingual education for majority language students? In T. K. Bhatia & W. Ritchie (Eds.), Handbook of bilingualism and multi­culturalism (pp. 547–576). Malden, MA: Blackwell. Genesee, F., Boivin, I., & Nicoladis, E. (1996). Talking with strangers: A study of bilingual children’s communicative competence. Applied Psycholinguistics, 17, 427–442. http://dx.doi.org/10.1017/ S0142716400008183 Genesee, F., & Bourhis, R. Y. (1988). Evaluative reactions to language choice strategies: The role of socio-structural factors. Language & Communication, 8, 229–250. http://dx.doi.org/ 10.1016/0271-5309(88)90020-1 Genesee, F., Nicoladis, E., & Paradis, J. (1995). Language differentiation in early bilingual development. Journal of Child Language, 22, 611–631. http://dx.doi.org/10.1017/S0305000900009971 Grüter, T., & Paradis, J. (Eds.). (2014). Input and experience in bilingual development. Philadelphia, PA: John Benjamins. http://dx.doi.org/10.1075/tilar.13 Leopold, W. (1939–1949). Speech development of a bilingual child: A linguist’s record (Vols. 1–4). Evanston, IL: Northwestern University. Macnamara, J. (1966). Bilingualism and primary education. Edinburgh, Scotland: Edinburgh University Press. Maneva, B., & Genesee, F. (2002). Bilingual babbling: Evidence for language differentiation in dual language acquisition. In B. Skarabela, S. Fish, & A. H.-J. Do (Eds.), The Proceedings of the 26th Boston University Conference on Language Development (pp. 383–392). Somerville, MA: Cascadilla Press. Meisel, J. M. (1989). Early differentiation of languages in bilingual children. In K. Hyltenstam & L. Obler (Eds.), Bilingualism across the lifespan: Aspects of acquisition, maturity, and loss (pp. 13–40). Cambridge, England: Cambridge University Press. http://dx.doi.org/10.1017/ CBO9780511611780.003 Mercier, J., Pivneva, I., & Titone, D. (2014). Individual differences in inhibitory control relate to bilingual spoken word processing. Bilingualism: Language and Cognition, 17, 89–117. http://dx.doi.org/10.1017/S1366728913000084 Paradis, J., Crago, M., Genesee, F., & Rice, M. (2003). Bilingual children with specific language impairment: How do they compare with their monolingual peers? Journal of Speech, Language, and Hearing Research, 46, 113–127. http://dx.doi.org/10.1044/1092-4388(2003/009) Paradis, J., & Genesee, F. (1996). Syntactic acquisition in bilingual children: Autonomous or interdependent? Studies in Second Language Acquisition, 18, 1–25. http://dx.doi.org/10.1017/ S0272263100014662 Paradis, J., Nicoladis, E., & Genesee, F. (2000). Early emergence of structural constraints on codemixing: Evidence from French–English bilingual children. Bilingualism: Language and Cogni­ tion, 3, 245–261. http://dx.doi.org/10.1017/S1366728900000365



Shifting Perspectives on Bilingualism 

 19

Peal, E., & Lambert, M. (1962). The relation of bilingualism to intelligence. Psychological Monographs: General and Applied, 76(27), 1–23. http://dx.doi.org/10.1037/h0093840 Pierce, L., Chen, J. K., Delcenserie, A., Genesee, F., & Klein, D. (2014). Persistent effects of the birth language on later neural processing. Unpublished manuscript. Pivneva, I., Palmer, C., & Titone, D. (2012). Inhibitory control and 12 proficiency modulate bilingual language production: Evidence from spontaneous monologue and dialogue speech. Frontiers in Psychology, 3, 57. http://dx.doi.org/10.3389/fpsyg.2012.00057 Ronjat, J. (1913). Le developpement du language observé chez un enfant bilingue. Paris, France: Edouard Champion. Saer, D. J. (1923). The effect of bilingualism on intelligence. British Journal of Psychology, 14, 25–38. Sauve, D., & Genesee, F. (2000). Grammatical constraints on child bilingual code mixing. Annual Conference of the American Association for Applied Linguistics, Vancouver, British Columbia, Canada. Scotton, C. M., & Ury, W. (1977). Bilingual strategies: The social functions of code-switching. Inter­ national Journal of the Sociology of Language, 13, 5–20. Tao, L., Marzecová, A., Taft, M., Asanowicz, D., & Wodniecka, Z. (2011). The efficiency of attentional networks in early and late bilinguals: The role of age of acquisition. Frontiers in Psychology, 2, 123. http://dx.doi.org/10.3389/fpsyg.2011.00123 Vihman, M. M. (1985). Language differentiation by the bilingual infant. Journal of Child Language, 12, 297–324. http://dx.doi.org/10.1017/S0305000900006450 Volterra, V., & Taeschner, T. (1978). The acquisition and development of language by bilingual children. Journal of Child Language, 5, 311–326. http://dx.doi.org/10.1017/S0305000900007492 Whitford, V., & Titone, D. (2012). Second-language experience modulates first- and second-language word frequency effects: evidence from eye movement measures of natural paragraph reading. Psychonomic Bulletin & Review, 19, 73–80. http://dx.doi.org/10.3758/s13423-011-0179-5 Whitford, V., & Titone, D. (2014). Eye movements and the perceptual span during first- and secondlanguage sentence reading in bilingual older adults. Manuscript submitted for publication. Whitford, V., & Titone, D. (2015). Second-language experience modulates eye movements during first- and second-language sentence reading: Evidence from a gaze-contingent moving window paradigm. Journal of Experimental Psychology: Learning, Memory, and Cognition, 41, 1118–1129. Zwanziger, E. E., Allen, S. E., & Genesee, F. (2005). Crosslinguistic influence in bilingual acquisition: Subject omission in learners of Inuktitut and English. Journal of Child Language, 32, 893–909. http://dx.doi.org/10.1017/S0305000905007129

I Early Bilingualism

Suzanne Quay and Simona Montanari

2 Early Bilingualism: From Differentiation to the Impact of Family Language Practices In recent years, the field of language acquisition has come to recognize that exposure to two languages in infancy is almost as common as monolingual exposure and that any theory of language development should also explain the early acquisition of two languages. Consequently, bilingual first-language acquisition, or BFLA (referring to both bilingual and multilingual development), has become the focus of much research around the world. Despite this interest, however, studies of bilingual development have been dominated, at least until very recently, by a monolingual perspective with bilingual speakers being typically seen as “two monolinguals in one.” Thus, one of the major issues that has preoccupied researchers has been that of language differentiation (i.e., whether and when young bilingual children differentiate language-specific sounds, words, and rules for each of their languages). This focus arose after Volterra and Taeschner (1978) proposed that bilingual children initially use their languages indiscriminately because of a unitary or fused system at the phonological, lexical, and syntactic levels (the unitary language system hypothesis). According to them, bilingual children differentiate their words into two separate lexicons only in a second stage, and around age 3—in a third stage—they differentiate their grammatical rules into two distinct sets, one for each language. Although many studies on BFLA followed Volterra and Taeschner’s proposal, investigators also began to look at children’s ability to differentiate languages in context, that is, while addressing speakers of distinct languages. The differentiated language system hypothesis, proposed by Genesee (1989), claimed that by the time bilingual-to-be children begin talking, they show signs of differentiating their phonology, lexicon, and language choices. Likewise, as soon as there is evidence for syntax, they use their two syntactic systems differentially. The proposal of an initially undifferentiated system in BFLA was the basis of bilingual–monolingual comparisons and implied that bilingually exposed babies experience prolonged language development during the process of learning two languages. However, there has been a trend in the past decade to move away from viewing, studying, and assessing bilingual children as two monolingual children in one. The literature reviewed in this chapter shows how studies related to BFLA have evolved over the years from an analysis of the emergence of separate linguistic systems to a focus on the particular features of bilingual speech or the characteristics of dual language environments and practices. After reviewing the work on differentiation and development in early bilingualism, this chapter discusses literature on the impact of the language environment in terms of family language policies and the role of input from parents, siblings and grandparents. Current perspectives that accept code-mixing http://dx.doi.org/10.1037/14939-003 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

24 

 Quay and Montanari

(the use of two or more languages in discourse) as a normal part of bilingual practices are also discussed. This chapter concludes with implications from emerging fields and from conceptual changes in studying early bilingual speakers that affect societal support for, and the future education of, young emergent bilingual speakers.

Differentiation and Development in Early Bilingualism Dual Lexical Resources The production of translation equivalents (words that have the same meaning across the two languages, such as book and its Spanish equivalent libro) is a clear indication of the development of two languages. A number of investigators in the last two decades have shown that bilingual-to-be children acquire translation equivalents both at early and at later stages of lexical development. These children thus seem to develop separate lexicons from early in development. Pearson, Fernández, and Oller (1995), for instance, documented the existence of equivalents in the early vocabularies (2–12 words) and later vocabularies (up to 500 words) of 27 Spanish–English bilingual children and found that the average percentage of doublets was around 30% at all observations. Holowka, Brosseau-Lapré, and Petitto (2002), who examined bilingual lexical development in children learning French and English and in children learning French and Langue des signes québecoise (Quebec sign language), found strikingly similar results for both groups. Other case studies of individual bilingual and trilingual children learning a variety of languages have also documented parallel percentages of equivalents at both early and later stages (Deuchar & Quay, 2000; Montanari, 2010; Nicoladis & Secco, 2000). Overall, these results suggest that developing multilingual children have language-specific lexical resources from the earliest stages of vocabulary development. Although some studies have documented few equivalents before age 1 year, 5 months (Montanari, 2010; Nicoladis & Secco, 2000), a rate of equivalents as high as 20% to 25%, which is typical after this age, has been interpreted as evidence of separate lexical systems starting from relatively early in development. Deuchar and Quay (2000), however, argued that the appearance of translation equivalents is a necessary but “not a sufficient condition for establishing lexical differentiation” (p. 64). Lexical differentiation can only be established when appropriate language choice is apparent, that is, when children are able to use lexical resources in a specific language with speakers of that language.

Language Choice and Mixing in Discourse Early theorists did not necessarily take into account the availability of dual lexical resources when they examined inappropriate language choices. Rather, they inter-



Early Bilingualism 

 25

preted any form of code-mixing—the use of a language other than the interlocutor’s language—as evidence that children could not differentiate languages pragmatically (i.e., in context). However, because inappropriate language choices, especially in the earliest stages of BFLA, may be due to a lack of lexical resources, more recent studies moved away from seeing code-mixing as a sign of an undifferentiated language system. Nicoladis and Genesee (1996), for instance, who found four French– English bilingual toddlers unable to choose language appropriately with their parents early on, suggested that limited vocabulary might have been responsible for this lack of differentiation. Indeed, appropriate language choice did emerge between ages 1 year, 9 months and 2 years, 4 months when children had more lexical resources in each language to use with each parent. Recent studies of language choice also interpret code-mixing as a typical and positive communicative strategy among young bilingual speakers who are sensitive to their interlocutors’ linguistic needs, in particular, their language skills, use, and preferences. For example, Comeau, Genesee, and Lapaquette (2003) found that 2- to 3-year-old French–English bilingual children were capable of varying their patterns of language choice following the lead of unfamiliar interlocutors who changed their rates of mixing over three sessions. This finding was interpreted as evidence that young bilingual speakers can also monitor language input on-line and adjust their language use accordingly. Studies of language choice have also been extended to children learning more than two languages simultaneously. Quay (2008) and Montanari (2009) showed, for instance, that even trilingual 2-year-olds can show interlocutor sensitivity when raised following the one-person-one-language approach (OPOL). Both the Japanese– Chinese–English 2-year-old in Quay (2008) and the Tagalog–Spanish–English toddler in Montanari (2009) modified their patterns of language choice when faced with multiple language users, following their interlocutors’ language preferences, proficiency, and expectations as to appropriate language use. Quay (2011) further found that two trilingual toddlers who were learning the majority language in the day care were even able to show discourse separation (the use of the interlocutor’s language in at least 90% of their utterances) in the day care environment. Overall, these studies point to young bilingual children’s ability to differentiate their languages in context from very early on in development, if not from their very first words. It is also now accepted among investigators that code-mixing tends to result from vocabulary gaps, namely, the lack—or inexistence—of specific words in the language of the interactional context.

Vocabulary Size The early lexicon—and the way words are used—reflect a complex interaction between input from the environment and vocabulary learning. That is, bilingual children cannot be expected to have translation equivalents for words they never hear in their

26 

 Quay and Montanari

environment or to make appropriate language choices with speakers who regularly code-mix. For these reasons, researchers in the last decade have moved away from the issue of lexical and pragmatic differentiation and have focused instead on the size of bilingual vocabularies as compared with monolingual vocabularies. Unfortunately, bilingual–monolingual comparisons have often considered each of the bilingual speaker’s languages separately, and they have not taken into account the link between language exposure and word production, mistakenly revealing a bilingual disadvantage. Thus, bilingual children have been typically found to possess smaller vocabularies in each language when compared with monolingual peers (as indicated by both productive and receptive measures of vocabulary size), both at early (Conboy & Thal, 2006; Hoff et al., 2012) and at later ages (Bialystok, Luk, Peets, & Yang, 2010; Oller & Eilers, 2002). Hoff et al. (2012) found, for instance, that 47 Spanish–English bilingual toddlers lagged behind monolingual toddlers of equivalently high socioeconomic status if assessed in only one language; that is, their English production vocabularies were smaller than those of monolingual peers. Similarly, using a large sample of 1,738 children, Bialystok et al. (2010) demonstrated that bilingual school-age children had lower English receptive vocabulary scores than monolingual speakers at all tested ages, although this vocabulary gap was evident only for words used in a home context but not for words used in a school context. Many recent studies of young bilingual children, however, have also shown that when the two languages are assessed in terms of total as well as conceptual vocabularies (total minus translation equivalents), these children perform as well as monolingual peers in their receptive and expressive lexicon (De Houwer, Bornstein, & Putnick, 2014; Hoff et al., 2012; Poulin-Dubois, Bialystok, Blaye, Polonia, & Yott, 2013; Poulin-Dubois, Blaye, Coutya, & Bialystok, 2011). Bialystok et al. (2010) pointed out, for instance, that the lack of home words in English was probably filled by knowledge of those words in the home language, “making it likely that the total vocabulary for bilingual children is in fact greater than that of monolinguals” (p. 530). Poulin-Dubois et al. (2011) further found that bilingual children performed better than monolingual children in tests that assessed cognitive skills like executive functions, and attributed this finding to bilingual children’s greater attentional control during language use because of simultaneous language activation. Poulin-Dubois et al. (2013) found, moreover, that bilingual 2-year-olds with a higher proportion of translation equivalents in their expressive vocabulary showed faster access to words in a comprehension task. Overall, all these studies indicate that bilingual children do not lag behind monolingual children when their total vocabulary knowledge is considered. Also, a clear relation exists between amount of language exposure and lexical development, a finding that emphasizes the importance of input for dual language acquisition. As put by De Houwer et al. (2014), bilingualism does not cause delay: Instead of focusing on bilingual-monolingual comparisons, examining variation in children’s language exposure, regardless of the number of languages they are acquiring, holds better promise for understanding the large variability in early lexical development. (p. 1209)



Early Bilingualism 

 27

Clearly, future studies of bilingual lexical development should move away from comparing bilingual to monolingual speakers and should examine instead the specific characteristics of language environments and practices that promote vocabulary learning in two languages.

Grammatical Development Although early theorists proposed that bilingual children go through an initial stage in which “the same syntactic rules are applied to both languages” (Volterra & Taeschner, 1978, p. 312), research in the last two decades has provided overwhelming evidence that young bilingual speakers acquire separate grammatical properties and word order patterns for each language from early in development (Meisel, 1994; Paradis & Genesee, 1996). De Houwer (2005), for instance, reviewing results from 68 studies on a wide variety of morphosyntactic topics, ranging from verb morphology (e.g., he jump-s; she talk-ed), determiners (e.g., the), negation (e.g., not), and word order, to name a few, found that young bilingual speakers acquire the grammatical and syntactic properties of each language as two fundamentally different systems, developing them “along two separate morpho-syntactic paths” (De Houwer, 2005, p. 39). For example, French– English bilingual toddlers learn verb morphology and negation in French and English at different ages, clearly indicating the emergence of separate grammars (Paradis & Genesee, 1996). The patterns produced by the bilingual speakers also appear to mirror, in part, those exhibited by monolingual peers, suggesting that bilingual speakers do not learn each language in a way that is fundamentally different from the case of monolingual acquisition. More evidence of the separate development of bilingual speakers’ morphosyntactic systems comes from studies that examine the connection between vocabulary size and grammatical learning in bilingual development. Conboy and Thal (2006), for example, examined the correlation between lexical and grammatical abilities in 64 Spanish– English bilingual toddlers and found that grammatical development in each language (as measured by utterance length and sentence complexity) was correlated to vocabulary size in the same language but not in the other. Similarly, Marchman, MartínezSussmann, and Dale (2004) found that within-language vocabulary–grammar associations were stronger than cross-language relationships in 113 Spanish–English toddlers. Overall, these results suggest not only that grammatical learning is specifically tied to lexical learning in a given language, but they provide further evidence that “in early simultaneous bilingualism, development in each language proceeds separately and in a language-specific manner” (Conboy & Thal, 2006, p. 727). Although investigators agree that bilingual children build separate grammatical systems, there is also consensus that these systems do not develop autonomously but interdependently, especially when grammatical constructions overlap at the surface level for both languages and when such constructions are at the interface between

28 

 Quay and Montanari

syntax and pragmatics (Paradis & Genesee, 1996; Serratrice, Sorace, Filiaci, & Baldo, 2012). As a result, much recent research has examined the nature and extent of this cross-language influence in a variety of constructions and language pairs. Overall, this research has shown that young bilingual speakers might use the word order patterns of one language in the other language, producing atypical constructions. For example, a Spanish–English bilingual child might place an adjective before a noun in Spanish following English word order even if Spanish does not always allow this construction—as in un rojo carro, literally, “a red car.” Young bilingual speakers may also omit elements that are obligatory in one language because their other language allows such omissions, or they might use a structure that exists in one language in contexts in which such use is not semantically or pragmatically appropriate (Silva-Corvalán & Montanari, 2008). Although such atypical constructions had led early theorists to postulate an early “unified syntax” for both languages, research has now shown that instances of cross-language interactions are infrequent (Chapter 14, this volume). Yet, they can persist well into the late elementary school years (Serratrice et al., 2012), suggesting that such interactions might be a permanent feature of bilingualism. Recent investigations of bilingual grammatical development have also been focused on bilingual–monolingual comparisons and, as in the case for lexical learning, they have found bilingual speakers to lag behind monolingual peers when measured in each language separately (Gathercole & Thomas, 2009; Hoff et al., 2012). However, results have also differed as a function of the social context in which the studies were conducted. For instance, Paradis, Crago, and Genesee (2005–2006) found French– English bilingual 3-year-olds in Montreal to be as accurate as French monolingual peers in the use of French object pronouns; on the other hand, Pérez-Leroux, Pirvulescu, and Roberge (2009) found same-age bilingual children from Toronto, an English-majority city, to lag behind monolingual peers in the same construct. Studies of older bilingual children have revealed parallel findings, documenting larger differences between bilingual and monolingual speakers in contexts that strongly favor monolingualism (e.g., the United States; see Oller & Eilers, 2002) rather than bilingualism (e.g., Wales; see Gathercole & Thomas, 2009). Although it is impossible to ascertain whether these effects are simply due to the children’s dominance or rather the context in which the studies were set, it is important that future research focuses on the extent to which characteristics of the social context such as the prestige of the languages involved and the opportunities to use each language can affect, enhance, or slow down bilingual grammatical development.

Phonological Differentiation and Development The first part of this chapter concludes with a discussion of bilingual speech sound acquisition because research in this area (which has focused on the emergence of



Early Bilingualism 

 29

different aspects of phonology in the two languages and the extent to which these elements are differentiated) has yielded inconclusive results with a multiplicity of findings across different languages and populations. For instance, studies focusing on the characteristics of bilingual children’s early vocalizations (such as the types of sounds and syllables produced during babbling) have been interpreted as evidence of both presence and lack of differentiation. In particular, some investigations have reported differentiated babbling depending on the context in which the babbling occurred (e.g., Maneva & Genesee, 2002, where an infant’s babbling with his French-speaking father displayed French-specific patterns whereas his babbling with his English-speaking mother was more “English-like”). Others, however, have found that babbling reflected the features of only one language irrespective of linguistic context (Poulin-Dubois & Goodz, 2001). Likewise, analyses of the sound characteristics of early words yielded inconclusive results. Some studies of the sound inventories and error patterns produced by individual bilingual and trilingual children speaking a variety of languages reported signs of differentiation despite influence between phonological systems (e.g., Montanari, 2011). For example, children displayed sound inventories and substitutions (e.g., replacing /r/ with /w/ in the word red) that were fundamentally language-specific, that is, different between languages. In contrast, Vihman (2002) proposed that bilingual speakers start “without a system” with phonetic patterns and individual words learned holistically and with similar word templates across both languages. Like Vihman, Goldstein and Gildersleeve-Neumann (2012) also proposed that simultaneous bilingual children before age two do not contrast similar phonemes or word shapes in their two languages. Studies of the acoustic characteristics of specific sounds in the two languages have also revealed the same mixed results, with some studies reporting differentiated patterns between languages (like the child in Deuchar & Quay, 2000, who learned the voicing contrast— distinguishing /p/ vs. /b/—in Spanish and English at different ages) while others conclude that early lack of differentiation, delay, and un-uniform development are all possible paths in bilingual phonological acquisition (e.g., Kehoe, Lleó, & Rakow, 2004). Overall, the consensus is that individual children may show different developmental trajectories while establishing separate phonological systems. Therefore, researchers in the last ten years have moved away from the issue of phonological differentiation and have been more concerned with the rate of phonological development in bilingual children and the extent to which the phonological systems interact in development. Interestingly, research on the rate of development has also shown different outcomes. Focusing primarily on Spanish–English bilingual speakers in the United States, some studies have shown evidence of delayed acquisition of sounds in one or both languages compared with monolingual speakers (Fabiano-Smith & Goldstein, 2010; Gildersleeve-Neumann, Kester, Davis, & Peña, 2008). Yet, other comparable studies have found no difference (Burrows & Goldstein, 2010; Fabiano-Smith & Barlow, 2010) or even an accelerated rate of acquisition in bilingual speakers compared with monolingual peers (Goldstein & Bunta, 2012; Grech & Dodd, 2008).

30 

 Quay and Montanari

Although there is a lack of agreement as to whether bilingual children develop speech sounds at a slower, comparable, or faster rate than monolingual peers, investigators agree that there are qualitative differences and more variation in the speech production of bilingual speakers. Researchers have interpreted differences from monolingual patterns as evidence of interaction between early bilingual speakers’ phonological systems, leading to negative transfer of elements across languages. That is, children might use sounds that are specific to only one language in the other language’s productions, producing phonological patterns that deviate from monolingual speech. Several studies have indeed documented larger error rates in English words for Spanish–English bilingual children in the United States compared with monolingual children (Fabiano-Smith & Goldstein, 2010; Gildersleeve-Neumann et al., 2008). At the same time, however, studies that have controlled for language use and proficiency and involved slightly older children have found evidence of positive transfer, with higher accuracy rates and fewer phonological and articulation errors in bilingual speech than in monolingual speech (Goldstein & Bunta, 2012; Grech & Dodd, 2008). It was argued, in these studies, that bilingual speakers transferred phonetically similar sounds from one language to the other, facilitating the acquisition and accurate production of these sounds in both languages. Yet, although it is universally accepted that phonological systems interact in bilingual speakers’ development, the amount and direction of transfer have been found to vary considerably. In sum, contrasting findings and large individual differences are hallmarks of the research on bilingual phonological development. It is possible that the heterogeneity within the population of bilingual children per se—that is, their different language exposure patterns, proficiency levels and phonological learning style—might explain these contrasting results. It is also possible that findings are further confounded by differences between languages in sound inventories, sound complexity, and sound frequency as language pairs characterized by phonological similarity may less easily yield evidence of differentiation—or of negative transfer—than languages with less phonological overlap. Also, societal and cultural factors may play a crucial, yet understudied, role. The course of bilingual phonological development might indeed differ depending on the characteristics of the social context (e.g., the prestige of the languages, the extent of contact between speakers, the amount of code-switching by and language skills of input providers) as well as the child’s socialization process (i.e., the amount and type of social interactions and the use of infant-directed speech; see the following section). Future studies should examine how all these factors interact with individual variation, possibly producing different patterns of development and interaction.

Summary The studies reviewed here indicate that separation of the two languages is evident, perhaps with the exception of phonology, from very early in development. The course



Early Bilingualism 

 31

of bilingual acquisition does not differ considerably from the course of language development in children learning only one language. That is, first-language acquisition is fundamentally the same for all typically developing children whether they are raised monolingually or bilingually. However, qualitative differences between monolingual and bilingual language acquisition exist. Bilingual children, unlike mono­lingual children, alternate their languages in discourse from their first words for reasons that range from filling in vocabulary gaps to fulfilling sophisticated communicative functions (see Chapter 5, this volume). Bilingual speakers may also transfer features from one language to the other, a process that may facilitate or delay the acquisition of these elements in either one or both languages (see Chapter 14). In addition, because the environment might better support the development of one language over the other, the bilingual speaker’s two languages might not develop in a parallel manner (see Chapter 7). Finally, social, cultural, and input-related factors may assume disproportionately high significance in influencing the course and rate of bilingual development as compared with their influence in the case of monolingualism. The next section addresses the importance of these factors, in particular, the crucial role of home and family language practices for successful bilingual outcomes.

The Importance of Home and Family Language Practices Recent research has begun to focus on the characteristics of the social and inter­ actional contexts in which the language learning of young bilingual speakers takes place as well as on the specific parental/familial strategies that might affect such development. In this line of research, the socialization process—and the home language environment where it occurs—becomes the focus of study as it provides a “community of practice” where children learn to speak, act, and believe according to the family’s own norms. This section focuses on the critical role that the input from immediate caregivers and family members has in forming the child’s linguistic environment and hence of promoting bilingual development.

Family Language Policies The research field of family language policy (FLP) is relatively new and underexplored (King, Fogle, & Logan-Terry, 2008). It is a growing field that connects child language acquisition studies with language policy to examine how parents’ language decisions, practices, and beliefs influence child outcomes (e.g., Fogle & King, 2013). King et al. (2008) defined FLP as conscious planning that takes into account parental language goals, attitudes, or intentions in relation to language use in the family. Having an FLP,

32 

 Quay and Montanari

however, does not guarantee success as external factors outside the home such as community and societal values and educational policies may affect the good intentions of families to raise bilingual children. Work on FLP tries to close the gap between macro­ issues (language ideology and language attitudes) and microissues (interactional patterns in bilingual homes; King et al., 2008), bringing together ideology (beliefs about language use), management (intervention), and practice (daily language choice). In a review of the theoretical perspectives in the study of FLP, Schwartz and Verschik (2013) discussed the family as an important domain for studying language policy. They pointed out that Fishman’s model for reversing language shift indicates that at the earliest stages of reviving or making languages sustainable, the family is critical for the maintenance of ethnic languages. Without intergenerational language transmission between parents and children in the home, it is not effective to campaign for the use of a language in education, the work place, government, and mass media (the later stages of Fishman’s model). The family and home community provide the initial stage in children’s language socialization and form their closest language ecology. Spolsky’s (2007) language policy model also discusses how the home language ecology is quickly affected by external domains such as the language practices and beliefs of peers in the neighborhood and of teachers in the school context. According to Spolsky (2007), “schools reflect the ideological position of those who control them,” and the “school domain is the one most likely to be influenced externally, whether from ‘below’ (home, religion, neighborhood) or ‘above’ (levels of government), and to be most often the target of activist intervention in support of one variety or another” (p. 11). Clearly, the language environment children are in is variable and part of this variation is organized by social and political factors. The interdisciplinary nature of FLP studies also encompasses earlier inter­ actional studies, such as Lanza (1997), that focused on the role of parental language teaching strategies in early bilingual development. Recent studies have looked as well at the psychological and emotional aspects of FLP. Kopeliovich (2013), for example, discussed the code-switching policies of her family in Israel to demonstrate that the family was at an intermediate level between the individual and the community. In particular, she recounted how she went against the community’s pressure for linguistic purism for the sake of raising four “happy lingual” Russian–Hebrew children. Tannenbaum (2012) further argued that FLP should address the emotional aspects of home language maintenance and loss more seriously and that language decisions made by immigrant families can be reconceptualized as defense and coping mechanisms. Such studies show FLP to be a complex and interdisciplinary research domain.

Input in the Home—Parents, Siblings, and Grandparents Why is there so much variability in the ability and use of bilingual children’s languages? De Houwer (2011) argued convincingly that such variability results from dif-



Early Bilingualism 

 33

ferences in language input environments (e.g., parental language use patterns, relative and absolute frequencies of input for each language, sociointeractional strategies). She discussed the findings from the literature on early second-language acquisition (ESLA) in comparison with BFLA. Notably, ESLA children with less exposure to the SL input do not develop at the same rate as BFLA children who have heard the same language from birth. De Houwer (2011) pointed out that even BFLA children do not all end up speaking two languages equally well and may even speak only one of their languages. In a survey of 1,899 bilingual families in Flanders, De Houwer (2007) found that 25% of children reared bilingually from early on failed to speak two languages between the ages of 6 and 10 years. Because both their parents used the majority language, Dutch, at home, with one parent also using the minority language, the researchers surmised that not enough input in the minority language had been provided for children to become active bilingual speakers. Yamamoto (2001) also found that majority language use in the home by parents affected active bilingualism in Japanese–English families in Japan. Place and Hoff (2011), who collected detailed diary records and MacArthur-Bates inventories from the mothers of 29 Spanish–English 25-month-olds, further confirmed that the relative amount of exposure to each language predicted bilingual outcomes. Interestingly, not just the amount but specific properties of the input, such as different sources and native speakers, improved the English outcomes for these 2-year-olds in South Florida. It was found that parents who used their weaker language (English in this case) for communication at home exposed their children to less language as well as to less complex language, which in turn resulted in their children producing a lower number of English words (Place & Hoff, 2011). Adding to this line of research, Mayr and Montanari (2015) found that the number and type of input providers affected phonological development in each language differently in trilingual children. The sisters in the study showed different developmental patterns in the two minority languages—Italian and Spanish—that have virtually identical voiceless stops. In particular, their Italian phonological system showed interaction with English, but the Spanish sound system was unaffected by it. These differences were attributed to the nature of the input in each language: Italian was being provided exclusively by bilingual Italian–English speakers—some being native and some nonnative. Input in Spanish, on the other hand, was solely provided by a single, monolingual, and native speaker, who might have provided the children with a more consistent and unambiguous input from which to extract language-specific information and avoid cross-language interaction. In other words, specific characteristics of the input affected development in subtle and unexpected ways. In addition to amount of exposure and the number and type of input providers, sociocultural factors, such as use of infant-directed speech (IDS), have been shown to affect language development. IDS is the kind of speech that adults typically produce when talking to infants, characterized by higher and more varied pitch, longer pauses, and shorter phrases than typical adult-to-adult speech. In IDS, phonetic distinctions are exaggerated, making important phonetic information (e.g., single words, stressed

34 

 Quay and Montanari

syllables, specific vowels) more salient. Many studies of monolingual children learning a variety of languages have shown that infants are not only more attracted to infant- than to adult-directed speech but they better learn certain language elements when exposed to IDS (Liu, Kuhl, & Tsao, 2003). However, in the case of bilingualism, infants might only hear IDS in one language or they might hear it to different extents in each language. These early differences in input characteristics may set bilingual infants on different growth trajectories, making development in one language look different from the other language or from that of monolingual speakers. RamírezEsparza, García-Sierra, and Kuhl (2010) found, for instance, that English monolingual and Spanish–English bilingual infants produced more babbling with the caregiver who used more IDS than with the one who did not, suggesting that differences in the extent to which caregivers use this register may affect BFLA. Recent research also suggests that social engagement is very important, if not crucial, for early language learning. Kuhl, Tsao, and Liu (2003) showed, for instance, that monolingual English-learning infants exposed to Mandarin in naturalistic play sessions over 2 months were able to learn a Mandarin sound contrast similarly to monolingual Mandarin-learning infants. However, English-learning infants who watched videos or listened to audio recordings of these play sessions failed to learn this contrast. Other studies suggest that joint attention episodes—where both infant and adult are attending together to an object or event—are particularly conducive to language learning (Conboy, Brooks, Meltzoff, & Kuhl, 2015). These findings suggest that characteristics of the social context along with infants’ ability to engage with caregivers strongly influence language learning. In the case of bilingual speakers, infants may have different degrees of social engagement with their input providers, and this may result in different learning experiences in each language. Parental discourse strategies are also important in promoting bilingual development. Much of the research in early bilingualism has focused on the OPOL approach (e.g., Juan-Garau & Pérez-Vidal, 2001) where one parent uses a majority language and the other parent a minority one. Some research has also reported on families using two minority home languages according to the OPOL approach while raising their children in a third-language environment (Montanari, 2009; Quay, 2012). There are also OPOL cases of minority language use in the home (Deuchar & Quay, 2000; Kennedy & Romo, 2013) and cases in which one parent uses a nonnative minority language in the home (King et al., 2008; Nakamura & Quay, 2012). Although no studies have reported unsuccessful outcomes for majority language development, the results are variable for the acquisition of minority languages. Successful bilingual development appears to be based on parents’ consistent language choice and use of parental discourse strategies to encourage or ensure the use of the minority language in the home (Lanza, 1997; see Quay, 2012, for a review). Parents speaking the minority language who used strategies like minimal grasp (pretending not to understand the majority language used by their children) were successful in helping children become active bilingual speakers. Parents who allowed a bilingual context with strategies



Early Bilingualism 

 35

like move on and code-switching that demonstrated their own bilingual abilities and their understanding and acceptance of their children’s language choice were less successful. In sum, it appears that active BFLA results more often when parental discourse strategies make use of and enforce the minority language. Interactions in families are not just unidirectional from adults to children but also from children to family members (Fogle & King, 2013). Pearson (2007) pointed out that children can invite more input of a certain language by using this language themselves. Children’s language preferences are often based on interactions in the larger speech community outside the home. In a similar way, the language environment of the child outside the family can also indirectly influence the mother’s language use with her child. Prevoo, Mesman, Van IJzendoorn, and Pieper (2011) found, for instance, that Turkish immigrant mothers in the Netherlands increased their use of Dutch while addressing their toddlers when these children began attending a playgroup or day care center. The increase was particularly significant when the families lived in a neighborhood with a low percentage of Turkish immigrants. Therefore, environmental factors and children’s own language preferences have a strong impact on parents’ language use in spite of FLP. As Prevoo et al. (2011) suggested, parents need to be made aware of this finding so that they can “find an appropriate balance between the use of the ethnic and the host language” (p. 574). Besides parents, the presence of older children and sibling position can also affect the transmission of home languages (Schwartz, 2010). The influence of siblings’ language use on an individual child and the family is related to factors beyond birth order, such as relationship, language use between siblings, gender, age, family size, and so on (Barron-Hauwaert, 2011). Older siblings tend to bring the societal language into the home even when it is against FLP (Caldas & Caron-Caldas, 2002; Kopeliovich, 2013). For example, Kopeliovich (2013), describing the Russian–Hebrew development of four siblings from birth to late childhood, reported that the parents had to change their FLP of using only Russian in their home in Israel once the second child, 3 years younger than the first, was born. With the birth of each sibling, the older ones preferred to use Hebrew among themselves and to their younger siblings, and the parents felt they had to accommodate their children’s language preference for the sake of family harmony. Bridges and Hoff (2014) also found that older siblings used the societal language more to talk to toddlers under age 2 years, 6 months in the United States. The toddlers with older school-age siblings were more advanced in the societal language, English, whereas those without older school-age siblings were more advanced in the home language, Spanish. The presence of school-age older siblings also increased the mothers’ use of the societal language with their toddlers. In contrast to these findings, Kennedy and Romo (2013) described a case in which parents fostered emotional ties to a minority language and asked explicitly for assistance from an older sibling in transmitting this language to a younger one. More detailed studies on actual language interactions between siblings are necessary (Schwartz, 2010), as their effects on FLP are not yet well understood.

36 

 Quay and Montanari

Intergenerational learning also takes place between children and grandparents, who can play an important role in child care. Gregory, Arju, Jessel, Kenner, and Ruby (2007) and Curdt-Christiansen (2013) focused on the important (but often unacknowledged) role that grandparents can play in the language and literacy practices of primary school–age children. For example, Gregory et al. found that a Bengali grandmother socialized her grandchildren not only in reciting choras (rhymes in Bengali with a moral or religious meaning) but also by telling them European traditional tales like Snow White in Bengali, which helped the children better understand the stories that they had heard in English at school. Similarly, Curdt-Christiansen found that Chinese grandmothers in Montreal and Singapore facilitated their grandchildren’s process of becoming literate in multiple languages through their modeling and implicit demonstration of literacy use for a variety of purposes, from wordplay involving writing and guessing riddles to structured literacy practice using workbooks. Ruby (2012) reported on a Bangladeshi grandmother in London who tried to coax the use of the home language from her 6-year-old granddaughter through the practice of code-mixing and informal teaching of Bangla. In a study of a preschooler, Quay (2013) further showed how another code-switching grandmother helped her grandson between ages 3 and 5 to shift dominance from the home language, Chinese, to the societal language, English, in preparation for formal schooling in Canada. In this case, the grandmother decided not to follow the parents’ wish to expose the child mainly to the home language before school because she feared that her grandson would be at a disadvantage when he entered kindergarten. The grandmother provided a bilingual context with her code-mixing practices, which resulted, in turn, in the child’s production of code-mixed utterances that were longer and more complex than those he could produce in each language separately. This study implies that emergent bilingual speakers can be more expressive when allowed to use all their linguistic resources in discourse (see also Chapter 5, this volume).

Current Perspectives and New Directions Although the scientific understanding of bilingualism has evolved considerably during the last decades, and BFLA is now considered to be a normal and advantageous practice in families with two or more languages, social and political realities around the world do not necessarily challenge monolingual and fixed understandings of language and language use. Characteristics of bilingual speech—such as code-mixing—are indeed stigmatized in the community at large, and they continue to be seen as a sign of confusion among bilingual speakers. Even in the educational context, being bilingual has been shown to negatively affect the perception and assessment of not only linguistic skills but also of general academic potential (Berthele, 2012). This is due to a persistent monolingual perspective that views bilingualism as a sort of “double monolingualism” whereby languages should be



Early Bilingualism 

 37

used as separate entities. Such a monolingually oriented social context deeply influences emergent bilingual children’s language development, their language choices, and their educational outcomes. The trend to study BFLA as part of FLP is expected to increase awareness that the varied learning environments in which bilingual children are raised in the home in the early years and outside the home in child care facilities and educational institutions strongly affect their language and academic learning. For instance, the research is only just starting to examine the effects of interlocutors other than parents (e.g., siblings, grandparents) and varied language contexts (from monolingual to bilingual ones) on bilingual development. Investigators have also begun to discuss the challenges parents face when maintaining bilingualism in children who enter school or become adolescents (Schwartz, 2010). What parents can accomplish in terms of FLP in the early years may indeed be lost in the later years when teenagers refuse to speak the minority language because peer influence and societal attitudes play a more important role in their lives (Caldas & Caron-Caldas, 2002). Thus, raising bilingual children after the early years becomes increasingly challenging when societal and external factors intervene. Recent studies of BFLA point toward a bilingual/multilingual lens rather than a monolingual one to view bilingual children. With this lens, being raised with two languages is just as typical as being raised with one, and alternating languages in discourse is seen as a natural and effective means of communication. De Houwer (2013) also proposed a new area of investigation, that of harmonious bilingual development, a situation where “children and their families do not experience any interpersonal problems because of the language contact situation, or have a clearly positive experience with bilingualism” (p. 2). The well-being of young bilingual children (cf. Kopeliovich, 2013) is the focus of this area of investigation, as bilingual child rearing may not lead to successful bilingual outcomes when negative attitudes exist on the part of child care professionals, educators, and peers toward early bilingualism or any of the languages involved. How negative attitudes from society at large and input factors interact in dual language development is an area that still needs further investigation. Future studies should specifically focus on the ever-changing characteristics of the social context, the language environment, and family practices that affect whether children become and remain productive bilingual speakers from infancy to adulthood.

References Barron-Hauwaert, S. (2011). Bilingual siblings: Language use in families. Bristol, England: Multilingual Matters. Berthele, R. (2012). The influence of code-mixing and speaker information on perception and assessment of foreign language proficiency: An experimental study. The International Journal of Bilingualism, 16, 453–466. http://dx.doi.org/10.1177/1367006911429514

38 

 Quay and Montanari

Bialystok, E., Luk, G., Peets, K. F., & Yang, S. (2010). Receptive vocabulary differences in monolingual and bilingual children. Bilingualism: Language and Cognition, 13, 525–531. http://dx.doi.org/ 10.1017/S1366728909990423 Bridges, K., & Hoff, E. (2014). Older sibling influences on the language environment and language development of toddlers in bilingual homes. Applied Psycholinguistics, 35, 225–241. http:// dx.doi.org/10.1017/S0142716412000379 Burrows, L., & Goldstein, B. A. (2010). Whole word measures in bilingual children with speech sound disorders. Clinical Linguistics & Phonetics, 24, 357–368. http://dx.doi.org/10.3109/ 02699200903581067 Caldas, S., & Caron-Caldas, S. (2002). A sociolinguistic analysis of the language preferences of adolescent bilinguals: Shifting allegiances and developing identities. Applied Linguistics, 23, 490–514. http://dx.doi.org/10.1093/applin/23.4.490 Comeau, L., Genesee, F., & Lapaquette, L. (2003). The Modeling Hypothesis and child bilingual codemixing. The International Journal of Bilingualism, 7, 113–126. http://dx.doi.org/10.1177/ 13670069030070020101 Conboy, B. T., Brooks, R., Meltzoff, A. N., & Kuhl, P. K. (2015). Social interaction in infants’ learning of second-language phonetics: An exploration of brain-behavior relations. Developmental Neuropsychology, 40, 216–229. http://dx.doi.org/10.1080/87565641.2015.1014487 Conboy, B. T., & Thal, D. J. (2006). Ties between the lexicon and grammar: Cross-sectional and longitudinal studies of bilingual toddlers. Child Development, 77, 712–735. http://dx.doi.org/ 10.1111/j.1467-8624.2006.00899.x Curdt-Christiansen, X. L. (2013). Implicit learning and imperceptible influence: Syncretic literacy of multilingual Chinese children. Journal of Early Childhood Literacy, 13, 348–370. http://dx.doi.org/ 10.1177/1468798412455819 De Houwer, A. (2005). Early bilingual acquisition: Focus on morphosyntax and the separate development hypothesis. In J. F. Kroll & A. M. B. de Groot (Eds.), Handbook of bilingualism: Psycholinguistic approaches (pp. 30–48). New York, NY: Oxford University Press. De Houwer, A. (2007). Parental language input patterns and children’s bilingual use. Applied Psycholinguistics, 28, 411–424. http://dx.doi.org/10.1017/S0142716407070221 De Houwer, A. (2011). Language input environments and language development in bilingual acquisition. Applied Linguistics Review, 2, 221–240. http://dx.doi.org/10.1515/9783110239331.221 De Houwer, A. (2013). Harmonious bilingual development: Young families’ well-being in language contact situations. International Journal of Bilingualism. Advance online publication. http:// dx.doi.org/10.1177/1367006913489202 De Houwer, A., Bornstein, M., & Putnick, D. L. (2014). A bilingual–monolingual comparison of young children’s vocabulary size: Evidence from comprehension and production. Applied Psycholinguistics, 35, 1189–1211. http://dx.doi.org/10.1017/S0142716412000744 Deuchar, M., & Quay, S. (2000). Bilingual acquisition: Theoretical implications of a case study. Oxford, England: Oxford University Press. Fabiano-Smith, L., & Barlow, J. A. (2010). Interaction in bilingual phonological acquisition: Evidence from phonetic inventories. International Journal of Bilingual Education and Bilingualism, 13, 81–97. http://dx.doi.org/10.1080/13670050902783528 Fabiano-Smith, L., & Goldstein, B. A. (2010). Phonological acquisition in bilingual Spanish–English speaking children. Journal of Speech, Language, and Hearing Research, 53, 160–178. http:// dx.doi.org/10.1044/1092-4388(2009/07-0064)



Early Bilingualism 

 39

Fogle, L. W., & King, K. A. (2013). Child agency and language policy in transnational families. Issues in Applied Linguistics, 19, 1–25. Gathercole, V. M., & Thomas, E. M. (2009). Bilingual first-language development: Dominant language takeover, threatened minority language take-up. Bilingualism: Language and Cognition, 12, 213–237. http://dx.doi.org/10.1017/S1366728909004015 Genesee, F. (1989). Early bilingual development: One language or two? Journal of Child Language, 16, 161–179. http://dx.doi.org/10.1017/S0305000900013490 Gildersleeve-Neumann, C. E., Kester, E. S., Davis, B. L., & Peña, E. D. (2008). English speech sound development in preschool-aged children from bilingual English–Spanish environments. Language, Speech, and Hearing Services in Schools, 39, 314–328. http://dx.doi.org/ 10.1044/0161-1461(2008/030) Goldstein, B., & Bunta, F. (2012). Positive and negative transfer in the phonological systems of bilingual speakers. The International Journal of Bilingualism, 16, 388–401. http://dx.doi.org/ 10.1177/1367006911425817 Goldstein, B., & Gildersleeve-Neumann, C. (2012). Phonological development and disorders. In B. Goldstein (Ed.), Bilingual language development and disorders in Spanish-English speakers (2nd ed., pp. 285–310). Baltimore, MD: Brookes. Grech, H., & Dodd, B. (2008). Phonological acquisition in Malta: A bilingual language learning context. The International Journal of Bilingualism, 12, 155–171. http://dx.doi.org/10.1177/ 1367006908098564 Gregory, E., Arju, T., Jessel, J., Kenner, C., & Ruby, M. (2007). Snow White in different guises: Interlingual and intercultural exchanges between grandparents and young children at home in East London. Journal of Early Childhood Literacy, 7(1), 5–25. http://dx.doi.org/10.1177/ 1468798407074831 Hoff, E., Core, C., Place, S., Rumiche, R., Señor, M., & Parra, M. (2012). Dual language exposure and early bilingual development. Journal of Child Language, 39, 1–27. http://dx.doi.org/ 10.1017/S0305000910000759 Holowka, S., Brosseau-Lapré, F., & Petitto, L. A. (2002). Semantic and conceptual knowledge underlying bilingual babies’ first signs and words. Language Learning, 52, 205–262. http://dx.doi.org/ 10.1111/0023-8333.00184 Juan-Garau, M., & Pérez-Vidal, C. (2001). Mixing and pragmatic parental strategies in early bilingual acquisition. Journal of Child Language, 28, 59–86. http://dx.doi.org/10.1017/ S0305000900004591 Kehoe, M., Lleó, C., & Rakow, M. (2004). Voice onset time in bilingual German–Spanish children. Bilingualism: Language and Cognition, 7, 71–88. http://dx.doi.org/10.1017/S1366728904001282 Kennedy, K. D., & Romo, H. D. (2013). “All colors and hues”: An autoethnography of a multi­ethnic family’s strategies for bilingualism and multiculturalism. Family Relations: An Interdisciplinary Journal of Applied Family Studies, 62, 109–124. http://dx.doi.org/10.1111/j.1741-3729.2012.00742.x King, K. A., Fogle, L., & Logan-Terry, A. (2008). Family language policy. Language and Linguistics Compass, 2, 907–922. http://dx.doi.org/10.1111/j.1749-818X.2008.00076.x Kopeliovich, S. (2013). Happylingual: A family project for enhancing and balancing multi­lingual development. In M. Schwartz & A. Verschik (Eds.), Successful family language policy: Parents, children and educators in interaction (pp. 249–275). Dordrecht, the Netherlands: Springer. http://dx.doi.org/10.1007/978-94-007-7753-8_11 Kuhl, P. K., Tsao, F. M., & Liu, H. M. (2003). Foreign-language experience in infancy: Effects of short-term exposure and social interaction on phonetic learning. Proceedings of the National

40 

 Quay and Montanari

Academy of Sciences of the United States of America, 100, 9096–9101. http://dx.doi.org/ 10.1073/pnas.1532872100 Lanza, E. (1997). Language mixing in infant bilingualism: A sociolinguistic perspective. Oxford, England: Clarendon Press. Liu, H. M., Kuhl, P. K., & Tsao, F. M. (2003). An association between mothers’ speech clarity and infants’ speech discrimination skills. Developmental Science, 6, F1–F10. http://dx.doi.org/ 10.1111/1467-7687.00275 Maneva, B., & Genesee, F. (2002). Bilingual babbling: Evidence for language differentiation in dual language acquisition. In B. Skarabela, S. Fish, & A. H.-J. Do (Eds.), Proceedings of the 26th Annual Boston University Conference on Language Development (pp. 383–392). Somerville, MA: Cascadilla Press. Marchman, V. A., Martínez-Sussmann, C., & Dale, P. S. (2004). The language-specific nature of grammatical development: Evidence from bilingual language learners. Developmental Science, 7, 212–224. http://dx.doi.org/10.1111/j.1467-7687.2004.00340.x Mayr, R., & Montanari, S. (2015). Cross-linguistic interaction in trilingual phonological development: The role of the input in the acquisition of the voicing contrast. Journal of Child Language, 42, 1006–1035. http://dx.doi.org/10.1017/S0305000914000592 Meisel, J. M. (Ed.) (1994). Bilingual first language acquisition. French and German grammatical development. Amsterdam, the Netherlands: John Benjamins. Montanari, S. (2009). Pragmatic differentiation in early trilingual development. Journal of Child Language, 36, 597–627. http://dx.doi.org/10.1017/S0305000908009112 Montanari, S. (2010). Translation equivalents and the emergence of multiple lexicons in early trilingual development. First Language, 30, 102–125. http://dx.doi.org/10.1177/ 0142723709350528 Montanari, S. (2011). Phonological differentiation before age two in a Tagalog–Spanish–English trilingual child. International Journal of Multilingualism, 8, 5–21. http://dx.doi.org/ 10.1080/14790711003671846 Nakamura, J., & Quay, S. (2012). The impact of caregivers’ interrogative styles in English and Japanese on early bilingual development. International Journal of Bilingual Education and Bilingualism, 15, 417–434. http://dx.doi.org/10.1080/13670050.2012.665827 Nicoladis, E., & Genesee, F. (1996). A longitudinal study of language differentiation in young bilingual children. Language Learning, 46, 439–464. http://dx.doi.org/10.1111/ j.1467-1770.1996.tb01243.x Nicoladis, E., & Secco, G. (2000). The role of a child’s productive vocabulary in the language choice of a bilingual family. First Language, 20, 3–28. http://dx.doi.org/10.1177/014272370002005801 Oller, D. K., & Eilers, R. (2002). Language and literacy in bilingual children. Clevedon, England: Multilingual Matters. Paradis, J., Crago, M., & Genesee, F. (2005–2006). Domain-general versus domain-specific accounts of specific language impairment: Evidence from bilingual children’s acquisition of object pronouns. Language Acquisition: A Journal of Developmental Linguistics, 13, 33–62. http://dx.doi.org/ 10.1207/s15327817la1301_3 Paradis, J., & Genesee, F. (1996). Syntactic acquisition in bilingual children: Autonomous or independent? Studies in Second Language Acquisition, 18, 1–25. http://dx.doi.org/10.1017/ S0272263100014662 Pearson, B. Z. (2007). Social factors in childhood bilingualism in the United States. Applied Psycholinguistics, 28, 399–410. http://dx.doi.org/10.1017/S014271640707021X



Early Bilingualism 

 41

Pearson, B. Z., Fernández, S., & Oller, D. K. (1995). Cross-language synonyms in the lexicons of bilingual infants: One language or two? Journal of Child Language, 22, 345–368. http://dx.doi.org/ 10.1017/S030500090000982X Pérez-Leroux, A. T., Pirvulescu, M., & Roberge, Y. (2009). Bilingualism as a window into the language faculty: The acquisition of objects in French-speaking children in bilingual and monolingual contexts. Bilingualism: Language and Cognition, 12, 97–112. http://dx.doi.org/10.1017/ S136672890800391X Place, S., & Hoff, E. (2011). Properties of dual language exposure that influence 2-year-olds’ bilingual proficiency. Child Development, 82, 1834–1849. http://dx.doi.org/10.1111/ j.1467-8624.2011.01660.x Poulin-Dubois, D., Bialystok, E., Blaye, A., Polonia, A., & Yott, J. (2013). Lexical access and vocabulary development in very young bilinguals. The International Journal of Bilingualism, 17, 57–70. http://dx.doi.org/10.1177/1367006911431198 Poulin-Dubois, D., Blaye, A., Coutya, J., & Bialystok, E. (2011). The effects of bilingualism on toddlers’ executive functioning. Journal of Experimental Child Psychology, 108, 567–579. http://dx.doi.org/ 10.1016/j.jecp.2010.10.009 Poulin-Dubois, D., & Goodz, N. (2001). Language differentiation in bilingual infants: Evidence from babbling. In J. Cenoz & F. Genesee (Eds.), Trends in bilingual acquisition (pp. 95–106). Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1075/tilar.1.06pou Prevoo, M. J. L., Mesman, J., Van IJzendoorn, M. H., & Pieper, S. (2011). Bilingual toddlers reap the language they sow: Ethnic minority toddlers’ childcare attendance increases maternal host language use. Journal of Multilingual and Multicultural Development, 32, 561–576. http:// dx.doi.org/10.1080/01434632.2011.609279 Quay, S. (2008). Dinner conversations with a trilingual two-year-old: Language socialization in a multilingual context. First Language, 28, 5–33. http://dx.doi.org/10.1177/0142723707083557 Quay, S. (2011). Trilingual toddlers at daycare centers: The role of caregivers and peers in language development. International Journal of Multilingualism, 8, 22–41. http://dx.doi.org/10.1080/ 14790711003671853 Quay, S. (2012). Discourse practices of trilingual mothers: Effects on minority home language development in Japan. International Journal of Bilingual Education and Bilingualism, 15, 435–453. http://dx.doi.org/10.1080/13670050.2012.665828 Quay, S. (2013, June). Shifting dominance through language mixing: Caregiver input and a preschooler’s bilingual discourse. Paper presented at the International Symposium on Bilingualism 9, Nanyang Technological University, Singapore. Ramírez-Esparza, N., García-Sierra, A., & Kuhl, P. (2010). Naturalistic social communication and speech development in monolingual and bilingual infants. The Journal of the Acoustical Society of America, 128, 2459. http://dx.doi.org/10.1121/1.3508809 Ruby, M. (2012). The role of a grandmother in maintaining Bangla with her granddaughter in East London. Journal of Multilingual and Multicultural Development, 33, 67–83. http://dx.doi.org/ 10.1080/01434632.2011.638075 Schwartz, M. (2010). Family language policy: Core issues of an emerging field. Applied Linguistics Review, 1, 171–192. http://dx.doi.org/10.1515/9783110222654.171 Schwartz, M., & Verschik, A. (2013). Achieving success in family language policy: Parents, children and educators in interaction. In M. Schwartz & A. Verschik (Eds.), Successful family language policy: Parents, children and educators in interaction (pp. 1–20). Dordrecht, the Netherlands: Springer. http://dx.doi.org/10.1007/978-94-007-7753-8_1

42 

 Quay and Montanari

Serratrice, L., Sorace, A., Filiaci, F., & Baldo, M. (2012). Pronominal objects in English–Italian and Spanish–Italian bilingual children. Applied Psycholinguistics, 33, 725–751. http://dx.doi.org/ 10.1017/S0142716411000543 Silva-Corvalán, C., & Montanari, S. (2008). The acquisition of ser, estar (and be) by a Spanish– English bilingual child: The early stages. Bilingualism: Language and Cognition, 11, 341–360. http://dx.doi.org/10.1017/S136672890800357X Spolsky, B. (2007). Towards a theory of language policy. Working Papers in Educational Linguistics, 22, 1–14.

Tannenbaum, M. (2012). Family language policy as a form of coping or defence mechanism. Journal of Multilingual and Multicultural Development, 33, 57–66. http://dx.doi.org/10.1080/ 01434632.2011.638074 Vihman, M. M. (2002). Getting started without a system: From phonetics to phonology in bilingual development. The International Journal of Bilingualism, 6, 239–254. http://dx.doi.org/ 10.1177/13670069020060030201 Volterra, V., & Taeschner, T. (1978). The acquisition and development of language by bilingual children. Journal of Child Language, 5, 311–326. http://dx.doi.org/10.1017/S0305000900007492 Yamamoto, M. (2001). Language use in interlingual families: A Japanese–English sociolinguistic study. Clevedon, England: Multilingual Matters.

Christopher T. Fennell, Angeline Sin-Mei Tsui, and Tamara M. Hudon

3 Speech Perception in Simultaneously Bilingual Infants Two languages in one head? No one can live at that speed! —Eddie Izzard

The above quote reflects how comedian Eddie Izzard (1999) interpreted the British reaction to the norm of European multilingualism. Although this is a humorous take on sociocultural approaches to multilingualism, there is truth in the observation. Individuals from monolingual cultures may look on bilingualism with a mix of awe and confusion. When the topic turns to simultaneous bilingualism in infancy, the natural extension of that point of view is to assume that babies would not be able to handle the supposed challenges of acquiring two languages. One can see this in the advice sometimes given to parents to speak only one language to an infant, under the assumption that multiple language exposure may lead to language confusion or delay. But, as with much child-rearing advice, one needs to separate myth and misinformation from evidence (see Byers-Heinlein & Lew-Williams, 2013). In this chapter, we outline established scientific research on the development of speech perception in simultaneous bilingual speakers, a skill that primarily transpires over infancy. As such, speech perception is the foundation on which later language is built. The research presented herein can help to address whether and how early bilingualism impacts language development in its initial stages.

The Beginnings of Speech Perception: Discriminating Rhythm One fascinating study of early bilingual speech perception revealed that bilingualism can even impact speech perception prenatally. Byers-Heinlein, Burns, and Werker (2010) examined neonates’ discrimination of and preference for rhythmically distinct languages. Languages fall along a rhythmic continuum on the basis of the relative distributions and variability of vocalic and consonantal intervals (Ramus, Nespor, & Mehler, 2000). This continuum can be broadly divided into three categories: stresstimed languages (e.g., English, Dutch), syllable-timed languages (e.g., French, Tagalog), and mora-timed languages (e.g., Japanese). At birth, monolingual infants attend more This work was supported by a Natural Sciences and Engineering Research Council of Canada Grant. http://dx.doi.org/10.1037/14939-004 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Eds.) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

44 

 Fennell, Tsui, and Hudon

to their native language than a language from a different rhythmic class, but they do not discriminate languages from the same rhythmic class (e.g., Mehler et al., 1988). For example, an infant only exposed to English prenatally will attend more to that language than Japanese, but not Dutch. Byers-Heinlein et al. examined whether neonates who heard two rhythmically distinct languages prenatally would prefer both languages. Using a high amplitude sucking procedure, they tested newborns whose mothers had regularly spoken only English or both English and Tagalog during pregnancy. This task takes advantage of the congenitally organized behavior of sucking, which neonates can control. Increased sucking to a stimulus type can therefore indicate interest or preference. An analysis of the number of strong sucks produced by the neonates in response to English and Tagalog speech revealed that monolingual neonates produced significantly fewer such sucks to Tagalog than to English, but bilingual neonates’ responses to the two languages did not significantly differ. The bilingual neonates also produced more sucks to Tagalog than English monolingual neonates. These data indicate that bilingual newborns prefer both of their languages at birth. However, an alternate explanation is that perhaps bilingual neonates, unlike their monolingual peers, may not discriminate broad rhythmic language distinctions because of their more variable linguistic experiences prenatally. Therefore, their preference for both languages may be a preference for one mixed language system, rather than two separable languages. To refute this, Byers-Heinlein et al. (2010) habituated English monolingual and Tagalog–English bilingual newborns to English and then switched the language to Tagalog at test (or vice versa). Both groups showed significant increases in strong sucks when the language changed, thus demonstrating language discrimination. Coupled with their attentional focus on the languages heard prenatally, these discrimination skills should give those infants learning languages that differ in rhythm a strong foundation for learning two language systems (e.g., French–English infants in Canada). Indeed, bilingual infants’ separation of and attention to rhythm allow them to eventually learn language-specific structures that can differ across their two languages. For example, word order (Gervain & Werker, 2013) and lexical stress (BijeljacBabic, Serres, Höhle, & Nazzi, 2012) distinctions across languages have differing associated prosodic cues. Although a number of infants are raised in bilingual cultures whose languages share a rhythmic class (e.g., Catalonia: Catalan–Spanish), no language discrimination study has examined newborns with prenatal exposure to rhythmically similar languages. However, infants learning two syllable-timed languages are able to discriminate their languages in familiarization tasks at the earliest ages tested: 3 to 4 months of age (Catalan–Spanish in Bosch & Sebastián-Gallés, 2001; Basque–Spanish in Molnar, Gervain, & Carreiras, 2014). It appears that infants rely on small rhythmic differences rather than segments (i.e., phones) to separate their similar native languages at this young age. For instance, 3- to 4-month-old bilingual infants can still successfully dis-



Speech Perception in Simultaneously Bilingual Infants 

 45

criminate low-pass filtered Basque from Spanish, an audio transformation that removes segmental information while retaining rhythmic properties (Molnar et al., 2014). Despite bilingual infants’ successful separation of their languages early in development, they may not be attending to their native languages in the same manner as monolingual infants. Over a set of experiments, 4-month-old monolingual (Catalan or Spanish) and bilingual infants (Catalan–Spanish) heard sentences alternating between two speakers, one to the right and one to the left (Bosch & Sebastián-Gallés, 1997). Half the sentences were from a native language (Catalan or Spanish) and half from a nonnative language. For some infants, the nonnative language was of the same rhythmic class as their native language (Spanish or Catalan for the monolingual infants, Italian for the bilingual infants), or of a different rhythmic class (English). Monolingual infants oriented quicker to their native language in comparison with nonnative languages (e.g., Catalan monolingual infants oriented quicker toward Catalan than toward Spanish or English). Bilingual infants treated their native languages in the same manner (i.e., equal orientation times to Spanish and Catalan). However, their reaction to the nonnative languages was opposite to that of monolingual infants: they oriented quicker to English and Italian than to their native languages. Bosch and Sebastián-Gallés (1997) proposed that bilingual infants may be slower to orient to native languages because their language discrimination process involves two steps: (a) recognize the language as familiar, and (b) identify which familiar language it is. On the other hand, monolingual infants simply need to reject a nonnative language as unfamiliar, a quicker, one-step process. In support of this hypothesis, Molnar et al. (2014) found that Spanish–Basque bilingual infants also attended to their native languages longer than monolingual infants of those languages. Thus, although the end result of within-rhythmic class language discrimination is similar between infants learning one versus two languages, the processes underlying that ability may differ between the groups. We will return to this distinction between milestones and processes later, but now we turn to bilingual infants’ discrimination of languages via another modality. Speech is multimodal. Visual cues from talking faces transmit important linguistic information in conjunction with the auditory stream, and infants are sensitive to these visual cues. English-learning monolingual infants of 4 and 6 months of age differentiate English from French merely from silent talking faces, but they lose this ability by 8 months because French is absent in their language environment. However, French–English bilingual babies of 8 months retain the ability to visually discriminate English from French, supporting the claim that this perceptual skill is maintained via specific language exposure in infants’ environment (Weikum et al., 2007). A follow-up study explored whether bilingual infants’ prolonged ability to visually discriminate languages may generalize to nonnative languages. Sebastián-Gallés, Albareda-Castellot, Weikum, and Werker (2012) tested whether 8-month-old infants

46 

 Fennell, Tsui, and Hudon

learning Spanish and Catalan could visually differentiate two nonnative languages: English and French. Their results showed that bilingual, but not monolingual, infants succeeded. The researchers argued that this was because of enhanced visual attention to talking faces driven by the infants’ regular experience of visually separating two languages. Taken together, these results show that simultaneous bilingual infants may possess a perceptual advantage in both native and nonnative visual speech discrimination. We now return to auditory speech perception and its next major ontogenetic development: phonetic refinement.

From Nonnative to Native Phonetic Perception Over the first year of life, monolingual infants move from a language-general sensitivity to both native and nonnative phonemes to a heightened language-specific perception of native-language sounds (e.g., Werker & Tees, 1984), a process termed perceptual narrowing. For example, English-learning infants can initially discriminate both English phonemic contrasts (e.g., labial and alveolar stops: [b] vs. [d]) and Hindi sounds that are not phonemic in English (e.g., dental and retroflex stops: [d̪] vs. [ɖ]). However, they only discriminate the English contrast by 12 months. Although bilingual infants can broadly distinguish their native languages, they may have difficulty discriminating close native phonemic contrasts because of the complexity inherent in learning two sound inventories. As Byers-Heinlein and Fennell (2014) argued, bilingual infants need to refine more phonological categories than monolingual infants and must do so in the face of imperfectly overlapping phonetic distributions across their two languages. The paragraphs below highlight such cases of imperfect phonetic alignment. Bosch and Sebastián-Gallés (2003) conducted the first major study examining phonetic refinement in bilingual infants. Using the same familiarization procedure previously used to test language discrimination, the researchers tested monolingual (Spanish or Catalan) and bilingual (Spanish–Catalan) infants’ discrimination of a close vowel contrast (/e/–/e/) that exists in Catalan but not in Spanish. Spanish monolingual infants demonstrated the standard developmental pattern of perceptual narrowing to their native language, discriminating the nonnative Catalan contrast at 4 months of age and failing at 8 months. Catalan monolingual infants also adhered to their expected pattern: discriminating their native phonemic contrast at both ages. However, Spanish– Catalan infants produced unexpected results. They discriminated the native Catalan contrast at 4 and 12 months, but failed to distinguish the vowels at 8 months. This unique U-shaped developmental pattern had not been seen in monolingual studies of phoneme discrimination. Unsurprisingly, bilingual infants’ intriguing results inspired multiple follow-up studies to determine the source of their difficulty discriminating native contrasts at 8 months.



Speech Perception in Simultaneously Bilingual Infants 

 47

One of the first assumptions was that bilingual infants may have difficulty only with contrasts that do not have distinct distributions in their speech input. Maye, Werker, and Gerken (2002) proposed that distributional learning, whereby infants track the frequency of phonetic information in the input, is a mechanistic explanation for how infants acquire and refine phonetic categories. Maye et al. trained infants 6 to 8 months of age on a unimodal distribution of phones or a bimodal distribution. Infants exposed to the bimodal distribution successfully discriminated the end points of the continuum at test (i.e., the contrast), whereas those infants trained on a unimodal distribution did not. The /e/–/e/ contrast used in the Bosch and Sebastián-Gallés (2003) study may not have a clear bimodal distribution. There is a reduced frequency of this vowel contrast in bilingual infants’ input, as it is only found in one of their languages (Catalan) and is infrequent in that language. Furthermore, there is a Spanish /e/ vowel that is much more frequent and falls directly between the Catalan targets in a F1 by F2 plot (Bosch & Sebastián-Gallés, 2003). If the bilingual Spanish–Catalan infants are not receiving a clear bimodal distribution of the /e/–/e/ contrast because of this overlap with a frequent Spanish vowel, essentially collapsing the less frequent Catalan vowels into its distribution, it is perhaps no surprise they go through a period wherein they have difficulty determining the boundaries of the target vowels. Follow-up studies initially focused on the frequency aspect of the above argument, with the hypothesis that bilingual infants may be able to discriminate close contrasts that frequently occur in their input. Burns, Yoshida, Hill, and Werker (2007) examined Canadian French–English bilingual infants’ ability to discriminate three highly frequent phones that fall along a close voice onset time continuum across the two languages. Although French and English possess two labial stops that contrast in voicing, the phonemes have different phonetic realizations (aspiration and voicing) across the languages. Speakers of both languages perceive the [b] phone as /b/ and the [ph] phone as /p/. However, because of the cross-linguistic differences, the [p] phone is perceived as /b/ by English monolingual speakers and as /p/ by French monolingual speakers. To successfully learn both languages, bilingual infants should discriminate all three phones. Burns et al. tested such discrimination by habituating English monolingual and French–English bilingual infants to the median [p] and testing them on a change to [b] and to [ph]. Although both groups had similar results at 6 to 8 months of age, bilingual infants over 10 months discriminated all three phones, whereas English monolingual infants this age only discriminated their native distinction: [p] from [ph]. Similarly, Sundara, Polka, and Molnar (2008) found that Canadian French–English bilingual infants of 10 to 12 months are able to discriminate the French dental production of /d/ from the English alveolar production of the same phoneme in a habituation paradigm. Although the acoustic-phonetic distribution of /d/ across the two languages highly overlaps, this phoneme is highly frequent in both languages, and therefore this finding lends support to the frequency argument.

48 

 Fennell, Tsui, and Hudon

Although the data above appear to support a frequency argument, SebastiánGallés and Bosch (2009) wished to explicitly test this hypothesis by replicating their original study with a frequent and acoustically similar vowel contrast with similar productions in both Spanish and Catalan: /o/–/u/. Monolingual infants of 4 and 8 months had no difficulty distinguishing /o/ from /u/; however, infants learning Spanish and Catalan again failed to discriminate the vowels at 8 months, despite succeeding at 4 and 12 months. Because frequency did not appear to explain these results, Sebastián-Gallés and Bosch turned to a striking dissimilarity between their studies and the work demonstrating bilingual success: the languages being acquired. Whereas Spanish–Catalan bilingual infants have a transient difficulty with a close contrast, work with French–English bilingual infants shows successful discrimination of target contrasts. It is unlikely that sociocultural factors are contributing to this difference, as bilingualism is valued in the respective regions of Spain and Canada where the infants were tested. Instead, it may be something about the relationship between the languages that is driving this effect. Sebastián-Gallés and Bosch (2009; see also Sebastián-Gallés, 2011) argued that bilingual infants’ apparent impairments in phoneme discrimination tasks are artifacts of differential processing of speech changes. Bilingual infants may discriminate phonemes to the same extent as monolingual infants. However, their language environment may promote a greater acceptance of acoustic variation in phonemes—small changes are normal and therefore not surprising. This is because of two factors in their acoustic environment that differ from monolingual infants: the higher probability of hearing accented speech (e.g., Bosch & Ramon-Casas, 2011) and the presence of cognates—two words with similar acoustic forms and meanings across two languages (e.g., porta and puerta mean door in Catalan and Spanish, respectively). This latter factor would specifically affect Spanish–Catalan bilingual infants more than French– English bilingual infants because the number of cognate words across Spanish and Catalan is much greater than across French and English. Therefore, bilingual infants learning languages with more lexical form overlap may be more prone to accept small acoustic–phonetic changes. One piece of evidence in support of this argument is that 8-month-old Spanish–English bilingual infants successfully dishabituate to a change in the same vowel contrast (/e/–/e/; which also exists in English) that causes difficulty for Spanish–Catalan bilingual infants (Sundara & Scutellaro, 2011). Sundara and Scutellaro (2011) hypothesized that bilingual infants’ success may be because of their ability to track phonemes across rhythmically different languages (Spanish– English) better than across rhythmically similar ones (Spanish–Catalan). However, an equally valid hypothesis is that infants learning languages with fewer cognates (Spanish and English) have stronger reactions to phoneme changes than infants learning languages with a large amount of lexical overlap (Spanish and Catalan). Of course, an alternate explanation is that the presence of lexical overlap in the input leads to less clear phonological boundaries for bilingual infants (Curtin, ByersHeinlein, & Werker, 2011). To support the argument that Spanish–Catalan bilingual



Speech Perception in Simultaneously Bilingual Infants 

 49

infants’ past failures were because of greater acceptance of phonetic changes and not “fuzzier” representations, Albareda-Castellot, Pons, and Sebastián-Gallés (2011) sought to demonstrate that these infants possessed discriminable phonetic representations of the Catalan vowels that caused them difficulty in previous research. To do so, they used a measure that did not rely on a dishabituation response (i.e., a surprise reaction to a phoneme change). They instead used the anticipatory eye movement (AEM) task to test Spanish–Catalan infants’ discrimination of the Catalan /e/–/e/ contrast at 8 months. Infants watched a cartoon face disappear behind the bottom of a T-shaped occluder. The vowel that played while the face was occluded predicted the side it would reemerge from at the top (e.g., /e/ predicted right; /e/ predicted left). If infants can discriminate the two vowels, they will begin to anticipate the face’s eventual location on hearing a specific vowel. Both monolingual Catalan and bilingual Spanish–Catalan infants correctly anticipated where to look, but Spanish mono­ lingual infants did not because they cannot perceive the nonnative contrast. The authors thus argued that bilingual infants have similar native-language phonetic representations to monolingual infants and the bilingual difficulties seen in past research are attributable to greater acceptance of phoneme changes. However, one issue that complicates the above argument is that Spanish–Catalan infants are only “impaired” in habituation tasks at 8 months, but not at 4 or 12 months. If they have increased tolerance for phoneme changes, why would this flexibility only be present at one specific age? Perhaps bilingual infants do have weaker phonetic representations when initially modifying phonetic boundaries to native-language categories, especially when acquiring languages with lexical overlap, and the AEM task detected these weaker representations. In AEM tasks, infants’ trials are scored as correct even if they look minutely longer to the correct over incorrect side. Thus, it is a highly sensitive task. Further, this task may actually give infants a training boost for the target contrast. Yeung and Werker (2009) showed that monolingual infants could be trained to distinguish a nonnative contrast if the members of the contrast were each paired with a distinct visual cue (i.e., two different objects). In AlbaredaCastellot et al. (2011), one vowel is always paired with the left side of the occluder and the other with the right side. So, bilingual infants’ phonetic representations may have been enhanced within the task. Results from electrophysiological studies of bilingual infants’ phoneme discrimination may help to elucidate whether their difficulties are because of weaker phoneme representations or differential attention to phoneme changes. Event-related potential (ERP) research has demonstrated that phoneme discrimination is related to a component in the electroencephalogram (EEG) signal called a mismatch response (MMR). The MMR is generated in response to a deviant phoneme produced in a string of standard phonemes (e.g., deviant “da” sounds randomly embedded in a string of standard “ba” syllables). When adults discriminate a phoneme contrast, deviants evoke a negative deflection in the EEG signal (Näätänen, 1990). However, children under 4 years, and especially infants, can instead demonstrate a positive wave in response

50 

 Fennell, Tsui, and Hudon

to phoneme deviants. This positive MMR can be seen as an immature version of the adult negative MMR (Morr, Shafer, Kreuzer, & Kurtzberg, 2002). To date, only two ERP studies, both with Spanish–English American infants, have examined bilingual infants’ phoneme discrimination around 8 months of age. One showed no MMR in 6- to 9-month-olds (Garcia-Sierra et al., 2011), even though monolingual infants this age produce a negative MMR to the same stimuli (Rivera-Gaxiola, Silva-Pereyra, & Kuhl, 2005). The other showed a developmentally transient positive MMR at 6 months over fewer electrode sites than in monolingual infants (Shafer, Yu, & Datta, 2011). Note that the bilingual infants tested in these studies were learning languages that share few cognates and yet, they had potentially weaker neural responses to phoneme changes in early infancy. Nevertheless, although these studies appear to indicate weaker native phonetic representations in bilingual infants, both studies had sampling issues: The former tested mostly low socioeconomic status infants, who may thus have poorer language skills, and the latter had small sample sizes. Further research will be needed to determine the cause of bilingual difficulties with some phoneme contrasts. Perhaps counterintuitively considering the above research, one final aspect of bilingual infants’ development of phoneme perception that is generating interest in the field of psycholinguistics is the possibility that bilingual infants may have an extended period of nonnative phoneme discrimination. This idea was partially inspired by the findings discussed earlier that bilingual infants remain sensitive to language differences in the visual modality, whereas monolingual infants do not. If native-language experience is motivating the decline in sensitivity to nonnative phonemes, then an extended period of nonnative sensitivity might be predicted for those infants with less language exposure to their native languages than monolingual infants (Curtin et al., 2011). Further, the presence of extra phonetic categories across their two languages, in comparison with monolingual infants, could enable discrimination of contrasts that are not native to either language but are close to native phonemes. Finally, bilingual infants as young as 7 months have better attentional control than their monolingual peers (Kovács & Mehler, 2009), which may allow them to attend to subtler linguistic distinctions. Only two studies have explored nonnative phoneme discrimination in bilingual infants. Petitto et al. (2012) tested 4- to 6-month-old and 10- to 12-month-old infants learning only English and those learning English and another language on a native and a nonnative contrast. Using functional near-infrared spectroscopy (fNIRS), a brain imaging technique that uses infrared light to noninvasively measure activation via cortical blood flow, they discovered group differences in infants’ activation to the contrasts in the left inferior frontal cortex, a brain region associated with language processing. Whereas monolingual infants showed an age-related increase in activation in response to the native contrast, bilingual infants’ activation decreased with age. For the nonnative contrast, bilingual infants showed a smaller decrease in activation with age, and the monolingual infants had similar activation across the two ages. Monolingual infants’ dissimilar activation patterns to the two contrasts were thought to



Speech Perception in Simultaneously Bilingual Infants 

 51

reflect their known perceptual narrowing to native contrasts. The authors interpreted bilingual infants’ similar brain responses to native and nonnative contrasts (decreasing activation with age) as evidence for similar perception, and thus openness to nonnative differences. However, two aspects of the results complicate that interpretation. First, all infants showed a significant brain response to the phoneme change in both contrasts. Second, bilingual infants’ activation to the native contrast decreased over development, which may reflect bilingual infants’ transient difficulties with some native distinctions. Thus, the similar decrease to nonnative stimuli may also reflect difficulty with that contrast. Despite the potential issues noted above, a study of nonnative tone discrimination in infants acquiring two nontonal languages may support the neurophysiological findings. Using a habituation design, Liu and Kager (2013) found that mono­ lingual Dutch-learning infants discriminated a Mandarin phonological tone change at 6 months, but not at 9 or 12 months. Infants learning Dutch and another nontonal language discriminated the nonnative contrast at 6 and 12 months but failed to do so at 9 months. Monolingual infants thus showed the standard perceptual narrowing response, but bilingual infants demonstrated the U-shaped curve previously seen with some native contrasts. Although these data do not show bilingual sensitivity to nonnative contrasts across all ages, and may indeed support general bilingual difficulties around 8 months of age, they reflect similar processing of native and nonnative phonemes with the end result being nonnative discrimination by bilingual infants at 12 months. In summary, bilingual and monolingual infants performed similarly in the majority of phoneme discrimination studies presented in this section. One notable exception is bilingual infants’ transient difficulty with phoneme discrimination around 8 months. However, that age-limited difficulty may also be restricted to particular contrasts and to infants learning languages with lexical overlap. There are also hints that bilingual infants may remain sensitive to nonnative distinctions, but further research is necessary to determine whether bilingual infants’ enhanced sensitivity to nonnative phonetic contrasts is a stable phenomenon.

Phonotactics The phonotactic rules of a language determine the allowable combinations of phonemes within words. For example, the phoneme sequence /str/ is acceptable in English, but not in Japanese. These rules permit infants to determine permissible word candidates and also serve as cues to the boundaries between words (e.g., /kb/ cannot occur at the beginning or end of an English word and therefore is a cue to a word boundary, like “take back”). Monolingual infants are sensitive to the nativelanguage phonotactics by 9 months. They prefer listening to lists of words that contain

52 

 Fennell, Tsui, and Hudon

high frequency legal sequences over lists of words that contain low frequency or illegal phoneme sequences (e.g., Friederici & Wessels, 1993). As with determining the correct distributions of phonetic variability for accurate phoneme perception, bilingual infants face an increased computational load in phonotactic development. Not only do they need to determine which sound combinations are more probable in their input, they may also have to determine whether those combinations are legal in one of their languages and illegal in the other. Only one study has examined phonotactic development in bilingual infants. SebastiánGallés and Bosch (2002) tested 10-month-old infants learning Spanish and Catalan on their preference for legal versus illegal Catalan sequences. As expected, Catalan, but not Spanish, monolingual infants preferred legal over illegal Catalan words. The results from the bilingual group illustrated how language exposure levels impacted the use of language-specific phonotactic rules. Catalan-dominant bilingual infants, with dominance determined by the mother’s dominant language, performed similarly to Catalan monolingual infants, showing a preference for legal over illegal Catalan words. Although Spanish-dominant bilingual infants did not significantly prefer legal over illegal Catalan words, their preference scores were higher than those of Spanish monolingual infants, showing some effects of exposure. Thus, with enough exposure to the target patterns, bilingual and monolingual infants have parallel development of phonotactic rules.

Recognition of Word Forms To date, very little work has examined word form recognition in bilingual infants. Word forms are distinguished from words in that they refer only to the sound sequence of a word and not to a word-referent pairing. It is possible for an infant to recognize the acoustic form of a word without yet understanding its meaning. Further, no visual referents for word forms are presented in such studies. These studies therefore reflect infants’ ability to segment acoustic forms from the speech stream, either in their language environment or during lab-based training, and subsequently recognize these acoustic forms. As such, this research belongs more to the speech perception tradition of psycholinguistics than to the field of semantics. Using the headturn preference procedure (HPP), Jusczyk and Aslin (1995) familiarized English-learning infants to two words (e.g., dog and cup). Infants of 7.5, but not 6, months recognized familiar word forms in continuous speech at test by orienting (via a headturn) longer toward a visual stimulus while hearing passages containing the familiarized word forms over those with novel word forms (e.g., feet and bike). These results also held if infants were trained on passages and tested on isolated words. Jusczyk and Aslin also found that similar sounding, but phonemically different, words were not a sufficient match to produce a preferential response at 7.5 months (e.g., trained on gike and tested on bike). This suggests that monolingual



Speech Perception in Simultaneously Bilingual Infants 

 53

English infants recognize familiar words in fluent speech by 7.5 months, and they encode word forms in sufficient detail to detect phonemic mispronunciations. Bilingual infants’ word form recognition has also been examined using the HPP. English and Welsh monolingual 9- to 12-month-olds and Welsh–English bilingual 11-month-olds heard lists of words that contained either frequent or phonologically similar infrequent words from their natural learning environment (Vihman, Thierry, Lum, Keren-Portnoy, & Martin, 2007). Pilot research had demonstrated that English infants preferred listening to frequent over infrequent words at 11 months (i.e., longer headturns to frequent words), but Welsh infants did not exhibit this preference until 12 months of age. Welsh words undergo consonant mutation for a variety of morphological and syntactic reasons (e.g., possessive adjectives produce initial consonant mutation so that the word cath [cat] transforms into ei gath [his cat], ei chath [her cat], eu cath [their cat], etc.). It is therefore unsurprising that Welsh infants take longer to distinguish frequent words from phonologically similar infrequent ones, as they need to track a word’s identity throughout changes in its surface form. Vihman et al. (2007) replicated their earlier monolingual results: bilingual infants recognized frequent words in English on par with their monolingual English-learning peers (11 months). Interestingly, they recognized frequent words in Welsh at this same age, one month earlier than their Welsh-learning peers. The researchers found parallel results in an ERP experiment. English monolingual infants and Welsh–English bilingual infants produced differential ERP responses to frequent versus infrequent words, but Welsh monolingual infants did not. Thus, Welsh–English bilingual infants appeared to be at an advantage over their monolingual Welsh-learning peers in that their exposure to English, a more phonologically “stable” language, supported their ability to recognize words in the comparatively more “unstable” Welsh (Vihman et al., 2007). Singh and Foong (2012) extended this research to younger bilingual infants and tested a finer level of phonological detail. Using the HPP, they found that Mandarin– English bilingual infants were sensitive to phonemic changes in Mandarin tone (meaningful pitch changes) at 7.5 and 11 months. At 9 months, however, bilingual infants appeared to generalize words across differences in lexical tone. These data follow the U-shaped developmental pattern seen in some phoneme discrimination work, supporting the hypothesis that bilingual infants may develop a temporary tolerance around 8 months for phonemic substitutions that signal a change in a word’s meaning. One recent study asked whether bilingual infants’ hypothesized increased tolerance for phonemic changes in word forms is adaptive (Hudon, 2013). Specifically, bilingual infants’ accommodation of accents was tested using the HPP. Instead of a phonemic change to familiarized words at test, infants were tested on an accent change. Monolingual infants successfully recognize a familiarized word at test when it is presented in a different accent at 13 months, but they fail to do so at 9 months (e.g., Schmale & Seidl, 2009). Bilingual infants may show enhanced accent accommodation because of their greater exposure to phonetic variability. Hudon (2013) tested French–English bilingual infants of 9 and 13 months on their recognition of

54 

 Fennell, Tsui, and Hudon

familiarized words that changed accents at test. Infants heard changes from the local English accent to French-accented or Mandarin-accented English. Bilingual infants succeeded in accommodating the familiar French-accented English at both 9 and 13 months, thus showing earlier accent accommodation than monolingual infants. This has important ramifications for lexical acquisition, as it allows them to recognize a word as the same across familiar accents. However, they failed to accommodate the unfamiliar Mandarin-accented English at 9 months. Thus, bilingual infants are not open to any accent change; their phonetic flexibility is restricted to variability consistent with their native languages.

Detecting Phoneme Changes in Words: Testing Phonology The direct links between the refinement of native language sound patterns in the first year of life and the mapping of those sounds to meaning (i.e., word learning) have been the focus of many studies over the past two decades. However, very few of these emergent phonology studies involve infants acquiring two languages. Infants’ phonological perception is usually tested in one of two ways. First, infants can be taught a novel word, or words, in the lab and then tested on mispronunciations of the newly acquired word. Second, infants can be tested on mispronunciations of words they have naturally learned in their language environment. Fennell, Byers-Heinlein, and Werker (2007) took the first approach. Bilingual infants of 14, 17, and 20 months who were acquiring English and another language were habituated to two novel word–object pairings. Although the objects were visually distinct, their labels formed a minimal pair, only differing by the initial consonant (Object 1 paired with /bI/; Object 2 with /dI/). An English monolingual female speaker produced the target minimal pair tokens. To test their use and perception of pho­ nemes, infants were given two test trials after habituation. In one trial infants experienced a familiar pairing (e.g., Object 1–/bI/) and in the other they heard the wrong label for the presented object (e.g., Object 1–/dI/). English-learning infants begin to successfully use English consonants when learning new words at 17 months of age (i.e., they do not confuse minimally different object labels; see Werker, Fennell, Corcoran, & Stager, 2002). However, Fennell et al. found that bilingual infants did not successfully use those same consonants to disambiguate similar-sounding novel words until 20 months. In case the results were because of the linguistic heterogeneity of the bilingual sample, they repeated the experiment with two homogenous groups: French–English and (Cantonese or Mandarin) Chinese–English bilingual infants. Both groups replicated the initial findings: bilingual infants did not succeed in distinguishing the minimal pair until three months later than mono­ lingual infants.



Speech Perception in Simultaneously Bilingual Infants 

 55

However, another study using the same task as Fennell et al. showed that bilingual infants could succeed in learning minimal pair words at 17 months. Mattock, Polka, Rvachew, and Krehm (2010) tested 17-month-olds learning French and English on two minimally different object labels (/bos/–/gos/). They recorded a bilingual speaker producing the target words in an English accent and a French accent. Infants were presented the words in both accents throughout training and test phases, hearing three English and three French productions of the target label per trial. At test, bilingual infants detected the phoneme change in the object label in this mixedaccent condition. However, monolingual infants of the same age failed with these mixed stimuli. When stimuli were restricted to monolingual infants’ native accent (e.g., French monolingual infants only hearing French-accented tokens at training and test), monolingual infants succeeded in distinguishing the minimal pair. But monolingual infants again failed to notice the change in target word when tested on stimuli from their nonnative accent (e.g., French monolingual infants hearing Englishaccented tokens). Fennell et al. (2007) postulated that bilingual infants may have failed in their study because of weaker phoneme representations, an argument that was on the basis of the research showing bilingual infants’ poorer phoneme discrimination in their first year. Mattock et al. (2010) argued that bilingual infants flexibly process more phonetic variability because they experience such variability in their everyday language environments, whereas monolingual infants’ lack of such variability leads them to fail to process unfamiliar targets. However, Fennell and Byers-Heinlein (2014) hypothesized that there may be neither a bilingual delay nor advantage in the task. Bilingual adults, even highly proficient simultaneous bilingual adults, possess slightly different accents in their native languages from monolingual adults of those languages (Antoniou, Best, Tyler, & Kroos, 2010). Infants are vulnerable to speech that differs from their home accent when learning words (Schmale, Hollich, & Seidl, 2011), and bilingual infants tend to have at least one bilingual parent and monolingual infants tend to have monolingual parents. Because the bilingual infants in the study by Fennell et al. (2007) only heard words produced by a monolingual speaker and the bilingual infants in Mattock et al. (2010) were never tested on monolingualaccented tokens, their respective results may be because of matches or mismatches between the stimuli and their home accents, not to multilingualism. Indeed, Fennell and Byers-Heinlein (2014) found that 17-month-old bilingual infants succeeded in distinguishing a minimal pair when hearing a bilingual adult produce the tokens, but failed when hearing a monolingual adult, and that monolingual infants showed the opposite pattern. Despite the difference in what stimuli the two groups process more efficiently, the main conclusion of this study is that monolingual and bilingual infants follow the same pattern of development: both groups succeed with minimal pairs at 17 months when listening to a speaker who sounds like people in their environment. Research testing mispronunciations of known words commonly uses preferential looking tasks. Infants see two pictures side by side (e.g., a doll and a car) and a target

56 

 Fennell, Tsui, and Hudon

object is either named correctly (e.g., doll) or incorrectly (e.g., goll). Monolingual infants as young as 11 months have reduced proportion looking to the target when hearing mispronunciations in comparison with correct pronunciations, suggesting that they encode words in full phonological detail (Swingley, 2005). The mispronunciation effect found in word recognition studies has not, however, held across bilingual samples. Unlike their Catalan monolingual peers, Spanish–Catalan bilingual 17‑ to 24-month-olds did not detect a mispronunciation in familiar words when the vowel change corresponded to the Catalan-only /e/–/e/ contrast: looking to the target object equally whether it was pronounced correctly or incorrectly (RamonCasas, Swingley, Sebastián-Gallés, & Bosch, 2009). This was not because of a general inability to hear the distinction between these sounds, as early discrimination problems resolve by 12 months. Their failure, however, does seem to be restricted to the Catalan-specific contrast, because mispronunciations consisting of contrasts common to both Spanish and Catalan hindered word recognition. Further, Ramon-Casas and Bosch (2010) noted that the Ramon-Casas et al. (2009) study used only cognate words, which may have driven bilingual infants’ acceptance of mispronunciations. Indeed, they found that Spanish–Catalan bilingual infants of the same age detected /e/–/e/ mispronunciations when target words were not cognates. Taken together, these results are similar to the arguments presented earlier for bilingual infants’ difficulties in phoneme discrimination: the amount or degree of phonological overlap (at the phonemic and lexical level) between languages influences bilingual infants’ reactions to mispronunciations.

Conclusion Despite having more complex and variable input than their monolingual peers, bilingual infants do not differ in the developmental timing of most speech perception milestones, such as language discrimination, phonetic refinement of vowels and consonants (in most cases), acquisition of phonotactic rules, and the perception and use of phonological categories in novel and familiar (noncognate) words (see Table 3.1). All of the above emergent skills are hypothesized to rely on infants’ detection of patterns in their language input. To reach language targets at the same developmental time as monolingual infants, bilingual infants need to track and maintain separation of the patterns present in each of their languages. The early emerging skill of detecting rhythmic differences between languages is in all probability the perceptual gateway to bilingual infants’ language separation. However, Kandhadai, Danielson, and Werker (2014) recently proposed an additional sociocultural argument for language separation in early infancy. Kandhadai et al. (2014) began their argument by highlighting bilingual infants’ ability to visually discriminate languages from talking faces. This attention to faces reflects the social nature of language. Extending from that starting point, it may not be

Byers-Heinlein et al. (2010) Molnar et al. (2014) Bosch & Sebastián-Gallés (1997) Bosch & Sebastián-Gallés (2001) Weikum et al. (2007) Gervain & Werker (2013) Sebastián-Gallés et al. (2012) Bijeljac-Babic et al. (2012) Shafer et al. (2011) Petitto et al. (2012) Sebastián-Gallés & Bosch (2009) Sebastián-Gallés & Bosch (2009) Sundara & Scutellaro (2011) Bosch & Sebastián-Gallés (2003) Sundara et al. (2008) Garcia-Sierra et al. (2011) Liu & Kager (2013) Burns et al. (2007) Albareda-Castellot et al. (2011) Sebastián-Gallés & Bosch (2002) Singh & Foong (2012) Hudon (2013) Vihman et al. (2007) Fennell et al. (2007) Fennell & Byers-Heinlein (2014) Mattock et al. (2010) Ramon-Casas et al. (2009) Ramon-Casas & Bosch (2010)

Rhythmic discrimination

Canada Spain Spain Spain Canada Canada Spain France United States Canada Spain Spain United States Spain Canada United States Netherlands Canada Spain Spain United States Canada Wales Canada Canada Canada Spain Spain

Country English and Tagalog Basque and Spanish Spanish and Catalan Spanish and Catalan English and French English and other Spanish and Catalan French and other English and Spanish English and other Spanish and Catalan Spanish and Catalan English and Spanish Spanish and Catalan English and French English and Spanish Dutch and others English and French Spanish and Catalan Spanish and Catalan English and Mandarin English and French English and Welsh English and others English and French English and French Spanish and Catalan Spanish and Catalan

Languages 0 3.5 4 4 4, 6, 8 7 8 10 3 to 36 4–6, 10–12 4, 8 4, 8 4, 8 4, 8, 12 6–8, 10–12 6–9, 10–12 6, 9, 12 7, 11, 14–20 8 10 7.5, 9, 11 9, 13 9–12 14, 17, 20 17 17 17–24 24

Age (months)

No No No No No No Nonnative discrimination No Neural differences at some ages Neural differences At 8 months only At 8 months only No At 8 months only No At 6–9 months only At 12 months only No No No At 9 months only No No At 17 months only No Phonetic variability advantage For cognates No

Did bilingual infants differ from monolingual infants in macrostructure?

Note: When monolingual infants of Language A and bilingual infants of Languages A and B differed in their performance, but that difference was because of bilingual infants’ perception of a property from Language B, this was not considered a difference in macrostructure, as one would expect the same behavior from monolingual infants of Language B.

Phonological discrimination

Phonotactics Word form recognition

Phonetic discrimination

Study

Speech perception skill

Tab. 3.1: Summary of Infant Speech Perception Studies Involving Bilingual Infants

58 

 Fennell, Tsui, and Hudon

a stretch to state that bilingual infants, while attending to the faces of the speakers in their environment, also attend to the cues that surround those speakers. When Mom speaks Japanese and interacts with other speakers in Japanese cultural contexts (e.g., wearing traditional clothes, attending temple, eating certain foods, interacting with ethnically Japanese people, etc.), her infant may be associating that language with the host of cultural cues that come along for free. When Mom is speaking English, there may be a different set of cultural cues present. Kandhadai et al. applied the concept of acquired distinctiveness to further highlight the potential power of associated sociocultural cues to aid the infant in separating their two languages. Acquired distinctiveness refers to infants’ enhanced ability to perceive subtle differences in minimally contrasting stimuli if those stimuli are individually associated with different salient cues (e.g., Yeung & Werker, 2009, where members of a nonnative phonetic contrast were paired with two distinct objects). The salient sociocultural cues associated with each language in infants’ learning environment may actually be enhancing the structural linguistic differences present, allowing for the rapid acquisition of both languages. Although bilingual and monolingual infants may reach gross developmental targets at similar ages, the processes underlying those abilities may still differ. Werker, Byers-Heinlein, and Fennell (2009) highlighted this distinction as shared advances in the macrostructure of language development (i.e., major milestones), but differences in its microstructure (i.e., detailed workings of acquisition mechanisms). The former highlights the robustness of language acquisition, and the latter reveals how the infant brain adapts to more complex and variable language environments. Examples of these processing differences have been highlighted throughout this chapter. Although monolingual and bilingual infants detect rhythmic differences between languages on the same developmental timetable (macrostructure), bilingual infants react slower than monolingual infants to a native language in comparison with a nonnative language (microstructure), which is thought to reflect a two-step identification process (i.e., determine whether it is a native language, then which native language). Most phonetic discrimination studies show similar successes for infants learning one or two languages (macrostructure), even for between-language productions of the same phoneme. But the few studies demonstrating bilingual infants’ lack of phoneme discrimination may be reflective of an adaptation to learning two languages with a large amount of lexical overlap (microstructure). This phonological leniency is also seen in infants’ greater acceptance of mispronunciations of cognate words over noncognate words. A final example concerns infants’ use of phonological detail in novel words. There is no developmental discrepancy in their first use of such detail (macrostructure) after accounting for preferential processing of phonological productions that match infants’ learning environment (microstructure). Finally, we would like to highlight some issues illustrated in the exhaustive list of peer-reviewed publications presented in Table 3.1. First, bilingual infant speech perception is an extremely new area of psycholinguistic research, with the oldest study



Speech Perception in Simultaneously Bilingual Infants 

 59

published less than 20 years ago and two thirds of the research published in the last 5 years. We have yet to answer many questions concerning speech perception in infants learning two languages; but, the rate of emerging research is extremely encouraging. Second, research with more sampling diversity is needed, both in terms of languages and sociocultural backgrounds. The bulk of research has been conducted in Spain and Canada with Spanish–Catalan and French–English learners respectively. Although research using these populations has answered fundamental questions, these families face fewer cultural challenges (e.g., low socioeconomic status) than in other areas, as bilingualism is valued in both cultures (which is also why such research is funded and conducted in these locations). Further, they only represent a tiny fraction of bilingual speakers’ possible language combinations. Research on bilingual infants in cultures where bilingualism is not the norm may highlight challenges not seen in the populations tested thus far and research with more linguistic diversity may reveal intriguing, language-specific interactions affecting infant speech perception processes. Studies with bilingual infants provide us with a deeper understanding of early language development and its underlying processes in all infants. Further, the findings emerging from this research have practical implications for parents and educators, revealing how and with what ease young, simultaneous bilingual infants acquire the most basic parts of language. To address the point raised in the first paragraph of this chapter: Two languages fit quite easily into infants’ heads.

References Albareda-Castellot, B., Pons, F., & Sebastián-Gallés, N. (2011). The acquisition of phonetic categories in bilingual infants: New data from an anticipatory eye movement paradigm. Developmental Science, 14, 395–401. http://dx.doi.org/10.1111/j.1467-7687.2010.00989.x Antoniou, M., Best, C. T., Tyler, M. D., & Kroos, C. (2010). Language context elicits native-like stop voicing in early bilinguals’ productions in both L1 and L2. Journal of Phonetics, 38, 640–653. http://dx.doi.org/10.1016/j.wocn.2010.09.005 Bijeljac-Babic, R., Serres, J., Höhle, B., & Nazzi, T. (2012). Effect of bilingualism on lexical stress pattern discrimination in French-learning infants. PLoS ONE, 7, e30843. http://dx.doi.org/ 10.1371/journal.pone.0030843 Bosch, L., & Ramon-Casas, M. (2011). Variability in vowel production by bilingual speakers: Can input properties hinder the early stabilization of contrastive categories? Journal of Phonetics, 39, 514–526. http://dx.doi.org/10.1016/j.wocn.2011.02.001 Bosch, L., & Sebastián-Gallés, N. (1997). Native-language recognition abilities in 4-month-old infants from monolingual and bilingual environments. Cognition, 65, 33–69. http:// dx.doi.org/10.1016/S0010-0277(97)00040-1 Bosch, L., & Sebastián-Gallés, N. (2001). Evidence of early language discrimination abilities in infants from bilingual environments. Infancy, 2, 29–49. http://dx.doi.org/10.1207/ S15327078IN0201_3

60 

 Fennell, Tsui, and Hudon

Bosch, L., & Sebastián-Gallés, N. (2003). Simultaneous bilingualism and the perception of a language-specific vowel contrast in the first year of life. Language and Speech, 46, 217–243. http://dx.doi.org/10.1177/00238309030460020801 Burns, T. C., Yoshida, K. A., Hill, K., & Werker, J. F. (2007). The development of phonetic representation in bilingual and monolingual infants. Applied Psycholinguistics, 28, 455–474. http://dx.doi.org/10.1017/S0142716407070257 Byers-Heinlein, K., Burns, T. C., & Werker, J. F. (2010). The roots of bilingualism in newborns. Psychological Science, 21, 343–348. http://dx.doi.org/10.1177/0956797609360758 Byers-Heinlein, K., & Fennell, C. T. (2014). Perceptual narrowing in the context of increased variation: Insights from bilingual infants. Developmental Psychobiology, 56, 274–291. http://dx.doi.org/ 10.1002/dev.21167 Byers-Heinlein, K., & Lew-Williams, C. (2013). Bilingualism in the early years: What the science says. LEARNing Landscapes, 7(1), 95–112. Curtin, S., Byers-Heinlein, K., & Werker, J. F. (2011). Bilingual beginnings as a lens for theory development: PRIMIR in focus. Journal of Phonetics, 39, 492–504. http://dx.doi.org/10.1016/ j.wocn.2010.12.002 Fennell, C. T., & Byers-Heinlein, K. (2014). You sound like Mommy: Bilingual and monolingual infants learn words best from speakers typical of their language environments. International Journal of Behavioral Development, 38, 309–316. http://dx.doi.org/10.1177/0165025414530631 Fennell, C. T., Byers-Heinlein, K., & Werker, J. F. (2007). Using speech sounds to guide word learning: The case of bilingual infants. Child Development, 78, 1510–1525. http://dx.doi.org/10.1111/ j.1467-8624.2007.01080.x Friederici, A. D., & Wessels, J. M. I. (1993). Phonotactic knowledge of word boundaries and its use in infant speech perception. Perception & Psychophysics, 54, 287–295. http://dx.doi.org/ 10.3758/BF03205263 Garcia-Sierra, A., Rivera-Gaxiola, M., Percaccio, C. R., Conboy, B. T., Romo, H., Klarman, L., . . . Kuhl, P. K. (2011). Bilingual language learning: An ERP study relating early brain responses to speech, language input, and later word production. Journal of Phonetics, 39, 546–557. http://dx.doi.org/ 10.1016/j.wocn.2011.07.002 Gervain, J., & Werker, J. F. (2013). Prosody cues word order in 7-month-old bilingual infants. Nature Communications, 4, 1490. http://dx.doi.org/10.1038/ncomms2430 Hudon, T. E. (2013). Bilingual infants’ accommodation of accented speech (Unpublished doctoral dissertation). University of Ottawa, Ottawa, Quebec, Canada. Izzard, E. (Producer). (1999). Eddie Izzard: Dress to Kill [DVD]. Jusczyk, P. W., & Aslin, R. N. (1995). Infants’ detection of the sound patterns of words in fluent speech. Cognitive Psychology, 29, 1–23. http://dx.doi.org/10.1006/cogp.1995.1010 Kandhadai, P., Danielson, D., & Werker, J. (2014). Culture as a binder for bilingual acquisition. Trends in Neuroscience and Education, 3, 24–27. http://dx.doi.org/10.1016/j.tine.2014.02.001 Kovács, A. M., & Mehler, J. (2009). Cognitive gains in 7-month-old bilingual infants. Proceedings of the National Academy of Sciences of the United States of America, 106, 6556–6560. http://dx.doi.org/ 10.1073/pnas.0811323106 Liu, L., & Kager, R. (2013). How bilingualism alter infants’ tone perception during perceptual reorganization. In S. Baiz, N. Goldman, & R. Hawkes (Eds.), Proceedings of the 37th Annual Boston University Conference on Language Development (pp. 231–240). Boston, MA: Cascadilla.



Speech Perception in Simultaneously Bilingual Infants 

 61

Mattock, K., Polka, L., Rvachew, S., & Krehm, M. (2010). The first steps in word learning are easier when the shoes fit: Comparing monolingual and bilingual infants. Developmental Science, 13, 229–243. http://dx.doi.org/10.1111/j.1467-7687.2009.00891.x Maye, J., Werker, J. F., & Gerken, L. (2002). Infant sensitivity to distributional information can affect phonetic discrimination. Cognition, 82, B101–B111. http://dx.doi.org/10.1016/ S0010-0277(01)00157-3 Mehler, J., Jusczyk, P., Lambertz, G., Halsted, N., Bertoncini, J., & Amiel-Tison, C. (1988). A precursor of language acquisition in young infants. Cognition, 29, 143–178. http://dx.doi.org/10.1016/ 0010-0277(88)90035-2 Molnar, M., Gervain, J., & Carreiras, M. (2014). Within-rhythm class native language discrimination abilities of Basque-Spanish monolingual and bilingual infants at 3.5 months of age. Infancy, 19, 326–337. http://dx.doi.org/10.1111/infa.12041 Morr, M. L., Shafer, V. L., Kreuzer, J. A., & Kurtzberg, D. (2002). Maturation of mismatch negativity in typically developing infants and preschool children. Ear and Hearing, 23, 118–136. http:// dx.doi.org/10.1097/00003446-200204000-00005 Näätänen, R. (1990). The role of attention in auditory information processing as revealed by eventrelated potentials and other brain measures of cognitive function. Behavioral and Brain Sciences, 13, 201–233. http://dx.doi.org/10.1017/S0140525X00078407 Petitto, L. A., Berens, M. S., Kovelman, I., Dubins, M. H., Jasinska, K., & Shalinsky, M. (2012). The “perceptual wedge hypothesis” as the basis for bilingual babies’ phonetic processing advantage: New insights from fNIRS brain imaging. Brain and Language, 121, 130–143. http://dx.doi.org/ 10.1016/j.bandl.2011.05.003 Ramon-Casas, M., & Bosch, L. (2010). Are non-cognate words phonologically better specified than cognates in the early lexicon of bilingual children? In M. Ortega-Llebaria (Ed.), Selected Proceedings of the 4th Conference on Laboratory Approaches to Spanish Phonology (pp. 31–36). Somerville, MA: Cascadilla. Ramon-Casas, M., Swingley, D., Sebastián-Gallés, N., & Bosch, L. (2009). Vowel categorization during word recognition in bilingual toddlers. Cognitive Psychology, 59, 96–121. http://dx.doi.org/ 10.1016/j.cogpsych.2009.02.002 Ramus, F., Nespor, M., & Mehler, J. (2000). Correlates of linguistic rhythm in the speech signal. Cognition, 75, AD3–AD30. http://dx.doi.org/10.1016/S0010-0277(00)00101-3 Rivera-Gaxiola, M., Silva-Pereyra, J., & Kuhl, P. K. (2005). Brain potentials to native and non-native speech contrasts in 7- and 11-month-old American infants. Developmental Science, 8, 162–172. http://dx.doi.org/10.1111/j.1467-7687.2005.00403.x Schmale, R., Hollich, G., & Seidl, A. (2011). Contending with foreign accent in early word learning. Journal of Child Language, 38, 1096–1108. http://dx.doi.org/10.1017/S0305000910000619 Schmale, R., & Seidl, A. (2009). Accommodating variability in voice and foreign accent: Flexibility of early word representations. Developmental Science, 12, 583–601. http://dx.doi.org/10.1111/ j.1467-7687.2009.00809.x Sebastián-Gallés, N. (2011). Bilingual language acquisition: Where does the difference lie? Human Development, 53, 245–255. http://dx.doi.org/10.1159/000321282 Sebastián-Gallés, N., Albareda-Castellot, B., Weikum, W. M., & Werker, J. F. (2012). A bilingual advantage in visual language discrimination in infancy. Psychological Science, 23(9), 994–999. http://dx.doi.org/10.1177/0956797612436817

62 

 Fennell, Tsui, and Hudon

Sebastián-Gallés, N., & Bosch, L. (2002). Building phonotactic knowledge in bilinguals: Role of early exposure. Journal of Experimental Psychology: Human Perception and Performance, 28, 974–989. http://dx.doi.org/10.1037/0096-1523.28.4.974 Sebastián-Gallés, N., & Bosch, L. (2009). Developmental shift in the discrimination of vowel contrasts in bilingual infants: Is the distributional account all there is to it? Developmental Science, 12, 874–887. http://dx.doi.org/10.1111/j.1467-7687.2009.00829.x Shafer, V. L., Yu, Y. H., & Datta, H. (2011). The development of English vowel perception in monolingual and bilingual infants: Neurophysiological correlates. Journal of Phonetics, 39, 527–545. http://dx.doi.org/10.1016/j.wocn.2010.11.010 Singh, L., & Foong, J. (2012). Influences of lexical tone and pitch on word recognition in bilingual infants. Cognition, 124, 128–142. http://dx.doi.org/10.1016/j.cognition.2012.05.008 Sundara, M., Polka, L., & Molnar, M. (2008). Development of coronal stop perception: Bilingual infants keep pace with their monolingual peers. Cognition, 108, 232–242. http://dx.doi.org/10.1016/ j.cognition.2007.12.013 Sundara, M., & Scutellaro, A. (2011). Rhythmic distance between languages affects the development of speech perception in bilingual infants. Journal of Phonetics, 39, 505–513. http:// dx.doi.org/10.1016/j.wocn.2010.08.006 Swingley, D. (2005). 11-month-olds’ knowledge of how familiar words sound. Developmental Science, 8, 432–443. http://dx.doi.org/10.1111/j.1467-7687.2005.00432.x Vihman, M. M., Thierry, G., Lum, J., Keren-Portnoy, T., & Martin, P. (2007). Onset of word form recognition in English, Welsh, and English–Welsh bilingual infants. Applied Psycho­linguistics, 28, 475–493. http://dx.doi.org/10.1017/S0142716407070269 Weikum, W. M., Vouloumanos, A., Navarra, J., Soto-Faraco, S., Sebastián-Gallés, N., & Werker, J. F. (2007). Visual language discrimination in infancy. Science, 316, 1159–1159. http://dx.doi.org/ 10.1126/science.1137686 Werker, J. F., Byers-Heinlein, K., & Fennell, C. T. (2009). Bilingual beginnings to learning words. Philosophical Transactions of the Royal Society of London: Series B. Biological Sciences, 364, 3649–3663. http://dx.doi.org/10.1098/rstb.2009.0105 Werker, J. F., Fennell, C. T., Corcoran, K. M., & Stager, C. L. (2002). Infants’ ability to learn phonetically similar words: Effects of age and vocabulary size. Infancy, 3, 1–30. http://dx.doi.org/ 10.1207/S15327078IN0301_1 Werker, J. F., & Tees, R. C. (1984). Cross-language speech perception: Evidence for perceptual reorganization during the first year of life. Infant Behavior & Development, 7, 49–63. http://dx.doi.org/ 10.1016/S0163-6383(84)80022-3 Yeung, H. H., & Werker, J. F. (2009). Learning words’ sounds before learning how words sound: 9-month-olds use distinct objects as cues to categorize speech information. Cognition, 113, 234–243. http://dx.doi.org/10.1016/j.cognition.2009.08.010

Barbara T. Conboy and Simona Montanari

4 Early Lexical Development in Bilingual Infants and Toddlers A central aspect of language acquisition is building a lexicon: a mental dictionary of words that includes their phonological forms, meanings, and grammatical functions in sentences. Infants learning a spoken language make significant progress in lexical learning in their first year, as they become familiar with native speech sound patterns, begin to recognize words in ongoing speech, and begin to associate those words with their referents. They subsequently expand their vocabularies, become more efficient at processing words, and learn to use words in sentences. Multiple research groups have shown that infants achieve milestones in lexical development at similar ages whether they are raised with bilingual or monolingual language input (for reviews, see Pearson, 2013; Werker, 2012). Researchers and practitioners have considered this ability of bilingual infants to learn at the same pace as monolingual infants something of a paradox because of the lesser exposure bilingual infants receive to each language and because the cues for word learning often conflict across languages. However, several studies have begun to shed light on the foundational skills that allow infants to learn word forms and meanings from two distinct sets of input, and how that learning further shapes infants’ processing abilities. This chapter reviews research on lexical development in bilingual infants and toddlers, focusing on the mechanisms involved in extracting and recognizing words from the speech stream, learning what those words mean, and encoding information about those words in memory. Several variables that have been linked to individual differences in rates of lexical acquisition, including the unique skills that develop in bilingual infants and toddlers in response to their input, are also discussed.

Segmentation of Words From the Input One of the first problems infants have to solve in learning words is identifying those words in the input. Because caregivers rarely speak to infants in single-word utterances (Brent & Siskind, 2001), infants must first segment words from the ongoing speech stream before they can learn the meanings of those words. The segmentation of words from ongoing speech is challenging because words are usually not separated with pauses or other obvious acoustic cues to their boundaries (see Jusczyk, 1999). This task is even more complex in bilingual situations because infants must learn to segment words from two different sets of input. http://dx.doi.org/10.1037/14939-005 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

64 

 Conboy and Montanari

As reviewed in Chapter 3 of this volume, infants use a variety of mechanisms— including statistical, prosodic, and phonetic cues—to segment words from the input. For instance, infants as young as 6 to 8 months of age use familiar words in continuous speech to identify boundaries (onsets or offsets) with surrounding words that are unknown and thus segment the unknown words. Another basis for word segmentation is statistical learning—the ability to compute and remember statistical regularities in sequences of stimuli, a basic ability present in newborn infants (e.g., Teinonen, Fellman, Näätänen, Alku, & Huotilainen, 2009). For example, an infant might notice that the syllable “ba” is frequently followed by the syllable “by” in the input, as in the word baby; therefore, this sequence with a high transitional probability will be recognized as a word as opposed to sequences of sounds or syllables with lower transitional probabilities (see Saffran, 2003). A further cue to word segmentation that is accessible to infants is prosody; stress patterns in words (realized as pitch, intensity, and/or duration variations) may cue word onsets and offsets in languages in reliable ways, and infants as young as 8 to 9 months of age have been shown to rely on this strategy for word segmentation (Thiessen & Saffran, 2003). Finally, toward the end of their first year, infants also use the phonetic regularities of their native language to segment words from the speech stream, including allophonic variations (Christophe, Dupoux, Bertoncini, & Mehler, 1994; Jusczyk, Hohne, & Bauman, 1999; among others) and phonotactic patterns (what sound sequences are permissible within words; Gonzalez-Gomez & Nazzi, 2013; Jusczyk, Cutler, & Redanz, 1993; see also Chapter 3, this volume). Data on word segmentation in bilingual infants are only beginning to emerge, but the finding that infants raised bilingually develop receptive vocabulary (when their total learning across languages is considered) on the same schedule as monolingual infants from the same communities and socioeconomic status (SES) suggests that bilingual infants can segment words from different sets of input (e.g., De Houwer, Bornstein, & Putnick, 2014; Pearson, 2013). On experimental tasks, infants learning two languages from the same rhythmic (prosodic) class (Catalan and Spanish) segment monosyllabic words from both languages at the same ages (6 and 8 months) as monolingual infants learning each of those languages (Bosch, Figueras, Teixidó, & Ramon-Casas, 2013), and infants learning two rhythmically distinct languages (English and French) segment bisyllabic words from both languages at the same ages (8 months) as monolingual infants (Polka & Sundara, 2003; see also Chapter 3, this volume). Tracking two sets of regularities in speech input—in the bilingual case— might be challenging for infants. Monolingual adults succeeded in segmenting words by simultaneously tracking two sets of statistics (transitional probabilities) on an artificial language learning task only when different voices were used for each language (Weiss, Gerfen, & Mitchel, 2009). Thus, dual-language learning infants might have difficulty tracking the statistics of each language if their input from both languages comes from the same caregivers. More research on this topic is needed; however, De Houwer (2007) documented that infants who hear two different input languages from



Early Lexical Development 

 65

the same caregivers learn words at similar rates to infants who hear each language from different caregivers. An ability that might allow bilingual infants to learn separate sets of regularities for each language when the input comes from the same speakers is the perceptual separation of the two languages (see Chapter 3, this volume). When infants acquire two languages from different rhythmic classes, they can use prosodic cues to separate those languages early in development. For example, in a study that used highamplitude sucking, in which rates of nonnutritive sucks on a pacifier are compared across different stimulus types, newborn infants who heard two rhythmically different languages from their mothers in utero (English and Tagalog) differentiated those languages (Byers-Heinlein, Burns, & Werker, 2010). When the two languages bilingual infants acquire are from the same rhythmic class, such as Catalan and Spanish, phonological differences between those languages allow infants to separate the languages by 4 months of age (Bosch & Sebastián-Gallés, 2001). At 4, 6, and 8 months, bilingual infants differentiated video clips (with no sound) of the same person speaking two different languages (English or French) using purely visual cues; monolingual infants the same age did this at 4 and 6 but not at 8 months (Weikum et al., 2007). Thus, bilingual infants appear to maintain for a longer period of time than monolingual infants a perceptual ability that could help them keep their input languages separate. The way that stress is used to make lexical contrasts in each of bilingual infants’ input languages also influences perception. For example, French does not use stress patterns to contrast word meanings, whereas in many other languages (e.g., English, Spanish), changing the syllable that is stressed changes the meaning of a word. Compared with monolingual French-learning infants the same age, 10-month-old bilingual infants learning French and a language with lexical stress contrasts (e.g., English) showed enhanced discrimination of prosodic contrasts in novel word stimuli that were pronounced in a language they had never heard before (German); moreover, bilingual infants whose dominant language was the language that used lexical stress contrasts showed more sensitivity to prosody than bilingual infants with more balanced input (Bijeljac-Babic, Serres, Höhle, & Nazzi, 2012). Exposure to a language that uses lexical stress thus appears to heighten the sensitivity of infants to such stress patterns. There is some evidence that the experiences bilingual infants have in perceptually switching between their languages while tracking dual sets of regularities enhance cognitive and perceptual abilities that in turn facilitate further language development. For example, 12-month-old bilingual infants showed an advantage in the simultaneous learning of multiple speech repetition structures on an artificial language learning task (e.g., ABA and AAB patterns, where A represents one syllable and B a different syllable; see Chapter 13). In this research, eye-tracking measures indicated that bilingual infants learned both sequences after a few minutes of exposure, but monolingual infants the same age did not, suggesting that practice hearing two different

66 

 Conboy and Montanari

sets of structural regularities refines this ability in bilingual infants (Kovács & Mehler, 2009a). Bilingual advantages have also been found in the ability to control attention to conflicting nonlinguistic stimuli in 7-month-old infants (Kovács & Mehler, 2009b) and toddlers (Poulin-Dubois, Blaye, Coutya, & Bialystok, 2011), and older children from a variety of backgrounds (see Chapters 13 and 14, this volume); however, they have not been found in all studies. It is not yet known whether enhanced cognitive control abilities facilitate the segmentation of words from bilingual input. In summary, infants use statistical, prosodic, and phonetic/phonological information to segment words from ongoing speech in their input, and their use of these cues varies with age but also varies according to the language the infant is acquiring. Although more research is needed, some studies have suggested that bilingual infants succeed with the added complexity of segmenting words from dual language input. This may be possible because they are able to separate the input, and because they develop enhanced cognitive control skills.

Word Recognition Segmentation abilities allow infants to recognize words in their input, a first step in word learning. Under experimental conditions, modifications of the input, such as the use of accentuation and repetition, have been shown to enhance word recognition (Männel & Friederici, 2013). Such modifications may simulate real-world speech input to infants. Infant-directed speech, which infants prefer over adultdirected speech and which often contains repetitions, accentuation, and other acoustic modifications, has been shown to facilitate word segmentation (Singh, Nestor, Parikh, & Yull, 2009) and recognition (Thiessen, Hill, & Saffran, 2005; Zangl & Mills, 2007). Mandel, Jusczyk, and Pisoni (1995) showed that infants as young as 4.5 months of age recognize a few words used frequently in their input (e.g., mommy, their own name) on experimental tasks that use the head-turn preference procedure. In this procedure, repetitions of familiar and unfamiliar stimuli are played from a loudspeaker and the relative amounts of time infants turn their head in the direction of the loudspeaker, before losing interest and looking away, are recorded as measures of familiarity. By 10 to 11 months infants’ ability to recognize new words becomes more robust. When 11-month-old infants are presented with familiar and unfamiliar words from their environments one word at a time, they show differences in brain activity (event-related potentials [ERPs]) before all of the word is heard (Thierry, Vihman, & Roberts, 2003). In the ERP procedure, amplitudes brain activity (indexing the firing of large populations of neurons) at particular moments in time are measured using electrodes placed on the infant’s or child’s scalp, and compared across conditions (e.g., familiar vs. unfamiliar words the infant passively listens to). Moreover, infants as young as 10 months show an ERP familiarity effect for words they have been exposed



Early Lexical Development 

 67

to experimentally in continuous speech, even when the words are also presented in continuous speech at test (Junge, Cutler, & Hagoort, 2014). Infants raised with bilingual input recognize familiar words at least by 11 months; however, this may vary for each language of the same infants. In addition to the relative amounts of exposure infants have had to each language, phonological properties of the two languages may influence how quickly word onsets are recognized. For example, the initial consonants of Welsh words change depending on whether the word is feminine or masculine; this variation in cues, along with the way in which initial syllables in Welsh words are stressed, may explain why Welsh monolingual infants do not recognize words as early as English monolingual infants, and why Welsh–English bilingual infants show a weaker recognition effect for Welsh words than for English words (Vihman, Thierry, Lum, Keren-Portnoy, & Martin, 2007). There is evidence that talker-specific information such as the fundamental frequency (pitch) of one’s voice facilitates the encoding of words into infants’ long-term memory (e.g., Houston & Jusczyk, 2003). Monolingual children typically learn language from multiple speakers, but bilingual children may learn one of their languages from only one speaker. It is not yet known how such speaker variables affect how bilingual infants recognize word forms in each language.

Word Comprehension and Processing When we say that an infant comprehends a word, we mean, at a minimum, that the infant has made an association between the word and any meaning to which it refers. Although the term comprehension is used by many language specialists to refer to an understanding of the full range of meanings corresponding to a word across a variety of contexts, this section reviews the early stages of word comprehension that occur before infants learn the exact meanings of words and how to use words appropriately across multiple contexts. Research on how infants process the words they comprehend is also reviewed. Large-scale studies that have used caregiver report questionnaires, such as the MacArthur-Bates Communicative Development Inventory (MB-CDI; Fenson et al., 1993), have suggested that infants comprehend many words before they begin to say them. The MB-CDI “Words and Gestures” form (normed on monolingual American English-learning 8- to 18-month-old infants) allows parents and other caregivers to select from a list those words they think an infant understands only, or both understands and says. The MB-CDI norms indicate that American English-learning infants comprehend dozens of words by 12 months of age, whereas infants typically say few, if any, words at that age (Fenson et al., 2007). Similar patterns have been reported across several languages and dialects for which adaptations of the MB-CDI have been applied to large samples (for a review, see Law & Roy, 2008). To date, no such largescale studies have been conducted with bilingual infants, though several smaller

68 

 Conboy and Montanari

scale studies have indicated that bilingual infants may learn fewer words in each of their languages than monolingual infants learning those languages (for a review, see Pearson, 2013). The finding of smaller vocabulary sizes in each language in bilingual compared with monolingual groups of infants the same age is not surprising, given that bilingual infants have their total language input divided across two languages. However, multiple input factors (quantity, quality, and the attitudes toward each language in the community) can impact how many words individual children learn (De Houwer, 2007). Bilingual infants often have more exposure to one language and know more words in that stronger, or more dominant language than in their weaker, or nondominant language (Pearson, Fernández, Lewedeg, & Oller, 1997). Moreover, even when exposure amounts are balanced, bilingual infants often learn each language in different settings, and therefore do not always have the opportunity to learn two different names (i.e., doublets) for the same concept (Pearson, Fernández, & Oller, 1993). Pearson and her colleagues suggested that composite measures of vocabulary size based on caregiver reports such as the MB-CDI be calculated taking infants’ knowledge in both languages into account (for a review, see Pearson, 2013). Separate inventories for each language may be completed, and the scores on both inventories added together, for a total vocabulary score that includes doublets as well as words known in only one language. Alternatively, the number of concepts to which the infant has attached a word may be calculated, so that doublets are counted only once (i.e., total conceptual vocabulary). When composite methods are used, the vocabulary sizes of bilingual infants are generally in the same ranges as those of monolingual infants the same ages (see Pearson, 2013). Although there are valid reasons for using each of these measures of vocabulary, a total vocabulary score that reflects all of the word forms an infant knows, across both languages, may be a better estimate of infants’ total lexical learning than a total conceptual score (Core, Hoff, Rumiche, & Señor, 2013). Many researchers and practitioners are concerned about using caregiver reports to measure word comprehension, because caregivers may confuse word comprehension with word familiarity and overestimate infants’ word knowledge. A related problem is that caregivers sometimes underestimate infants’ knowledge. In the bilingual case, the primary caregiver is often not the primary source of input in both languages, and thus multiple caregivers should be involved in completing these reports. Researchers have developed experimental tasks to more directly measure comprehension in infants and toddlers. One method involves having the infant or toddler touch or point to a real object, or to a picture of an object or action on a computer screen, given two or more choices; this method has also been used to capture relative speeds of word processing across two languages in bilingual toddlers (Poulin-Dubois et al., 2013). To test very young infants who cannot reliably identify items by pointing or touching, intermodal preferential looking (PL) tasks have been used (i.e., infants are presented with two pictures while a word is played simultaneously, and their eye gaze is recorded). Studies using PL tasks have shown that infants as young as



Early Lexical Development 

 69

6 months of age can associate common words with their referents (e.g., Bergelson & Swingley, 2012). Online methods that measure the time course of activation have allowed researchers to study how efficiently infants access meaningful words in both languages. These include ERP paradigms in which brain activity to words is used to measure the speed of lexical access as well as behavioral (eye-gaze tracking) methods. For example, Conboy and Mills (2006) compared ERPs to known and unknown words in 20-monthold bilingual toddlers acquiring English and Spanish. More rapid word processing occurred in the children’s dominant vs. nondominant language, and differences in processing across children were linked to composite vocabulary sizes. These results were interpreted as showing that more “mature” patterns of processing known words are associated not only with brain maturation, but also with language experience. Behavioral (eye-gaze tracking) online methods have also indicated that the efficiency of word processing is linked to the experience bilingual toddlers have with each language. Using a PL paradigm, Marchman, Fernald, and Hurtado (2010) measured the speed of word processing in bilingual 30-month-old toddlers learning English and Spanish. The speed with which the toddlers correctly identified words in each language (i.e., oriented visually toward the appropriate picture to go with the word they heard) was linked to expressive vocabulary size in that language, but not to vocabulary size in the other language. Processing speed in each language also predicts later language scores, and experience with each language predicts both these skills (Hurtado, Grüter, Marchman, & Fernald, 2014). Using a computer-administered picture-pointing (i.e., touchscreen) task, Poulin-Dubois and her colleagues (2013) compared the language comprehension skills of bilingual and monolingual 24-month-old toddlers who were matched on total expressive vocabulary sizes (using English and French adaptations of the MB-CDI). There were no group differences in word identification accuracy or reaction time, but within the bilingual group, children with higher proportions of doublets (i.e., two names for the same thing) in their expressive vocabularies had faster reaction times than those who knew fewer doublets.

Processes Involved in Linking Word Forms to Meanings Although bilingual infants do not always know words from both languages for the same referent, MB-CDI data suggest that they are capable of learning cross-language synonyms (doublets) from the earliest points in development, if given the opportunities in their environment, and that infants with more advanced lexical comprehension understand more doublets (De Houwer, Bornstein, & De Coster, 2006). Infants have also been shown to reject new words as labels for objects they have already associated with a word, a learning behavior known as mutual exclusivity (ME), which is thought to facilitate word learning (Markman, 1994). Bilingual infants adapt their word learning behaviors to meet the demands of learning more than one language. For example,

70 

 Conboy and Montanari

whereas monolingual 17- to 22-month-old infants used ME on an experimental wordlearning task, bilingual infants of the same age and with the same vocabulary size did not use ME (Houston-Price, Caloghiris, & Raviglione, 2010). In other research, bilingual 17- to 18-month-old infants used ME to a lesser extent than monolingual infants, and trilingual infants the same age did not use ME (Byers-Heinlein & Werker, 2009); moreover, bilingual infants who understood many doublets did not use ME but bilingual infants the same age who did not know doublets used ME to the same extent as monolingual peers (Byers-Heinlein & Werker, 2013). This latter finding suggests that the ways that infants learn new word meanings may be influenced by what they have already learned. Older (3-year-old) monolingual and bilingual children have also differed in the extent to which they used ME on experimental tasks (e.g., Davidson, Jergovic, Imami, & Theodos, 1997; Kalashnikova, Mattock, & Monaghan, 2015). However, other research has shown no differences between monolingual and bilingual 27- and 35-month-old children in the use of ME and no association between the extent to which bilingual children used ME and the amounts of doublets they knew (Frank & Poulin-Dubois, 2002). When infants learn to associate new words with their referents, they must also encode how the words sound in memory. Several studies have investigated whether bilingual and monolingual infants differ in their ability to encode the sounds in newly learned words using detailed phonological representations. In these studies, infants were presented with a switch task, in which they were presented simultaneously with a pseudoword and a picture of a novel object, allowing them to associate the word with a particular meaning. This was done for two different word–object pairings in which the objects were visually distinct but their labels were phonetically similar or dissimilar. When the infants habituated to the first word–object pairing (i.e., lost interest in looking at the object on the screen), the word from the second pairing was presented with the picture of the same object, and looking times were measured for evidence of dishabituation (i.e., recovery of eye gaze to the screen, a sign of discrimination between the habituated word and the new word). Monolingual infants typically succeed with the switch task at 14 months when the words sound very different but not until 17 months of age when the words are phonetically similar (e.g., bih and dih; Stager & Werker, 1997). Bilingual infants have shown different patterns, depending on the mode of presentation of the words. In one study, bilingual infants did not differentiate phonetically similar words as referents for different objects until 20 months, indicating a delay compared to monolingual infants (Fennell, Byers-Heinlein, & Werker, 2007). However, bilingual infants do not have more difficulty on this task than monolingual infants when the words are phonetically dissimilar, indicating that they learn to make word-object associations at the same ages (Byers-Heinlein, Fennell, & Werker, 2013). Moreover, Mattock, Polka, Rvachew, and Krehm (2010) found that 17-month-old bilingual infants succeeded on this task when the phonetically similar words were produced in a bilingual manner— with pronunciations from both languages (English and French). A follow-up study



Early Lexical Development 

 71

confirmed that bilingual infants failed to discriminate phonetically similar words produced in a monolingual manner (i.e., in one language’s phonology) and monolingual infants the same age failed when the words were produced in a bilingual manner (Fennell & Byers-Heinlein, 2014). Together, these findings show that bilingualism per se does not hinder infants’ encoding of the phonetic properties of new words that they are mapping to meaningful referents, but that monolingual and bilingual infants younger than 20 months of age respond differently to changes in word pronunciation because they do not yet have robust phonological representations of words.

Unique Bilingual Word Processing Demands The studies on word comprehension reviewed thus far have shown that bilingual infants learn the words of each of their languages in many of the same ways as monolingual infants, with slight differences emerging as they learn to manage two lexical systems. Bilingual infants must learn different phonological forms for the same concept (doublets) and access those forms from two lexicons that may be simultaneously activated. Evidence that bilingual infants develop interconnected lexicons comes from studies of phonological and semantic priming, which probe the cross-language activation of related lexical items. Phonological priming effects were found in bilingual German– English 21- to 43-month-old toddlers (Von Holzen & Mani, 2012). The children were tested using a PL paradigm in which they first heard a carrier phrase in their second language (English) such as “I saw a slide”, and subsequently heard a target word in their first language (German) that was either phonologically related to the last word of the carrier phrase (e.g., kleid), phonologically related through translation (e.g., the English word leg was paired with the German word stein because leg translates to bein in German), or unrelated. At the same time, the children were shown two competing pictures, one that depicted the target word and a distracter. Children’s looking toward the target was facilitated by the phonological primes, and there was interference from the priming that occurred through translation, showing that both languages were activated. Semantic priming was demonstrated by Singh (2014), who presented 30-monthold bilingual (Mandarin–English) toddlers with carrier phrases ending with a concrete noun (e.g., cat) and then presented the children with a semantically related, or target, word (e.g., dog), at the same time that the children were shown two competing pictures, one that depicted the target word and a distracter (e.g., a picture of a car). The prime and target were either in the same language or in different languages. The children showed a priming effect (faster visual orientation to the target picture) both across and within languages when the primes were presented in children’s dominant language, indicating that bilingual children this age use information from one language when accessing words in the other language. Competition between words that results from the simultaneous activation of both languages may lead to some of the enhancements in cognitive control documented in several studies of bilingual infants and toddlers (reviewed earlier in this chapter

72 

 Conboy and Montanari

and in Part IV of this volume). It is not known how cognitive control abilities relate to specific language abilities of bilingual infants or toddlers. In one study, 3-year-old bilingual children showed enhanced attentional control on a nonlinguistic task compared to monolingual peers, and within the bilingual group, attentional control was positively associated with success on a novel word-learning task that required children to inhibit one word interpretation in favor of a different interpretation (Yoshida, Tran, Benitez, & Kuwabara, 2011). In addition to managing interconnected and simultaneously activated lexicons, bilingual-language learners often encounter the uniquely bilingual situation of processing switches between languages. When toddlers process words in a mixed language condition, additional cognitive load may occur because of the activation of both lexicons and because of the need to use cognitive control processes. The ERPs of 20-month-old bilingual toddlers as they processed words they did or did not understand in each language suggested that processing in mixed conditions was slightly slower than processing in single-language conditions and that different distributions of brain activity occurred across these conditions, possibly reflecting the use of different cognitive processes (Conboy & Mills, 2006). Kuipers and Thierry (2012) specifically explored how ERPs varied in bilingual and monolingual toddlers as they processed words on mixed (language switching) and single language (nonswitching) trials during a picture-word task. In this research, children were presented with pictures accompanied by a word that either matched or did not match the picture; the trials were mostly presented in English, but occasionally were presented in Welsh. The bilingual (Welsh–English), but not monolingual, toddlers showed a unique ERP effect to the switched trials, suggesting that bilingual children’s language experience led to the development of a unique processing ability.

Word Production The expressive vocabulary skills of bilingual infants and toddlers have primarily been studied using caregiver report questionnaires and naturalistic language samples. Caregiver report methods allow researchers to avoid the pitfalls of eliciting, transcribing, and analyzing spontaneous speech samples from infants and toddlers and facilitate the collection of data from larger samples of children. Similar to the results reported above for receptive vocabulary, caregiver report measures such as a bilingual Spanish– English version of the Language Development Survey (Patterson, 2000) and composite scores derived from different language adaptations of the MB-CDI (Mancilla-Martinez, Pan, & Vagh, 2011; Marchman & Martínez-Sussmann, 2002) have been shown to be valid for measuring the expressive vocabulary skills of bilingual infants and toddlers, including children from families of low SES. Composite expressive vocabulary scores obtained for bilingual toddlers using MB-CDIs are in similar ranges as for monolingual children of the same ages and SES levels (e.g., De Houwer et al., 2014; Hoff et al., 2012;



Early Lexical Development 

 73

Pearson, 2013; Pearson et al., 1993). Moreover, if bilingual toddlers receive sufficient input to one of their languages (e.g., at least 60% of their input), their scores are in the same range as monolingual toddlers learning that language (Cattani et al., 2014). Bilingual infants developing typically achieve basic developmental milestones in expressive vocabulary at the same ages as monolingual infants. They produce their first words at approximately 1 year of age, learn their first 50 words before their second birthday, and show rapid acceleration in expressive vocabulary growth between their first and second birthdays (e.g., Core et al., 2013; Hoff et al., 2012; Patterson, 1998; Pearson, 2013; Pearson et al., 1993). Data obtained from MB-CDIs have also indicated that the expressive vocabularies of bilingual toddlers in each language resemble those of monolingual learners in their composition, with an early preponderance of social words that gives way to a dominance in referential words (e.g., noun object labels), and more gradual increases in the proportions of verbs, adjectives, and grammatical function words between 20 and 30 months (Conboy & Thal, 2006).

Input Factors in Bilingual Lexical Development In both bilingual and monolingual infants, vocabulary sizes are associated with the amounts of exposure children have to language; in bilingual infants, relative amounts of learning in each language are linked to amounts of exposure to that language (e.g., David & Wei, 2008; García-Sierra et al., 2011; Hoff et al., 2012; Hurtado et al., 2014; Parra, Hoff, & Core, 2011; Place & Hoff, 2011). However, other extrinsic variables may affect learning. There is evidence that when there is not sufficient exposure to a language that has minority status in a community, many children fail to learn to speak that language well (see Chapters 6 and 7, this volume). Aspects of input that potentially affect relative rates of learning across the languages of bilingual children are not well understood. Bilingual infants may receive input in each of their languages from interlocutors with distinct interaction styles and in distinct contexts, and certain styles and contexts may facilitate vocabulary learning to a greater degree than others (Patterson, 2002; Song, Tamis-LeMonda, Yoshikawa, Kahana-Kalman, & Wu, 2012; Tamis-LeMonda et al., 2014). Whether or not parents switch languages when speaking with their infants (e.g., intrasentential code-switching) has been considered a variable that could potentially impact early lexical development, because of changes in the reliability of the statistical cues and other regularities that facilitate lexical learning. In one large-scale study, a small negative association was found between parental use of intrasentential code-switching and infants’ vocabulary sizes in one language (Byers-Heinlein, 2013). Other research has not found a relationship between parent rates of mixing and infants’ composite vocabulary sizes, although those studies used smaller sample sizes and some included other forms of mixed input in addition to intrasentential code-switching (Bail, Morini, & Newman, 2015; David & Wei, 2008; Place & Hoff, 2011).

74 

 Conboy and Montanari

In addition to mixed input, a unique feature of bilingual input is that bilingual infants sometimes receive primary input in one language from nonnative speakers (Bosch & Ramon-Casas, 2011). Although the recognition of word forms by both bilingual and monolingual infants is affected by the extent to which the dialect (pronunciation) of the speaker matches that of the infant’s community (e.g., Fennell & ByersHeinlein, 2014; Mattock et al., 2010), multidialectal and bilingual infants appear to show greater flexibility in ignoring variations in pronunciation of words than monodialectal infants (Durrant, Delle Luche, Cattani, & Floccia, 2014; Ramon-Casas, Swingley, Sebastián-Gallés, & Bosch, 2009). Differences across infants in processing efficiency could also determine how input is processed. Bilingual toddlers with more exposure to English have larger vocabulary sizes in English and process words more efficiently in English; those with more exposure to Spanish and larger vocabulary sizes in Spanish process words more efficiently in Spanish (Conboy & Mills, 2006; Marchman et al., 2010). Vocabulary size in each language is also related to phonological memory as well as relative amounts of exposure (Parra et al., 2011).

Conclusion This chapter reviewed some patterns and basic mechanisms of early lexical development in infants and toddlers raised with dual language exposure. When research on bilingual infants and toddlers was not available, hypotheses that consider the unique circumstances of bilingual lexical development were discussed. The available research has shown that although many variables—external and internal to infants— influence bilingual lexical development, early dual language exposure offers infants an opportunity to become bilingual.

References Bail, A., Morini, G., & Newman, R. S. (2015). Look at the gato! Code-switching in speech to toddlers. Journal of Child Language, 42, 1073–1101. Bergelson, E., & Swingley, D. (2012). At 6–9 months, human infants know the meanings of many common nouns. Proceedings of the National Academy of Sciences of the United States of America, 109, 3253–3258. http://dx.doi.org/10.1073/pnas.1113380109 Bijeljac-Babic, R., Serres, J., Höhle, B., & Nazzi, T. (2012). Effect of bilingualism on lexical stress pattern discrimination in French-learning infants. PLoS ONE, 7, e30843. http://dx.doi.org/ 10.1371/journal.pone.0030843 Bosch, L., Figueras, M., Teixidó, M., & Ramon-Casas, M. (2013). Rapid gains in segmenting fluent speech when words match the rhythmic unit: Evidence from infants acquiring syllable-timed languages. Frontiers in Psychology, 4, 106. http://dx.doi.org/10.3389/fpsyg.2013.00106



Early Lexical Development 

 75

Bosch, L., & Ramon-Casas, M. (2011). Variability in vowel production by bilingual speakers: Can input properties hinder the early stabilization of contrastive categories? Journal of Phonetics, 39, 514–526. http://dx.doi.org/10.1016/j.wocn.2011.02.001 Bosch, L., & Sebastián-Gallés, N. (2001). Evidence of early language discrimination abilities in infants from bilingual environments. Infancy, 2, 29–49. http://dx.doi.org/10.1207/S15327078IN0201_3 Brent, M. R., & Siskind, J. M. (2001). The role of exposure to isolated words in early vocabulary development. Cognition, 81, B33–B44. http://dx.doi.org/10.1016/S0010-0277(01)00122-6 Byers-Heinlein, K. (2013). Parental language mixing: Its measurement and the relation of mixed input to young bilingual children’s vocabulary size. Bilingualism: Language and Cognition, 16, 32–48. http://dx.doi.org/10.1017/S1366728912000120 Byers-Heinlein, K., Burns, T. C., & Werker, J. F. (2010). The roots of bilingualism in newborns. Psycho­ logical Science, 21, 343–348. http://dx.doi.org/10.1177/0956797609360758 Byers-Heinlein, K., Fennell, C. T., & Werker, J. F. (2013). The development of associative word learning in monolingual and bilingual infants. Bilingualism: Language and Cognition, 16, 198–205. http://dx.doi.org/10.1017/S1366728912000417 Byers-Heinlein, K., & Werker, J. F. (2009). Monolingual, bilingual, trilingual: Infants’ language experience influences the development of a word-learning heuristic. Developmental Science, 12, 815–823. http://dx.doi.org/10.1111/j.1467-7687.2009.00902.x Byers-Heinlein, K., & Werker, J. F. (2013). Lexicon structure and the disambiguation of novel words: Evidence from bilingual infants. Cognition, 128, 407–416. http://dx.doi.org/10.1016/ j.cognition.2013.05.010 Cattani, A., Abbot-Smith, K., Farag, R., Krott, A., Arreckx, F., Dennis, I., & Floccia, C. (2014). How much exposure to English is necessary for a bilingual toddler to perform like a monolingual peer in language tests? International Journal of Language & Communication Disorders, 49, 649–671. http://dx.doi.org/10.1111/1460-6984.12082 Christophe, A., Dupoux, E., Bertoncini, J., & Mehler, J. (1994). Do infants perceive word boundaries? An empirical study of the bootstrapping of lexical acquisition. Journal of the Acoustical Society of America, 95, 1570–1580. http://dx.doi.org/10.1121/1.408544 Conboy, B. T., & Mills, D. L. (2006). Two languages, one developing brain: Event-related potentials to words in bilingual toddlers. Developmental Science, 9, F1–F12. http://dx.doi.org/10.1111/ j.1467-7687.2005.00453.x Conboy, B. T., & Thal, D. J. (2006). Ties between the lexicon and grammar: Cross-sectional and longitudinal studies of bilingual toddlers. Child Development, 77, 712–735. Core, C., Hoff, E., Rumiche, R., & Señor, M. (2013). Total and conceptual vocabulary in SpanishEnglish bilinguals from 22 to 30 months: Implications for assessment. Journal of Speech, Language, and Hearing Research, 56, 1637–1649. http://dx.doi.org/10.1044/ 1092-4388(2013/11-0044) David, A., & Wei, L. (2008). Individual differences in the lexical development of French–English bilingual children. International Journal of Bilingual Education and Bilingualism, 11, 598–618. http://dx.doi.org/10.2167/beb478.0 Davidson, D., Jergovic, D., Imami, Z., & Theodos, V. (1997). Monolingual and bilingual children’s use of the mutual exclusivity constraint. Journal of Child Language, 24, 3–24. http://dx.doi.org/ 10.1017/S0305000996002917 De Houwer, A. (2007). Parental language input patterns and children’s bilingual use. Applied Psycho­ linguistics, 28, 411–424. http://dx.doi.org/10.1017/S0142716407070221

76 

 Conboy and Montanari

De Houwer, A., Bornstein, M. H., & De Coster, S. (2006). Early understanding of two words for the same thing: A CDI study of lexical comprehension in infant bilinguals. The International Journal of Bilingualism, 10, 331–347. http://dx.doi.org/10.1177/13670069060100030401 De Houwer, A., Bornstein, M. H., & Putnick, D. L. (2014). A bilingual–monolingual comparison of young children’s vocabulary size: Evidence from comprehension and production. Applied Psycholinguistics, 35, 1189–1211. http://dx.doi.org/10.1017/S0142716412000744 Durrant, S., Delle Luche, C., Cattani, A., & Floccia, C. (2014). Monodialectal and multidialectal infants’ representation of familiar words. Journal of Child Language, 42, 447–465. http:// dx.doi.org/10.1017/S0305000914000063 Fennell, C., & Byers-Heinlein, K. (2014). You sound like Mommy: Bilingual and monolingual infants learn words best from speakers typical of their language environments. International Journal of Behavioral Development, 38, 309–316. http://dx.doi.org/10.1177/0165025414530631 Fennell, C. T., Byers-Heinlein, K., & Werker, J. F. (2007). Using speech sounds to guide word learning: The case of bilingual infants. Child Development, 78, 1510–1525. http://dx.doi.org/10.1111/ j.1467-8624.2007.01080.x Fenson, L., Dale, P., Reznick, S., Thal, D., Bates, E., Hartung, J., . . . Reilly, J. (1993). MacArthur Communicative Development Inventories: User’s guide and technical manual. San Diego, CA: Singular. Fenson, L., Marchman, V., Thal, D. J., Dale, P. S., Reznick, J. S., & Bates, E. (2007). MacArthur-Bates Communicative Development Inventories: User’s guide and technical manual. Baltimore, MD: Brookes. Frank, I., & Poulin-Dubois, D. (2002). Young monolingual and bilingual children’s responses to violation of the Mutual Exclusivity Principle. The International Journal of Bilingualism, 6, 125–146. http://dx.doi.org/10.1177/13670069020060020201 García-Sierra, A., Rivera-Gaxiola, M., Percaccio, C., Conboy, B. T., Romo, H., Klarman, L., . . . Kuhl, P. K. (2011). Bilingual language learning: An ERP study relating early brain responses to speech, language input, and later word production. Journal of Phonetics, 39, 546–557. http://dx.doi.org/ 10.1016/j.wocn.2011.07.002 Gonzalez-Gomez, N., & Nazzi, T. (2013). Effects of prior phonotactic knowledge on infant word segmentation: The case of nonadjacent dependencies. Journal of Speech, Language, and Hearing Research, 56, 840–849. http://dx.doi.org/10.1044/1092-4388(2012/12-0138) Hoff, E., Core, C., Place, S., Rumiche, R., Señor, M., & Parra, M. (2012). Dual language exposure and early bilingual development. Journal of Child Language, 39, 1–27. http://dx.doi.org/10.1017/ S0305000910000759 Houston, D. M., & Jusczyk, P. W. (2003). Infants’ long-term memory for the sound patterns of words and voices. Journal of Experimental Psychology: Human Perception and Performance, 29, 1143–1154. http://dx.doi.org/10.1037/0096-1523.29.6.1143 Houston-Price, C., Caloghiris, Z., & Raviglione, E. (2010). Language experience shapes the development of the mutual exclusivity bias. Infancy, 15, 125–150. http://dx.doi.org/10.1111/ j.1532-7078.2009.00009.x Hurtado, N., Grüter, T., Marchman, V. A., & Fernald, A. (2014). Relative language exposure, processing efficiency and vocabulary in Spanish–English bilingual toddlers. Bilingualism: Language and Cognition, 17, 189–202. http://dx.doi.org/10.1017/S136672891300014X Junge, C., Cutler, A., & Hagoort, P. (2014). Successful word recognition by 10-month-olds given continuous speech both at initial exposure and test. Infancy, 19, 179–193. http://dx.doi.org/ 10.1111/infa.12040



Early Lexical Development 

 77

Jusczyk, P. W. (1999). How infants begin to extract words from speech. Trends in Cognitive Sciences, 3, 323–328. http://dx.doi.org/10.1016/S1364-6613(99)01363-7 Jusczyk, P. W., Cutler, A., & Redanz, N. J. (1993). Infants’ preference for the predominant stress patterns of English words. Child Development, 64, 675–687. http://dx.doi.org/ 10.2307/1131210 Jusczyk, P. W., Hohne, E. A., & Bauman, A. (1999). Infants’ sensitivity to allophonic cues for word segmentation. Perception & Psychophysics, 61, 1465–1476. http://dx.doi.org/10.3758/BF03213111 Kalashnikova, M., Mattock, K., & Monaghan, P. (2015). The effects of linguistic experience on the flexible use of mutual exclusivity in word learning. Bilingualism: Language and Cognition, 18, 626–638. Kovács, Á. M., & Mehler, J. (2009a). Cognitive gains in 7-month-old bilingual infants. Proceedings of the National Academy of Sciences of the United States of America, 106, 6556–6560. http:// dx.doi.org/10.1073/pnas.0811323106 Kovács, Á. M., & Mehler, J. (2009b). Flexible learning of multiple speech structures in bilingual infants. Science, 325, 611–612. http://dx.doi.org/10.1126/science.1173947 Kuipers, J. R., & Thierry, G. (2012). Event-related potential correlates of language change detection in bilingual toddlers. Developmental Cognitive Neuroscience, 2, 97–102. http://dx.doi.org/ 10.1016/j.dcn.2011.08.002 Law, J., & Roy, P. (2008). Parental report of infant language skills: A review of the development and application of the Communicative Development Inventories. Child and Adolescent Mental Health, 13, 198–206. http://dx.doi.org/10.1111/j.1475-3588.2008.00503.x Mancilla-Martinez, J., Pan, B. A., & Vagh, S. B. (2011). Assessing the productive vocabulary of Spanish–English bilingual toddlers from low-income families. Applied Psycholinguistics, 32, 333–357. http://dx.doi.org/10.1017/S0142716410000433 Mandel, D. R., Jusczyk, P. W., & Pisoni, D. B. (1995). Infants’ recognition of the sound patterns of their own names. Psychological Science, 6, 314–317. http://dx.doi.org/10.1111/ j.1467-9280.1995.tb00517.x Männel, C., & Friederici, A. D. (2013). Accentuate or repeat? Brain signatures of developmental periods in infant word recognition. Cortex, 49, 2788–2798. http://dx.doi.org/10.1016/ j.cortex.2013.09.003 Marchman, V. A., Fernald, A., & Hurtado, N. (2010). How vocabulary size in two languages relates to efficiency in spoken word recognition by young Spanish–English bilinguals. Journal of Child Language, 37, 817–840. http://dx.doi.org/10.1017/S0305000909990055 Marchman, V. A., & Martínez-Sussmann, C. (2002). Concurrent validity of caregiver/parent report measures of language for children who are learning both English and Spanish. Journal of Speech, Language, and Hearing Research, 45, 983–997. http://dx.doi.org/10.1044/ 1092-4388(2002/080) Markman, E. M. (1994). Constraints on word meaning in early language acquisition. Lingua, 92, 199–227. http://dx.doi.org/10.1016/0024-3841(94)90342-5 Mattock, K., Polka, L., Rvachew, S., & Krehm, M. (2010). The first steps in word learning are easier when the shoes fit: Comparing monolingual and bilingual infants. Developmental Science, 13, 229–243. http://dx.doi.org/10.1111/j.1467-7687.2009.00891.x Parra, M., Hoff, E., & Core, C. (2011). Relations among language exposure, phonological memory, and language development in Spanish–English bilingually developing 2-year-olds. Journal of Experimental Child Psychology, 108, 113–125. http://dx.doi.org/10.1016/ j.jecp.2010.07.011

78 

 Conboy and Montanari

Patterson, J. L. (1998). Expressive vocabulary development and word combinations of Spanish– English bilingual toddlers. American Journal of Speech-Language Pathology, 7, 46–56. http:// dx.doi.org/10.1044/1058-0360.0704.46 Patterson, J. L. (2000). Observed and reported expressive vocabulary and word combinations in bilingual toddlers. Journal of Speech, Language, and Hearing Research, 43, 121–128. http:// dx.doi.org/10.1044/jslhr.4301.121 Patterson, J. L. (2002). Relationships of expressive vocabulary to frequency of reading and television experience among bilingual toddlers. Applied Psycholinguistics, 23, 493–508. http://dx.doi.org/ 10.1017/S0142716402004010 Pearson, B. Z. (2013). Distinguishing the bilingual as a late talker from the late talker who is bilingual. In L. A. Rescorla & P. S. Dale (Eds.), Late talkers: Language development, interventions, and outcomes (pp. 67–87). Baltimore, MD: Brookes. Pearson, B. Z., Fernández, S. C., Lewedeg, V., & Oller, D. K. (1997). The relation of input factors to lexical learning by bilingual infants. Applied Psycholinguistics, 18, 41–58. http://dx.doi.org/ 10.1017/S0142716400009863 Pearson, B. Z., Fernández, S. C., & Oller, D. K. (1993). Lexical development in bilingual infants and toddlers: Comparison to monolingual norms. Language Learning, 43, 93–120. http://dx.doi.org/ 10.1111/j.1467-1770.1993.tb00174.x Place, S., & Hoff, E. (2011). Properties of dual language exposure that influence 2-year-olds’ bilingual proficiency. Child Development, 82, 1834–1849. http://dx.doi.org/10.1111/ j.1467-8624.2011.01660.x Polka, L., & Sundara, M. (2003, August). Word segmentation in monolingual and bilingual infant learners of English and French. In Proceedings of the 15th International Congress of Phonetic Sciences (pp. 1021–1024). Barcelona, Spain: Causal Productions. Poulin-Dubois, D., Bialystok, E., Blaye, A., Polonia, A., & Yott, J. (2013). Lexical access and vocabulary development in very young bilinguals. The International Journal of Bilingualism, 17, 57–70. http://dx.doi.org/10.1177/1367006911431198 Poulin-Dubois, D., Blaye, A., Coutya, J., & Bialystok, E. (2011). The effects of bilingualism on toddlers’ executive functioning. Journal of Experimental Child Psychology, 108, 567–579. http://dx.doi.org/ 10.1016/j.jecp.2010.10.009 Ramon-Casas, M., Swingley, D., Sebastián-Gallés, N., & Bosch, L. (2009). Vowel categorization during word recognition in bilingual toddlers. Cognitive Psychology, 59, 96–121. http://dx.doi.org/ 10.1016/j.cogpsych.2009.02.002 Saffran, J. R. (2003). Statistical language learning mechanisms and constraints. Current Directions in Psychological Science, 12, 110–114. http://dx.doi.org/10.1111/1467-8721.01243 Singh, L. (2014). One world, two languages: Cross-language semantic priming in bilingual toddlers. Child Development, 85, 755–766. http://dx.doi.org/10.1111/cdev.12133 Singh, L., Nestor, S., Parikh, C., & Yull, A. (2009). Influences of infant-directed speech on early word recognition. Infancy, 14, 654–666. http://dx.doi.org/10.1080/15250000903263973 Song, L., Tamis-LeMonda, C. S., Yoshikawa, H., Kahana-Kalman, R., & Wu, I. (2012). Language experiences and vocabulary development in Dominican and Mexican infants across the first 2 years. Developmental Psychology, 48, 1106–1123. http://dx.doi.org/10.1037/a0026401 Stager, C. L., & Werker, J. F. (1997). Infants listen for more phonetic detail in speech perception than in word-learning tasks. Nature, 388, 381–382. http://dx.doi.org/10.1038/41102 Tamis-LeMonda, C. S., Song, L., Luo, R., Kuchirko, Y., Kahana-Kalman, R., Yoshikawa, H., & Raufman, J. (2014). Children’s vocabulary growth in English and Spanish across early development and



Early Lexical Development 

 79

associations with school readiness skills. Developmental Neuropsychology, 39, 69–87. http:// dx.doi.org/10.1080/87565641.2013.827198 Teinonen, T., Fellman, V., Näätänen, R., Alku, P., & Huotilainen, M. (2009). Statistical language learning in neonates revealed by event-related brain potentials. BMC Neuroscience. Advance online publication. http://dx.doi.org/10.1186/1471-2202-10-21 Thierry, G., Vihman, M., & Roberts, M. (2003). Familiar words capture the attention of 11-month-olds in less than 250 ms. NeuroReport, 14, 2307–2310. http://dx.doi.org/10.1097/ 00001756-200312190-00004 Thiessen, E. D., Hill, E. A., & Saffran, J. R. (2005). Infant-directed speech facilitates word segmentation. Infancy, 7, 53–71. http://dx.doi.org/10.1207/s15327078in0701_5 Thiessen, E. D., & Saffran, J. R. (2003). When cues collide: Use of stress and statistical cues to word boundaries by 7- to 9-month-old infants. Developmental Psychology, 39, 706–716. http:// dx.doi.org/10.1037/0012-1649.39.4.706 Vihman, M. M., Thierry, G., Lum, J., Keren-Portnoy, T., & Martin, P. (2007). Onset of word form recognition in English, Welsh, and English–Welsh bilingual infants. Applied Psychol­inguistics, 28, 475–493. http://dx.doi.org/10.1017/S0142716407070269 Von Holzen, K., & Mani, N. (2012). Language nonselective lexical access in bilingual toddlers. Journal of Experimental Child Psychology, 113, 569–586. http://dx.doi.org/10.1016/j.jecp.2012.08.001 Weikum, W. M., Vouloumanos, A., Navarra, J., Soto-Faraco, S., Sebastián-Gallés, N., & Werker, J. F. (2007). Visual language discrimination in infancy. Science, 316, 1159–1159. http://dx.doi. org/10.1126/science.1137686 Weiss, D. J., Gerfen, C., & Mitchel, A. D. (2009). Speech segmentation in a simulated bilingual environment: A challenge for statistical learning? Language Learning and Development, 5, 30–49. http://dx.doi.org/10.1080/15475440802340101 Werker, J. (2012). Perceptual foundations of bilingual acquisition in infancy. Annals of the New York Academy of Sciences, 1251(1), 50–61. http://dx.doi.org/10.1111/j.1749-6632.2012.06484.x Yoshida, H., Tran, D. N., Benitez, V., & Kuwabara, M. (2011). Inhibition and adjective learning in bilingual and monolingual children. Frontiers in Psychology, 2, 210. http://dx.doi.org/10.3389/ fpsyg.2011.00210 Zangl, R., & Mills, D. L. (2007). Increased brain activity to infant-directed speech in 6- and 13-monthold infants. Infancy, 11, 31–62. http://dx.doi.org/10.1207/s15327078in1101_2

W. Quin Yow, Ferninda Patrycia, and Suzanne Flynn

5 Code-Switching in Childhood

Code-switching refers to the practice of alternating between two or more languages in the context of a single conversation (e.g., Lanza, 1992; Milroy & Muysken, 1995; MyersScotton, 1993; Ruan, 2003). Some researchers distinguish the term code-switching from code-mixing, where the former refers to the juxtaposition of two codes (languages or dialects) in a discourse that are grammatical and constrained by regularities, and the latter is more commonly used to refer to the mixing of two or more languages in a word or within a sentence (e.g., Brice & Anderson, 1999; Faltis, 1989; Meisel, 1989, 1994; Muysken, 2000; Nicoladis & Genesee, 1997; Poplack, 2001). For example, Meisel (1989) referred to code-switching as a language choice “that is constrained by the properties of the linguistic system” (p. 13) and code-mixing as “the fusion of two grammatical systems” (p. 37). Code-mixing is sometimes used exclusively to refer to young learners’ speech during the early stages of bilingual development. In particular, bilingual children are found to mix as early as 2 years old, but they apparently acquire the more subtle rules governing the use of language only at 5 years old or older. Hence, it is argued that young children code-mix rather than code-switch because they do not have the same abilities as adults to do so (Meisel, 1994). Although these researchers attempt to differentiate between code-switching and code-mixing on the basis of whether proficiency in both languages is already acquired and whether the use of the two languages is deliberate and systematic, other researchers use the terms code-switching and code-mixing interchangeably (e.g., Bokamba, 1989; Clyne, 1987; Genesee, 1989). For example, Poplack (2001) referred to code-switching as the “mixing, by bilinguals (or multilinguals), of two or more languages in a discourse with no change of interlocutor or topic [that] may take place at any linguistic structure” (p. 1). Some researchers prefer to focus not on details of constituent structure but on the discourse and interactional functions that code-switching performs for speakers. These researchers refer to code-switching and code-mixing as situational shifting (switching triggered by a change in situation), where switching/mixing is “an element in a socially agreed matrix of contextualization cues and conventions used by speakers to alert addressees, in the course of ongoing interaction, to the social and situational context of the conversation” (Gumperz, 1982, pp. 132–152; Pakir, 1989). We are grateful to the children and parents who participated and to the teachers and staff of Red School House and Creative O Preschoolers’ Bay. We thank Yvonne Yong, Qi Xuan Yap, Xiaoqian Li, and Wanyu Hung for their help in this study. This work was partially supported by the SUTD SRG under Grant SRG HASS 2011 011 and the Singapore–MIT International Design Center (IDC) under Grant IDG31100106 and IDD41100104 to the first author. http://dx.doi.org/10.1037/14939-006 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

82 

 Yow, Patrycia, and Flynn

The field has not yet reached a consensus concerning the use of this terminology. Thus, it is imperative for one to define the terms to be used at the outset of any discussion on this topic. In this chapter, we use the terms code-switching and code-mixing synonymously to mean the practice of a person’s use of two or more languages in a single unit of discourse, such as a word, an utterance, or a conversation without regard to grammatical and pragmatic constraints (Genesee, Paradis, & Crago, 2004; Meisel, 1989). We first describe the results of several relevant previous studies on code-switching with children; then we describe a study that we have conducted that examined the relationship between code-switching exposure/behaviour and lexical development in children from a multilingual society where code-switching is a common practice among the bilingual speakers.

Is Code-Switching a Sign of Confusion, an Indication of Language Loss, or a Mark of Language Competency? There exists a wealth of literature on code-switching in bilingual adults concerning the grammatical constraints evident in their speech (e.g., Cantone, 2007; MacSwan, 2014; MacSwan & McAlister, 2010; Meisel, 1994; Poplack, 1980) as well as the social and pragmatic functions served by such utterances (e.g., Gumperz, 1971; McClure, 1977; Valdes-Fallis, 1978; see also Chapter 10, this volume). In contrast, studies dealing with code-switching in the speech of children is more limited, being predominantly focused on the triggers for and the communicative functions of code-switching at this age (e.g., Grosjean, 1982; Saunders, 1988). Early research on code-switching in bilingual children interpreted instances of code-switching as evidence of undifferentiated phonological, lexical, and syntactic subsystems, and as a cause for concern as it might suggest that a bilingual child is not developing typically, is confused, or is linguistically incompetent (Labov, 1972; Lindholm & Padilla, 1978; see also Chapter 2, this volume). Proponents of this position believed that young bilingual children switch or mix words and other elements from their known languages in the same utterance because they do not have the ability to discriminate between these languages, or do so mainly to fill a lexical gap, especially in the early stages of language development (King & Fogle, 2006; Perani et al., 2003; Portocarrero, Burright, & Donovick, 2007). For example, it has been argued that bilingual children (3.5 years old or younger) borrow words from their less developed language when they do not yet have a translation equivalent and thus insert such lexical items into their dominant language (i.e., the language that the children are most fluent in), or use domain-specific words in one language only (see Nicoladis & Genesee, 1997, for a review). However, other research suggests that bilingual children’s code-switching is normal, is systematic, and conforms to the grammatical constraints imposed by each of the two participating languages and thus cannot be considered a sign of confusion



Code-Switching in Childhood 

 83

of their two languages (e.g., MacSwan, 1999; Meisel, 1994; Nicoladis & Genesee, 1997; Paradis, Nicoladis, & Genesee, 2000; van Gelderen & MacSwan, 2008). Case studies, for example, suggest that bilingual children’s code-switching is largely a performancebased phenomenon; however, their code-switches do indeed conform to the linguistic and pragmatic constraints of both languages. In such cases, the switches typically do not violate the grammatical principles and other such constraints of either language. In this way, these code-switches reflect the children’s communicative and linguistic competence in both languages (Genesee, 2002). In other words, bilingual children switch or mix words because they have the (equivalent) competency to do so in both languages and not because they are confused, incompetent, or do so randomly. For example, Meisel (1994) examined the utterances of two German–French bilingual children 1.5 to 3 years old and found that code-switching was used from early on and did not violate the grammatical properties of the languages involved. Paradis, Nicoladis, and Genesee (2000) examined the interactions of 15 French– English bilingual children and their parents at 6-month intervals from the ages of 2 to 3.5 years and found that children’s code-switched utterances adhered to adult-like grammatical constraints a significant portion of the time. This suggests that bilingual children code-switch because they possess the knowledge of how to merge two languages together in one utterance and the syntactic knowledge of each language even during an early period of language development. For further discussion of research on code-switching in mature speakers, see MacSwan (2013, 2014; see also Chapter 10, this volume). In contrast, several other researchers have suggested that code-switching may be related to language loss in children. In this context, language loss refers to the failure of children to attain the expected proficiency in their first language while children’s second language skills continue to develop usually because of the discrepant exposure to the two languages (Anderson, 2004; Schiff-Myers, 1992; Sharwood Smith & Van Buren, 1991). These situations usually involve sequential bilingual language learning in which children are first exposed to one language and then are introduced to a new target language. In these cases, children have been argued to begin to mix the two languages in their utterances as they are increasingly exposed to the new second language; their utterances often indicate an increase in error patterns in their first language, which has been argued to eventually lead to a loss of the first language (Kaufman & Aronoff, 1991). However, Guiberson, Barrett, Jancosek, and Yoshinaga-Itano (2006) studied 10 children of Mexican immigrant parents over the course of 3 years and found that children’s grammatical errors were predictive of language loss but code-switching was not. This and other related studies documented that children’s code-switch utterances are not related to language loss, suggesting that other variables (e.g., motivation, language aptitude, learning style) may influence whether children will maintain or lose their first language instead (Brice & Anderson, 1999; Guiberson et al., 2006; Moore, Beatty, & Perez-Mendez, 1995).

84 

 Yow, Patrycia, and Flynn

Code-Switching and the Larger Language and Social Environment The language and social environment that a bilingual speaker is in play an important role when deciding what, when, and how much to code-switch. Code-switching may be a well-accepted practice in some communities, but it hardly exist in others, resulting in variation between communities with respect to the prevalence of code-switching (Poplack & Sankoff, 1988). For example, in Montreal, many people are bilingual, but the languages are kept relatively separate compared with other bilingual communities (Heller, 1985; Nicoladis & Genesee, 1996). In this community, children between the ages of 2 and 3.5 years typically learn the pragmatic rule valid in Montreal that dictates avoiding code-switching (Nicoladis & Genesee, 1997). But even within this same community, there are variations in the practice of code-switching as some families are strict about keeping the languages separate whereas others are relaxed about switching and mixing. Children tend to develop and match their own rates of mixing to those of their interlocutors from their social community. For example, Comeau, Genesee, and Lapaquette (2003) found that 2.5-year-old bilingual children adapted their own rates of mixing to match their interlocutor who varied her rates of mixing from relatively low (15%) to relatively high (40%) and back to relatively low (15%) again. Similarly, TabouretKeller (1963) found that 2-year-old children developed rates of mixing that matched those of their parents. Although Goodz (1989) reported significant correlations between child and parent code-switching in half of the eight cases, Nicoladis (1995) found significant correlations between child and parent code-switching in only two of 12 cases in the same community in Montreal, suggesting variations in the extent of engagement in code-switching behaviour between families within the same community. Although the majority of the previous research focused on child–parent interactions in young bilingual children, studies on children’s code-switching behavior in their larger language environment (e.g., child care, preschool) are lacking. Furthermore, the effects of code-switching on language development in bilingual children in a sociolinguistic context where code-switching is the norm has not yet been adequately examined. In this chapter, we report a naturalistic observation study with bilingual children interacting with their teachers and peers in Singapore, where codeswitching is a common practice among its people. Singapore is a multilingual, multicultural society with English as the official language and three other languages as official mother tongues (Mandarin, Bahasa Melayu, Tamil). About 74% of its population is ethnic Chinese, whereas the remaining population is 13% Malays, 9% Indians, and 3% other (Singapore Department of Statistics, 2012). English is the most frequently spoken language at home with children between the ages of 5 and 9 years (51%) compared with Mandarin (28%), Bahasa Melayu (13%), Indian languages (6%), and others (2%; Singapore Department of Statistics, 2010). Because of its unique bilingual education policy, all children enrolled in



Code-Switching in Childhood 

 85

Singapore schools have to learn two languages: English and a mother tongue. Ethnic Chinese children typically learn English and Mandarin, whereas ethnic Malay children typically learn English and Bahasa Melayu, and ethnic Indian children typically learn English and Tamil from preschool onwards. Thus, English is the shared language amongst Singapore children in school. Although some research has been conducted to study code-switching in Singapore, which predominantly focused on adult bilingual code-switching, none of the studies examined the effects of exposure to and engagement in code-switching on children’s language development (Baetens Beardsmore, 1998; Goh, Zhao, & Liu, 2007; Ong & Zhang, 2010). Thus, the first objective of our study is to examine the impact of code-switching exposure and behavior on Singapore children’s linguistic proficiency in English and Mandarin. However, even within this population of bilingual speakers, the extent of code-switching may differ across families or school environments. Our second objective of the study is to compare the code-switching behavior among children from two Singapore schools that differ in their sociolinguistic context where one school appears to have a more active code-switching environment than the other. We propose that there are two important factors that would influence a bilingual child’s language development and code-switching behavior: (a) the child’s home language environment and (b) the child’s larger social environment (with peers).

The Study Participants A naturalistic observation study was conducted with 36 Mandarin–English children between the ages of 5 years, 5 months and 6 years, 7 months who came from two private schools in Singapore (16 from School M and 20 from School E; 15 girls, 21 boys) as part of a larger study on language acquisition. Eight additional participants’ data were not included in the analyses because of their low attendance during the observation days that resulted in very little recording time (less than 5% of the total recording time in the school) and one additional participant’s data were excluded because she only spoke eight utterances throughout the whole observation session.

Materials and Measures Demographic Information, Language Background, and Parental Reports on Code-Switching Behavior We obtained information about the children’s demographic information using a questionnaire that was sent to the parents via the school. The questionnaire asked for

86 

 Yow, Patrycia, and Flynn

the age and gender of the child, the child’s language exposure at home (e.g., “For a typical week, estimate how much time (%) your child hears and/or speaks”), and socioeconomic status (SES) as measured by parents’ educational status (range from 0 for no formal education to 5 for postgraduate degree). The questionnaire also contained items regarding parents’ and children’s codeswitching behavior in their communication with each other (Rodriguez-Fornells, Krämer, Lorenzo-Seva, Festman, & Münte, 2011). The first section consisted of four questions that asked parents how often they code-switch when speaking with their child or with each other in the presence of their child (e.g., “How often do you switch between languages within a sentence when speaking to your child?”) The second section consisted of three questions that asked how often parents observe their child code-switch when speaking to them (e.g., “How often does your child switch between languages within a sentence when speaking to you?”). Parents rated the frequency of their child and their own code-switching behaviors on a 5-point frequency scale (1 for never to 5 for always). A mean score of the parents’ code-switching behavior and a mean score of the child’s code-switching behavior for each child were calculated.

Peabody Picture Vocabulary Test, Fourth Edition (PPVT-IV; Dunn & Dunn, 2007) This is a norm-referenced standardized test of American English receptive vocabulary for individuals 2.5 years old or older. There was no equivalent approved Mandarin version that could be used in this study. Children were tested individually in this task. Each child was asked to select one picture from a set of four that depicts the word spoken by the experimenter. The test continued until the child made eight or more errors in any set of 12 items. Six children from School M and three children from School E did not participate in this task because of a lack of parental consent.

Measures of Children’s Spontaneous Speech The observation sessions of children’s naturalistic interactions at school were transcribed in accordance with the Codes for the Human Analysis of Transcript in the Child Language Data Exchange System (MacWhinney, 2000). The transcriptions were analyzed using the Computerized Language Analysis (CLAN; MacWhinney, 2000). One of our objectives of the study is to examine children’s code-switching behavior and its impact on language development. To this end, we computed several variables from the transcriptions that have been commonly studied in past related research: the number of different word roots, mean length of utterance, the number of code-switch utterances, and the number of pure utterances. The basic unit of our analysis of the children’s spontaneous speech is an utterance, which is defined as “a



Code-Switching in Childhood 

 87

word or group of words with a single intonation contour” (Lanza, 1992, p. 638). Routines (e.g., nursery rhymes; standardized greeting before meals and lessons; songs; games, such as “scissors-paper-stone”) are not included in all analyses. Ambiguous communicators or interjections that can used either in an English or Mandarin utterance (e.g., uh/哦, ah/啊, oh/噢), Singlish1 particles (e.g., meh, la, na, see Rubdy, 2007), and onomatopoeia (imitation of sounds, e.g., woof woof) were marked as nonwords in English and Mandarin contexts in our transcriptions. Therefore, these elements were also automatically excluded from all analyses.

Number of Different Word Roots (NDWR) The number of different word roots, also known as word types, is computed as a measure of the children’s expressive lexical diversity. Different words that are from the same word roots are considered as one word type (e.g., eat, ate, eaten). NDWR has been found to be positively associated with children’s vocabulary by other measures of vocabulary (Condouris, Meyer, & Tager-Flusberg, 2003), and it is commonly used as an indicator of children’s lexical development (Hewitt, Hammer, Yont, & Tomblin, 2005; Thordardottir, 2005; Watkins, Kelly, Harbers, & Hollis, 1995). We used the English and Mandarin lexicon database and guidelines from the CLAN website to determine what a word is in English and Mandarin. As such, words other than English and Mandarin (e.g., chaota, which means burnt in Hokkien), proper nouns (e.g., Rapunzel, which is a movie title), and unintelligible words were excluded from the computation of this measure. All words used in complete and incomplete utterances were included in this measure. To control for the differences in duration of recording among the children, we divided NDWR by each child’s duration of recording in minutes (Aukrust & Rydland, 2011).

Mean Length of Utterances (MLU) We computed MLU (ratio of morphemes over utterances; Brown, 1973) of pure English utterances and pure Mandarin utterances as a measure of general language competence in each language (Mishina-Mori, 2011). Unintelligible words in an utterance were automatically excluded from the morpheme count in CLAN, but the other intelligible words in that same utterance were included in this analysis. This is a common approach to calculate MLU in free play samples of recording, as noises from toys and the surroundings decrease the intelligibility of the children’s recorded utterances (Thordardottir, 2005). The classrooms in both schools had an “open-concept”, which resulted in the interference of the voices of children from other classroom or other groups in the same classroom. 1 Singlish, also known as Singapore Colloquial English, is a creolized form of English spoken in Singapore (Platt, 1975).

88 

 Yow, Patrycia, and Flynn

Number of Code-Switch Utterances An overall amount of code-switch utterances (intrasentential, intersentential, and interutterance switches; see the following section) was computed for each child. As children differed in how much they spoke throughout the observation period, we divided the number of code-switch utterances by the total number of utterances to obtain a percentage of code-switch utterances for each child. Intrasentential switch. An utterance is considered an intrasentential switch if it contains at least one morphological or lexical element from both languages within the same sentence or clause (Genesee, Nicoladis, & Paradis, 1995). Proper nouns in other languages, translation, and imitation were not considered as intrasentential switches in this analysis. The following are two examples of utterances that we coded as intrasentential switches: A: 我要 (I want to) test 这个 (this one). B: 他只怕被警察抓去 (He is afraid that the police will arrest him to the) zoo 给大家看或在 museum 里 (or museum to let everybody see).

Intersentential switch. Intersentential switch, which occurs outside the sentence or the clause level, is when a speaker’s utterance in one language is followed by an utterance in a different language (Genesee et al., 1995), either immediately or after a gap (pause, or other children’s conversation). We did not consider a change of language as an intersentential switch if the gap is too long (i.e., more than a 5-minute gap). Turn-taking between interlocutors was also not counted as an intersentential switch (e.g., Child A speaks Mandarin and Child B replies in English). Here is one example of an intersentential switch from a transcript: A: No, yes, my mommy said no table manners cannot sit with us. B: 你要改去坐在我们桌子上吗? (Do you want to change to sit on our table?)

Interutterance switch. An utterance is considered an interutterance switch when a pure utterance is preceded or followed by an intrasentential switch. We decided to include this category of coding because we found that the children in our study who produced intra- or intersentential switches also often switched between a pure utterance and an intrasentential switch utterance, or vice versa. No previous studies have reported such switches, but we believe it is important to count these switches to better understand the children’s full repertoire of code-switching behavior. For example, the following is a sample of intersentential switching: A: 我要 (I want) the short piece. B: I need a short piece, any color.



Code-Switching in Childhood 

 89

Number of Pure Utterances in English and Mandarin We calculated the number of pure utterances in English and Mandarin by excluding routines, single proper nouns, intrasentential switches, and utterances that contain translations and imitation of other languages. We also divided the number of pure utterances in English and Mandarin by the total number of utterances for each child to get a percentage of pure utterances for English and Mandarin, respectively.

Procedure Parents were informed about the study and asked to complete the language background questionnaire via the schools. The observation was conducted in each of the two schools for about 3 hours per day over 5 days. This was done so that children’s behavior in various activity sessions could be recorded, such as free play, mealtime, language lessons, and group project time (e.g., arts and craft, group writing activities). In each observation session, two research assistants recorded two different groups of children at the same time, each using a set of video camera and audio recorder to document the children’s self-talk and interactions with peers and teachers. At any one time, the camera captured the interactions of one small group (usually consisting of two to five children), as the children were split into small groups for most of their activities. The video recordings were used for transcribing the conversations, whereas the audio recordings were used to cross-check the transcriptions, especially in situations in which the conversations were unclear from the video recordings. At least two coders, trained on our criteria listed earlier, reviewed each section of the transcripts to ensure reliability and consistency of the coding. Finally, children were subsequently invited to participate in a one-to-one session in which we administered the PPVT task.

Results A total duration of 10:32:41 hours and 15:20:13 hours of observation in School M and School E, respectively, was transcribed and analyzed. In both schools, children were assigned into small groups facilitated by a teacher for most activities (except shower time, nap time, circle time, etc.). Both schools conducted English and Mandarin lessons every day, and children were exposed to each language about half of the time they spent in school. In spite of these similarities between the two schools, children from the two schools differed in the way they interacted with each other. In School M, children interacted with each other in English and Mandarin when they were in a group facilitated by a Mandarin teacher, and spoke only in Mandarin with their Mandarin teacher. However, in School E, children interacted with each other mostly in English even when a Mandarin teacher facilitated the group, although they, too, spoke in Mandarin with their Mandarin teacher. During mealtime

90 

 Yow, Patrycia, and Flynn

and tea time, although children from both schools spoke Mandarin with their Mandarin teacher and the school cook, School M children spoke English and Mandarin with each other, but School E children spoke only English with each other. Thus, even though both schools conducted and facilitated activities and lessons in English and Mandarin, and both groups of children were exposed to English and Mandarin equally in school, School E children conversed in English with each other most of the time.

Demographic and Language Background The demographic and language background information of the children involved in the study are summarized in Table 5.1. The parents’ average highest education attained was a college degree (a rating of 4 in the questionnaire). The children were reported to have an average exposure of 59.1% English and 37.1% Mandarin at home, with the remaining exposure in various Chinese dialects or other languages (e.g., a child was reported to have an exposure of 2% Japanese). Children from School E were reported to have more exposure to English at home than children from School M (see Table 5.1, t(29) = -2.90, p = .007; Bonferroni adjusted alpha level of .05/6 = .008 per test), but they did not differ significantly in the amount of code-switching exposure they received or engaged in at home, ps > .008. The children from both schools also did not differ in parental education (SES), p > .008. School E children were, on average, 5 months younger than School M children, and this difference was significant (t(33) = 4.14, p < .001).

Tab. 5.1: Demographic Information and Language Background

Agea Parental report of English exposure (%)a Parental report of Mandarin exposure (%) Parental report of code-switching  exposure at homeb Parental report of child’s code-switching  frequency at homec Average of parents’ education (SES)d

Overall M (SD)

School M M (SD)

School E M (SD)

6.05 (0.36) 59.10 (16.99) 37.13 (15.95) 2.43 (0.71)

6.29 (0.29) 50.36 (20.04) 44.14 (18.98) 2.68 (0.43)

5.87 (0.30) 66.29 (9.64) 31.35 (10.27) 2.30 (0.81)

2.36 (0.76)

2.72 (0.60)

2.16 (0.78)

3.97 (0.50)

4.14 (0.57)

3.82 (0.39)

Note: SES = socioeconomic status. a t-tests conducted between schools, significant at Bonferroni adjusted alpha level, p < .008. b Parents rated the frequency of their own code-switching behaviors with their children on a 5-point frequency scale (1 = never to 5 = always). cParents rated the observed frequency of their child’s code-switching behaviors on a 5-point frequency scale (1 = never to 5 = always). dSES measured by education level (0 = none or no formal education to 5 = postgraduate degree).



Code-Switching in Childhood 

 91

Children’s Spontaneous Speech An average of 525.31 utterances per child was recorded (see Table 5.2). The average number of observed switches (intrasentential, intersentential, and interutterance switches) in our sample of children constituted a small percentage of their total utterances (M = 14.98%, SD = 19.42). The children tended to produce more pure English utterances than pure Mandarin utterances or intrasentential switch utterances (M = 79.75% vs. 10.74% and 5.27%, respectively). Of the code-switched utterances, the children tended to engage in a greater amount of intrasentential switches and interutterance switches compared to intersentential switches (M = 5.27% and 5.98% vs. 3.75%, respectively). The Kolmogorov-Smirnov test of normality suggested that the data relating to the various variables in children’s spontaneous speech were not normally distributed.

Tab. 5.2: Measures of Children’s Spontaneous Speech and PPVT Standardized Score

Total number of all utterancesa Total number of all code-switched  utterances Total number of intrasentential switches Total number of intersentential switches Total number of interutterance switches Total number of pure English utterances Total number of pure Mandarin  utterances Percentage of all code-switched  utterancesb Percentage of intrasentential switchesb Percentage of intersentential switchesb Percentage of interutterance switchesb Percentage of pure English utterancesb Percentage of pure Mandarin utterancesb English NDWR (per minute)b Mandarin NDWR (per minute)b MLU pure English utterances MLU pure Mandarin utterancesb PPVT standardized score

Overall M (SD)

School M M (SD)

School E M (SD)

525.31 (427.91) 83.33 (152.31)

481.88 (354.60) 174.81 (194.58)

560.05 (484.95) 10.15 (12.49)

30.58 (62.00) 17.83 (25.61) 34.92 (68.39) 418.25 (397.16) 56.64 (110.98)

65.19 (81.57) 35.75 (29.80) 73.88 (89.22) 288.19 (226.94) 120.00 (144.81)

2.90 (3.61) 3.50 (4.51) 3.75 (5.57) 522.30 (473.57) 5.95 (6.98)

14.98 (19.42)

31.83 (18.19)

1.49 (1.74)

5.27 (7.73) 3.75 (4.48) 5.98 (8.25) 79.75 (21.63) 10.74 (15.45) 2.05 (0.76) 0.67 (0.85) 4.87 (0.65) 3.60 (2.18) 102.96 (12.82)

11.35 (8.26) 7.70 (4.03) 12.78 (8.31) 63.80 (22.88) 22.77 (16.62) 2.45 (0.88) 1.44 (0.73) 5.01 (0.58) 5.46 (1.17) 102.50 (15.35)

0.41 (0.49) 0.59 (0.73) 0.53 (0.78) 92.51 (7.96) 1.13 (1.62) 1.73 (0.45) 0.05 (0.05) 4.76 (0.70) 2.11 (1.57) 103.24 (11.58)

Note: PPVT = Peabody Picture Vocabulary Test; NDWR = number of different word roots; MLU = mean length of utterances. a The total number of all utterances is the sum of intrasentential switch utterances, pure English utterances, pure Mandarin utterances, and other utterances such as single proper nouns, translation, and imitation. Note that intersentential switch utterances comprise of only pure utterances whereas interutterance switches include pure and intrasentential switch utterances.bMann-Whitney U-tests conducted between schools, significant at Bonferroni adjusted alpha level, p < .005.

92 

 Yow, Patrycia, and Flynn

Therefore, we used Mann-Whitney U-tests to examine how School M children differ from School E children in the speech data. Code-switching behavior differed significantly between children from the two schools. Children from School M generally codeswitched more than children from School E (mean = 31.83%, range = 8.12%–60.99% vs. mean = 1.49%, range = 0%–5.93%, respectively, Z = -5.11, p < .001; Bonferroni adjusted alpha level of .05/11=.005 per test). More specifically, there were five children from School E who did not code-switch at all and seven children from School M who code-switched more than 30% of the time. Here is a sample of a transcript of two children from School M who were drawing a house together: ZHI: Anthony 我不让你, 我真的不让你住在我的房间 (I do not let you; I really do not let you stay   in my room). ANT: 我跟老师说 (I tell teacher). ANT: I can tell your mommy and your daddy. ZHI: I trick you lah, mister. ANT: Actually 这里是我们出口对吗 (Actually here is our exit, right)?

Here is a sample of a transcript of two children from School E who were playing with LEGOs: DAV: I need a short piece. MON: Where’s the bad guys? DAV: 我要 (I want) the short piece. DAV: I need a short piece, any color. DAV: Found it, I just found it.

School M children’s proportion of pure Mandarin utterances, Mandarin NDWR, and MLU of pure Mandarin utterances were also found to be significantly greater than School E children (Z = -5.04, p < .001, Z = -5.10, p < .001, and Z = -4.95, p < .001, respectively). Interestingly, though School M children showed a smaller proportion of pure English utterances, they showed a larger English NDWR than School E children (Z = 4.39, p < .001 and Z = -3.18, p < .001, respectively), but were comparable in English PPVT standardized scores (Z = -4.27, p > .10). As the children from School M were significantly older than children from School E, we ran a nonparametric analysis of covariance (Quade’s test; Quade, 1967) with the above same variables controlling for age, and we obtained the same pattern of results. It is interesting to note that although School M children had greater exposure to Mandarin, produced more Mandarin utterances, and had greater Mandarin lexical



Code-Switching in Childhood 

 93

diversity than School E children, they were actually more proficient in English than Mandarin themselves, as seen in their higher percentage of pure English utterances and number of different English word roots compared to their pure Mandarin utterances and number of different Mandarin word roots.

Correlational Analyses We conducted nonparametric partial correlational analyses controlled for age (Spearman’s rho; Bonferroni adjusted alpha level of .05/9 = .006 per test) with the various variables of interest measured in the study: percentage of code-switched utterances, English and Mandarin NDWR, MLU of pure English and Mandarin utterances, English PPVT, and home language exposure (see Table 5.3). Preliminary analyses revealed that our results did not differ between the three types of code-switch utterances (intrasentential, intersentential, and interutterance switches); therefore, we combined the three types of utterances into one (termed as code-switched utterances). We found that the amount of code-switched utterances (as a percentage of overall utterances) was positively correlated with Mandarin NDWR and MLU of pure Mandarin utterances (rs = .95, rs = .77, p < .001, respectively). Thus, children’s frequency of code-switching was positively related to their language competence in Mandarin (i.e., the number of different word roots and mean length of utterances). In addition, home exposure to Mandarin had a significant relationship with Mandarin MLU, such that the greater the exposure, the larger the MLU (rs =.53, p = .003). Similar relationships were not observed with the English measures. In addition, the amount of code-switched utterances was not significantly related to the English PPVT standard scores, p >. 10. Tab. 5.3: Spearman Correlation Table—Overall (Controlled for Age) Measures Mandarin

English

1 1. Percent of code-switched utterances 2. Mandarin NDWR (per minute) 3. MLU of pure Mandarin utterances 4. Mandarin home exposure (%)

1. Percent of code-switched utterances 2. English NDWR (per minute) 3. MLU of pure English utterances 4. English home exposure (%) 5. PPVT

.95** .77** .43*

2

.85** .46*

3

4

.53**

-

1

2

3

4

.12 -.28 -.45* -.33

.37* -.07 -.05

.21 .29

.30

Note: NDWR = number of different word roots; MLU = mean length of utterances; PPVT = Peabody Picture Vocabulary Test. *p < .05. **p < .006.

94 

 Yow, Patrycia, and Flynn

Tab. 5.4: Spearman Correlation Table for School M Measures

1

2

3

4

5

6

1. Percent of code-switched utterances 2. Mandarin NDWR (per minute) 3. MLU of pure Mandarin utterances 4. Mandarin home exposure (%) 5. English home exposure (%) 6. English NDWR (per minute) 7. MLU of pure English utterances

.87** .44+ .17 -.34 -.49+ -.19

.64** .32 -.61* -.36 -.28

.48+ -.66* -.34 -.11

-.75** -.19 -.24

-.01 .16

.61*

Note: NDWR = number of different word roots; MLU = mean length of utterances. +p < .10, *p < .05. **p < .01. Spearman Correlation Table for School E Measures

1

2

3

4

5

6

1. Percent of code-switched utterances 2. Mandarin NDWR (per minute) 3. MLU of pure Mandarin utterances 4. Mandarin home exposure (%) 5. English home exposure (%) 6. English NDWR (per minute) 7. MLU of pure English utterances

.86** .45* .28 -.13 -.31 -.42+

.70** .21 -.011 -.24 -.36

.16 .002 -.18 -.01

-.94** -.46+ -.14

.44+ .17

.26

Note: NDWR = number of different word roots; MLU = mean length of utterances. +p < .10, *p < .05. **p < .01.

Earlier we noted that children from School M tended to have larger Mandarin NDWR and MLU, and code-switched more than children from School E. Table 5.4 presents a more detailed comparison between children from the two schools. Results showed similar positive relationships between code-switching frequency, Mandarin NDWR, and MLU of pure Mandarin utterances in both groups of children. However, although School M children’s Mandarin NDWR and MLU of pure Mandarin utterances were positively related to the exposure of Mandarin at home and negatively related to the exposure of English at home, School E children’s Mandarin NDWR and MLU did not show similar relationships with language exposure at home.

Conclusion This chapter focused on code-switching in childhood. Researchers have sometimes used bilingual children’s code-switching as evidence of linguistic incompetence or language confusion. We explored why this is not the case; bilingual children, in fact, code-switch because they have the linguistic lexical competency in both languages to



Code-Switching in Childhood 

 95

do so. We reported a naturalistic study that examined the prevalence of code-switching behaviour in Mandarin–English children from two different school environments in Singapore, and explored how code-switching and social environment may be associated with bilingual children’s language development. The amount of code-switching behavior spontaneously exhibited by children in a natural setting (classroom), as well as parental reports of code-switching exposure at home, did not negatively correlate with their English receptive vocabulary, lexical diversity (number of different word roots), and linguistic complexity (mean length of utterance), suggesting that code-switching between English and another language (i.e., Mandarin) does not adversely impact children’s English receptive and expressive linguistic competency. On the contrary, there were positive correlations between the amount of code-switching and Mandarin lexical diversity and linguistic complexity. Thus, bilingual children who code-switched more demonstrated a greater level of expressive linguistic competency in Mandarin while showing no sign of a weaker level of competency in English compared to those who code-switched less, indicating that code-switching may be a mark of linguistic competency. Although increased home exposure to Mandarin may have positively impacted children’s linguistic competency in Mandarin, code-switching behavior with peers in school may play a comparable, if not more significant, role in the development of Mandarin, especially for those who have less exposure to the language at home (such as children from School E). It is plausible that code-switching may have encouraged the use of Mandarin (a language that is less frequently used at home) in a school environment with peers. In particular, for children who have less chance to use Mandarin at home, code-switching may provide opportunities for them to practice and use Mandarin in a peer-to-peer social setting, which may, in turn, promote the linguistic development of Mandarin in these children. Finally, the social environment largely influences children’s language behavior. In the context of a multilingual population where English is the shared language and code-switching is a common practice, differences in code-switching behavior between children from two school environments were observed, such that children from one school tended to converse more in pure Mandarin utterances, had richer Mandarin lexicons, produced more complex Mandarin utterances, and code-switched more than children from another school who predominantly converse in English among themselves. This suggests that the children’s immediate language and social environment has a strong influence on their code-switching behavior with other children and the development of the languages (particularly the language that is typically used less at home). We speculate that the same study with different language combinations (e.g., English and Spanish, and others as discussed in this volume) would yield similar results when factors such as SES, language status, and so forth, are similarly controlled for. This is because what we have isolated here reflects a more general phenomenon related to the requirements of code-switching. Just as code-switching among children in Singapore is a sign of growing linguistic competence in Mandarin

96 

 Yow, Patrycia, and Flynn

(the language they use less at home), code-switching among minority children in the United States is also a sign of growing linguistic competence in English (the language they use less at home). In sum, there is no evidence that code-switching negatively impacts English linguistic competency in bilingual children. Instead, children’s code-switching frequency at school is strongly related to their linguistic competency in Mandarin, a language that is less frequently used at home. The immediate language environment (home and school) played a significant role in the bilingual children’s language behavior and development.

References Anderson, R. (2004). First language loss in Spanish-speaking children, patterns of loss and implications for clinical practice. In B. Goldstein (Ed.), Bilingual language development & disorders in Spanish-English speakers (pp. 187–212). Baltimore, MD: Brookes. Aukrust, V. G., & Rydland, V. (2011). Preschool classroom conversations as long-term resources for second language and literacy acquisition. Journal of Applied Developmental Psychology, 32, 198–207. http://dx.doi.org/10.1016/j.appdev.2011.01.002 Baetens Beardsmore, H. (1998). Language shift and cultural implications in Singapore. In S. Gopinathan, H. W. Kam, A. Pakir, & V. Saravanan (Eds.), Language, education and society in Singapore—Issues and trends (2nd ed., pp. 85–98). Singapore: Times Academic Press. Bokamba, E. (1989). Are there syntactic constraints on code-mixing? World Englishes, 8, 277–292. http://dx.doi.org/10.1111/j.1467-971X.1989.tb00669.x Brice, A., & Anderson, R. (1999). Code mixing in a young bilingual child. Communication Disorders Quarterly, 21, 17–22. http://dx.doi.org/10.1177/152574019902100103 Brown, R. (1973). A first language: The early stages. Cambridge, MA: Harvard University Press. http://dx.doi.org/10.4159/harvard.9780674732469 Cantone, K. F. (2007). Code-switching in bilingual children. Dordrecht, the Netherlands: Springer. Clyne, M. (1987). Constraints on code switching: How universal are they? Linguistics, 25, 739–764. http://dx.doi.org/10.1515/ling.1987.25.4.739 Comeau, L., Genesee, F., & Lapaquette, L. (2003). The modeling hypothesis and child bilingual codemixing. The International Journal of Bilingualism, 7, 113–126. http://dx.doi.org/10.1177/ 13670069030070020101 Condouris, K., Meyer, E., & Tager-Flusberg, H. (2003). The relationship between standardized measures of language and measures of spontaneous speech in children with autism. American Journal of Speech-Language Pathology, 12, 349–358. http://dx.doi.org/10.1044/ 1058-0360(2003/080) Dunn, L. M., & Dunn, D. M. (2007). Peabody Picture Vocabulary Test (4th ed.). Minneapolis, MN: NCS Pearson. Faltis, C. (1989). Code-switching and bilingual schooling: An examination of Jacobson’s new concurrent approach. Journal of Multilingual and Multicultural Development, 10, 117–127. http:// dx.doi.org/10.1080/01434632.1989.9994368 Genesee, F. (1989). Early bilingual development: One language or two? Journal of Child Language, 16, 161–179. http://dx.doi.org/10.1017/S0305000900013490



Code-Switching in Childhood 

 97

Genesee, F. (2002). Portrait of the bilingual child. In V. Cook (Ed.), Portrait of the L2 user (pp. 170–196). New York, NY: Multilingual Matters. Genesee, F., Nicoladis, E., & Paradis, J. (1995). Language differentiation in early bilingual development. Journal of Child Language, 22, 611–631. http://dx.doi.org/10.1017/S0305000900009971 Genesee, F., Paradis, J., & Crago, M. (2004). Dual language development and disorders: A handbook on bilingualism and second language learning. Baltimore, MD: Brookes. Goh, H. H., Zhao, C., & Liu, Y. (2007). Code-switching in the classroom: A taboo norm. SingTeach, 9. Retrieved from http://singteach.nie.edu.sg/issue09-ideas01/ Goodz, N. S. (1989). Parental language mixing in bilingual families. Infant Mental Health Journal, 10, 25–44. http://dx.doi.org/10.1002/1097-0355(198921)10:13.0.CO;2-R Grosjean, F. (1982). Life with two languages: An introduction to bilingualism. Cambridge, MA: Harvard University Press. Guiberson, M., Barrett, K. C., Jancosek, E. G., & Yoshinaga-Itano, C. (2006). Language maintenance and loss in young preschool age children of Mexican immigrants: Longitudinal study. Communication Disorders Quarterly, 28, 4–17. http://dx.doi.org/10.1177/15257401060280010601 Gumperz, J. J. (1971). Bilingualism, bidialectalism and classroom interaction. In A. S. Dil (Ed.), Language in social groups (pp. 311–339). Stanford, CA: Stanford University Press. Gumperz, J. J. (1982). Discourse strategies. New York, NY: Cambridge University Press. http://dx.doi.org/ 10.1017/CBO9780511611834 Heller, M. (1985). Ethnic relations and language use in Montréal. In N. Wolfson & J. Manes (Eds.), Language of inequality (pp. 75–90). The Hague, the Netherlands: Mouton. http://dx.doi.org/ 10.1515/9783110857320.75 Hewitt, L. E., Hammer, C. S., Yont, K. M., & Tomblin, J. B. (2005). Language sampling for kindergarten children with and without SLI: Mean length of utterance, IPSYN, and NDW. Journal of Communication Disorders, 38, 197–213. http://dx.doi.org/10.1016/j.jcomdis.2004.10.002 Kaufman, D., & Aronoff, M. (1991). Morphological disintegration and reconstruction in first language attrition. In H. W. Seliger & R. M. Vago (Eds.), First language attrition (pp. 175–188). Cambridge, England: Cambridge University Press. http://dx.doi.org/10.1017/ CBO9780511620720.012 King, K. A., & Fogle, L. (2006). Raising bilingual children: Common parental beliefs and current research. Washington, DC: Center for Applied Linguistics. Labov, W. (1972). Some principles of linguistic methodology. Language in Society, 1, 97–120. http:// dx.doi.org/10.1017/S0047404500006576 Lanza, E. (1992). Can bilingual two-year-olds code-switch? Journal of Child Language, 19, 633–658. http://dx.doi.org/10.1017/S0305000900011600 Lindholm, K. J., & Padilla, A. M. (1978). Language mixing in bilingual children. Journal of Child Language, 5, 327–335. http://dx.doi.org/10.1017/S0305000900007509 MacSwan, J. (1999). A minimalist approach to intrasentential code switching. New York, NY: Garland. MacSwan, J. (2013). Code switching and linguistic theory. In T. K. Bhatia & W. Ritchie (Eds.), Handbook of bilingualism and multilingualism (2nd ed., pp. 223–350). Oxford, England: Blackwell. MacSwan, J. (Ed.). (2014). Grammatical theory and bilingual codeswitching. Cambridge, MA: MIT Press. MacSwan, J., & McAlister, K. T. (2010). Naturalistic and elicited data in grammatical studies of codeswitching. Studies in Hispanic and Lusophone Linguistics, 3, 521–532. http://dx.doi.org/ 10.1515/shll-2010-1085

98 

 Yow, Patrycia, and Flynn

MacWhinney, B. (2000). The CHILDES Project: Tools for analyzing talk. Mahwah, NJ: Erlbaum. McClure, E. (1977). Aspects of code-switching in the discourse of bilingual Mexican-American children. Washington, DC: Georgetown University Roundtable on Languages and Linguistics. Meisel, J. M. (1989). Early differentiation of languages in bilingual children. In K. Hyltenstam & L. Obler (Eds.), Bilingualism across the lifespan: Aspects of acquisition, maturity, and loss (pp. 13–40). Cambridge, England: Cambridge University Press. http://dx.doi.org/10.1017/ CBO9780511611780.003 Meisel, J. M. (1994). Code switching in young bilingual children: The acquisition of grammatical constraints. Studies in Second Language Acquisition, 16, 413–439. http://dx.doi.org/ 10.1017/S0272263100013449 Milroy, L., & Muysken, P. (1995). Introduction: Code-switching and bilingualism research. In L. Milroy & P. Muysken (Eds.), One speaker, two languages: Cross-disciplinary perspectives on code-switching (pp. 1–14). Cambridge, England: Cambridge University Press. http://dx.doi.org/ 10.1017/CBO9780511620867.001 Mishina-Mori, S. (2011). A longitudinal analysis of language choice in bilingual children: The role of parental input and interaction. Journal of Pragmatics, 43, 3122–3138. http://dx.doi.org/ 10.1016/j.pragma.2011.03.016 Moore, S. M., Beatty, J., & Perez-Mendez, C. (1995). Developing cultural competence in early childhood assessment. Boulder: University of Colorado Press. Muysken, P. (2000). Bilingual speech: A typology of code-mixing. Cambridge, England: Cambridge University Press. Myers-Scotton, C. (1993). Common and uncommon ground: Social and structural factors in codeswitching. Language in Society, 22, 475–503. http://dx.doi.org/10.1017/ S0047404500017449 Nicoladis, E. (1995). Code-mixing in young bilingual children (Unpublished doctoral dissertation). McGill University, Montreal, Quebec, Canada. Nicoladis, E., & Genesee, F. (1996). A longitudinal study of pragmatic differentiation in young bilingual children. Language Learning, 46, 439–464. http://dx.doi.org/10.1111/ j.1467-1770.1996.tb01243.x Nicoladis, E., & Genesee, F. (1997). Language development in preschool bilingual children. Journal of Speech-Language Pathology and Audiology, 21, 258–270. Ong, K. K. W., & Zhang, L. J. (2010). Metalinguistic filters within the bilingual language faculty: A study of young English–Chinese bilinguals. Journal of Psycholinguistic Research, 39, 243–272. http://dx.doi.org/10.1007/s10936-009-9137-z Pakir, A. (1989). Linguistic alternants and code selection in Baba Malay. World Englishes, 8, 379–388. http://dx.doi.org/10.1111/j.1467-971X.1989.tb00676.x Paradis, J., Nicoladis, E., & Genesee, F. (2000). Early emergence of structural constraints on codemixing: Evidence from French-English bilingual children. Bilingualism: Language and Cognition, 3, 245–261. http://dx.doi.org/10.1017/S1366728900000365 Perani, D., Abutalebi, J., Paulesu, E., Brambati, S., Scifo, P., Cappa, S. F., & Fazio, F. (2003). The role of age of acquisition and language usage in early, high-proficient bilinguals: An fMRI study during verbal fluency. Human Brain Mapping, 19, 170–182. http://dx.doi.org/10.1002/hbm.10110 Platt, J. T. (1975). The Singapore English speech continuum and its basilect “Singlish” as a “Creoloid”. Anthropological Linguistics, 17, 363–374.



Code-Switching in Childhood 

 99

Poplack, S. (1980). “Sometimes I’ll start a sentence in Spanish y termino el espaniol”: Toward a typology of code-switching. Linguistics, 18, 581–618. http://dx.doi.org/10.1515/ ling.1980.18.7-8.581 Poplack, S. (2001). Code-switching (linguistic). In N. J. Smelser & P. B. Baltes (Eds.), International encyclopedia of the social and behavioral sciences (pp. 2062–2065). London, England: Elsevier Science. http://dx.doi.org/10.1016/B0-08-043076-7/03031-X Poplack, S., & Sankoff, D. (1988). Code-switching. In U. Ammon, N. Dittmar, & K. J. Mattheier (Eds.), Sociolinguistics: An international handbook of the science of language and society (Vol. 2, pp. 1174–1180). Berlin, Germany: Walter de Gruyter. Portocarrero, J. S., Burright, R. G., & Donovick, P. J. (2007). Vocabulary and verbal fluency of bilingual and monolingual college students. Archives of Clinical Neuropsychology, 22, 415–422. http:// dx.doi.org/10.1016/j.acn.2007.01.015 Quade, D. (1967). Rank analysis of covariance. Journal of the American Statistical Association, 62, 1187–1200. http://dx.doi.org/10.1080/01621459.1967.10500925 Rodriguez-Fornells, A., Krämer, U. M., Lorenzo-Seva, U., Festman, J., & Münte, T. F. (2011). Selfassessment of individual differences in language switching. Frontiers in Psychology, 2, 388. Ruan, J. (2003). A study of bilingual Chinese/English children’s code switching behavior. Academic Exchange Quarterly, 7, 142–145. Rubdy, R. (2007). Singlish in school: Impediment or a resource? Journal of Multilingual and Multicultural Development, 28, 308–324. Saunders, G. (1988). Bilingual children: From birth to teens. Clevedon, England: Multilingual Matters. Schiff-Myers, N. B. (1992). Considering arrested language development and language loss in the assessment of second language learners. Language, Speech, and Hearing Services in Schools, 23, 28–53. http://dx.doi.org/10.1044/0161-1461.2301.28 Sharwood Smith, M., & Van Buren, P. (1991). First language attrition and the parameter setting model. In H. W. Seliger & Robert M. Vago (Eds.), First language attrition (pp. 17–30). Cambridge, England: Cambridge University Press. Singapore Department of Statistics (DOS). (2010). Singapore census of population 2010: Statistical release 1—Demographic characteristics, education, language and religion. Retrieved from http://www.singstat.gov.sg/publications/publications-and-papers/cop2010/cop2010-sr1 Singapore Department of Statistics (DOS). (2012). Population trends, 2013. Retrieved from http:// www.singstat.gov.sg/publications/publications-and-papers/population-and-populationstructure/population-trends Tabouret-Keller, A. (1963). L’acquisition du langage parlé chez un petit enfant en milieu bilingue [The acquisition of spoken language in bilingual infants]. In J. de Ajuriaguerra, F. Bresson, P. Fraisse, B. Inhelder, P. Oleron, & J. Piaget (Eds.), Problèmes de psycholinguistique (pp. 205–219). Paris, France: Presses Universitaires de France. Thordardottir, E. T. (2005). Early lexical and syntactic development in Quebec French and English: Implications for cross-linguistic and bilingual assessment. International Journal of Language & Communication Disorders, 40, 243–278. http://dx.doi.org/10.1080/ 13682820410001729655 Valdes-Fallis, G. (1978). Code-switching as a deliberate verbal strategy: A micro-analysis of direct and indirect requests among bilingual Chicano speakers. In R. Duran (Ed.), Latino Language and Communicative Behavior (pp. 95–108). Norwood, NJ: Ablex.

100 

 Yow, Patrycia, and Flynn

van Gelderen, E., & MacSwan, J. (2008). Interface conditions and code-switching: Pronouns, lexical DPs, and checking theory. Lingua, 118, 765–776. http://dx.doi.org/10.1016/ j.lingua.2007.05.003 Watkins, R. V., Kelly, D. J., Harbers, H. M., & Hollis, W. (1995). Measuring children’s lexical diversity: Differentiating typical and impaired language learners. Journal of Speech & Hearing Research, 38, 1349–1355. http://dx.doi.org/10.1044/jshr.3806.1349

II Factors Affecting Bilingualism Across the Lifespan

Sharon Unsworth

6 Quantity and Quality of Language Input in Bilingual Language Development One of the defining characteristics of bilingual language development is variability, in terms of individual children’s outcomes in their two (or more) languages and in terms of children’s language learning experience. While individual variation also exists in monolingual language development, and many of the factors contributing to experi­ ential variation in bilingual contexts, such as complexity of maternal speech, are also at play in monolingual contexts, there also exist characteristics that are more typical of bilingual language experience. These include exposure to input from nonnative speak­ ers, the variety of speakers providing input in a given language, and the existence of multilingual utterances in the input. A growing body of research from the fields of psychology, linguistics, and education seeks to address the extent to which such expe­ riential factors relate to bilingual children’s developing language skills. The purpose of this chapter is to provide an overview of the burgeoning literature examining the role of input quantity and quality on bilingual language development, with an emphasis on the most recent findings. In the present context, bilingual language development is understood in a broad sense, and the studies under review here examine language acquisition in simultaneous and successive (or sequential) bilingual speak­ ers. (For a recent review of the sparse literature on input effects in trilingual acquisition, see Unsworth, 2013b.) Almost all studies concern children between the ages of 2 and 10 years. The chapter is organized as follows. The first part considers different sources of input in the bilingual child’s language environment (following Gathercole, 2014). This section also considers the role of the child’s own language output. The second part of the chapter reviews studies focusing on and comparing specific linguistic domains, most usually vocabulary and morphosyntax. Finally, implications of some of these findings are considered for theories of language acquisition and for parenting and education.

Variability in Bilingual Children’s Language Experience: Sources of Input Children growing up with more than one language usually obtain their input for each language from different sources. They may have relatively balanced exposure, or this may be skewed in favor of one language over the other. In comparison with mono­ lingual children, bilingual children are generally assumed to hear comparatively less http://dx.doi.org/10.1037/14939-007 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

104 

 Sharon Unsworth

input per language (cf. De Houwer, 2014). There may be multiple sources of input for a language, or this may be restricted to just one person or place. This section reviews the extant literature for a range of these sources, including home and parents, sib­ lings and peers, native speakers, and preschool and day care.

Home and Parents Most studies examining the role of input on bilingual language development focus on children not yet in school and for whom home is the most important source of lan­ guage input. Typically, researchers make use of relative measures (e.g., 70% Spanish– 30% English), on the basis of parental report. These are either direct estimates made by the parents or indirect estimates calculated by the researcher on the basis of detailed parental questionnaires. Bilingual children are by definition exposed to a language other than the societal (or majority) language at home; how much children heard in each depends in part of the specific family constellation (see Chapter 2, this volume). The impact of home language use on children’s developing language skills has been examined in numerous studies. For example, work by Paradis and colleagues on French–English bilingual children has shown that variation in home language input/use affects children’s rate of acquisition of vocabulary and morphosyntax (Paradis, Nicoladis, Crago, & Genesee, 2011). Similar findings have been observed for Welsh–English bilingual children growing up in Wales and Spanish–English bilingual children in the United States (Gathercole & Thomas, 2009; Oller & Eilers, 2002). There are also studies where amount of language input at home in the societal language was not found to relate to children’s language out­ comes, most likely because the parents’ proficiency level in that language was so low (Chondrogianni & Marinis, 2011; Golberg, Paradis, & Crago, 2008). Interestingly, the impact of amount of societal language use at home may be greater for children’s development in the minority language. Hoff, Rumiche, Burridge, Ribot, and Welsh (2014) found that in a group of bilingual Spanish–English toddlers growing up in the United States, English use at home was a positive predictor of English expres­ sive vocabulary skills for children growing up in one parent, one language families, whereas it was a weak and nonsignificant predictor of Spanish vocabulary scores. In contrast, in families with two Spanish-speaking parents, English use at home was a weak and nonsignificant predictor for English vocabulary but a strong negative predic­ tor for Spanish vocabulary, in line with earlier findings by Hammer, Davison, Lawrence, and Miccio (2009). In a series of studies on bilingual Welsh–English children and adults, Gathercole and colleagues have repeatedly shown the impact of minority language use at home on the rate and end state of acquisition (e.g., Gathercole & Thomas, 2009; Thomas, Williams, Jones, Davies, & Binks, 2014). In a large-scale survey of family language use in Belgium, De Houwer (2007) found that children were most likely to speak the



Quantity and Quality of Language Input 

 105

minority and the societal language when both parents spoke the minority language and at most one parent spoke the societal language at home. In other words, in fam­ ilies where both parents spoke the societal language as well as (or instead of) the home language, children were significantly less likely to speak the minority language. The positive contribution of parental use of the minority language at home to chil­ dren’s development in that language has been observed elsewhere, and this seems especially important once children start schooling in the majority language (Dixon, Zhao, Quiroz, & Shin, 2012; Duursma et al., 2007; Willard, Agache, Jäkel, Glück, & Leyendecker, 2015). It may not, however, necessarily guarantee children’s continuing minority language development (Sheng, Lu, & Kan, 2011); other social, linguistic, and educational factors may be more relevant.

Siblings and Peers Parents are, of course, not the only source of children’s language input at home. Anec­ dotally, older siblings in families in which both parents speak the minority language are often reported as the source of increased societal language use within the home. There is to date little systematic research addressing the influence of siblings on bilingual children’s language development (but see Barron-Hauwaert, 2011). In one of the few studies on the topic, Bridges and Hoff (2014) found that Spanish–English bilingual tod­ dlers growing up in the United States with older siblings attending English-language school did indeed hear more English input at home—from their older siblings and from their mothers—than children without school-age siblings. This, in turn, was related to children’s reported vocabulary and grammatical complexity scores, with toddlers with school-age siblings showing more advanced development in English than toddlers without school-age siblings, and the reverse pattern holding for Spanish. These findings suggest that siblings should not only be seen as sources of language input themselves but also as potential agents of change in the language use of other family members. Input from other children is also available from peers in the form of friends and classmates. The limited research available on this topic suggests that peers may be a useful and important source of input in the home and societal language (e.g., Jia & Fuse, 2007). In the school context, Palermo et al. (2014) found frequency of peer inter­ action to be a better predictor of Spanish-speaking preschoolers’ receptive and expres­ sive vocabulary in their second language (L2) English than the frequency of teacher interaction.

Native Speakers The discussion thus far has illustrated the clear relationship between the amount of input to which bilingual children are exposed and their rate of acquisition (at least in

106 

 Sharon Unsworth

certain linguistic domains—see the following sections for further discussion). Unlike most monolingual children, bilingual children may hear input from native and non­ native speakers, who in turn may vary in their level of proficiency (Fernald, 2006). Whereas in the aforementioned studies no relation was found between input quantity and language outcomes (e.g., Paradis, 2011), it was suggested that the low proficiency of the input-providers may be responsible. Similarly, exposure to nonnative input has also been put forward as an explanation for bilingual children’s apparent fossilization in the acquisition of grammatical gender in Dutch (Cornips & Hulk, 2008). That non­ native speakers may be a less effective source of language input than native speakers finds support in a study by Place and Hoff (2011; see also Driessen, van der Slik, & De Bot, 2002), who observed proportion of native input to be a significant predictor of bilingual Spanish–English toddlers’ vocabulary, even after controlling for input quan­ tity (see also Driessen et al., 2002). As the authors noted, their data indicated that non­ native speech may not support language acquisition as effectively as native speech, but why this should be the case remains unclear. Furthermore, the more general ques­ tions of whether native input should always be preferred over nonnative input, and at what level of proficiency nonnative speakers should be encouraged to provide input in a minority language, remain unanswered. This is clearly an area for future research.

Variety and Richness Place and Hoff (2011) found that the number of different speakers providing input in English was also a significant predictor of children’s vocabulary scores in that language. More specifically, following work on early phonological and lexical learn­ ing, the authors suggested that exposure to a variety of sources may be necessary “to extract the categories that will support later recognition and production” (Place & Hoff, 2011, p. 1847). It is also possible that an increase in the number of different people providing input in a given language may simply increase the variation in lexical items to which the child is exposed, especially if associated with different con­ texts, and this may be (part of) what contributes to higher vocabulary scores. Work by Jia and colleagues (e.g., Jia & Fuse, 2007) on the L2 English of Chinese and Korean children in the United States incorporated what is essentially a measure of variety (e.g., number of predominantly L2-speaking friends) in their “richness” vari­ able. Other elements included hours of TV viewing in the L2, number of books read in the L2, and the extent to which the L2 was spoken at home (see also Paradis, 2011).

School, Preschool, and Day Care The discussion thus far has almost exclusively focused on language input within the home. School, and for younger children, preschool or day care, is of course another important source of language input. In fact, for children from homes where only



Quantity and Quality of Language Input 

 107

the minority language is spoken, attendance at some kind of child care or at school may be the main or perhaps even the only contact with the societal language. In the context of bilingual education, school may also be an important source of a different, more academic type of input in the minority language, as well as giving the child the opportunity to meet same-language peers (see Chapter 11, this volume). The focus here lies on the role of language input provided by teachers to children learning the school language as a second language. A limited number of studies have examined the extent to which specific linguis­ tic properties of teachers’ speech relate to monolingual children’s developing school language skills (e.g., Huttenlocher, Vasilyeva, Cymerman, & Levine, 2002), but there is hardly any research on this question for bilingual children. One exception is a study on the vocabulary growth in preschool children learning English as L2 in a mono­ lingual classroom by Bowers and Vasilyeva (2011). They found that the overall amount of speech produced by teachers was a significant predictor of growth in children’s Peabody Picture Vocabulary Test scores within one school year. The lexical diversity of teacher talk was not related to the children’s scores, whereas their mean length utterances (MLUs) showed a negative relation whereby an increase in teachers’ number of words per utterance corresponded to a decrease in children’s vocabulary growth. Bowers and Vasilyeva explained these results as follows. The positive rela­ tion between children’s scores and total amount of speech is in-line with previous research (e.g., Huttenlocher et al., 2002); the negative relation with MLU may reflect the fact that even though these 4- to 5-year-old children were likely able to hold in memory the longer and complex sentences produced by teachers, these utterances were at this stage of their L2 development still too difficult to deconstruct, and this may have complicated the word learning process. Finally, the failure to find any relation with lexical diversity may also relate to the relatively low proficiency of these L2 chil­ dren: the teacher’s use of more sophisticated vocabulary was likely too advanced and hence unlikely to facilitate lexical growth. In a similar study, Gámez and Levine (2013) examined the impact of the quality of language input provided by teachers in transitional bilingual Spanish–English kindergarten classes on children’s first language (L1) Spanish. The results showed that children’s gains in expressive language were related to the rate of complexity (i.e., the proportion of Spanish language utterances with more than one clause divided by total number of Spanish language utterances) and number of unique word types in teachers’ input. Importantly, this observation held after controlling for total amount of teacher talk and other interclass differences; in addition, there was no relation between child’s level at start of year and measures of teacher input, providing evi­ dence for the independent effect of teacher input throughout the year. The authors speculated that the difference between their and Bowers and Vasilyeva’s (2011) results with respect to the effect of structural complexity may be because of the different way in which each study operationalized this variable (i.e., the use of multiclause utter­ ances vs. MLU).

108 

 Sharon Unsworth

Output and Other Factors One final factor relating to variability in bilingual children’s language experience and the amount of exposure they may hear is their own language output. Whereas input and output are often combined into one language use score (e.g., Bedore et al., 2012), some recent studies have suggested that output may be a significant predictor of children’s outcomes in its own right. For example, in a study on the early semantic and morphosyntactic development of Spanish–English bilingual speakers, Bohman, Bedore, Peña, Mendez-Perez, and Gillam (2010) found that output is a significant pre­ dictor of both domains in both languages, whereas input was only relevant for both domains in English. The authors suggested that “using a language (i.e., output) forces the learner to process the language in a way that only hearing it (i.e., input) does not” (Bohman et al., 2010, p. 339; see also Sheng et al., 2011, for differential effects of output across languages and age groups). In a study on the acquisition of verbal morphology and vocabulary in Dutch–English bilingual children in the Netherlands between the ages of 3 and 17 years, Unsworth (2015) also observed that output was a better predictor of children’s scores on the minority language, English in this case, which is in line with earlier findings for heritage language development (Montrul, 2008). Output/use may also be important simply because it invites more input (as in the input–proficiency–use cycle proposed by Pearson, 2007). A recent study in this context is that of Ribot and Hoff (2014), who showed that amount of input at home was related to children’s code-switching behavior. Specifically, Spanish–English toddlers (aged 2.5 years) who switched to English when speaking Spanish heard more English input at home than children who switched in the other direction or children who switched infrequently. The reverse was also found to hold. Additionally, children with no asym­ metry in code-switching behavior had more balanced input. In short, then, emerging evidence suggests that how much bilingual children use a language may contribute to their overall development in that language in a different way than how much exposure they get, although why exactly this should be so is not yet completely understood. There are various other factors that may affect the quantity and the quality of the input to which bilingual children are exposed and that have been claimed to predict children’s language outcomes. These include factors that are also relevant to mono­ lingual acquisition, such as socioeconomic status, usually indexed by maternal edu­ cation (Hoff, 2006), as well as factors that are characteristic of the bilingual language experience, including the number of input-providers with whom the child exclusively speaks the language in question (Place & Hoff, 2011) and the amount of mixing in the input (Byers-Heinlein, 2013).

Summary Most studies investigating the role of input in bilingual language development have focused on the effect of varying amounts of exposure within the home and have



Quantity and Quality of Language Input 

 109

shown that more input generally leads to quicker rates of acquisition. These effects may however be moderated by the proficiency level of the input-providers and they may differ for the bilingual speaker’s two languages. Other factors shown to affect bilingual children’s language experience and consequently influence development include children’s language use; the variety of input sources available to them; and the existence of older, school-age siblings.

Variability in Bilingual Children’s Language Outcomes: Input Effects Across Linguistic Domains Children growing up with more than one language vary considerably in their language outcomes (see Chapter 2, this volume). The literature reviewed in the first part of this chapter suggests that at least in part, this is likely because of certain characteristics of their language experience which, in turn, lead to variation in quantity and the quality of the input to which they are exposed. In the second part of this chapter examines the relationship between input quantity and quality and language outcomes relate more closely, considering the extent to which input effects holds across different lin­ guistic domains.

Vocabulary One of the most robust findings concerning differential rates of acquisition as a con­ sequence of variability in bilingual children’s language experience is for vocabulary development. The general observation is as follows: When both languages are taken into account, young bilingual children’s total vocabularies are comparable to those of their monolingual peers, but when only one language is considered, bilingual children typically show a slower rate of development than children acquiring the same lan­ guage as their only language (e.g., Marchman, Fernald, & Hurtado, 2010; Patterson, 2004; Pearson, Fernández, & Oller, 1993). This slower single language growth has been related to the distributed characteristic of bilingual knowledge, or the fact that bilinguals often acquire and use their languages in different contexts means that their vocabulary is distributed across two languages (Oller, Pearson, & Cobo-Lewis, 2007). Evidence for this claim comes from a large-scale study by Bialystok, Luk, Peets, and Yang (2010) employing the widely used Peabody Picture Vocabulary Test (Dunn & Dunn, 2007). Whilst persistent bilingual– monolingual differences were observed across a wide age range, separating the words into those related to home and those related to school revealed a striking difference: the significant bilingual–monolingual difference persisted for home words, whereas on school words, the two groups were much more comparable.

110 

 Sharon Unsworth

Recent studies, however, have challenged the claim that monolingual speakers consistently outperform bilingual speakers in their single language vocabulary devel­ opment (De Houwer, Bornstein, & Putnick, 2014). Smithson, Paradis, and Nicoladis (2014) suggested that the more positive wider sociocultural context and higher status of the home language may account for the differences in their findings for French– English bilinguals in (a French minority context in) Canada when compared with earlier findings of other bilingual populations. One further factor that may also con­ tribute to higher vocabulary scores for certain language combination is the existence of cognates (Stadthagen-González, Gathercole, Pérez-Tattam, & Yavas, 2013). Bilingual children’s rate of vocabulary acquisition is also strongly influenced by variation in amount of exposure (e.g., Scheele et al., 2010). This relationship does not appear to be direct, however. Rather, it has been suggested that once a certain thresh­ old has been reached, more input is not required (Gathercole, 2007). For example, in a study on the vocabulary development of bilingual French–English preschoolers in Montréal, Canada, Thordardottir (2011) observed that on the group level, 40% to 60% exposure was enough for the bilingual children to attain comparable scores to their monolingual peers on receptive vocabulary. Interestingly, for productive vocabulary, the threshold was higher, at more than 60%. These findings have since been replicated for younger children growing up bilingually in the United Kingdom (Cattani et al., 2014; see also Hoff et al., 2012, for comparable findings). The studies reviewed thus far investigated how differences in input quantity and quality relate to offline measures of bilingual children’s vocabulary development. In a series of recent studies, researchers at Stanford University have explored whether dif­ ferential input also affects bilingual children’s language processing skills. First, in a study on 30-month-old simultaneous bilingual Spanish–English children, Marchman, Fernald, and Hurtado (2010) found that processing efficiency, as measured by reaction times in a looking-while-listening procedure, was related to vocabulary size, thereby replicating an earlier finding for monolingual children. Subsequently, Hurtado, Grüter, Marchman, and Fernald (2014) examined the relation between relative amount of expo­ sure to relative vocabulary size and relative processing efficiency in 30- and 36-monthyear old bilingual children from the same population. The results showed a signifi­ cant relation between amount of exposure and vocabulary size, on the one hand, and between amount of exposure and processing speed, on the other. Furthermore, process­ ing speed and amount of exposure at 30 months each predicted unique variance in chil­ dren’s vocabulary scores at 36 months, suggesting that “increased processing efficiency enables children to more effectively take advantage of whatever amount of Spanish they hear to gain expressive vocabulary knowledge in Spanish” (Hurtado et al., 2014, p. 199).

Grammar and Morphosyntax Properties of bilingual children’s bilingual experience, and in particular, differential input, have been related to their rate of acquisition of a range of morphosyntactic



Quantity and Quality of Language Input 

 111

phenomena, including verbal morphology (Blom, 2010; Blom, Paradis, & Duncan, 2012; Nicoladis, Palmer, & Marentette, 2007; Paradis et al., 2011), nominal morphology (Thomas et al., 2014), and grammatical gender (Gathercole & Thomas, 2009; Montrul & Potowski, 2007; Unsworth, 2013a), as well as more comprehensive assessments of children’s grammatical abilities (Chondrogianni & Marinis, 2011; Hoff et al., 2012). Similar to the findings for vocabulary reported above, Thordardottir (2015) found evidence for a nonlinear relationship between amount of exposure and the acquisi­ tion of morphosyntax, that is, children with approximately 50% exposure were indis­ tinguishable from their monolingual peers. Not all measures included in the study were equally sensitive to input effects, however; the diversity of morphemes which children used correctly was particularly sensitive to differences in amount of exposure. Further­ more, input effects differed between languages. Whereas MLU was significantly asso­ ciated with amount of input for English, this was not the case for French. Similar non­ linear patterns of association between input quantity and morphosyntactic outcomes have been found for Spanish–English bilingual children (e.g., Bedore et al., 2012) and for Dutch–English bilingual children (Unsworth, 2015). In this latter study, on pat­ terns of language dominance, it was observed that children classified as being Dutchdominant, where dominance was operationalized in terms of MLU and number of dif­ ferent verbs, all had at least 65% exposure to that language, whereas exposure levels for balanced children were significantly lower. Whereas some studies show that—given sufficient input—bilingual children perform comparably with their monolingual peers, others suggest that for certain linguistic properties, or in circumstances of limited input, bilingual children may fail to catch up. For example, Thomas et al. (2014) found that in their acquisition of plural mor­ phology in Welsh, children from English-only homes were still struggling with most plural forms at age 11, whereas children from Welsh-only homes were approaching adult norms on some of these at the same age. In fact, even adults who had grown up with Welsh and English from birth were not as accurate in their production of plural morphology as adults who had grown up with Welsh only at home and who had acquired English at a later age. As Thomas noted, these findings suggest that for complex phenomena, such as Welsh plural morphology, exposure from birth seems to be insufficient for complete acquisition to take place as the effects of reduced input appear to be long lasting (see also Gathercole & Thomas, 2009). The finding that variation in amount of exposure may affect the acquisition of linguistic properties differently as result of the level of opacity or complexity of that property is in line with other work (Blom et al., 2012; Gathercole, 2002; Paradis, 2010). The claim here is that, as in monolingual acquisition, bilingual acquisition is affected by factors such as type/token frequency and the transparency of form-to-function mappings, but as a result of having to split their time across two languages, the impact of some of these factors may be greater for bilingual speakers than mono­ lingual speakers, at least in the early stages of development. It is also possible that differences in input quantity may affect aspects of the same target language property differently. In a study on the acquisition of grammatical gender

112 

 Sharon Unsworth

by simultaneous bilingual Dutch–English children, Unsworth (2013a) observed input effects for gender-marking on definite determiners, but for gender-marking on adjec­ tives such effects were absent. This was expected for the following reasons. In Dutch, which has a two-way gender system, distinguishing between common and neuter, the form of the definite determiner is the main source/marking of gender information for any given noun; furthermore, properties of the Dutch grammatical gender system con­ spire to make the gender of neuter nouns difficult to detect. Therefore, input effects are to be expected. Gender-marking on adjectives, on the other hand, involves the applica­ tion of a rule. Whereas a certain amount of input will of course be necessary to acquire this rule of gender agreement, once it is acquired, it should be applied consistently, as long as the child knows the gender of the noun in question. This means that once chil­ dren’s knowledge of gender attribution is taken into account, input effects on gender agreement should be minimal, and this turned out to be the case. Unsworth (2013a) argued that for phenomena such as gender-marking on defi­ nite determiners in Dutch (i.e., properties where considerable input effects are to be expected), bilingual children can better be matched to monolingual children not on chronological age, as is usually the case, but rather on their cumulative length of expo­ sure, that is, the approximate number of years of exposure taking into account variation in amount of exposure from year to year (see Bohman et al., 2010; Gutiérrez-Clellen & Kreiter, 2003; Thordardottir et al., 2006, for similar calculations, albeit used for dif­ ferent purposes). The logic of the argument is as follows: If one wants to address the question of whether reduced input or bilingualism is the relevant predictor of bilingual children’s lower accuracy on a given phenomenon, simultaneous bilingual children who are 8-years-old at time of testing and who have heard the target language for on average 75% of their exposure time since birth can better be compared with 6-yearold monolinguals than with 8-year-old monolinguals. Although such a comparison no doubt requires a certain degree of oversimplification, and it inevitably introduces another confound, namely that the comparison monolingual group will always be younger (Long & Rothman, 2014), it arguably offers a potentially informative dimension to monolingual–bilingual comparisons and the exploration of input effects therein.

Other Domains Relatively few studies have directly assessed the role of input in bilingual acquisi­ tion from domains other than vocabulary and morphosyntax, and the results are mixed. Bohman et al. (2010) included a measure semantic fluency and found that children’s input and their output were significant predictors of scores in the home language, Spanish (but not in the societal language, English). Unsworth (2014), on the other hand, investigated the acquisition of direct object scrambling in Dutch in Dutch–English bilingual speakers and found no effects of input. This was predicted given that the input provides insufficient information concerning the nature of this



Quantity and Quality of Language Input 

 113

phenomenon. Similarly, in a study on the narrative abilities of L2 English children in Canada, Paradis and Kirova (2014) found that children who were born in Canada and who were therefore exposed to more English at home and for a longer period of time did not score any differently than children who were born outside of Canada. Possible explanations the authors suggested for this finding is that English exposure for all children came from native speakers, and the two groups did not differ in terms of the richness of their English-language environment outside of school. In other words, the comparable quality of input for these two groups of children may have compensated for the differences in quantity of input. Finally, a few recent studies have investigated input effects on children’s pho­ nological memory, as measured by nonword repetition (NWR) tasks. Thordardottir (2014) found that children’s scores on NWR were either not associated or only weakly associated with amount of exposure. She suggested that this was because NWR relies less on knowledge of the language in question in comparison with the ability to understand or produce real words, for example. These results contrast with findings from Parra, Hoff, and Core (2010), who found that phonological memory (as mea­ sured by NWR) not only correlated with amount of exposure in each of their bilingual toddlers two languages but also mediated the effect of amount of exposure on chil­ dren’s reported vocabulary and grammar scores.

Cross-Domain Comparisons We have seen that there is an increasing body of research showing that variation in bilingual children’s language experience, in the form of quantitative and qualitative differences between bilingual children and in some cases between bilingual children and their monolingual peers, impacts on their development in a range of linguistic domains. To date, however, relatively few studies have examined the effect of dif­ ferential input across various linguistic domains within the same children, yet such cross-domain comparisons are essential if we are to fully understand the nature of these effects and their implications. As mentioned above, bilingual children’s vocabulary knowledge is assumed to take on a distributed characteristic, and it has been suggested that this may lead to profile effects whereby bilingual children’s scores on tests of vocabulary will be system­ atically lower than those testing knowledge which is not tied to a particular context, such as phonics (Oller et al., 2007; Verhoeven, 1994). Some support for the notion of profile effects was found in a study of L1 Turkish children acquiring English by Chondrogianni and Marinis (2011): Participants were less accurate on receptive vocab­ ulary and complex morphosyntax (articles, passives, wh-questions) than on tests of general grammar comprehension and the production of tense-marking morphology. In addition to the distributed characteristic of bilingual vocabulary, Chondrogianni and Marinis suggested that L1 transfer may also play a role in explaining these findings;

114 

 Sharon Unsworth

indeed, L1 transfer (or crosslinguistic influence) is also a factor that needs to be taken into account in considering how bilingual language acquisition interacts with input factors. In their study on the narrative abilities of L2 English children, Paradis and Kirova (2014) found that children’s scores were lower for subskills which depended more on specific knowledge of English grammar and vocabulary (e.g., MLU, lexical diversity) than those relating to story grammar, a skill they argue “could be considered a shared as opposed to a distributed skill” (p. 343, emphasis in original). This study is thus in line with Oller and colleagues’ (2007) proposal concerning profile effects. Another set of studies comparing children’s performance across a range of linguis­ tic domains focuses not on their scores (“Are children more advanced in one domain vs. another?”) but on the existence of input effects (“Are there input effects in one domain vs. another?”). For example, in Thordardottir’s (2011, 2014, 2015) series of studies on bilingual French–English preschoolers in Montreal, a similar nonlinear relationship between amount of exposure and language development was established for vocabu­ lary and morphosyntax. In their study on Spanish–English toddlers in Florida, Hoff and colleagues (2012) also observed similar patterns in the relationship between rate of acquisition and vocabulary and grammar scores. Likewise, Paradis (2011) observed that the same input factors, namely length of L2 English exposure and richness of the English environment, emerged as significant predictors of children’s scores on vocabulary and verbal morphology. These findings contrast with those of Unsworth (2014), who compared the acqui­ sition of linguistic properties taken from two different domains, namely, gendermarking on definite determiners, a morphosyntactic property of Dutch with a considerable lexical component, (as previously discussed) and the acquisition of meaning restrictions on different word orders (scrambling), a property involving com­ positional semantic and syntactic processes. In short, Unsworth argued that because the information needed to acquire these meaning restrictions is simply unavailable in the input (a “poverty of the stimulus” problem), differences between bilingual chil­ dren in terms of amount of exposure should be largely irrelevant, and hence no input effects are expected. The results were consistent with this prediction: bilingual chil­ dren with varying degrees of weekly exposure to Dutch acquired the relevant inter­ pretive constraints within the same timeframe as their monolingual peers, unlike gender-marking on definite determiners.

Summary Faster rates of development in bilingual children have been shown to correlate with relative amount of input, and this has been observed for the acquisition of vocabulary and certain aspects of morphosyntax, most typically verbal morphology. This relation­



Quantity and Quality of Language Input 

 115

ship is, however, nonlinear in nature. Research into input effects in other domains is sparse and the results are mixed. Limited number of studies examining input effects across domains suggest that there are profile effects, that is, children may be com­ paratively more advanced in one area than another. More research is however needed to establish whether input effects emerge in a similar fashion across domains.

Some Implications This section draws together the findings discussed in this chapter to highlight some of the implications for theories of language acquisition, and for parenting and education.

Theories of Language Acquisition Hoff et al. (2012), Paradis (2011), and Thordardottir (2015) have all argued that finding similar input effects across different domains is consistent with usage-based or emer­ gentist approaches (e.g., Lieven & Tomasello, 2008). As Paradis (2011) noted, what is relevant about these theories in the context of input effects in bilingual acquisition is the emphasis on input properties as determinants of rate of acquisition, and the assumption that the same learning mechanisms dictate lexical and morphosyntac­ tic acquisition (see also Gathercole, 2007). The aforementioned findings do indeed align well with the spirit of such an empiricist approach. To further corroborate such claims, however, more specific predictions are needed about exactly where and when input effects are expected (see Blom et al., 2012, for a recent attempt). Furthermore, as Unsworth’s (2014) findings illustrate, to gain a complete understanding of the inter­ action between input and bilingual acquisition, studies also need to include prop­ erties of language that are either arguably absent from the input or at least highly infrequent, and to compare these with properties such as vocabulary and verbal morphology, for which input effects are easier to account for on any approach. Indeed, it is worth pointing out that nativist approaches to language acquisition do not claim that there is no role for input, nor are they incompatible with the observa­ tion that variation in bilingual—or, for that matter, monolingual—children’s language experience impacts on rates of acquisition. Nativist theories such as universal grammar are, however, theories about the nature of the linguistic knowledge acquired (“prop­ erty” theories) rather than how the learner moves from state of knowledge to another (“transition” theories; Gregg, 1996). To capture the largely input-dependent differ­ ences in rate of acquisition observed in much of the literature reviewed here, a theory of universal grammar must be combined with some kind of transitional component (e.g., Yang, 2002). Only then can specific predictions be made concerning the precise effect of variation in input on bilingual children’s developing linguistic systems.

116 

 Sharon Unsworth

Parenting and Education Given the large and growing number of children in the world growing up with more than one language, a better understanding of bilingual language acquisition in general, and the effect of certain input properties on this process more in particular, is crucial for educators to successfully support their students’ development in both of languages. Given that early language experience, and in particular, school-entry levels of language development, have been shown to predict later academic achievement (see Chapter 11, this volume), knowing how best to promote language learning in the early years is of critical importance. For example, the results of Place and Hoff’s (2011) study suggest that encouraging nonnative parents of bilingual toddlers to access native-speaker input and to increase the number of interlocutors providing native-speaker input is likely to promote children’s societal language development. In addition, encouraging parents, especially those with low proficiency, to speak the societal rather than the minority language to their children is not to be advised (Hammer et al., 2009; Paradis, 2011). Not only is this unlikely to lead to better L2 performance, removing vital input in the home language will decrease the likelihood of children becoming fully-fledged bilingual speakers, with the consequence that they may miss out on many of the advantages which this entails. Finally, there is some evidence to suggest that teachers should be encouraged to use syntactically complex utterances (i.e., using multiple clauses) and a range of different words (Gámez & Levine, 2013).

Conclusion and Future Research Bilingual children’s language experience is characterized by variability, as are their language outcomes. The extant literature reviewed in this chapter suggests that the two are clearly related in some meaningful way. This relationship is complex, however, and its intricacies are far from well understood. Variation in input quantity has been found to impact on children’s rate of acquisition in vocabulary and syntax, although not for all age groups and not in all contexts. In addition, it appears that input beyond a certain level (around 50%–60%) does not facilitate further development, suggesting that bilingual children are able to achieve as much as monolingual children but with less (Thordardottir, 2011, 2015). This interpretation of the findings provides a positive twist to a debate all too often couched in terms of delay and deviance from the mono­ lingual norm prevalent in most of the countries in which this research takes place. All input is not equal, however. Recent research has demonstrated the impor­ tance of input quality as well as quantity. More specifically, factors such as exposure from native speakers and from a range of different sources have been shown to have a positive impact on children’s developing language skills, at least in the early years. Despite the ongoing thriving research enterprise on the topic, most studies examining the role of input quantity and quality in bilingual acquisition are still



Quantity and Quality of Language Input 

 117

rather limited in scope. To fully understand this relationship, more comparisons are needed across multiple linguistic domains and across multiple age groups, and these should preferably be longitudinal in nature. Furthermore, the child’s complete social network, including siblings and peers, as well as the wider social context needs to be considered. Finally, more sophisticated statistical techniques (e.g., the polynomial fitting procedures used in Thordardottir, 2011, 2014; the cluster analysis in Cattani et al., 2014) as well as more careful consideration of how we quantify input properties (Blom et al., 2012; Grüter et al., 2014) will surely also further our understanding of the array of outstanding issues on this topic.

References Barron-Hauwaert, S. (2011). Bilingual siblings: Language use in families. Clevedon, England: Multilingual Matters. Bedore, L. M., Peña, E. D., Summers, C. L., Boerger, K. M., Resendiz, M. D., Greene, K., . . . Gillam, R. B. (2012). The measure matters: Language dominance profiles across measures in Spanish–English bilingual children. Bilingualism: Language and Cognition, 15, 616–629. http://dx.doi.org/ 10.1017/S1366728912000090 Bialystok, E., Luk, G., Peets, K. F., & Yang, S. (2010). Receptive vocabulary differences in monolingual and bilingual children. Bilingualism: Language and Cognition, 13, 525–531. http://dx.doi.org/ 10.1017/S1366728909990423 Blom, E. (2010). Effects of input on the early grammatical development of bilingual children. The International Journal of Bilingualism, 14, 422–446. http://dx.doi.org/10.1177/ 1367006910370917 Blom, E., Paradis, J., & Duncan, T. S. (2012). Effects of input properties, vocabulary size, and L1 on the development of third person singular –s in child L2 English. Language Learning, 62, 965–994. http://dx.doi.org/10.1111/j.1467-9922.2012.00715.x Bohman, T. M., Bedore, L. M., Peña, E. D., Mendez-Perez, A., & Gillam, R. B. (2010). What you hear and what you say: Language performance in Spanish–English bilinguals. International Journal of Bilingual Education and Bilingualism, 13, 325–344. http://dx.doi.org/10.1080/13670050903342019 Bowers, E. P., & Vasilyeva, M. (2011). The relation between teacher input and lexical growth of preschoolers. Applied Psycholinguistics, 32, 221–241. http://dx.doi.org/10.1017/ S0142716410000354 Bridges, K., & Hoff, E. (2014). Older sibling influences on the language environment and language development of toddlers in bilingual homes. Applied Psycholinguistics, 35, 225–241. http:// dx.doi.org/10.1017/S0142716412000379 Byers-Heinlein, K. (2013). Parental language mixing: Its measurement and the relation of mixed input to young bilingual children’s vocabulary size. Bilingualism: Language and Cognition, 16, 32–48. http://dx.doi.org/10.1017/S1366728912000120 Cattani, A., Abbot-Smith, K., Farag, R., Krott, A., Arreckx, F., Dennis, I., & Floccia, C. (2014). How much exposure to English is necessary for a bilingual toddler to perform like a monolingual peer in language tests? International Journal of Language & Communication Disorders, 49, 649–671. http://dx.doi.org/10.1111/1460-6984.12082

118 

 Sharon Unsworth

Chondrogianni, V., & Marinis, T. (2011). Differential effects of internal and external factors on the development of vocabulary, tense morphology, and morpho-syntax in successive bilingual children. Linguistic Approaches to Bilingualism, 1, 318–345. http://dx.doi.org/10.1075/ lab.1.3.05cho Cornips, L., & Hulk, A. (2008). Factors of success and failure in the acquisition of grammatical gender in Dutch. Second Language Research, 24, 267–295. http://dx.doi.org/10.1177/ 0267658308090182 De Houwer, A. (2007). Parental language input patterns and children’s bilingual use. Applied Psycholinguistics, 28, 411–424. http://dx.doi.org/10.1017/S0142716407070221 De Houwer, A. (2014). The absolute frequency of maternal input to bilingual and mono­lingual children: A first comparison. In T. Grüter & J. Paradis (Eds.), Input and experience in bilingual development (pp. 37–58). Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1075/tilar.13.03deh De Houwer, A., Bornstein, M., & Putnick, D. L. (2014). A bilingual–monolingual comparison of young children’s vocabulary size: Evidence from comprehension and production. Applied Psycholinguistics, 35, 1189–1211. http://dx.doi.org/10.1017/S0142716412000744. Dixon, L. Q., Zhao, J., Quiroz, B. G., & Shin, J. (2012). Home and community factors influencing bilingual children’s ethnic language vocabulary development. The International Journal of Bilingualism, 16, 541–565. http://dx.doi.org/10.1177/1367006911429527 Driessen, G., van der Slik, F., & De Bot, K. (2002). Home language and language proficiency: A large-scale longitudinal study in Dutch primary schools. Journal of Multilingual and Multicultural Development, 23, 175–194. http://dx.doi.org/10.1080/01434630208666464 Dunn, L. M., & Dunn, D. M. (2007). Peabody picture vocabulary test (PPVT-4) (4th ed.). Minneapolis, MN: Pearson. Duursma, E., Romero-Contreras, S., Szuber, A., Proctor, P., Snow, C. E., August, D., & Caldéron, M. (2007). The role of home literacy and language environment on bilinguals’ English and Spanish vocabulary development. Applied Psycholinguistics, 28, 171–190. http://dx.doi.org/ 10.1017/S0142716406070093 Fernald, A. (2006). When infants hear two languages: Interpreting research on early speech perception by bilingual children. In E. Hoff & P. McCardle (Eds.), Childhood bilingualism: Research on infancy through school age (pp. 19–29). Clevedon, England: Multilingual Matters. Gámez, P. B., & Levine, S. C. (2013). Oral language skills of Spanish-speaking English language learners: The impact of high-quality native language exposure. Applied Psycholinguistics, 34, 673–696. http://dx.doi.org/10.1017/S0142716411000919 Gathercole, V. C. M. (2002). Command of the mass/count distinction in bilingual and mono­lingual children: An English morphosyntactic distinction. In D. K. Oller & R. E. Eilers (Eds.), Language and literacy in bilingual children (pp. 175–206). Clevedon, England: Multilingual Matters. Gathercole, V. C. M. (2007). Miami and North Wales, so far and yet so near: Constructivist account of morpho-syntactic development in bilingual children. International Journal of Bilingual Education and Bilingualism, 10, 224–247. http://dx.doi.org/10.2167/beb442.0 Gathercole, V. C. M. (2014). Bilingualism matters: One size does not fit all. International Journal of Behavioral Development, 38, 359–366. http://dx.doi.org/10.1177/0165025414531676 Gathercole, V. C. M., & Thomas, E. M. (2009). Bilingual first-language development: Dominant language takeover, threatened minority language take-up. Bilingualism: Language and Cognition, 12, 213–237. http://dx.doi.org/10.1017/S1366728909004015



Quantity and Quality of Language Input 

 119

Golberg, H., Paradis, J., & Crago, M. (2008). Lexical acquisition over time in minority L1 children learning English as L2. Applied Psycholinguistics, 29, 41–65. Gregg, K. R. (1996). The logical and developmental problems of second language acquisition. In W. C. Ritchie & T. K. Bhatia (Eds.), Handbook of second language acquisition (pp. 49–81). San Diego, CA: Academic Press. http://dx.doi.org/10.1016/B978-012589042-7/50004-9 Grüter, T., Hurtado, N., Marchman, V., & Fernald, A. (2014). Language exposure and online processing efficiency: Relative versus absolute measures. In T. Grüter & J. Paradis (Eds.), Input and experience in bilingual development (pp. 15–36). Amsterdam, the Netherlands: John Benjamins. Gutiérrez-Clellen, V. F., & Kreiter, J. (2003). Understanding child bilingual acquisition using parent and teacher reports. Applied Psycholinguistics, 24, 267–288. http://dx.doi.org/10.1017/ S0142716403000158 Hammer, C. S., Davison, M. D., Lawrence, F. R., & Miccio, A. W. (2009). The effect of maternal language on bilingual children’s vocabulary and emergent literacy development during Head Start and kindergarten. Scientific Studies of Reading, 13, 99–121. http://dx.doi.org/10.1080/ 10888430902769541 Hoff, E. (2006). How social contexts support and shape language development. Developmental Review, 26, 55–88. http://dx.doi.org/10.1016/j.dr.2005.11.002 Hoff, E., Core, C., Place, S., Rumiche, R., Señor, M., & Parra, M. (2012). Dual language exposure and early bilingual development. Journal of Child Language, 39, 1–27. http://dx.doi.org/10.1017/ S0305000910000759 Hoff, E., Rumiche, R., Burridge, A., Ribot, K. M., & Welsh, S. N. (2014). Expressive vocabulary development in children from bilingual and monolingual homes: A longitudinal study from two to four years. Early Childhood Research Quarterly, 29, 433–444. http://dx.doi.org/10.1016/ j.ecresq.2014.04.012 Hurtado, N., Grüter, T., Marchman, V., & Fernald, A. (2014). Relative language exposure, processing efficiency and vocabulary in Spanish–English bilingual toddlers. Bilingualism: Language and Cognition, 17, 189–202. http://dx.doi.org/10.1017/S136672891300014X Huttenlocher, J., Vasilyeva, M., Cymerman, E., & Levine, S. (2002). Language input and child syntax. Cognitive Psychology, 45, 337–374. http://dx.doi.org/10.1016/S0010-0285(02)00500-5 Jia, G., & Fuse, A. (2007). Acquisition of English grammatical morphology by native Mandarinspeaking children and adolescents: Age-related differences. Journal of Speech, Language, and Hearing Research, 50, 1280–1299. http://dx.doi.org/10.1044/1092-4388(2007/090) Lieven, E. V. M., & Tomasello, M. (2008). Children’s first language acquisition from a usage-based perspective. In P. Robinson & N. Ellis (Eds.), Handbook of cognitive linguistics and second language acquisition (pp. 168–196). London, England: Routledge. Long, D., & Rothman, J. (2014). Some caveats to the role of input in the timing of child bilingualism. Linguistic Approaches to Bilingualism, 4, 351–356. http://dx.doi.org/10.1075/lab.4.3.09lon Marchman, V. A., Fernald, A., & Hurtado, N. (2010). How vocabulary size in two languages relates to efficiency in spoken word recognition by young Spanish–English bilinguals. Journal of Child Language, 37, 817–840. http://dx.doi.org/10.1017/S0305000909990055 Montrul, S. A. (2008). Incomplete acquisition in bilingualism. Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1075/sibil.39 Montrul, S. A., & Potowski, K. (2007). Command of gender agreement in school-age Spanish–English bilingual children. The International Journal of Bilingualism, 11, 301–328. http://dx.doi.org/10. 1177/13670069070110030301

120 

 Sharon Unsworth

Nicoladis, E., Palmer, A., & Marentette, P. (2007). The role of type and token frequency in using past tense morphemes correctly. Developmental Science, 10, 237–254. http://dx.doi.org/10.1111/ j.1467-7687.2007.00582.x Oller, D. K., & Eilers, R. E. (Eds.). (2002). Language and literacy in bilingual children. Clevedon, England: Multilingual Matters. Oller, D. K., Pearson, B. Z., & Cobo-Lewis, A. B. (2007). Profile effects in early bilingual language and literacy. Applied Psycholinguistics, 28, 191–230. http://dx.doi.org/10.1017/S0142716407070117 Palermo, F., Mikulski, A. M., Fabes, R. A., Hanish, L. D., Martin, C. L., & Starget, L. E. (2014). English exposure in the home and classroom: Predictions to Spanish-speaking preschoolers’ English vocabulary skills. Applied Psycholinguistics, 35, 1163–1187. Paradis, J. (2010). Bilingual children’s acquisition of English verb morphology: Effects of language exposure, structure complexity, and task type. Language Learning, 60, 651–680. http://dx.doi.org/ 10.1111/j.1467-9922.2010.00567.x Paradis, J. (2011). Individual differences in child English second language acquisition: Comparing child-internal and child-external factors. Linguistic Approaches to Bilingualism, 1, 213–237. http://dx.doi.org/10.1075/lab.1.3.01par Paradis, J., & Kirova, A. (2014). English second-language learners in preschool: Profile effects in their English abilities and the role of home language environment. International Journal of Behavioral Development, 38, 342–349. http://dx.doi.org/10.1177/0165025414530630 Paradis, J., Nicoladis, E., Crago, M., & Genesee, F. (2011). Bilingual children’s acquisition of the past tense: A usage-based approach. Journal of Child Language, 38, 554–578. http://dx.doi.org/ 10.1017/S0305000910000218 Parra, M., Hoff, E., & Core, C. (2010). Relations among language exposure, phonological memory, and language development in Spanish–English bilingually developing 2-year-olds. Journal of Experimental Child Psychology, 108, 113–125. http://dx.doi.org/10.1016/j.jecp.2010.07.011 Patterson, J. L. (2004). Comparing bilingual and monolingual toddlers’ expressive vocabulary size: Revisiting Rescorla and Achenbach (2002). Journal of Speech, Language, and Hearing Research, 47, 1213–1215. http://dx.doi.org/10.1044/1092-4388(2004/089) Pearson, B. Z. (2007). Social factors in childhood bilingualism in the United States. Applied Psycholinguistics, 28, 399–410. http://dx.doi.org/10.1017/S014271640707021X Pearson, B. Z., Fernández, S. C., & Oller, D. K. (1993). Lexical development in bilingual infants and toddlers: Comparison to monolingual norms. Language Learning, 43, 93–120. http://dx.doi.org/ 10.1111/j.1467-1770.1993.tb00174.x Place, S., & Hoff, E. (2011). Properties of dual language exposure that influence 2-year-olds’ bilingual proficiency. Child Development, 82, 1834–1849. http://dx.doi.org/10.1111/ j.1467-8624.2011.01660.x Ribot, K. M., & Hoff, E. (2014). “¿Cómo estas?” “I’m good.” Conversational code-switching is related to profiles of expressive and receptive proficiency in Spanish–English bilingual toddlers. International Journal of Behavioral Development, 38, 333–341. http://dx.doi.org/10.1177/ 0165025414533225 Scheele, A. F., Leseman, P. P. M., & Mayo, A. Y. (2010). The home language environment of monolingual and bilingual children and their language proficiency. Applied Psycholinguistics, 31, 117–140. http://dx.doi.org/10.1017/S0142716409990191 Sheng, L., Lu, Y., & Kan, P. F. (2011). Lexical development in Mandarin–English bilingual children. Bilingualism: Language and Cognition, 14, 579–587. http://dx.doi.org/10.1017/ S1366728910000647



Quantity and Quality of Language Input 

 121

Smithson, L., Paradis, J., & Nicoladis, E. (2014). Bilingualism and receptive vocabulary achievement: Could sociocultural context make a difference? Bilingualism: Language and Cognition, 17, 810–821. Stadthagen-González, H., Gathercole, V. C. M., Pérez-Tattam, R., & Yavas, F. (2013). Vocabulary assessment of bilingual adults: To cognate or not to cognate. In V. C. M. Gathercole (Ed.), Solutions for the assessment of bilinguals (pp. 125–145). Bristol, England: Multilingual Matters. Thomas, E. M., Williams, N., Jones, L. A., Davies, S., & Binks, H. (2014). Acquiring complex structures under minority language conditions: Bilingual acquisition of plural morphology in Welsh. Bilingualism: Language and Cognition, 17, 478–494. http://dx.doi.org/10.1017/ S1366728913000497 Thordardottir, E. (2011). The relationship between bilingual exposure and vocabulary development. The International Journal of Bilingualism, 15, 426–445. http://dx.doi.org/10.1177/ 1367006911403202 Thordardottir, E. (2014). The typical development of simultaneous bilinguals: Vocabulary, morphosyntax and language processing in two age groups of Montreal preschoolers. In T. Grüter & J. Paradis (Eds.), Input and experience in bilingual development (pp. 141–160). Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1075/tilar.13.08tho Thordardottir, E. (2015). The relationship between bilingual exposure and morphosyntactic development. International Journal of Speech-Language Pathology, 17, 97–114. http://dx.doi.org/ 10.3109/17549507.2014.923509 Thordardottir, E., Rothenberg, A., Rivard, M. E., & Naves, R. (2006). Bilingual assessment: Can overall proficiency be estimated from separate measurement of two languages? Journal of Multilingual Communication Disorders, 4, 1–21. http://dx.doi.org/10.1080/14769670500215647 Unsworth, S. (2013a). Assessing the role of current and cumulative exposure in simultaneous bilingual acquisition: The case of Dutch gender. Bilingualism: Language and Cognition, 16, 86–110. http://dx.doi.org/10.1017/S1366728912000284 Unsworth, S. (2013b). Current issues in multiple first language acquisition. Annual Review of Applied Linguistics, 33, 21–50. http://dx.doi.org/10.1017/S0267190513000044 Unsworth, S. (2014). Comparing the role of input in bilingual acquisition across domains. In T. Grüter & J. Paradis (Eds.), Input and experience in bilingual development (pp. 181–201). Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1075/tilar.13.10uns Unsworth, S. (2015). Amount of exposure as a proxy for dominance in bilingual language acquisition. In C. Silva-Corvalan & J. Treffers-Daller (Eds.), Language dominance in bilinguals: Issues of measurement and operationalization (pp. 156–173). Cambridge, England: Cambridge University Press. Verhoeven, L. (1994). Transfer in bilingual development: The linguistic interdependence hypothesis revisited. Language Learning, 44, 381–415. http://dx.doi.org/10.1111/j.1467-1770.1994.tb01112.x Willard, J. A., Agache, A., Jäkel, J., Glück, C. W., & Leyendecker, B. (2015). Family factors predicting vocabulary in Turkish as a heritage language. Applied Psycholinguistics, 36, 875–898. Yang, C. (2002). Knowledge and learning in natural language. Oxford, England: Oxford University Press.

Virginia C. Mueller Gathercole

7 Factors Moderating Proficiency in Bilingual Speakers What factors determine or contribute to bilingual speakers’ proficiency, or their overall knowledge of and fluency in each of their languages? This chapter highlights the multiple factors influencing proficiency. This discussion adopts the position, from the start, that proficiency should be kept separate from language ability (Peña, Bedore, & Fiestas, 2013); the latter has to do with the language-learning capacity itself, whereas the former refers to a speaker’s sum-total knowledge and facility in using the particular language in question. This distinction is important because it is often difficult to tease apart level of proficiency in a language from language ability, and that is an issue that speech and language therapists, educators, and parents must cope with. The remedies for difficulties in ability and proficiency can be drastically different (see, e.g., Gathercole, 2013a, 2013b). When the distinction is not recognized, bilingual children may be misdiagnosed as having speech problems (abilities) and be sent for speech therapy simply because they are struggling with a second language (L2 [proficiency]; Letts, 2013; Peña, Bedore, & Fiestas, 2013), or they may be overlooked for speech problems because of implicit assumptions sometimes made about acquisition in bilingual speakers (O’Toole & Hickey, 2013). Kindergartners or young grade schoolers whose first language (L1) is distinct from the dominant language of the classroom may be viewed as academically inferior, again simply because an inability to navigate the academic subjects in the L2 (because of a low level of proficiency) is interpreted as an inability to handle the content of the material (because of low ability). This chapter takes the position that for normally developing bilingual speakers, we can assume that language abilities are commensurate with those of monolingual speakers. Proficiency, on the other hand, as observed in performance in the language-specific use of vocabulary or grammar, may vary from one child to another or one group to another because it is a product, in monolingual and bilingual speakers, of one’s range of experience, exposure, and use of the language in question. The purpose of this chapter is to examine what factors affect and modulate the language proficiency observed in bilingual children and adults. It will lay out the contributions of the quantity and quality of input in the home and in the community, and the role of the sources of that input; discuss the distributed characteristic, profile effects in bilingual speakers, and the role of linguistic complexity; touch on inter­action between This work was supported in part by ESRC Grant RES-062-23-0175 and by ESRC/HEFCW Grant RES-535-30-0061. http://dx.doi.org/10.1037/14939-008 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

124 

 Virginia C. Mueller Gathercole

the two languages; and discuss socioeconomic variables that may affect exposure and attitudes toward the languages (see Gathercole, 2014). The focus here is on simultaneous and early sequential bilingual speakers. Given a starting point at which bilingual children and monolingual children are assumed to have equal abilities with language learning, it becomes easier to delineate what might contribute to level of proficiency with each language in question. An important point to raise at the outset, and which will become apparent soon, is that it is unlikely that bilingual speakers’ proficiency in any of their languages will parallel exactly the proficiency observed in corresponding monolingual speakers. This is, simply stated, because bilingual speakers are learning two languages. First, the contexts of learning can be different from those of monolingual speakers in each language. Second, the knowledge of each language has the potential of influencing the knowledge of the other language, in that bilingual speakers may naturally draw linkages between components of the two languages at some points of development and at some points of the linguistic structures. Where and when such linkages occur are fundamental questions in themselves, and although they are not the focus of this chapter, I touch on them below insofar as such linkages can affect bilingual speakers’ use of the two languages.

Input and Exposure First and foremost, bilingual speakers’ acquisition of their two languages is closely tied to input (see Chapter 6, this volume). The fundamental fact that one necessarily must be exposed to a language to learn it does not tell us what quantity, quality, or range of distribution is necessary for exposure to allow acquisition to occur, nor whether children with distinct levels or types of exposure learn language at the same pace and in the same way.

Quantity in the Home It is clear that the relative quantity of input in each language can affect the timing of acquisition of aspects of that language. This has been documented for a variety of language pairs (e.g., Gathercole, Thomas, Roberts, Hughes, & Hughes, 2013; Hoff et al., 2012; Oller & Eilers, 2002). In our own work, we have looked extensively at children’s acquisition of Welsh and English in the North Wales context and have discovered that children’s home language profiles can predict timing of acquisition for each language. When bilingual children who are growing up in only Welsh homes (OWH), Welsh and English homes (WEH), and only English homes (OEH) are compared, those who have more input in Welsh at home gain knowledge of vocabulary and grammatical structures in Welsh prior to those with less input in Welsh, and, conversely, those who



Factors Moderating Proficiency in Bilingual Speakers 

 125

have more input in English at home gain knowledge of these in English earlier than those with less input in English (Gathercole, Pérez-Tattam, Stadthagen-González, & Thomas, 2014; Gathercole & Thomas, 2007, 2009; Gathercole, Thomas, & Laporte, 2001; Gathercole et al., 2013; Thomas & Gathercole, 2007). The relative differences are most notable at younger ages, and as children gain facility with the given language, differences across home language groups tend to get neutralized, with all groups catching up with those who had an initial head start. This is especially true for grammatical knowledge, which appears to depend crucially on the compilation of a critical mass of experience with a given structure to allow the child to command the patterns in question on a productive basis (Gathercole, 2002). The effects of input in the home appear to hold even in adulthood, especially for the minority language. We found that when adults were tested for English and Welsh vocabulary in Wales, those who had grown up in OWH homes had higher vocabularies in Welsh if they were partnered with another adult who had grown up in an OWH home. This was true even in contrast with similar OWH adults whose partner had grown up either in a WEH home or an OEH home (Gathercole, 2007a). It is probable that the first group is using relatively more Welsh in the home as adults than the other two groups, so they are hearing more Welsh on a daily basis and are reminded of less common words; or it is possible that their OWH partners have higher vocabularies in Welsh than the others’ WEH and OEH partners, and these OWH partners are promoting continued vocabulary acquisition by providing less common words in the input. Some have suggested that this overall picture may depend on whether one is examining expressive vocabulary or receptive vocabulary (Poulin-Dubois, Bialystok, Blaye, Polonia, & Yott, 2013), what instrument is being used (parental reports? elicited production? forced-choice receptive task?), and whether translational equivalents are included or not in vocabulary tests. And some have observed no differences in performance, as suggested by De Houwer, Bornstein, and Putnick’s (2014) examination of vocabulary knowledge in 13-month-old and 20-month-old Dutch–French bilingual and monolingual children. Such results revealing similarity of performance across groups may hinge on optimal exposure to both languages and the presence of multiple translation equivalents in the two languages (David & Wei, 2008; Pearson, Fernández, Lewedeg, & Oller, 1997; see Poulin-Dubois et al., 2013). For example, Bosch and Ramon-Casas (2014), who reported an advantage in Catalan–Spanish 18-month-old bilingual infants’ vocabularies over those of monolingual infants, noted that Catalan and Spanish share many lexical items; these authors attributed the advantage in the bilingual infants to the prevalence of cognates in the two languages, which, they argued, allows the child to “double up,” so to speak, on vocabulary acquisition (to be discussed in a later section of this chapter). In the absence of such conditions, such results are rare: In a recent examination of bilingual children between 24 and 36 months of age growing up with seven language pairs, Gatt and O’Toole (2013) reported that in total vocabulary scores (in

126 

 Virginia C. Mueller Gathercole

which the words in the two languages are added together), the bilingual children had vocabularies similar to those of monolingual children; however, on total conceptual vocabulary scores, the bilingual children lagged behind the monolingual children. This difference in total conceptual vocabulary was more apparent in the older children (31–36 months) than in the younger group (24–30 months).

Fluidity in Home Language Input and Influences From the Community Language Outside the Home A range of factors also affect the level of input inside and outside the home, including the relative prominence of speakers of the two languages in the community (Gathercole & Thomas, 2009), changes in the home language profiles as children or their siblings bring a school language into the home, and sociocultural factors influencing the child’s uptake of each language. The dominant language of a community can “win out,” both in terms of the relative proportion of input experienced and in terms of the child’s own output (Fillmore, 2000), with social and affective dynamics also at play in the child’s uptake of the two languages (Jean & Geva, 2012; Oller, Jarmulowicz, Pearson, & Cobo-Lewis, 2011; Snow & Hakuta, 1992). In one recent study, Jia, Chen, Kim, Chan, and Jeung (2014) reported on Chinese– English and Korean–English bilingual children growing up in the United States. These children’s primary source of input in the heritage language was their parents, but as they got older, the proportional use of the heritage language with parents decreased from a high of 90% at the youngest ages to 70% use at older ages. Changes in language use with siblings and peers were more drastic, favoring the heritage language at the younger ages (60%–70% use) but English during the school years (80% use). Furthermore, input through the media (TV and leisure reading) was dominated by English from the very beginning (50%–70% English; see also Ribot & Hoff, 2014). These patterns correlated with performance on picture naming and lexical fluency in the two languages. Ribot and Hoff (2014) reported similar effects in relation to language choice in code-switching in Spanish–English bilingual speakers in South Florida: Language choice correlates highly with exposure and lexical proficiency. In similar studies, Hoff and her colleagues (Bridges & Hoff, 2014; Place & Hoff, 2011) have reported similar input and output effects in Spanish–English bilingual speakers in the South Florida area. Bridges and Hoff (2014) reported that 22- and 30-month-old bilingual children who had older siblings scored higher in English on vocabulary and grammatical measures; older siblings are acting as significant sources of the community language for younger siblings. Furthermore, the authors added that when children begin school, even the mother’s use of English in speech to her children increased dramatically.



Factors Moderating Proficiency in Bilingual Speakers 

 127

These results are consistent with our own work. In a study of fluent, educated bilingual adults in Miami, Stadthagen-González, Gathercole, Pérez-Tattam, and Yavas (2013) reported that input to bilingual children in Spanish declined as they got older, with parental input in the heritage language decreasing with age, to a low of about 35% in some groups. The proportion of Spanish input from siblings was lower, depending on the family profiles: Those bilingual children who came from an early immigrant group heard Spanish from older siblings 74% of the time and from younger siblings 66% of the time. But those who were born in Miami and grew up in homes in which parents initially spoke only Spanish to them used Spanish only 34% of the time to older siblings and 23% of the time to younger siblings. Those who grew up in homes in which parents spoke Spanish and English to them heard Spanish even less from older and younger siblings, only 7% to 10% of the time. These input profiles are complemented by output statistics for use as adults: Although most continue speaking Spanish with their parents, the use of Spanish in other contexts is overtaken by English, including in speech to siblings and friends. Even the immigrant group uses Spanish with siblings only 30% of the time, and with friends 20% of the time. Those born in the United States speak Spanish only 5% to 16% of the time to siblings and friends. These input patterns affect vocabulary and grammatical performance in the minority language, but not in the majority language (Pérez-Tattam, Gathercole, Yavas, Stadthagen-González, & Anrrich, 2013; Stadthagen-González et al., 2013). In Wales, we have found similar effects concerning the community language (e.g., Gathercole et al., 2013). Home language effects are modulated by the community language, especially for Welsh in those children who come from OEH homes (Thomas, Gathercole, & Hughes, 2013).

Source of the Language In addition to quantity of exposure, the particular variants to which the child is exposed are critical to what is being learned. Fennell and Byers-Heinlein (2014), for example, recently reported that monolingual and bilingual infants can perceive phonological contrasts in their languages, but only according to what they have heard in the input: monolingual children perceive distinctions only when monolingual adults are speaking; bilingual children perceive distinctions only when bilingual adults are speaking. This suggests that early learning is tied intimately with the contexts of learning and the particular linguistic qualities characteristic of the interlocutors’ speech. Place and Hoff’s (2011) work suggests that the degree to which a child can “intake” the input depends on whether the adult input comes from monolingual or bilingual speakers. They found that having exposure to exclusively English conversational partners, to English from a variety of sources, and to English from native speakers contributed to higher scores in English.

128 

 Virginia C. Mueller Gathercole

Distributed Characteristic and Profile Effects The contexts in which bilingual children hear the two languages are not isomorphic— they may overlap to some extent, but there are likely to be some contexts that favor one or the other language. Because of this, what children learn in the two languages can show a distributed characteristic (Oller, 2005; Oller & Pearson, 2002; Patterson & Pearson, 2004), in which they knows how to say some things in one language and some other things in the other. For example, children may know scientific terms in one language, but know sports terms in the other, depending on which language they experience in each of those contexts. Such a distributed characteristic can persist into adulthood, so that, for example, academic language may come easily in one language and not in the other. Oller, Pearson, and Cobo-Lewis (2007) argued that such a distributed characteristic in bilingual speakers leads to profile effects in bilingual children’s command of distinct components of the language. Profile effects, simply stated, are differences in children’s performance in distinct areas of the language they are learning or on tests measuring different aspects of knowledge. Bilingual children might show high abilities in reading in comparison to monolingual peers of the same age, but might show lower picture naming scores in comparison (Oller et al., 2007; see also Ribot & Hoff, 2014). Oller et al. (2007) provided evidence from bilingual children from their Miami study (Oller & Eilers, 2002) that (a) such profile effects are apparent across linguistic domains in bilingual children’s performance, (b) seem consistent across bilingual children (e.g., with all bilingual children showing higher reading scores than naming scores), (c) are more pronounced at younger ages (in their data, Grade 2) than at older ages (Grade 5), and (d) are more persistent in the less dominant language (Spanish) than in the dominant language (English). In their study, bilingual children performed at a higher level on phonics and reading than on picture naming, and performed on vocabulary reasoning (e.g., providing synonyms and antonyms) at a level intermediate between these two. Oller et al. (2007) attributed the differential performance to the role that the distributed characteristic of vocabulary knowledge plays in the various subskills. For grammatical structures that rely crucially on knowledge of the vocabulary items that participate in them, lower performance will be linked with lower vocabulary knowledge; for performance that relies less on knowledge of particular vocabulary items (e.g., phonological metalinguistic awareness), performance will excel once the child has gained the knowledge in question, regardless of vocabulary knowledge. Paradis and Kirova (2014) also considered profile effects in bilingual children in their data on English-language learning children’s narrative abilities in English. They differentiate performance on story grammar, first mentions, utterance length, sentence complexity, lexical diversity, and story length. Children generally performed within normal ranges for all of these except utterance length, but they performed better on story grammar and story length than on the other elements. Furthermore, there were sizable standard deviations, with many children below less than one standard



Factors Moderating Proficiency in Bilingual Speakers 

 129

deviation on some subscores. This was particularly true for utterance length, lexical diversity, and first mentions, for which 43%, 24%, and 33% of the children, respectively, fell more than one standard deviation below the norms. These substantial differences indicate that some aspects of the grammar are easier than others, leading the bilingual children’s performance to have an appearance of being optimal in some areas and less than optimal in others. The reasons for the differences, Paradis and Kirova pointed out, partly have to do with the fact that performance on some of these (e.g., how to express first mentions) is more dependent on knowledge of the specific language than others (e.g., how to put together story structure). That is, some skills, like story structure, can be seen as more “language neutral,” and, therefore seem less prone to profile effects. These profile effects are a reminder that bilingual children’s abilities are not all or none but instead show differential performance depending on what particular aspect of the language one is examining.

Linguistic Complexity The profile effects discussed by Oller et al. (2007) were more prevalent at less advanced stages in the language (second graders more than fifth graders, more in Spanish than in English), partly because what a child needs to gain a command of different aspects of the language can vary in complex ways. The same general principle can apply across a wide range of aspects of the grammar that the child must learn to become fluent in the language—in particular, some aspects demand a greater amount of exposure and experience than others to gain a command of those items, so those can be expected to be learned later. In particular, we can expect aspects of the language that are complex or involve opaque form-function mappings to take longer to learn than aspects that are simpler or involve transparent mappings. Differences in opacity are easiest to observe when structures are compared across languages. For example, Spanish grammatical gender is considerably transparent, whereas Welsh grammatical gender is extremely opaque (Gathercole, Thomas, & Laporte, 2001). Spanish that-trace structures are transparent, whereas English thattrace structures are opaque (Gathercole, 2002). Simple active sentences in a language are usually simpler in construction than sentences with embedded relative clauses. We can expect all children, whether monolingual or bilingual, in general, to acquire the simpler or more transparent structures more easily than related structures that are more complex or more opaque (Gathercole, 2007b; Smoczyńska, 1985). With more opaque structures, we can expect it to take more experience to gain whatever critical mass of input information is needed for the child to draw out regular patterns and use them productively. What this means in relation to a bilingual learner, then, is that, for simpler or more transparent structures, bilingual learners can perform at a level commensurate with their peers early on, so that differences across groups become neutralized earlier than

130 

 Virginia C. Mueller Gathercole

for more complex or more opaque structures. This will result in something similar to profile effects, in that the bilingual learners may perform similarly to monolingual learners and to one another on some structures, but perform distinctly on others. In fact, for some bilingual learners, some of the most complex or most opaque structures may never be fully acquired, as they may be timed “off the map.” For such cases, the learner will be observed to have incomplete learning. Such cases have been observed precisely in the types of structures mentioned above (e.g., that-trace structures in English [Gathercole, 2002] and grammatical gender in Welsh [Gathercole, Thomas, & Laporte, 2001; Thomas & Gathercole, 2007]). Some learners may never gain a full command of such structures, even though their command of the language otherwise appears very fluent. Thus, it is not necessarily the case that everything bilingual learners are learning falls on a line that runs parallel or even merges with development in the monolingual learners. For aspects of the language that are structurally simple or transparent, bilingual learners readily catch up with monolingual learners. For aspects of the language that are structurally complex or opaque, bilingual learners’ acquisition of those structures may remain incomplete for longer, simply because of the fact that the necessary critical mass of input needed to sort out the organization of the structure in question may take longer to achieve or, in some cases, may never be achieved.

Interactive Influences of One Language on the Other How might the acquisition, knowledge, or organization of the languages in bilingual speakers differ from their acquisition, knowledge, and organization in corresponding monolingual speakers? Can bilingual children’s knowledge of or performance in one language carry over to their proficiency in the other language? Whereas there is considerable evidence that simultaneous and early bilingual speakers’ linguistic systems, particularly at the morphosyntactic level, develop separately from very early (see Chapter 2, this volume), there is also ample evidence that the two systems must eventually be linked, as cross-linguistic priming studies show that forms from one language call up related forms in the speaker’s other language (Dijkstra & Van Hell, 2003; Dijkstra & Van Heuven, 1998; Grosjean, 1998, 2001). There is considerable debate on the important question of interaction, and one must clarify whether any observed linkages in children are linguistic links, cognitive links, or metalinguistic/ metacognitive links (Gathercole et al., 2014). A full discussion of this issue is beyond the scope of this chapter, but I touch on two areas here because the question of carry­ over from one language to the other is highly relevant to that of bilingual speakers’ ultimate proficiency and use of their languages. Two places in which linkages are apparent are worthy of mention: (a) situations in which the two languages share surface phonological shapes and (b) cross-language convergence in bilingual semantic organization.



Factors Moderating Proficiency in Bilingual Speakers 

 131

Shared Phonological Form There is growing evidence that when there are shared phonological shapes—cognates and near-cognates—children may be able to take advantage of the similarities. Bosch and Ramon-Casas’s (2014) above-mentioned examination of the vocabularies of monolingual and bilingual 18-month-old infants exposed to Catalan and Spanish revealed such an advantage. These authors suggested that whether bilingual infants’ vocabularies are seen to lag behind those of monolingual infants in each of their languages may be related to whether or not the child is learning dissimilar or similar languages. They argued that part of learning words is learning their phonological shapes, and shared phonological shapes may promote vocabulary development: Forms that are identical or similar in the two languages are heard twice as often as nonsimilar words in the two languages, with the same reference across the languages. Catalan and Spanish parents reported on their children’s expressive knowledge of translation equivalents in the two languages. Some words were identical in form (sí meaning yes) in the two languages (i.e., cognates), some were similar in form (gat in Catalan, gato in Spanish meaning cat; near-cognates), and some were dissimilar in form (ocell in Catalan, pájaro in Spanish meaning bird). These authors found that their bilingual infants’ scores were compatible with or exceeded those of the monolingual infants: Catalan monolingual infants’ total vocabulary was 43; Spanish monolingual infants’ total vocabulary was 50; and bilingual infants’ combined total vocabulary was 75.4 (Spanish: 33.5, Catalan: 41.9). The bilingual infants’ vocabularies included, on average, 18 cognates. The authors concluded that phonological proximity of words across bilingual infants’ two languages can facilitate lexical acquisition. Schelletter (2002) reported a similar advantage in the acquisition of words sharing phonological form in a very young German–English bilingual child. With older bilingual children, the evidence is more mixed. García (1991) reported that fourth-grade Spanish–English bilingual children seldom made use of cognate knowledge while reading, suggesting instead that a certain level of development and possibly explicit instruction in the use of cognates was needed for bilingual children to adopt this strategy. On the other hand, Nagy, García, Durgunoğlu, and HancinBhatt (1993) found that L1 vocabulary knowledge and ability to recognize cognates affected performance of children in fourth to sixth grades in a comprehension task that included cognates. Kelley and Kohnert (2012) similarly found a cognate advantage in 8- to 13-year-old L1 Spanish learners of English as an L2, although they reported considerable individual variation in the evidence for a cognitive advantage. Pérez, Peña, and Bedore (2010) conducted a study looking into vocabulary knowledge of kinder­garten and first-grade Spanish–English bilingual children and found that children who were dominant in Spanish performed better with cognates than with noncognates, whereas children who were dominant in English performed better with noncognates. For adults, our tests of the vocabulary knowledge in Miami Spanish–English fluent bilingual speakers (Stadthagen-González et al., 2013) showed better performance

132 

 Virginia C. Mueller Gathercole

in bilingual speakers on cognates and near-cognates than on noncognates, providing 10% to 12% more correct responses to cognates than to noncognates. It is possible that bilingual speakers are able to capitalize on surface phonological similarities in their two languages in different ways at different points in development. In infants, when children are first learning the mappings between individual words and referents, the recurrence of the same form in the input from the two languages in reference to the same extension may support the retention and recall of those forms. In older children, as words become more entrenched in their linguistic and nonlinguistic contexts of usage in each language, the linkages between the individual forms may be somewhat weaker. Then as bilingual speakers gain full fluency in the two languages, the links across languages may be strengthened through usage of the two languages in more and more similar contexts.

Shared Cognition and Semantic Understanding of Meaning Bilingual children are processing the input they are taking in from each language through a common cognitive understanding and through a common understanding of the social and pragmatic contexts in which the languages are used. Because children are developing cognitively at the same time as they are acquiring language, the understanding of the semantics of the languages being learned necessarily interacts with that cognitive understanding. Yet languages differ dramatically in how they carve up the semantic space, so that categories in one language can be divided up into distinct categories in the other or may crosscut those of the other, or some concepts may not be encoded at all (Elston-Güttler & Williams, 2008; Prior, MacWhinney, & Kroll, 2007; Tokowicz, Kroll, De Groot, & Van Hell, 2002). For example, hill in English corresponds to at least two different words in Welsh, bryn (a mini-mountain type hill) and (g)allt (an incline, as on a road); chwyrnu in Welsh corresponds to two verbal categories in English, snore and growl. Work has been providing clear evidence that in bilingual speakers, there is semantic influence from one language to the other in such cases. This is true for L2 learners, not only in the direction of the L1 to the L2 but also from the L2 to the L1 (Brown & Gullberg, 2008; Elston-Güttler & Williams, 2008; Jiang, 2002, 2004; Malt & Sloman, 2003; Pavlenko, 2003; Pavlenko & Malt, 2011; Wolff & Ventura, 2009), and for simultaneous and early sequential bilingual leaners (Ameel, Malt, Storms, & van Assche, 2009; Ameel, Storms, Malt, & Sloman, 2005; Gathercole & Moawad, 2010). For example, Ameel and colleagues (2005, 2009) have found that simultaneous Dutch–French bilingual speakers show convergence of categories of bottles and dishes across their two languages, so that their semantic groupings are distinct in both languages from what is observed for monolingual speakers. Such evidence will have natural consequences for bilingual speakers’ performance related to the linguistic encoding of such categories and concepts. What may look like a lack of knowledge or a lack of conceptual under-



Factors Moderating Proficiency in Bilingual Speakers 

 133

standing (e.g., if a Welsh-speaking child asserts in English that an incline in the road is not a hill or claims that her sleeping father is growling, or if an English-speaking child asserts in Welsh that an incline in the road is a bryn or denies that her sleeping father can chwyrnu) may be a simple consequence of the interaction of the speaker’s two languages within the semantic-conceptual domain. Such phonological and semantic cases of communication between the bilingual speakers’ two languages, and any others in which bilingual speakers’ two languages influence each other, constitute one of the primary realms in which a bilingual speakers’ knowledge and organization of their two languages will necessarily differ from the knowledge and organization in the corresponding monolingual speakers, even in the long run. One way of thinking about this is that monolingual speakers have only one set of words, one set of phonological shapes, one set of semantic categories that are linked together to form the language they speak, whereas bilingual speakers have a richer set of words, phonological patterns, and semantic categories that are linked together, so it is natural that those linkages have subtle influences across the two languages.

Socioeconomic Factors The final factor I consider here is socioeconomic level. How socioeconomic factors influence bilingual children’s acquisition of their two languages can be complex and is not necessarily straightforward. We know that children growing up in poverty can lag behind their peers in language performance (Hart & Risley, 1995). This may be in part because their parents may speak less to them, in part because of the more limited opportunities afforded to such children, and perhaps in part because of attitudinal and motivational factors separating parents across the socioeconomic strata. All of these could apply to children growing up bilingually, and one additional factor may be the linguistic knowledge of their parents, who may be highly dominant in one of the two languages. Evidence indicates that these factors can indeed affect language knowledge in bilingual children. In Oller and Eilers’s (2002) study of bilingual speakers in the Miami context, socioeconomic status (SES) was directly controlled for, so that half the children in each group came from high SES families, and half from low SES families. In that study, a consistent finding across linguistic measures was that for English, high SES children had the early advantage, but for Spanish, low SES children in two-way schools or from homes in which only Spanish was spoken had the early advantage. These results are likely related to input factors: For English, the range of opportunities available to higher SES children for exposure to the English-dominant community were in all probability greater than those available to lower SES children; for Spanish, the low SES parents from homes in which only Spanish was spoken rated their own

134 

 Virginia C. Mueller Gathercole

command of English as low, and, consequently, their children may have been exposed to Spanish to a greater extent than the children of parents from higher SES backgrounds. Gatt and O’Toole’s (2013) more recent examination of vocabulary performance of bilingual infants from seven language pairs, mentioned earlier, found a significant correlation between the fathers’ educational levels and total vocabulary scores (in which the scores in the two languages were added together). In another study, Chiat and colleagues (2013) tested L2 bilingual children on a sentence-repetition task for several language pairs. Their findings show that in the case of Russian–German bilingual children’s performance on German, 17% of high SES bilingual children performed less than one standard deviation below the mean, but fully 41% of low SES bilingual children fell into that category; for Turkish–English bilingual children’s performance on English, 13% of low SES monolingual children fell below 1 to 2 standard deviations below the mean, as did 18% of the bilingual children, but another 70% of the low SES bilingual children fell more than 2 standard deviations below the mean. Even in highly proficient bilingual adults, SES may have long-standing effects on performance. In our recent study in Miami in which we examined educated bilingual adults’ performance on English and Spanish receptive vocabularies, mentioned earlier (Stadthagen-González et al., 2013), we found that although mean scores in both languages were at or above the norms in both languages, SES of the family in which the adult had been raised had an effect. English scores correlated with parental education for those who had grown up in ESH homes (in which parents spoke to them in both English and Spanish), but not for those from only Spanish-speaking homes (OSH) or L1 Spanish–L2 English homes. Spanish scores for the group who grew up in ESH homes also showed a positive correlation with parental educational level (but not for those from OSH homes), and a negative correlation for the L1S-L2E group. It is possible that higher SES families from the ESH homes placed more value on keeping their children bilingual, whereas those from immigrant groups may have placed higher value on language integration into the majority language. Such results on normally developing bilingual speakers are striking and could have important ramifications for bilingual speakers growing up under less favorable socioeconomic conditions. To examine the effects of SES more closely, we recently reexamined data on Welsh–English bilingual speakers’ performance on receptive vocabulary and receptive grammar in their two languages in relation to SES (Gathercole, Kennedy, & Thomas, 2015). In that study, parental educational levels and professions were coded as proxies for participants’ SES levels. Correlational and regression analyses were conducted to gauge the relative contributions of age, home language background, and SES on performance on the vocabulary and grammar measures. The results revealed that home language input and SES were highly predictive of performance. However, the relative influence of these two factors on performance depended on the measure and the age group considered. For vocabulary, home language exposure appears to be the greatest predictor of performance at the earliest ages through the primary school ages; but SES is also significant at many ages, and appears to play a more predictive role than home language at the later ages. For English vocabulary, SES was the greatest predictor



Factors Moderating Proficiency in Bilingual Speakers 

 135

of performance at the teen years and among older adults; for Welsh vocabulary, SES never contributed as much as home language to performance, but its role was relatively high especially at the teen years. For grammar measures, home language and SES were similarly predictive of performance at the early, primary, and teen years. Interestingly, a general pattern that emerged was that although home language and SES can influence performance at any age, home language may be more influential at younger ages, when perhaps more fundamental aspects of the language are being learned, whereas SES may be more influential at later ages, when more subtle fine-tuning of the language may be taking place. The predictive power of home language appears to be more persistent for Welsh, in that home language is the best predictor of performance on Welsh vocabulary through the teen years and on the Welsh grammar through the primary school years. In contrast, for English vocabulary, SES becomes a more powerful predictor of performance at the older years, especially the teen years and the older adult years and is the most powerful predictor of performance on the English grammar task at the primary school and teen years. It is possible that as the differences across home language groups become leveled with growing proficiency with age, the effects of SES become more visible and are perhaps more long-lasting; that is, at younger ages, the predictive value of SES may be partially masked by the strength of home language as a predictor.

General Discussion Numerous factors contribute to bilingual speakers’ development of proficiency in each of their languages. I have touched on quantity and nature of exposure, in the home and in the community, patterns of distribution of contexts of usage, profile effects and the role of linguistic complexity, possible interactions between the two languages being learned, and the relative role that socioeconomic level may play. All of these are natural outgrowths of the contexts in which bilingual children are acquiring their two languages and the nature of the language learning process itself. With sufficient, continued exposure to the two languages, bilingual speakers can be expected to achieve levels of proficiency similar to those of monolingual speakers, with a few qualifications on the nature of their knowledge and the organization of their linguistic systems, because of the distributed characteristic of bilingual acquisition and to possible interactions between the bilingual speaker’s two languages. The relative role that SES level may play in bilingual speakers’ language acquisition is worthy of further comment. The evidence for the influence of SES factors on acquisition of vocabulary and grammar is striking. The participants in the Miami study and our recent Welsh study were all typically developing children or typical adults. They were not growing up in poverty, nor were they from immigrant groups more typical in today’s shifting geographical boundaries. The impact of SES in these latter groups of bilingual speakers is worthy of further in-depth study in the light of

136 

 Virginia C. Mueller Gathercole

the results here. If the results are representative of the pervasiveness of SES influence on performance, we might expect the role of SES to play an even more significant role in the performance of children growing up under less advantageous conditions. What is clear is that these many factors that influence performance and proficiency in bilingual speakers can affect each individual learner in distinct ways, so that subtle variations can be expected across bilingual speakers and across bilingual populations. All of these should be kept in mind when considering questions such as how bilingual speakers compare with monolingual speakers, how to approach bilingual children and their abilities in the two languages, and what to expect from them in contexts that demand reliance on linguistic knowledge for other tasks, such as academic learning.

References Ameel, E., Malt, B. C., Storms, G., & van Assche, F. (2009). Semantic convergence in the bilingual lexicon. Journal of Memory and Language, 60, 270–290. http://dx.doi.org/10.1016/ j.jml.2008.10.001 Ameel, E., Storms, G., Malt, B., & Sloman, S. (2005). How bilinguals solve the naming problem. Journal of Memory and Language, 53, 60–80. http://dx.doi.org/10.1016/j.jml.2005.02.004 Bosch, L., & Ramon-Casas, M. (2014). First translation equivalents in bilingual toddlers’ expressive vocabulary: Does form similarity matter? International Journal of Behavioral Development, 38, 317–322. http://dx.doi.org/10.1177/0165025414532559 Bridges, K., & Hoff, E. (2014). Older sibling influences on the language environment and language development of toddlers in bilingual homes. Applied Psycholinguistics, 35, 225–241. http:// dx.doi.org/10.1017/S0142716412000379 Brown, A., & Gullberg, M. (2008). Bidirectional crosslinguistic influence in L1-L2 encoding of manner in speech and gesture: A study of Japanese speakers of English. Studies in Second Language Acquisition, 30, 225–251. http://dx.doi.org/10.1017/S0272263108080327 Chiat, S., Armon-Lotem, S., Marinis, T., Polisˇenská, K., Roy, P., & Seeff-Gabriel, B. (2013). Assessment of language abilities in sequential bilingual children: The potential of sentence imitation tasks. In V. C. M. Gathercole (Ed.), Issues in the assessment of bilinguals (pp. 56–89). Bristol, England: Multilingual Matters. David, A., & Wei, L. (2008). Individual differences in the lexical development of French-English bilingual children. International Journal of Bilingual Education and Bilingualism, 11, 598–618. http://dx.doi.org/10.2167/beb478.0 De Houwer, A., Bornstein, M. H., & Putnick, D. L. (2014). A bilingual–monolingual comparison of young children’s vocabulary size: Evidence from comprehension and production. Applied Psycholinguistics, 35, 1189–1211. http://dx.doi.org/10.1017/S0142716412000744 Dijkstra, T., & Van Hell, J. G. (2003). Testing the language mode hypothesis using trilinguals. International Journal of Bilingual Education and Bilingualism, 6, 2–16. http://dx.doi.org/10.1080/ 13670050308667769 Dijkstra, T., & Van Heuven, W. J. B. (1998). The BIA-model and bilingual word recognition. In J. Grainger & A. Jacobs (Eds.), Localist connectionist approaches to human cognition (pp. 189–225). Hillsdale, NJ: Erlbaum.



Factors Moderating Proficiency in Bilingual Speakers 

 137

Elston-Güttler, K., & Williams, J. N. (2008). First language polysemy affects second language meaning interpretation: Evidence for activation of first language concepts during second language reading. Second Language Research, 24, 167–187. http://dx.doi.org/10.1177/ 0267658307086300 Fennell, C., & Byers-Heinlein, K. (2014). You sound like Mommy: Bilingual and monolingual infants learn words best from speakers typical of their language environments. International Journal of Behavioral Development, 38, 309–316. http://dx.doi.org/10.1177/0165025414530631 Fillmore, L. W. (2000). Loss of family languages: Should educators be concerned? Theory Into Practice, 39, 203–210. http://dx.doi.org/10.1207/s15430421tip3904_3 García, G. E. (1991). Factors influencing the reading test performance of Spanish-speaking Hispanic students. Reading Research Quarterly, 26, 371–392. http://dx.doi.org/10.2307/747894 Gathercole, V. C. M. (2002). Monolingual and bilingual acquisition: Learning different treatments of thattrace phenomena in English and Spanish. In D. K. Oller & R. E. Eilers (Eds.), Language and literacy in bilingual children (pp. 220–254). Clevedon, England: Multilingual Matters. Gathercole, V. C. M. (Ed.). (2007a). Language transmission in bilingual families in Wales. Cardiff, Wales: Welsh Language Board. Gathercole, V. C. M. (2007b). Miami and North Wales, so far and yet so near: Constructivist account of morpho-syntactic development in bilingual children. International Journal of Bilingual Education and Bilingualism, 10, 224–247. http://dx.doi.org/10.2167/beb442.0 Gathercole, V. C. M. (Ed.). (2013a). Issues in the assessment of bilinguals. Bristol, England: Multilingual Matters. Gathercole, V. C. M. (Ed.). (2013b). Solutions for the assessment of bilinguals. Bristol, England: Multilingual Matters. Gathercole, V. C. M. (2014). Bilingualism matters: One size does not fit all. International Journal of Behavioral Development, 38, 359–366. http://dx.doi.org/10.1177/0165025414531676 Gathercole, V. C. M., Kennedy, I., & Thomas, E. M. (2015). Socioeconomic levels and bilinguals’ performance on language and cognitive measures. Bilingualism: Language and Cognition. Advance online publication. http://dx.doi.org/10.1017/S1366728915000504 Gathercole, V. C. M., & Moawad, R. (2010). Semantic interaction in early and late bilinguals: All words are not created equally. Bilingualism: Language and Cognition, 13, 385–408. http:// dx.doi.org/10.1017/S1366728909990460 Gathercole, V. C. M., Pérez-Tattam, R., Stadthagen-González, H., & Thomas, E. M. (2014). Bilingual construction of two systems: To interact or not to interact? In E. M. Thomas & I. Mennen (Eds.), Advances in the study of bilingualism (pp. 63–89). Bristol, England: Multilingual Matters. Gathercole, V. C. M., & Thomas, E. M. (2007). Prawf Geirfa Cymraeg, Fersiwn 7-11 (Welsh Vocabulary Test, Version 7-11). Retrieved from http://www.pgc.bangor.ac.uk Gathercole, V. C. M., & Thomas, E. M. (2009). Bilingual first-language development: Dominant language takeover, threatened minority language take-up. Bilingualism: Language and Cognition, 12, 213–237. http://dx.doi.org/10.1017/S1366728909004015 Gathercole, V. C. M., Thomas, E. M., & Laporte, N. (2001). The acquisition of grammatical gender in Welsh. Journal of Celtic Language Learning, 6, 53–87. Gathercole, V. C. M., Thomas, E. M., Roberts, E., Hughes, C., & Hughes, E. K. (2013). Why assessment needs to take exposure into account: Vocabulary and grammatical abilities in bilingual children. In V. C. M. Gathercole (Ed.), Issues in the assessment of bilinguals (pp. 20–55). Bristol, England: Multilingual Matters.

138 

 Virginia C. Mueller Gathercole

Gatt, D., & O’Toole, C. (2013, May). Parent report of early lexical production in bilingual children across varied contexts. Paper presented at Child Language Impairment in Multilingual Contexts, Krakow, Poland. Grosjean, F. (1998). Studying bilinguals: Methodological and conceptual issues. Bilingualism: Language and Cognition, 1, 131–149. http://dx.doi.org/10.1017/S136672899800025X Grosjean, F. (2001). The bilingual’s language modes. In J. L. Nicol (Ed.), One mind, two languages: Bilingual language processing (pp. 1–22). Oxford, England: Blackwell. Hart, B., & Risley, T. R. (1995). Meaningful differences in the everyday experience of young American children. Baltimore, MD: Brookes. Hoff, E., Core, C., Place, S., Rumiche, R., Señor, M., & Parra, M. (2012). Dual language exposure and early bilingual development. Journal of Child Language, 39(1), 1–27. http://dx.doi.org/10.1017/ S0305000910000759 Jean, M., & Geva, E. (2012). Through the eyes and from the mouths of young heritage-language learners: How children feel and think about their two languages. TESL Canada Journal/Revue TESL Du Canada, 29, 49–80. Jia, G., Chen, J., Kim, H. Y., Chan, P.-S., & Jeung, C. (2014). Bilingual lexical skills of school-age children with Chinese and Korean heritage languages in the United States. International Journal of Behavioral Development, 38, 350–358. http://dx.doi.org/10.1177/0165025414533224 Jiang, N. (2002). Form-meaning mapping in vocabulary acquisition in a second language. Studies in Second Language Acquisition, 24, 617–638. http://dx.doi.org/10.1017/ S0272263102004047 Jiang, N. (2004). Semantic transfer and its implications for vocabulary teaching in a second language. Modern Language Journal, 88, 416–432. http://dx.doi.org/10.1111/ j.0026-7902.2004.00238.x Kelley, A., & Kohnert, K. (2012). Is there a cognate advantage for typically developing Spanishspeaking English-language learners? Language, Speech, and Hearing Services in Schools, 43, 191–204. http://dx.doi.org/10.1044/0161-1461(2011/10-0022) Letts, C. (2013). What are the building blocks for language acquisition? Underlying principles of assessment for language impairment in the bilingual context. In V. C. M. Gathercole (Ed.), Solutions for the assessment of bilinguals (pp. 36–56). Bristol, England: Multilingual Matters. Malt, B. C., & Sloman, S. A. (2003). Linguistic diversity and object naming by non-native speakers of English. Bilingualism: Language and Cognition, 6, 47–67. http://dx.doi.org/10.1017/ S1366728903001020 Nagy, W. E., García, G. E., Durgunog l� u, A. Y., & Hancin-Bhatt, B. (1993). Spanish–English bilingual students’ use of cognates in English reading. Journal of Reading Behavior, 25, 241–259. http://dx.doi.org/10.1080/10862969009547816 Oller, D. K. (2005). The distributed characteristic in bilingual learning. In J. Cohen, K. McAlister, K. Rolstad, & J. MacSwan (Eds.), ISB4: Proceedings of the 4th International Symposium on Bilingualism (pp. 1744–1749). Somerville, MA: Cascadilla Press. Oller, D. K., & Eilers, R. E. (Eds.). (2002). Language and literacy in bilingual children. Clevedon, England: Multilingual Matters. Oller, D. K., Jarmulowicz, L., Pearson, B. Z., & Cobo-Lewis, A. B. (2011). Rapid spoken language shift in early second-language learning: The role of peers and effects on the first language. In A. Durgunog l� u & C. Goldenberg (Eds.), Language and literacy development in bilingual settings (pp. 94–120). New York, NY: Guilford Press.



Factors Moderating Proficiency in Bilingual Speakers 

 139

Oller, D. K., & Pearson, B. Z. (2002). Assessing the effects of bilingualism: A background. In D. K. Oller & R. E. Eilers (Eds.), Language and literacy in bilingual children (pp. 3–21). Clevedon, England: Multilingual Matters. Oller, D. K., Pearson, B. Z., & Cobo-Lewis, A. B. (2007). Profile effects in early bilingual language and literacy. Applied Psycholinguistics, 28, 191–230. http://dx.doi.org/10.1017/ S0142716407070117 O’Toole, C., & Hickey, T. (2013). Diagnosing language impairment in bilinguals: Professional experience and perception. Child Language Teaching and Therapy, 29, 91–109. http://dx.doi.org/ 10.1177/0265659012459859 Paradis, J., & Kirova, A. (2014). English second-language learners in preschool: Profile effects in their English abilities and the role of home language environment. International Journal of Behavioral Development, 38, 342–349. http://dx.doi.org/10.1177/0165025414530630 Patterson, J., & Pearson, B. Z. (2004). Bilingual lexical development: Influences, contexts, and processes. In G. Goldstein (Ed.), Bilingual language development and disorders in SpanishEnglish speakers (pp. 77–104). Baltimore, MD: Paul Brookes. Pavlenko, A. (2003). Eyewitness memory in late bilinguals: Evidence for discursive relativity. The International Journal of Bilingualism, 7, 257–281. http://dx.doi.org/10.1177/13670069030070030301 Pavlenko, A., & Malt, B. C. (2011). Kitchen Russian: Cross-linguistic differences and first-language object naming by Russian–English bilinguals. Bilingualism: Language and Cognition, 14, 19–45. http://dx.doi.org/10.1017/S136672891000026X Pearson, B. Z., Fernández, S. C., Lewedeg, V., & Oller, D. K. (1997). The relation of input factors to lexical learning by bilingual infants. Applied Psycholinguistics, 18, 41–58. http://dx.doi.org/ 10.1017/S0142716400009863 Peña, E. D., Bedore, L. M., & Fiestas, C. (2013). Development of bilingual semantic norms: Can two be one? In V. C. M. Gathercole (Ed.), Solutions for the assessment of bilinguals (pp. 103–124). Bristol, England: Multilingual Matters. Pérez, A. M., Peña, E. D., & Bedore, L. M. (2010). Cognates facilitate word recognition in young Spanish-English bilinguals’ test performance. Early Childhood Services, 4, 55–67. Pérez-Tattam, R., Gathercole, V. C. M., Yavas, F., Stadthagen-González, H., & Anrrich, G. (2013). Measuring grammatical knowledge and abilities in bilinguals: Implications for assessment and testing. In V. C. M. Gathercole (Ed.), Issues in the assessment of bilinguals (pp. 111–129). Bristol, England: Multilingual Matters. Place, S., & Hoff, E. (2011). Properties of dual language exposure that influence 2-year-olds’ bilingual proficiency. Child Development, 82, 1834–1849. http://dx.doi.org/10.1111/ j.1467-8624.2011.01660.x Poulin-Dubois, D., Bialystok, E., Blaye, A., Polonia, A., & Yott, J. (2013). Lexical access and vocabulary development in very young bilinguals. The International Journal of Bilingualism, 17, 57–70. http://dx.doi.org/10.1177/1367006911431198 Prior, A., MacWhinney, B., & Kroll, J. F. (2007). Translation norms for English and Spanish: The role of lexical variables, word class, and L2 proficiency in negotiating translation ambiguity. Behavior Research Methods, 39, 1029–1038. http://dx.doi.org/10.3758/BF03193001 Ribot, K. M., & Hoff, E. (2014). “¿Cómo estas?” “I’m good.” Conversational code-switching is related to profiles of expressive and receptive proficiency in Spanish–English bilingual toddlers. International Journal of Behavioral Development, 38, 333–341. http://dx.doi.org/10.1177/ 0165025414533225

140 

 Virginia C. Mueller Gathercole

Schelletter, C. (2002). The effect of form similarity on bilingual children’s lexical development. Bilingualism: Language and Cognition, 5, 93–108. http://dx.doi.org/10.1017/S1366728902000214 Smoczyńska, M. (1985). The acquisition of Polish. In D. I. Slobin (Ed.), The crosslinguistic study of language acquisition, Vol. 1: The data (pp. 595–686). Hillsdale, NJ: Erlbaum. Snow, C., & Hakuta, K. (1992). The costs of monolingualism. In J. Crawford (Ed.), Language loyalties (pp. 384–394). Chicago, IL: University of Chicago Press. Stadthagen-González, H., Gathercole, V. C. M., Pérez-Tattam, R., & Yavas, F. (2013). Vocabulary assessment of bilingual adults: To cognate or not to cognate. In V. C. M. Gathercole (Ed.), Solutions for the assessment of bilinguals (pp. 125–145). Bristol, England: Multilingual Matters. Thomas, E. M., & Gathercole, V. C. M. (2007). Children’s productive command of grammatical gender and mutation in Welsh: An alternative to rule-based learning. First Language, 27, 251–278. http://dx.doi.org/10.1177/0142723707077056 Thomas, E. M., Gathercole, V. C. M., & Hughes, E. K. (2013). Sociolinguistic influences on the linguistic achievement of bilinguals: Issues for the assessment of minority language competence. In V. C. M. Gathercole (Ed.), Solutions for the assessment of bilinguals (pp. 175–193). Bristol, England: Multilingual Matters. Tokowicz, N., Kroll, J. F., De Groot, A. M. B., & Van Hell, J. G. (2002). Number-of-translation norms for Dutch–English translation pairs: A new tool for examining language production. Behavior Research Methods: Instruments & Computers, 34, 435–451. http://dx.doi.org/10.3758/ BF03195472 Wolff, P., & Ventura, T. (2009). When Russians learn English: How the semantics of causation may change. Bilingualism: Language and Cognition, 12, 153–176. http://dx.doi.org/10.1017/ S1366728909004040

Silvina Montrul

8 Age of Onset of Bilingualism Effects and Availability of Input in First Language Attrition One of the most salient linguistic characteristics of immigrant populations across the world is that first-language skills are often affected in language contact situations. After reaching some degree of proficiency in both languages, bilingual speakers in these situations might become more dominant in the second language, they might gradually reduce their first-language use, and eventually exhibit changes or lose some of their abilities in this language. This chapter addresses two main variables that affect the degree and structure of first-language attrition and loss: age of acquisition and availability of input.

What Is Language Attrition? Nowhere is the dynamic complexity of first- and second-language learning and forgetting more evident than in situations of mass migration. Adult immigrants with more than 10 years of immersion in a second-language environment can experience first-language attrition and forget words and structural properties of their native language and experience changes in their pronunciation. Schmid (2011) defined language attrition as “the (total or partial) forgetting of a language by a healthy speaker” (p. 3). Language attrition is the loss of language proficiency and skills due to reduced/ interrupted exposure to and use of the language, and this can occur in the first language (L1) and in the second language (L2) after a period of disuse. In general, when measured quantitatively, language attrition in immigrant adults in vocabulary, morpho­syntax, and pronunciation is relatively minor and subject to high individual variation (Hopp & Schmid, 2013; Schmid, 2014; Schmid & Jarvis, 2014). By contrast, the consequences of learning the second majority language early are more dramatic for the grammatical development and stability of their children’s native languages. It has been amply documented that many heritage speakers exposed to two languages from birth do not fully acquire their family language under pressure from the majority societal language, and many of the resulting grammatical characteristics of the heritage language in children of immigrants is very similar to nonnative patterns typical of second-language learners in a nonimmigrant situation (Montrul, 2008; in press). When some of these immigrant families return to their home countries, their prepuberty children can experience heritage language reversal and second language http://dx.doi.org/10.1037/14939-009 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

142 

 Silvina Montrul

attrition soon after arrival (Flores, 2010). What used to be the heritage language (e.g., Japanese in the United States) turns into the primary and dominant language in the home country (e.g., Japanese in Japan), whereas the primary and dominant language in the immigrant situation (e.g., English in the United States) now becomes the second and secondary language in the home country (e.g., English in Japan). In this chapter, I discuss empirical evidence drawn from language loss in immigrant situations and in international adoption to illustrate two central factors that affect the degree and extent of language attrition at the grammatical level in heritage speakers: (a) the age when input is reduced and (b) the degree of accessibility to some input.

Language Attrition in an Immigration Context Mature native speakers show full development of different components of their language and are fluent when they speak. Proficiency is actual grammatical ability and fluency in a language at the level of vocabulary, pronunciation, morphology, syntax, semantics, discourse and pragmatics (Montrul, in press). The Chomskyan view of language acquisition portrays becoming a native speaker as a fast and efficient process, with an effortless transition from child to adult grammar (Crain, 1991). An implicit assumption of this model has been that once the native language is acquired by the end of the preschool period, its structural integrity is stable. In contrast to this view, Berman (2004) considered that becoming a proficient native speaker takes a long time. Many linguistic forms, even those that emerge at early preschool age, have a long developmental history to become acquired and mastered. Berman (2004) showed that the language of 9- and 10-year-old children differs markedly from those of adults, not only in content but also in morphosyntax and lexicon. Crucial to understanding later language development is the distinction between emergence, acquisition, and mastery. For Berman (2004), a native speaker is somebody who has mastered and become proficient in several dimensions of knowledge and use of the native language through a constant interaction of competence and performance. Several general properties characterize changes from emergence to mastery in Berman’s view. For example, children may have acquired the grammar of number and gender agreement yet still be at an “item-based” phase of person marking (Tomasello, 2001)—what for Berman would be partial knowledge. Similarly, a child who produces a given form (i.e., tense marking) in 60% of obligatory contexts, whereas adults produce it with 100% accuracy can be considered to have incomplete knowledge of tense marking because he or she does not show full morphological mastery. Language attrition research has amply demonstrated that as they learn the L2 and become proficient in it, some adults immersed in a L2 environment for more than 10 years, if not longer, can lose linguistic ability in their own native, family language (Iverson, 2012; Major, 1992). Adults undergoing L1 attrition exhibit reduced fluency (Schmid, 2011): They tend to talk hesitantly and slowly, with many pauses, repetition of words, self-corrections, and word finding difficulties. They resort to code-switching to make up for lexical gaps, and they avoid complex constructions like embedded



Age of Onset of Bilingualism Effects 

 143

clauses or passives. They may exhibit a foreign accent and what many fluent native speakers would consider grammatical and lexical errors (Ecke & Hall, 2013). The following examples are from a 35-year-old Spanish speaker who demonstrated significant attrition; she was adopted from Guatemala at age 9 by an American family (Montrul, 2011). Her pronunciation in Spanish sounded native, with occasional interference from English. Yet, her speech was nonfluent and labored when it came to lexical retrieval and grammatical accuracy. She compensated for word-finding difficulties with codeswitching and borrowings from English, as seen in the examples below (underlined words represent grammatical errors, disfluencies, and code-switching): 1. Como die-, ok, te- tenía teng- tenía casi diez años cuando mis padres americanos me adoptaron, so viví ahí en Guatemala hasta hasta que tenía completo nueve años. [I was almost 10 years when my American parents adopted me, so I lived there in Guatemala until until I had complete nine years.] 2. Sí, un poquito, mi abuelito nos cuidan mí a mi hermana y yo, nosotros estamos adoptado juntas [Yes, a little bit, my grandaddy took care of me and my sister. We were adopted together.] An important question is whether attrition mainly occurs at the level of processing and linguistic memory, or whether it affects actual linguistic competence and representations of phonemes, inflectional and derivational morphology, and core aspects of syntax and semantics. At least for immigrant adults, who experience intense exposure to the L2 in adulthood well after the time linguistic maturity was reached (after age 18), as in Figure 8.1, attrition mostly affects the level of processing and activation rather than deeper levels of linguistic representation (Montrul, 2008).

100

Language Proficiency

90 80 70 60 50 40 30 20

before migration (childhood and adolescence)

aer migraon (adulthood)

first language (L1) second language (L2)

10 0 4 5 6 7 8 9 10 11 12 13 14 16 17 18 20 25 30 35 40 45 50 55 60 65 70 Age (in years) Fig. 8.1: Idealized bilingual development of an adult undergoing L1 attrition in the context of L2 acquisition.

144 

 Silvina Montrul

However, language attrition has been shown to affect aspects of syntax and the syntax-pragmatics interface, including overproduction of overt subjects in pragmatically infelicitous contexts in speakers of null subject languages (Sorace, 2000; Tsimpli, 2007), and in the interpretations of pronouns and anaphors (Gürel, 2004; Kim, Montrul, & Yoon, 2010). Perpiñán (2013) recently reported changes in word order in Spanish related to stylistic variation also regulated by the syntax-discourse interface. When these structures are measured through different tasks there are quantitative differences, if minor, between L1 attriters and native speakers who are not attrited. When the L1 grammar appears to undergo changes such as these—as with grammatical phenomena at the syntax-discourse interface—many of the effects seem to be induced by the L2. Bylund (2009) reported changes in the expression of motion events in the L1 Spanish of speakers living in Sweden, which seem to be influenced by the way the same events are syntactically expressed in Swedish, their L2. The Spanish speakers undergoing attrition in his study verbalized endpoints in Spanish like they would do in Swedish, and did not use the structures typically used by native speakers of Spanish to describe the same events. Gürel (2002) also attributed the changes with pronoun interpretation in Turkish speakers living in the United States and Canada as related to L2 influence on the L1. Schmid (2011) and Sorace (2000) maintained that not all changes found in attrition can be explained by L2 influence on the L1, and Perpiñán (2013) acknowledged that in her data it is hard to tease the role of the L2 from other cognitive and universal processing mechanisms. Ideally, attrition should be documented longitudinally, like fossilization in L2 acquisition (Lardiere, 2007), where it can be established that at Time 1 individual x had age-appropriate knowledge of language y, and at Time 2, 3, or 4 knowledge of a given feature of y is unstable and shows errors or said feature of y is no longer produced/ known as judged by native speaker norms. However, in attrition studies this ideal design has not been frequently adopted, and in most cases it is not even feasible. A recent exception is Ecke and Hall’s (2013) case study of lexical retrieval (see also Opitz, 2013). That is why, typically, adults who grew up in a predominantly mono­ lingual environment are an ideal population on which to test potential attrition, because it can be assumed that they reached linguistic maturity before attrition set in. The potentially attrited L1 is compared or measured against the nonattrited version of the L1 as spoken by adult fluent bilingual or monolingual speakers not living in an L2 environment (Major, 1992; Montrul & Sánchez-Walker, 2013). Another reason why longitudinal studies are not very common is because attrition is subject to high individual variation: Some speakers show signs of attrition and others do not, and it would be hard to tell in advance who is going to show signs of attrition if a study were to be planned. For example, Montrul and Sánchez-Walker (2013) found that in a group of 23 adult immigrants from Mexico, 13 had native knowledge of the fact that animate objects in Spanish must be marked by the preposition a (as in Juan vio a María [Juan saw Mary] vs. *Juan vio María; as measured in production and grammaticality judgment tasks), whereas 10 exhibited omission errors. Those



Age of Onset of Bilingualism Effects 

 145

who made errors were older than the immigrants who did not make errors, acquired English later in life, and had resided in the United States longer. The two immigrants with the lowest accuracy had been in the United States for over 30 years. In general, adults undergoing attrition in an immigrant setting retain the ability to understand and use the language at advanced levels. It appears that reduced input and even disuse of the language for several years in adults do not seem to affect the integrity of the native grammar substantially (Schmid, 2007). In many reported cases, the effects of L1 attrition in adults have been minimal: After more than 50 years of language disuse Schmid (2002) found that German Jews living in the United States exhibited some transfer from English but very few actual morphosyntactic errors that could be attributed to L1 attrition. No adult undergoing attrition in a bilingual environment has been shown to regress in their language to such an extent as to forget how to conjugate verbs, ask questions, or produce and discriminate native sounds (Keijzer, 2007). However, if immigration happens before puberty, the native language of these children is less likely to be preserved at native levels in the new environment, and attrition in these cases does affect significantly aspects of morphology and syntax. As we will see next, the extent of attrition varies significantly depending on the age of the speaker when input is reduced, as in heritage languages, or interrupted, as in international adoptees.

Age Effects and Availability of Input in L1 Attrition Heritage Languages The impact of age on language attrition is easily observed in heritage languages and across generations, as immigrants become proficient in the majority language, as shown in Table 8.1. First generation immigrant adults are mature, fluent native speakers of the languages spoken in their countries. Some of them may have some knowledge Tab. 8.1: Patterns of Language Dominance and Proficiency of Heritage Speakers and the Parental Generation Generation

Possible language characteristics

1st (parents)

Dominant in the native language

2nd (children)

Dominant in the majority language

3rd (grandchildren)

Dominant in the majority language

Nonnative proficiency in the majority language Low-to-high proficiency in the heritage language Ranges from intermediate-low proficiency in the heritage language to monolingual in the majority language

146 

 Silvina Montrul

of the language spoken in the host country, whereas others learn it gradually as a second language when they start their life in the new environment. The heritage speakers are the children of adult immigrants: Some move with their parents in childhood (immigrant children), whereas others are born in the new country and exposed to the heritage language and the majority language since birth (children of immigrants). These children are the second generation and frequently grow up bilingual. The children of the second generation and the grandchildren of the first generation are the third generation, and they are also considered heritage speakers if they still have some knowledge of the heritage language, although many are monolingual speakers of the majority language already. By the fourth and subsequent generations, the heritage language is rarely used in the family. This pattern of declining bilingualism within families is common in many parts of the world even when the heritage languages are supported by continuous immigration (Sills & Hall, 2005). Most heritage speakers are simultaneous bilingual children or early sequential bilingual children, that is, child second-language learners. The vast scholarship on age effects in L2 acquisition supports the beliefs that children are better L2 learners than adults and that starting acquisition of a L2 in childhood significantly enhances the chances of reaching nativelike ability in the L2 later on (assuming, of course, sufficient and sustained exposure and use of the language). Support for these beliefs comes from studies of adults at time of testing, who were immigrant children. For example, the participants in the foundational Johnson and Newport (1989) study were Korean immigrants or children of Korean immigrants in the United States. Similarly, the Chinese immigrants in Spain studied by Granena and Long (2013) and the Spanish-speaking immigrants in Sweden studied by Abrahamsson and Hyltenstam (2009) were immigrants or children of immigrants learning English, Spanish, or Swedish, the majority and national languages. Child L2 learners immersed in a situation in which the L2 is the majority language tend to lose linguistic ability in their own L1, especially when they do not receive academic instruction in their heritage languages during the school-age period (Jia & Aaronson, 2003; Merino, 1983; Montrul, 2008; Montrul & Potowski, 2007; Yeni-Komshian, Flege, & Liu, 2000). When these children start school in the majority, official language (the L2), input and use of the L1 decrease as input and use of the L2 substantially increase, progressively leading to language shift. Eventually, the L2 takes over in frequency and domains of language use. The patterns of language proficiency and dominance are such that the L2 continues developing, whereas linguistic ability in the L1 diminishes or stabilizes at suboptimal levels of linguistic development, and this has consequences for their native language maintenance. Although basic vocabulary, phonology, and basic syntax and semantics may be in place by 3 or 4 years of age, the road to native-language proficiency and mastery is longer, extending to adolescence (Berman, 2004). More specifically, grammatical properties have different developmental schedules: Some are acquired in infancy and others during the age of later language development at school. For example, across



Age of Onset of Bilingualism Effects 

 147

languages that mark grammatical case overtly (e.g., nominative, accusative, dative), children gain command of simple case marking relations between ages 2 and 3 years (Perdue & Bowerman, 1990). Hebrew-speaking children acquire the direct object market -et (accusative, definite) by age 2. Ravid (2004) showed that Hebrew speaking children know and use the passive participle –u by ages 3 to 4. Yet, the acquisition of passive morphology in Hebrew is not fully mastered until age 9. In a test administered to school-age children of different ages Ravid showed that 6-year-olds gave consistently nonpassive responses when presented with obligatory contexts for passive formation. Speakers of Hebrew tend to avoid the passive in colloquial speech. Only 11- and 12-year-olds regularly provided passive constructions where required on the same test. Therefore, the morphology and syntax acquired during early language development are not mastered until later. Another example comes from Spanish. The subjunctive mood emerges with some expressions and clauses very early, at 2 years of age (López Ornat, 1994). Yet, monolingual children do not develop full command of the subjunctive meanings and uses until they are about 12 years old (Blake, 1983). Given the nature of language development, it is difficult to establish a posteriori whether a simultaneous or sequential bilingual young adult shows attrition or incomplete acquisition (Montrul, 2008). Attrition implies that some grammatical property was fully acquired and then lost or weakened, as we exemplified with adults. Incomplete acquisition may mean that a given grammatical property was never actually acquired, or that it was acquired but not fully mastered, and therefore it weakened and was subsequently lost. An example of a longitudinal study that teases apart these potential outcomes with Spanish-heritage speakers in the United States is Anderson (1999). Anderson looked at the decline of gender morphology in two normally developing Spanish-speaking siblings—Beatriz and Victoria. The siblings came to live to the United States at the ages of 3.5 years (Beatriz) and 1.5 years (Victoria) with their parents. Data collection consisted of naturalistic observations and started almost 3 years after arrival, when the children were 6 years, 7 months and 4 years, 7 months old, and ended 2 years later, when the children were 8 years, 5 months and 6 years, 5 months old. The two parents were from Puerto Rico, and Spanish was spoken at home. At the beginning of the study, the siblings spoke Spanish and English with each other, but toward the end of the data collection period, the children shifted to using only English with each other and Spanish with their parents. When data collection started, the older sibling had some literacy skills in Spanish, and during the course of the study the two children developed literacy skills in English in day care and preschool. The siblings were videotaped in play sessions with their parents every 1 to 2 months over a period of 22 months. The recordings were later transcribed, coded, and analyzed. Anderson focused on the development of gender agreement in noun phrases, which monolingual children control with close to 100% accuracy by the age of 3 years (Hernández Pina, 1984; Pérez-Pereira, 1991), and found that the two siblings made errors with gender. The percentage of errors with gender increased markedly in time for the two children, as shown in Figure 8.2.

Mean Percentage of Errors

148 

20 18 16 14 12 10 8 6 4 2 0

 Silvina Montrul

18.6

8.5

Victoria (4–6 years) 5.8

0 first recording

Beatriz (6–8 years)

second recording (2 years later) Time

Fig. 8.2: Mean percentage of errors with gender agreement in Spanish noun phrases in two Spanish–English bilingual children followed longitudinally. From “Loss of Gender Agreement in L1 Attrition: Preliminary Results,” by R. T. Anderson, 1999, Bilingual Research Journal, 23, p. 396. Copyright 1999 by Routledge. Adapted with permission.

At the time of the first recording, Beatriz—the older sibling—produced gender agreement 100% correctly (0% error), like a typical monolingual child her age (see Montrul, 2004). Two years later, she was producing 5.8% errors, which is very high for a native speaker. Victoria produced 8% errors during the first recording, which suggests that she had not yet mastered gender agreement at age-appropriate levels. By the end of the data collection period, Victoria’s gender error rates increased steadily to 17.4%, 18.2%, and 25%, in the last three sessions. That is, her error rate almost tripled. There is measurable change in time in the heritage language with respect to gender agreement, and each child is the baseline or control for herself. It is also possible for bilingual children who receive less input in the heritage language than in the majority language to exhibit full acquisition of some areas of the grammar, attrition in other areas, and yet incomplete acquisition of others. SilvaCorvalán (2014) followed two simultaneous bilingual children longitudinally for almost 5 years (ages 1 year; 10 months to 5 years; 11 months), focusing in this study on the same grammatical areas investigated in her study with the three generations of Mexican immigrants and Mexican Americans in Los Angeles (Silva-Corvalán, 1994). The two siblings developed age-appropriate command of tense in English, including tense morphology, modals, and auxiliaries. Both siblings followed a similar path of development in their stronger language and achieved essential verb system milestones at about the same age as English monolingual speakers. However, the development of the tenseaspect and mood (TMA) system of Spanish was far from nativelike. During the period from 2 years, 0 months old to 2 years, 11 months old the siblings made the typical tense and person agreement errors characteristic of monolingual child language develop-



Age of Onset of Bilingualism Effects 

 149

ment (Fernández Martínez, 1994; Gathercole, Sebastián, & Soto, 1999). Yet, Bren, the child with lower exposure to Spanish, showed weaker control of person agreement than Nico. Silva-Corvalán (2014) found that the tenses that are relatively more complex, and also less frequent in the adult input were acquired later or not at all by age 6 years, 0 months. Bren, the sibling with less exposure to Spanish, developed a more reduced tense system than his older brother Nico. At age 3 years, 6 months he used the present subjunctive but by age 6 years, 0 months he no longer used it, and he never developed the other complex tenses in Spanish. When these bilingual Spanish-speaking children started kindergarten in the United States, they had not yet acquired the complete TMA system in Spanish, like monolingual peers. The children also showed full knowledge of other aspects of the Spanish grammar. Like these siblings studied by Silva-Corvalán (2014), many children from immigrant families learning English in the United States experience subtractive bilingualism, and thus do not usually become balanced bilingual speakers later in childhood or as adults (Oller & Jarmulowicz, 2009), even when in some cases they started as such (Silva-Corvalán, 2014). The heritage language exhibits attrition and incomplete acquisition of several properties. Figure 8.3 shows a common pattern in heritage speakers who were simultaneous bilingual speakers in early childhood. The figure shows that their language development was balanced during early infancy, like Silva-Corvalán (2014) documented in the two children she studied, but the development of the languages diverges after age 5 to 6 years when significant exposure to the L2 increases. Some of these heritage speakers seek to relearn the L1 language in the classroom when they reach college age, and in some cases they make progress in some areas as adults, as the figure attempts to capture. 100 90

infancy

adulthood

childhood and adolescence

Language Proficiency

80 70 60 50 40 30

heritage language (L1)

20

majority language (simultaneous bilinguals)

10

bi

Age (in years)

Fig. 8.3: Idealized bilingual development of a simultaneous bilingual undergoing minority language decline.

70

50 60

40

25

18

17

16

14

13

11 12

9 10

8

7

5

6

4

rth 12 m 18 m 24 m 30 m 36 m

0

150 

 Silvina Montrul

Figure 8.4 shows a typical scenario of a child who grew up monolingually in the heritage language during the preschool period, and began acquiring the L2 at school, also to the detriment of the development of the L1. If this is an immigrant child, the child may have had an opportunity to attend school in the country of origin and in his native language. The language shift is clear in this graph, as the two lines cross over. These general tendencies, however, do not imply that achieving relatively balanced bilingualism after age 4 is not possible; it is just apparently not very common in the United States, although it appears to happen in Europe (Kupisch, Akpinar, & Stöhr, 2013). Furthermore, in countries where there is institutional support of the two languages and the languages are used frequently in daily life, relatively balanced bilingualism may be more likely, as in some areas in Canada. What leads to L1 attrition in childhood is reduced input and lack of consistent and sustained exposure to and use of the L1 during a time when the native language is not fully developed in the brain, most likely before and around the closure of the critical period (puberty). The L1 is used less because children growing up in a L2 environment spend most of their waking hours using the L2 at school and with peers, at the expense of the L1. The studies documenting extensive effects of attrition at the lexical, phonological and morphosyntactic levels are about children (Kaufman & Aronoff, 1991; Turian & Altenberg, 1991) or about adults who immigrated in childhood (Polinsky, 2006; Vago, 1991), suggesting important differences because of age of intense exposure to the L2 and reduced use of the L1. Bylund (2009) and Montrul (2008) independently showed that the younger the individual when reduction to input and lack of use of the L1 take place, the more

100 90

infancy

adulthood

Language Proficiency

80 70 60 50 40

first language (L1)

30

second language (L2)

20 10

birth 12m 18m 24m 30m 36m 4 5 6 7 8 9 10 11 12 13 14 16 17 18 20 25 30 35 40

0 Age (in years) Fig. 8.4: Idealized bilingual development of an early sequential bilingual undergoing language shift.



Age of Onset of Bilingualism Effects 

 151

severe the extent of language loss at the grammatical level, such that the effects of L1 attrition in childhood are more dramatic than in adulthood. Montrul (2008) also showed that within childhood and in a minority language context, simultaneous bilingual children are more vulnerable to attrition than sequential bilingual children, because sequential bilingual children were exposed to their L1 for a longer period of time than simultaneous bilingual children (see also Montrul, 2002). Figure 8.5 shows the hypothesized function between proficiency level in the first language and age of onset of exposure to the L2 to illustrate degree of language loss in different bilingual populations. The study of L1 attrition during childhood challenges the claim that language acquisition is largely complete by age 3 or 4 because it shows that the acquired linguistic competence is not very stable at this early age. L1 attrition also raises the question of when exactly a child becomes a stable speaker of his or her language so that grammatical competence remains relatively unaffected by fluctuations in input and lack of use of the language for extended periods of time. Possible answers vary. Considering research on L1 and L2 acquisition, and on existing research on the interplay of the two languages in bilingual children, Montrul (2008) proposed middle childhood, between ages 8 and 10, as a likely age of language fixation (see also Köpke & Schmid, 2003). However, Bylund (2009) based his estimations on existing maturational effect trends for L2 acquisition, hypothesizing that age 12—the onset of puberty—is the critical age for L1 attrition. Thus, it appears that

high

adult L2 learners (1st generation immigrants)

Proficiency

sequential bilinguals

simultaneous bilinguals

low

foreign adoptees

younger

Age

older

Fig. 8.5: Positive correlation between age and proficiency for L1 loss. From Incomplete Acquisition in Bilingualism: Re-Examining the Age Factor (p. 266), by S. A. Montrul, 2008, Amsterdam, the Netherlands: John Benjamins. Copyright 2008 by John Benjamins. Reprinted with permission.

152 

 Silvina Montrul

the period before and after puberty is relevant for language entrenchment. This does not mean that if a child reaches age 8 or 9 when extensive exposure and use of the L2 begins he or she will retain the L1 intact the rest of his or her life irrespective of input. What it means is that vulnerability to significant attrition because of reduced L1 exposure and use of the language is less likely at this age than at a younger age, and that substantial retention of the language is therefore more likely after this age. More empirical research is needed to investigate these hypothetical ages for vulnerability to attrition and their interaction with other factors more directly. For example, literacy has also been shown to reduce the likelihood of L1 attrition in children (Zaretsky & Bar-Shalom, 2010). More recent support for the hypothesis that there are age effects for attrition in children was provided by Flores (2012), who investigated the effects of age and quantity of input in a case of heritage-language reversal. The target group consisted of Portuguese–German bilingual speakers of Portuguese heritage who grew up in a country where German is the majority language (Germany and Switzerland). The German heritage speakers returned to Portugal some time in childhood, and at the time of testing, Portuguese was now the majority language and German was the minority, or heritage, language in Portugal. All the returnees had attended school in German when they lived in Germany and Switzerland. The returnees were divided into two groups, based on age of return. In the first group (child returnees) there were speakers with a mean age of 16.8 years at time of testing and a mean of 8.4 years at the time of return (range 7–10). The second group (postpuberty returnees) consisted of speakers who returned to Portugal between the ages of 12 and 14 years (mean 12.6) and were between 18 and 22 years (mean 20.9) at the time of testing. All claimed to be more fluent and dominant in Portuguese than in German. The goal of the study was to document potential grammatical attrition of German (the L2 acquired naturalistically and the language of schooling) and focused on two properties: object expression (clitic pronoun vs. null object) and word order (verb second [V2]). It is possible to drop objects in Portuguese, especially when the object is a topic and must be identified by the discourse. In German, it is also possible to drop an object when it is a topic, but topic drop in German is far more restricted than in Portuguese and must obey syntactic and pragmatic conditions. With respect to verb placement, German is a V2 language and the verb in matrix sentences must be in second position. This is a strictly syntactic requirement. Portuguese is not a V2 language, having SVO order. Object expression and verb placement were elicited through an oral task. If the returnees dropped objects in German in contexts where topic drop is not allowed, and if they did not respect the V2 rule for verb placement, these would be signs of attrition (as reflected by influence of Portuguese on German). The study was not longitudinal, and there are no data on these returnees before their return. For baseline groups, Flores (2012) recruited a group of adult Portuguese immigrants with knowledge of German who had recently returned to Portugal, and



Age of Onset of Bilingualism Effects 

 153

child Portuguese heritage speakers in Germany, who would be the proxy for the returnees if they had been tested while in Germany as children. Table 8.2 summarizes the main findings. The results show that the two returnee groups produced significantly more illicit null objects in German (*null object) than the adult and child control groups, who still had contact with German. The returnees seem to have imposed the Portuguese pragmatic and syntactic conditions on object drop to objects in German. More revealing, however, are the findings of verb placement, where the difference in production of illicit orders between the returnees who returned in childhood versus those who returned after puberty is striking: the child returnees showed 47.3% and 45.4% errors with *XPSV and *Vnfinal, the postpuberty returnees 3.5% and 2.5%. These findings show that for core aspects of syntax, like V2, reduced input in childhood significantly affects the integrity of the core grammar in heritage speakers of German. When contact with the language is reduced or interrupted after puberty, the effects of attrition on the core grammar are relatively minor. Through the use of different age groups in this cross-sectional study, Flores (2012) confirmed age effects on L1 attrition, supporting maturational explanations for language loss (Bylund, 2009; Montrul, 2008). To conclude this section, children who immigrate with their parents—heritage speakers—continue to have exposure to the family language to different degrees. Some develop very advanced proficiency in the heritage language and some others show variable outcomes of nonnative acquisition. Even if the children do not use the language frequently, some of them may end up being receptive bilingual speakers (Au, Knightly, Jun, & Oh, 2002; Sherkina-Lieber, 2011) who do not necessarily speak the heritage language but have some understanding of it. In contrast with heritage speakers who may have access to their heritage language, internationally adopted children are often adopted by families who do not speak their language, and the degree of attrition they experience is much greater, as shown in Figure 8.4. Tab. 8.2: Mean Percentages of Object and Verb Placement Constructions in German Object expression Group

Clitic

Child returnees Postpuberty returnees Adult control (Portugal) Child control (Germany)

41.5% 50.6% 88% 90.2%

Illicit verb placement

*Null object

Topic-drop

*XPSV

*Vnfinal

47.3% 34.4% 1.9% 0%

11.2% 15% 10.1% 9.8%

47.3% 3.5% 0% 0%

45.4% 2.5% 0% 0%

Note: *XPSV are ungrammatical sentences with the V in the third position, following XP and S; *Vnfinal are ungrammatical sentences where the verb is not in final position in embedded clauses. From “Differential Effects of Language Attrition in the Domains of Verb Placement and Object Expression,” by C. Flores, 2012, Bilingualism: Language and Cognition, 15, pp. 559 and 562. Copyright 2011 by Cambridge University Press. Adapted with permission.

154 

 Silvina Montrul

International Adoptees A key difference between heritage speakers and international adoptees is that input in the first language is interrupted abruptly right after adoption for most adoptees. In general, 80% of internationally adopted children are infants, 10% are adopted between ages 4 and 5 years, and the remaining 10% at age 6 years or older (Glennen, 2005; Hyltenstam, Bylund, Abrahamsson, & Park, 2009). Existing studies suggest that these children tend to lose their language easily and faster than heritage language speakers who are immigrant children and in some cases with very limited language retention depending on age of adoption (Pallier et al., 2003; Pierce, Klein, Chen, Delcenserie, & Genesee, 2014). For this reason, internationally adopted children are rarely functionally bilingual, representing instead a unique case of language acquisition, by which one language is actually replaced by another as they go from monolingualism in the L1 to monolingualism in the L2. Language loss occurs more rapidly, within the first year after adoption, in infants than in toddlers, and more rapidly in toddlers than in older children (Gindis, 1999; Glennen, 2005). Two longitudinal case studies of a Chinese infant (Nicoladis & Grabois, 2002) and of a 9-year-old Russian girl (Isurin, 2000) confirm these age effect trends and speed in vocabulary loss. The Chinese infant in Nicoladis and Grabois (2002) was unable to produce and understand Chinese words after 3 months in Canada. The Russian girl studied by Isurin (2000) showed 60% decline in lexical retrieval after one year of residence in the United States. Although it seems that the girl still retained some knowledge of Russian, Isurin also reported dramatic decline in production and comprehension of the language during childhood. The Spanish-speaking adoptee described earlier was adopted at age 9 (Montrul, 2011). In recent years there has been some interest in the extent of potential language retention in speakers who were adopted internationally. Two intriguing recent studies addressing these questions are Pallier et al. (2003) and Ventureyra (2005). They reported that 20 Korean adults (age at testing ranging from 20 to 32 years) adopted by French families between the ages of 3 and 10 had become nativelike in French but had no recollection of Korean basic vocabulary or sounds. Neuroimaging and behavioral tests revealed no traces of Korean even in speech perception in these speakers, who did not differ from native speaker French controls in any of the behavioral or brain imaging measures. There is some suggestive evidence that some adoptees do retain remnants of that first language, especially if they were adopted after infancy. Hyltenstam et al. (2009) and Oh, Au, and Jun (2010) tested adopted Koreans who were learning Korean as an L2 in the classroom, and compared them to instructed L2 learners of Korean who had not been exposed to Korean before. Although the Korean adoptees had no productive ability in Korean before enrolling in these classes, both studies found that the Korean adoptees were significantly better at discriminating phonetic contrasts of voice onset time in Korean than the L2 learners with no previous knowledge of Korean. Hyltenstam



Age of Onset of Bilingualism Effects 

 155

et al. also found that the highest performing adoptees in Korean were those with the highest age of adoption (9 and 10 years). This advantage was only apparent in phono­ logical discrimination but not in grammaticality judgments of morphosyntactic aspects of Korean, where the L2 learners outperformed the Korean adoptees. The fact that reexposure effects were observed in the Hyltenstam et al. and in the Oh et al. studies suggests that remnants of the L1 must lie in the mind in some sort of representation, and that total L1 loss is unlikely in these cases. It is also possible that the apparent age of adoption effect on language retention may be because of the length of time elapsed between the interruption to L1 input and the reexposure, such that for those adoptees who experienced a shorter span between interruption and reexposure the degree of retention or recollection may be stronger. Another recent study showing retention of perceptual correlates of tones in Mandarin is Pierce et al. (2014). Using functional MRI, the researchers showed that internationally adopted children from China living in Quebec and exposed to French since adoption maintained neural representations of their birth language despite functionally losing the language and not having conscious recollection of it. Montrul (2011) presented a detailed analysis of a Spanish-speaking adoptee’s morphosyntactic competence in a variety of oral and written tasks. The subject of the case study was a 34-year-old woman, Alicia, from Guatemala, adopted at age 9 by a family from the United States. Alicia had not used Spanish since her adoption but regained some of its use in college. At the time of testing, she self-identified herself as a person of Hispanic origin and in her daily life had strong connections to other Spanish speakers in the community. Compared with most cases of adoption, Alicia showed substantial retention of Spanish and she can be considered a special case of heritage speaker. The examples shown earlier came from Alicia. Alicia completed a written proficiency test in Spanish and a written acceptability judgment task (she could read and write in Spanish). She also completed a cloze test in English and a version of the English grammaticality judgment task used by Johnson and Newport (1989) and DeKeyser (2000) in written form. Alicia scored with 93% and 98.9% accuracy in the English proficiency test and acceptability judgment task, within native speakers’ range, and 56% and 58% in the proficiency and acceptability judgment task in Spanish, well below the range of Spanish native speakers (91%–100%). In addition to these overall measures, Alicia returned for two more testing sessions and several other oral and written tasks in Spanish (story retelling task, picture description task, morphology recognition task, meaning judgment tasks, elicited written production tasks), targeting verbal and nominal morphosyntax, including tense, aspect and mood, gender agreement in nouns, and differential object marking. As in the proficiency measures reported, Alicia’s linguistic performance in these tasks was compared with the performance of a group of Spanish native speakers who served as comparison norms. In the tasks testing gender agreement, the native speakers obtained 99% to 100% accuracy scores whereas Alicia’s scores ranged from 60% to 85% accuracy, showing a high attrition rate of gender. She showed poor control of differential object

156 

 Silvina Montrul

marking (about 60% accuracy) and very poor accuracy on verbs in the subjunctive (21%) and in the imperfect (27%). Accuracy on the preterite was 71%. She could not discriminate between the semantic implications of preterite and imperfect forms in a meaning judgment task, unlike the comparison group of native speakers. On the assumption that 9-year-old children have already mastered gender agreement (Montrul & Potowski, 2007), differential object marking (Montrul & SánchezWalker, 2013), and verbal morphology (Montrul, 2004), and with the comparison of Alicia’s results with those of native speakers on all the measures tested, it is safe to conclude that the reason why Alicia’s grammar of Spanish is not at the level of a 9-year-old on all these grammatical areas is because of changes resulting from L1 attrition. This was most likely caused by lack of input in Spanish during a critical age of language development—9 to 14 years old—and uneven use and exposure to Spanish after that. Hence, this study also shows that age is a crucial factor in language loss. At the same time, this case study showed that interrupted linguistic input in childhood for several years has dramatic effects on the ability to develop a first language fully. Yet, those effects appear to be modulated by age: Alicia was adopted within the age period hypothesized to be when vulnerability to severe language loss is less likely (after age 8–9). She also attended school in Guatemala while she lived there. In the context of other existing studies that have reported almost total language loss regardless of the age of adoption, this case study contributes more evidence for the claim that the later the age of adoption, the less severe the degree of first language loss. It also underscores the effects of other socioaffective and educational factors which may also contribute indirectly to the degree of retention and effect relearning of the language after reexposure. Still, the degree of language loss exhibited by Alicia is more significant when compared with the linguistic abilities of heritage speakers who immigrated at the same age and lived with their parents all their lives: Montrul (2002) showed that heritage speakers who immigrated between ages 8 and 10 differed from monolingually raised native speakers on morphosyntactic and semantic measures of tense and aspect morphology in minor ways. Therefore, age of onset of L2 exposure plays a role in the extent of L1 attrition in childhood in adoption and nonadoption contexts. At the same time, the degree and speed of L1 loss are related to quantity of input: Adopted children show more severe and rapid language loss than nonadopted children, who retain functional ability in the L1 to different degrees.

Conclusion For a language to develop, stabilize, and not regress, a critical mass of input and use is required during an extended period of time, including the span of the language development period, which does not automatically end at age 4 and 5 but that continues throughout the period of schooling and later language development. Whether a native language is susceptible to loss and to what extent between the ages of 12 and



Age of Onset of Bilingualism Effects 

 157

18 has not been investigated in any depth. The study of heritage languages leads us to reconsider the importance of this period to appreciate what grammatical properties and linguistic skills remain undeveloped and underdeveloped in bilingual children. Heritage languages are vulnerable to attrition in immigrant families, but the degree and extent of attrition depend on the age of onset of bilingualism. Adults are less prone to attrition than children, who in turn can display low levels of proficiency in the heritage language as a result of incomplete acquisition and attrition because of reduced exposure and use of the language during a crucial period of language development. Furthermore, the status of a language as a heritage language can change in immigrants’ lifespan as a result of migration, and in cases of heritage language reversal, attrition of the heritage language is similarly subject to age effects. Interrupted exposure to linguistic input in childhood for several years has dramatic effects on the ability to develop a first language fully, as exemplified by the cases of international adoption. Yet, the effects and extent of attrition, which can actually result in major language loss, appear to be modulated by age. The younger the child when he or she is adopted the more severe the loss of the first language. The extent of language retention in older children depends on whether they continue to have some exposure and use of the first language. Thus, age and amount of input are two crucial factors in primary language attrition.

References Abrahamsson, N., & Hyltenstam, K. (2009). Age of onset and nativelikeness in a second language: Listener perception versus linguistic scrutiny. Language Learning, 59, 249–306. http://dx.doi.org/ 10.1111/j.1467-9922.2009.00507.x Anderson, R. T. (1999). Loss of gender agreement in L1 attrition: Preliminary results. Bilingual Research Journal, 23, 389–408. http://dx.doi.org/10.1080/15235882.1999.10162742 Au, T. K., Knightly, L. M., Jun, S. A., & Oh, J. S. (2002). Overhearing a language during childhood. Psychological Science, 13, 238–243. http://dx.doi.org/10.1111/1467-9280.00444 Berman, R. (2004). Between emergence and mastery: The long developmental route of language acquisition. In R. Berman (Ed.), Language development across childhood and adolescence (pp. 9–34). Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1075/ tilar.3.05ber Blake, R. (1983). Mood selection among Spanish-speaking children, ages 4 to 12. Bilingual Review, 10, 21–32. Bylund, E. (2009). Maturational constraints and first language attrition. Language Learning, 59, 687–715. http://dx.doi.org/10.1111/j.1467-9922.2009.00521.x Crain, S. (1991). Language acquisition in the absence of experience. Behavioral and Brain Sciences, 14, 597–612. http://dx.doi.org/10.1017/S0140525X00071491 DeKeyser, R. (2000). The robustness of critical period effects in second language acquisition. Studies in Second Language Acquisition, 22, 499–534.

158 

 Silvina Montrul

Ecke, P., & Hall, C. J. (2013). Tracking tip-of-the-tongue states in a multilingual speaker: Evidence of attrition or instability in lexical systems? International Journal of Bilingualism, 17, 734–751. Fernández Martínez, A. (1994). El aprendizaje de los morfemas verbales. Datos de un estudio longitudinal. [Learning verbal morphemes. Data from a longitudinal study] In S. López Ornat (Ed.), La adquisición de la lengua española (pp. 29–45). Madrid, Spain: Siglo XXI. Flores, C. (2010). The effect of age on language attrition: Evidence from bilingual returnees. Bilingualism: Language and Cognition, 13, 533–546. http://dx.doi.org/10.1017/ S136672890999054X Flores, C. (2012). Differential effects of language attrition in the domain of verb placement and object expression. Bilingualism: Language and Cognition, 15, 550–567. http://dx.doi.org/10.1017/ S1366728911000666 Gathercole, V. M., Sebastián, E., & Soto, P. (1999). The early acquisition of Spanish verbal morphology: Across the board or piecemeal knowledge? The International Journal of Bilingualism, 3, 133–182. http://dx.doi.org/10.1177/13670069990030020401 Gindis, B. (1999). Navigating uncharted waters: School psychologists working with inter­nationally adopted post-institutionalized children. Communiqué, 27, 20–27. Retrieved from http:// bgcenter.com/communique-article.htm Glennen, S. (2005). New arrivals: Speech and language assessment for internationally adopted infants and toddlers within the first months home. Seminars in Speech and Language, 26, 10–21. http://dx.doi.org/10.1055/s-2005-864212 Granena, G., & Long, M. (2013). Age of onset, length of residence, language aptitude, and ultimate attainment in three linguistic domains. Second Language Research, 29, 311–343. http:// dx.doi.org/10.1177/0267658312461497 Gürel, A. (2002). Linguistic characteristics of second language acquisition and first language attrition. Turkish overt versus null pronouns (Unpublished doctoral dissertation). McGill University, Montreal, Quebec, Canada. Gürel, A. (2004). Selectivity in L2 induced L1 attrition. A psycholinguistic account. Journal of Neurolinguistics, 17, 53–78. http://dx.doi.org/10.1016/S0911-6044(03)00054-X Hernández Pina, F. (1984). Teorías psicosociolingüísticas y su aplicación a la adquisición del español como lengua materna [Psycholinguistic theories and their application to the acquisition of Spanish as a mother tongue]. Madrid, Spain: Siglo XXI. Hopp, H., & Schmid, M. (2013). Perceived foreign accent in first language attrition and second language acquisition: The impact of age of acquisition and bilingualism. Applied Psycholinguistics, 34, 361–394. http://dx.doi.org/10.1017/S0142716411000737 Hyltenstam, K., Bylund, E., Abrahamsson, N., & Park, H.-S. (2009). Dominant language replacement: The case of international adoptees. Bilingualism: Language and Cognition, 12, 121–140. http:// dx.doi.org/10.1017/S1366728908004008 Isurin, L. (2000). Deserted island or a child’s first language forgetting. Bilingualism: Language and Cognition, 3, 151–166. http://dx.doi.org/10.1017/S1366728900000237 Iverson, M. (2012). Advanced language attrition of Spanish in contact with Brazilian Portuguese (Unpublished doctoral dissertation). University of Iowa, Iowa City, Iowa. Jia, G., & Aaronson, D. (2003). A longitudinal study of Chinese children and adolescents learning English in the United States. Applied Psycholinguistics, 24, 131–161. http://dx.doi.org/10.1017/ S0142716403000079



Age of Onset of Bilingualism Effects 

 159

Johnson, J. S., & Newport, E. L. (1989). Critical period effects in second language learning: The influence of maturational state on the acquisition of English as a second language. Cognitive Psychology, 21, 60–99. http://dx.doi.org/10.1016/0010-0285(89)90003-0 Kaufman, D., & Aronoff, M. (1991). Morphological disintegration and reconstruction in first language attrition. In H. Seliger & R. Vago (Eds.), First language attrition (pp. 175–188). Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1017/CBO9780511620720.012 Keijzer, M. (2007). Last in first out? An investigation of the regression hypothesis in Dutch emigrants in Anglophone Canada. Amsterdam, the Netherlands: Netherlands Graduate School of Linguistics. http://dx.doi.org/10.1075/ttwia.78.12kei Kim, J.-H., Montrul, S., & Yoon, J. (2010). Dominant language influence in acquisition and attrition of binding: Interpretation of the Korean reflexive caki. Bilingualism: Language and Cognition, 13, 73–84. http://dx.doi.org/10.1017/S136672890999037X Köpke, B., & Schmid, M. (2003). Language attrition. The next phase. In M. Schmid, B. Köpke, M. Keijzer, & L. Weilemar (Eds.), First language attrition: Interdisciplinary perspectives on methodological issues (pp. 1–14). Amsterdam, the Netherlands: John Benjamins. Kupisch, T., Akpinar, D., & Stöhr, A. (2013). Gender assignment and gender agreement in adult bilinguals and second language learners of French. Linguistic Approaches to Bilingualism, 3, 150–179. http://dx.doi.org/10.1075/lab.3.2.02kup Lardiere, D. (2007). Ultimate attainment in second language acquisition: A case study. Mahwah, NJ: Erlbaum. López Ornat, S. (1994). La adquisición de la lengua española [The acquisition of the Spanish language]. Madrid, Spain: Siglo XXI. Major, R. (1992). Losing English as a first language. Modern Language Journal, 76, 190–208. http:// dx.doi.org/10.1111/j.1540-4781.1992.tb01100.x Merino, B. (1983). Language loss in bilingual Chicano children. Journal of Applied Developmental Psychology, 4, 277–294. http://dx.doi.org/10.1016/0193-3973(83)90023-0 Montrul, S. (2002). Incomplete acquisition and attrition of Spanish tense/aspect distinctions in adult bilinguals. Bilingualism: Language and Cognition, 5, 39–68. http://dx.doi.org/10.1017/ S1366728902000135 Montrul, S. (2004). The acquisition of Spanish. Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1075/lald.37 Montrul, S. (2011). First language retention and attrition in an adult Guatemalan adoptee. Language, Interaction and Acquisition, 2, 276–311. http://dx.doi.org/10.1075/lia.2.2.05mon Montrul, S. (in press). The acquisition of heritage languages. Cambridge, England: Cambridge University Press. Montrul, S., & Potowski, K. (2007). Command of gender agreement in school-age Spanish–English bilingual children. The International Journal of Bilingualism, 11, 301–328. http://dx.doi.org/ 10.1177/13670069070110030301 Montrul, S., & Sánchez-Walker, N. (2013). Differential object marking in child and adult Spanish heritage speakers. Language Acquisition, 20, 109–132. http://dx.doi.org/10.1080/ 10489223.2013.766741 Montrul, S. A. (2008). Incomplete acquisition in bilingualism: Re-examining the age factor. Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1075/sibil.39 Nicoladis, E., & Grabois, H. (2002). Learning English and losing Chinese: A case study of a child adopted from China. The International Journal of Bilingualism, 6, 441–454. http://dx.doi.org/ 10.1177/13670069020060040401

160 

 Silvina Montrul

Oh, J. S., Au, T. K., & Jun, S. A. (2010). Early childhood language memory in the speech perception of international adoptees. Journal of Child Language, 37, 1123–1132. http://dx.doi.org/10.1017/ S0305000909990286 Oller, K., & Jarmulowicz, L. (2009). Language and literacy in bilingual children in the early school years. In E. Hoff & M. Shatz (Eds.), Blackwell handbook of language development (pp. 368–387). Malden, MA: Wiley-Blackwell. Opitz, C. (2013). A dynamic perspective on late bilinguals’ linguistic development in an L2 environment. The International Journal of Bilingualism, 17, 701–715. http://dx.doi.org/10.1177/ 1367006912454621 Pallier, C., Dehaene, S., Poline, J. B., LeBihan, D., Argenti, A. M., Dupoux, E., & Mehler, J. (2003). Brain imaging of language plasticity in adopted adults: Can a second language replace the first? Cerebral Cortex, 13, 155–161. http://dx.doi.org/10.1093/cercor/13.2.155 Perdue, C., & Bowerman, M. (1990). The structure of the simple clause in language acquisition: Introduction. Linguistics, 28, 1131–1133. http://dx.doi.org/10.1515/ling.1990.28.6.1131 Pérez-Pereira, M. (1991). The acquisition of gender: What Spanish children tell us. Journal of Child Language, 18, 571–590. http://dx.doi.org/10.1017/S0305000900011259 Perpiñán, S. (2013). Optionality in bilingual grammars. In M. Schmid & B. Köpke (Eds.), First language attrition (pp. 127–156). Amsterdam, the Netherlands: John Benjamins. Pierce, L. J., Klein, D., Chen, J.-K., Delcenserie, A., & Genesee, F. (2014). Mapping the unconscious maintenance of a lost first language. Proceedings of the National Academy of Sciences of the United States of America, 111, 17314–17319. http://dx.doi.org/10.1073/pnas.1409411111 Polinsky, M. (2006). Incomplete acquisition: American Russian. Journal of Slavic Linguistics, 14, 191–262. Ravid, D. (2004). Derivational morphology revisited: Later lexical development in Hebrew. In R. Berman (Ed.), Language development across childhood and adolescence (pp. 53–81). Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1075/tilar.3.07rav Schmid, M. (2002). First language attrition, use and maintenance: The case of German Jews in Anglophone countries. Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/ 10.1075/sibil.24 Schmid, M. (2007). The role of L1 use for L1 attrition. In B. Köpke, M. Schmid, M. Keijzer, & S. Dosterst (Eds.), Language attrition: Theoretical perspectives (pp. 135–153). Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1075/sibil.33.10sch Schmid, M. (2011). Language attrition. Cambridge, England: Cambridge University Press. http:// dx.doi.org/10.1017/CBO9780511852046 Schmid, M. (2014). The debate on maturational constraints in bilingual development: A perspective from first language attrition. Language Acquisition, 21, 386–410. http://dx.doi.org/10.1080/ 10489223.2014.892947 Schmid, M. S., & Jarvis, S. (2014). Lexical access and lexical diversity in first language attrition. Bilingualism: Language and Cognition, 17, 729–748. http://dx.doi.org/10.1017/ S1366728913000771 Sherkina-Lieber, M. (2011). Comprehension of Labrador Inuttitut functional morphology by receptive bilinguals (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Sills, R., & Hall, C. J. (2005). Cambios léxicos en el veneto de Chipilo, Puebla: una perspectiva psicolingüística [Lexical changes Chipilo Veneto, Puebla: A psycholinguistic perspective]. Lingüística Mexicana, 2, 277–293.



Age of Onset of Bilingualism Effects 

 161

Silva-Corvalán, C. (1994). Language contact and change: Spanish in Los Angeles. Oxford, England: Oxford University Press. Silva-Corvalán, C. (2014). Bilingual language acquisition: Spanish and English in the first six years. Cambridge, England: Cambridge University Press. http://dx.doi.org/10.1017/ CBO9781139162531 Sorace, A. (2000). Differential effects of attrition in the L1 syntax of near-native L2 speakers. Proceedings of the 24th Boston University Conference on Language Development, (pp. 719–725). Somerville, MA: Cascadilla Press. Tomasello, M. (2001). The item-based nature of children’s early syntactic development. In M. Tomasello & E. Bates (Eds.), Language development: The essential readings (pp. 169–186). Malden, MA: Blackwell. Tsimpli, I. (2007). First language attrition from a minimalist perspective: Interface vulnerability and processing effects. In B. Köpke, M. Schmid, M. Keijzer, & S. Dosterst (Eds.), Language attrition: Theoretical perspectives (pp. 83–98). Amsterdam, the Netherlands: John Benjamins. http:// dx.doi.org/10.1075/sibil.33.07tsi Turian, D., & Altenberg, E. (1991). Compensatory strategies of child first language attrition. In H. Seliger & R. Vago (Eds.), First language attrition (pp. 207–226). Cambridge, England: Cambridge University Press. http://dx.doi.org/10.1017/CBO9780511620720.014 Vago, R. (1991). Paradigmatic regularity in first language attrition. In H. Seliger & R. Vago (Eds.), First language attrition (pp. 241–252). Cambridge, England: Cambridge University Press. http:// dx.doi.org/10.1017/CBO9780511620720.016 Ventureyra, V. (2005). A la recherche de la langue perdue, Etude psycholinguistique de l’attrition de la première langue chez des coréens adoptés en France [In search of the lost language, psycholinguistics study of attrition of the first language among Korean adoptees in France] (Unpublished doctoral dissertation). Paris, France: Ecole des Hautes Etudes en Sciences Sociales. Yeni-Komshian, G. H., Flege, J. E., & Liu, S. (2000). Pronunciation proficiency in the first and second languages of Korean–English bilinguals. Bilingualism: Language and Cognition, 3, 131–149. http://dx.doi.org/10.1017/S1366728900000225 Zaretsky, E., & Bar-Shalom, E. G. (2010). Does reading in shallow L1 orthography slow attrition of language-specific morphological structures? Clinical Linguistics & Phonetics, 24, 401–415. http://dx.doi.org/10.3109/02699200903532532

David Birdsong and Jan Vanhove

9 Age of Second-Language Acquisition: Critical Periods and Social Concerns A classic topic in research on bilingualism across the lifespan is the relationship between the age at which learners start to acquire a second language (L2) and their ultimate level of proficiency in that language. Learning of an L2 that begins in infancy is typically associated with fluent speech, effortless language processing, and native accent. In contrast, late L2 learners tend to diverge from monolingual natives on measures of grammatical and lexical knowledge, processing speed, and acoustic properties of speech. Various classes of explanations for age effects in L2 acquisition—attitudinal, neurobiological, experiential, psychosocial, and cognitive—have been proposed in the literature. It is not the purpose of this chapter to examine these various accounts (for an overview, see Muñoz & Singleton, 2011). Similarly, with the exception of a brief discussion in the section on social concerns, this chapter does not consider research on the efficiency of early foreign language teaching in schools (for an overview, see Lambelet & Berthele, 2015). Rather, we are concerned with the hypothesis that L2 learning in naturalistic contexts is constrained by a critical period (CP). With its roots in the seminal works of Penfield and Roberts (1959) and Lenneberg (1967), the CP hypothesis for L2 acquisition (CPH/L2A) posits that nativelike attainment in the L2 from mere exposure is possible if learning begins within, but not after, a limited developmental span. Several recent overviews comprehensively summarize studies inspired by the CPH/L2A, and we do not intend to rehash these surveys. Instead, this chapter aims to provide readers with a technical toolkit to critically evaluate research on the divisive issue that is the CPH/L2A. To this end, we first discuss the CPH/L2A’s prediction that nativelikeness among learners with post-CP age of onset of acquisition (AoA) is impossible and highlights epistemological difficulties with this prediction. We then turn to the nature of the function that relates AoA to L2 attainment. The logic here is that a discontinuous function, but not a straight-line function, properly reflects the workings of a critical period. Here we illustrate how seemingly minor technical (statistical) caveats, which often risk being brushed under the rug as nit-picking, can fundamentally affect the conclusions of a study. In the final section, we expand our scope by considering the relevance of L2 acquisition theory to the social context of L2 learners and users. Specifically, we examine three issues relating to the notion that late L2 acquisition is inferior to first-language (L1) acquisition: early instruction of foreign languages in schools, the emphasis on deficits versus capacities, and societal prejudices against non-nativelikeness. http://dx.doi.org/10.1037/14939-010 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

164 

 Birdsong and Vanhove

Critical Period Hypothesis: Predictions, Evidence, and Criticisms Although often referred to as the critical period hypothesis for L2 acquisition, the CPH/ L2A is actually a conglomerate of partly overlapping, partly contradictory hypotheses (Long, 2005; Singleton, 2005). This variation is evident in the different posited time windows of the critical periods (see Muñoz & Singleton, 2011) as well in the language domains presumed to be affected by them (see Birdsong, 2006; Long, 2007; Scovel, 2006). In broad terms, however, empirical evaluations of CP accounts of L2 acquisition revolve around two key predictions. The first prediction is that nativelikeness is possible in young L2 acquirers but impossible in older (post-CP) acquirers. The second is that the function linking AoA to the learner’s ultimate attainment in the L2 is different before and after the closure of the hypothesized critical period. In what follows, we present a brief overview of behavioral evidence pertaining to these predictions and highlight epistemological and technical caveats concerning this evidence.

The Non-Nativelikeness Prediction Several CPH/L2As predict that L2 learners starting the acquisition process after the closure of the critical period cannot achieve nativelike competence in the target language (e.g., DeKeyser & Larson-Hall, 2005; Hyltenstam & Abrahamsson, 2003; Long, 1990). On the face of it, this non-nativelikeness prediction is falsifiable; finding a single nativelike post-CP L2 learner would prompt a reevaluation of the critical period concept (see Long, 1990). Studies examining this prediction typically involve L2 learners and L1 control participants engaging in a linguistic task. The L1 speakers’ scores are then used to establish an interval of nativelike performance on the task in question. Depending on the study, this interval is constructed as within one, two, or three standard deviations around the mean of the native speakers’ scores, or as the range of the native speakers’ scores. L2 learners with scores within this interval are considered to perform to nativelike standards. However, studies claiming to have observed nativelike performance by post-CP learners (e.g., Bongaerts, 1999; White & Genesee, 1996) have met with strong critiques. One recurring critique is that nativelike performance on a small battery of L2 tasks is not enough to establish across-the-board nativelikeness in real communicative situations and under close examination; had these learners been tested more thoroughly, their non-nativelikeness might have been discerned (see DeKeyser, 2006; DeKeyser & Larson-Hall, 2005; Long, 2005, 2007, for this and similar issues, but see Rothman, 2008, for counterarguments). For reasons of space, we will not address these studies in detail. Instead, we ask whether the concept of nativelikeness can fruitfully be used for addressing CPH/L2As.



Age of Second-Language Acquisition 

 165

Hyltenstam and Abrahamsson (2003) argued that researchers should conduct highly detailed linguistic analyses on learners’ performance. From their review they conclude that not only is there no convincing evidence that any claim of post-CP nativelikeness stands up to scrutiny, but that it is doubtful that any L2 learner, pre-CP learners included, can fully acquire all the subtleties of true nativeness. In an empirical follow-up to this idea, Abrahamsson and Hyltenstam (2008, 2009) found that L2 learners who pass for natives in an accent judgment task with naive native judges fare substantially worse when subjected to a battery of L2 tasks, with only a handful of early learners and no late learners passing the nativelikeness mark on all tasks. Although Hyltenstam and Abrahamsson’s emphasis on analytical rigor should be applauded, no constraints on its scope of application have been proposed. Thus, when L2 speakers meet the nativelikeness criterion on one task or measure, they may be subjected to unlimited additional tasks and more microscopic measures until a non-nativelike feature is detected (Birdsong, 2005b; Davies, 2004). Operationalizing and implementing the native standard, whether scrutinized or merely perceived, is therefore difficult and bound to be controversial. The issues involved range from which tasks to consider (e.g., are tasks on which L1 speakers show substantial variability at all suited?) and whom to use as control participants, to how to construct the interval of nativelike performance (e.g., is the criterion too stringent or too lenient?). For discussion of some of these issues, see Abrahamsson and Hyltenstam (2009), Andringa (2014), Birdsong and Gertken (2013), and Long (2005). Additionally, some researchers maintain that nativelike performance is observed only among extraordinarily gifted language learners, and that the performance of such individuals should not be taken as falsifying evidence for the CPH/L2A (Abrahamsson & Hyltenstam, 2008; DeKeyser, 2000). An additional consideration relates to the nature of bilingualism. Among active users of more than one language, no matter what the age of acquisition of the L2, the L2 influences a variety of features of the L1 just as the L1 influences features of the L2 (e.g., Cook, 1999; Grosjean, 1989; see also Chapter 7, this volume). (Note, though, that proponents of this position have never claimed that bilingualism effects alone can explain all departures from monolingual-likeness.) Because departures from monolingual-likeness (in both the L1 and the L2) are inevitable in early and late bilingualism alike, it is not reasonable to stipulate across-the-board nativelikeness as a criterion for falsifying the CPH/L2A (e.g., Cook, 1999; Ortega, 2009). Regarding this notion, the literature does reveal instances in which some early bilingual speakers are indistinguishable from native speakers on certain linguistic features. However, such findings highlight methodological concerns associated with empirically determining nativelikeness; if reports of nativelikeness in late learners can be critiqued for not stemming from sufficiently excruciating scrutiny across all linguistic domains, the evidence of nativelikeness among early L2 learners would have to come from similarly meticulous and exhaustive scrutiny of these young learners, too. Thus, the absence of evidence for non-nativelikeness in early learners does not amount to evidence for the absence of such nativelikeness.

166 

 Birdsong and Vanhove

Under the nature-of-bilingualism account, there should be no bilingual speakers, of any L2 AoA (including bilingual speakers from birth), who, under scrutiny and across all measurable domains, are identical to monolingual speakers. This account is supported by the fundamental observation that, among bilingual speakers of any L2 AoA—and even those with low L2 proficiency—the lexicons of both languages are activated in processing either language (e.g., Dijkstra & Van Heuven, 2002; Schwartz & Kroll, 2006). By definition, bilingual speakers do not process their lexicons like monolingual speakers, precisely because monolingual speakers do not have two languages. Additional evidence for the nature-of-bilingualism account comes from studies of voice onset time (VOT; e.g., Fowler, Sramko, Ostry, Rowland, & Hallé, 2008; MacLeod & Stoel-Gammon, 2005; Sundara, Polka, & Baum, 2006). In brief, these studies reveal that among speakers who are bilingual from birth and early bilingual learners, some VOT values in both the L1 and L2 of bilingual adults differ from those of monolingual adults. To summarize, the CPH/L2A and the nature-of-bilingualism argument make the same prediction of inevitable nonmonolingual nativelikeness among late L2 learners. However, on the nature-of-bilingualism view, nonmonolingual nativelikeness should be observed in the L2 of all active users of two languages, no matter what the AoA of the L2. For the CPH/L2A to prevail over the bilingualism account, absolute absence of bilingualism effects in some early L2 learners would have to be demonstrated. One might expect, however, that such endeavors would be unsatisfactory for the same reason that reports of nativelikeness in late L2 learners have been critiqued in the past: absence of evidence is not evidence of absence. Given the conceptual and empirical diffi­culties associated with the non-nativelikeness prediction, the critical period concept can perhaps be more fruitfully addressed via a second common prediction, which is discussed next.

The Discontinuity Prediction The second strong prediction made by a broad class of CPH/L2As is that the relationship between AoA and ultimate L2 attainment (UA) will not be linear. Instead, these CPH/L2As predict that the relationship between AoA and UA will be characterized by a bend in the function following the closure of the critical period (often referred to as a discontinuity). The discontinuity prediction is spelled out by DeKeyser (2012; DeKeyser & Larson-Hall, 2005; see also Long, 2013): The critical period concept implies a break in the AoA-proficiency function, i.e., an age . . . after which the decline of success rate in one or more areas of language is much less pronounced and/ or clearly due to different reasons. (DeKeyser, 2012, p. 445)

The logic behind the discontinuity prediction is that if heightened sensitivity to L2 input is confined to a temporally constrained critical period, then there should be no age-related differences because of maturational effects after the closure of



Age of Second-Language Acquisition 

 167

the critical period. Note that CP accounts do not necessarily predict the absence of age-related differences after the closure of the CP for reasons other than maturation: Age-related decreases in memory performance and test taking skills as well as changing socialization patterns may well result in a negative AoA–UA relationship in adult learners (DeKeyser, 2006, 2013). Such a pattern would not, however, suffice to falsify critical period accounts. In practice, then, most CPH/L2As still allow for some (less pronounced) post-CP declines in UA (e.g., Long, 2013). Both DeKeyser’s (2012) and Long’s (2013) prediction of a nonlinear relationship between AoA and UA correspond to the left and middle panels of Figure 9.1 (see Birdsong, 2005a). In the left panel, UA declines steeply up to the closure of the critical period, from which point onwards the AoA–UA function levels off (“stretched L”). In the middle pattern, maximum levels of L2 attainment are still possible during the first few years in the critical period but UA again drops steeply toward the closure of the CP (“stretched Z”). These patterns have in common a function that levels off after the closure of the critical period. A third possible pattern—inconsistent with both DeKeyser’s and Long’s predictions but sometimes encountered in CPH/L2A-related discussions (e.g., Hyltenstam & Abrahamsson, 2003; Yeni-Komshian, Flege, & Liu, 2000)—is the “stretched 7” pattern plotted in the right-hand panel. According to this conceptualization, UA remains at its peak level during the entire critical period and starts to decrease linearly thereafter. The remainder of this section discusses findings pertaining to predictions that the AoA–UA function levels off after the closure of the hypothesized critical period—that is, the common feature of the “stretched L” and “stretched Z” discontinuities. Census Studies One class of evidence addressing the discontinuity prediction comes from large-scale census studies. Stevens (1999), for instance, extracted self-reported English proficiency ‘Stretched Z’

‘Stretched 7’

Ultimate L2 Attainment

‘Stretched L’

Age of Acquisition Fig. 9.1: Three possible nonlinear relationships between age of acquisition and ultimate L2 attainment proposed in the literature. From “Age and Second Language Acquisition and Processing: A Selective Overview,” by D. Birdsong, 2006, Language Learning, 56, p. 15. Copyright 2006 by Wiley. Adapted with permission.

168 

 Birdsong and Vanhove

data of 8,046 foreign-born adults residing in the United States as well as the approximate ages at which they arrived in the United States from the 1990 U.S. Census. She found that the probability with which these immigrants considered that they spoke English “very well” decreased nonlinearly as a function of their AoA, but she noted that the data suggested no “abruptly defined ‘critical’ or sensitive period in L2 learning” (Stevens, 1999, p. 569). However, both nonlinear trends and the absence of abrupt discontinuities follow naturally from the ordinal logistic regression model that was used to analyze these data (i.e., under this model Stevens could not not have found nonlinear continuous trends; see Vanhove, 2013). In other words, Stevens’s results can be taken neither in support of a critical period account nor as evidence against it. In a similar study, Bialystok and Hakuta (1999) made use of a regression model that is able to capture abrupt changes in the function’s slope (“locally weighted scatterplot smoothing”) to model the relationship between self-reported English proficiency data from nearly 64,000 Chinese- and Spanish-speaking immigrants to the United States and their approximate ages of arrival. They found that this relationship was captured by a linearly declining function. Comparable results were obtained in even larger studies (Chiswick & Miller, 2008; Hakuta, Bialystok, & Wiley, 2003). In summary, these large-scale census studies did not confirm the discontinuity prediction. Instead, they point to a continual linear decrease of L2 proficiency levels with increasing AoA. These results have, however, been criticized on the grounds that the underlying data are not suited for testing the CPH/L2As (see DeKeyser, 2006; DeKeyser & Larson-Hall, 2005; Long, 2013; Stevens, 2004). Specifically, UA levels were extracted using coarse self-evaluations on a four- or five-level scale, and the AoA variable consisted of rough approximations. According to the critics, both of these procedures could have obscured a critical period effect in the data (but see Wiley, Bialystok, & Hakuta, 2005). Given the debated validity of these large-scale census studies, we turn our attention to studies designed specifically to address the discontinuity prediction next.

Task-Based Studies Comprehensive reviews of task-based studies have been provided by Birdsong (2005a), DeKeyser (2012), and DeKeyser and Larson-Hall (2005). We will not rehash these surveys but instead discuss four studies in some detail to illustrate common designs, touch on recurring points of criticism, and raise some less obvious but nonetheless important technical caveats about these studies.

Johnson and Newport (1989) Johnson and Newport (1989) recruited 46 Chinese- or Korean-speaking immigrants to the United States whose ages at arrival ranged from 3 to 39 years. Of these participants,



Age of Second-Language Acquisition 

 169

23 had arrived to the United States before age 15 (early arrivals) and 23 had arrived after age 17 (late arrivals). These participants had been living in the United States for 3 to 26 years at the time of testing.1 Their proficiency in English was tested with an aurally presented 276-item grammaticality judgment task (GJT). Johnson and Newport (1989) reported a global correlation between the participants’ ages of arrival and their GJT scores of r = -.77. Additionally, they also computed separate correlations for the early and for the late learners. For the early learners, GJT performance was strongly correlated with age (r = -.87), whereas for the late learners, the correlation was appreciably weaker (r = -.16). Johnson and Newport concluded from this finding that the AoA–UA function flattens off from puberty onwards, suggesting the workings of maturational constraints in L2 acquisition.2 Since Johnson and Newport’s study, several more studies have been conducted in similar vein, including those we turn to next.

Birdsong and Molis (2001) Birdsong and Molis (2001) presented the same grammaticality judgment task as Johnson and Newport (1989) to a sample of 61 Spanish-speaking immigrants to the United States. Their ages of arrival ranged from 3 to 44 years. Twenty-nine were early arrivals (AoA ≤ 16); 32 were late arrivals (AoA ≥ 17). Like Johnson and Newport (1989), Birdsong and Molis (2001) found a global correlation of r = -.77 between the participants’ ages of arrival and their GJT scores. Contrary to Johnson and Newport, however, Birdsong and Molis found that the subgroup correlation was stronger in the late (r = -.69) than in the young arrivals (r = -.24). As we illustrate below, however, differences between subgroup correlation coefficients do not necessarily indicate discontinuities in the AoA–UA function. Such discontinuities can more appropriately be identified using regression techniques. Birdsong and Molis therefore carried out regression analyses on Johnson and Newport as well as on their own data, the main results of which are presented in Figure 9.2. Note that whereas the Johnson and Newport data seem to be captured by a CPH/L2A-consistent “stretched L” shape, the Birdsong and Molis data form the inverse pattern (“stretched 7”). In sum, Birdsong and Molis (2001) failed to reproduce the results that Johnson and Newport (1989) had taken in support of a critical period account of L2 acquisition and,

1 Note that other studies featuring learners with short lengths of residence (e.g., Bialystok & Miller, 1999) have been dismissed by reviewers (e.g., DeKeyser & Larson-Hall, 2005; Long, 2013) on the grounds that their results do not speak to ultimate attainment but rather to rate of learning. DeKeyser (2013) recommended that length of residence be at least 10 years to ensure that learners are at asymptote. 2 Johnson and Newport (1989) additionally computed a series of t-tests between AoA-defined subgroups to determine whether the AoA–UA relationship exhibits discontinuities. For a critique of the use of subgroup mean comparisons to identify discontinuities, see Vanhove (2013).

170 

 Birdsong and Vanhove

Fig. 9.2: Scatterplot showing the relationship between age of arrival and GJT scores for studies by both Johnson and Newport (1989; dotted lines) and Birdsong and Molis (2001; solid lines). The regression lines were computed separately for early (AoA ≤ 16) and late (AoA ≥ 17) arrivals. From “On the Evidence for Maturational Constraints in Second-Language Acquisition,” by D. Birdsong and M. Molis, 2001, Journal of Memory and Language, 44, p. 240. Copyright 2001 by Elsevier. Reprinted with permission.

in fact, ended up with results that form the mirror image of Johnson and Newport’s findings. Long (2007) critiqued the Birdsong and Molis (2001) study for having used too easy a test, which he claimed gave rise to ceiling effects (i.e., the early plateau in Figure 9.2). With Birdsong and Molis, however, we note the need for criteria other than a post-hoc interpretation of the results for accepting an instrument in the original study and rejecting it in the replication. DeKeyser and Larson-Hall (2005) additionally suggested that the Birdsong and Molis results were tainted by outliers, notably the participant with AoA 35 and a test score of 166. Reanalyzing their results without the three late arrivals with the lowest scores, however, Birdsong and Molis still found a negative correlation. Furthermore, note again the need for criteria other than posthoc interpretations for rejecting data points: the Johnson and Newport (1989) study featured a participant with AoA 18 and a test score of 163 who arguably constituted a larger outlier (see Figure 9.2). Without external criteria, it seems difficult to defend critiquing Birdsong and Molis’s findings without similarly critiquing Johnson and Newport’s results.



Age of Second-Language Acquisition 

 171

DeKeyser, Alfi-Shabtay, and Ravid (2010) DeKeyser, Alfi-Shabtay, and Ravid (2010) set out to test the CPH/L2A’s discontinuity prediction in two parallel studies. The first sample consisted of 76 Russianspeaking immigrants to the United States and Canada; the second sample consisted of 62 Russian-speaking immigrants to Israel. The participants’ ages of arrival ranged from 4 to 71 years, and the lengths of residence ranged from 8 to 28 years. UA levels were assessed using a 200-item GJT (an adaptation of the Johnson & Newport, 1989, instrument). Global correlations computed for the relationship between AoA and UA closely resembled those reported by Johnson and Newport (1989) and by Birdsong and Molis (2001) in both studies (r = -.80 and -.79 for the North America and Israel studies, respectively). However, the correlation was stronger for early arrivals (AoA < 18; r = -.69 and -.48) than for middle-age arrivals (18 ≤ AoA ≤ 40; r = -.45 and -.37). DeKeyser et al. (2010) interpreted these differences as indicative of a discontinuity in the AoA–UA relationship and hence as cross-linguistic confirmation of the CPH/L2A’s discontinuity prediction. DeKeyser et al. also reported correlations between AoA and UA in late arrivals (AoA > 40; r = -.27 and -.53), which they attributed to the late arrivals’ old age at the time of testing. DeKeyser et al.’s (2010) conclusions were called into question on statistical grounds by Vanhove (2013). In CPH/L2A-related discussions, differences between subgroup correlation coefficients are often taken as evidence of changes in the slope of the AoA–UA function (e.g., DeKeyser, 2012, p. 448). Such a conclusion conflates correlation and regression. Briefly, correlation coefficients express how closely the data map on a line, whereas regression analysis seeks to estimate the functional characteristics of this line. It is possible for two relationships to be characterized by the same correlation coefficient but to have different functional forms, and vice versa (see, e.g., Vanhove, 2013). As CPH/L2As are currently formulated, however, the discontinuity prediction is concerned with the functional characteristics of the AoA-UA relationship, more specifically with changes in its slope. In the Johnson and Newport (1989) and Birdsong and Molis (2001) studies, correlational and regressional analyses sketch broadly the same picture (as illustrated by Figure 9.2). However, to illustrate how conflating correlation coefficients with regression coefficients can lead to diametrically opposed conclusions about the CPH/L2A, Vanhove (2013) carried out a reanalysis of the DeKeyser et al. (2010) data. Specifically, the data were modeled using piecewise regression models. These are comparable to ordinary regression models with the difference that the slope of the function relating the predictor to the outcome is allowed to change at a specified inflection point. Figure 9.3 graphically summarizes the piecewise regression models with inflection points at AoA = 18 for the DeKeyser et al. North America and Israel studies and also shows simple regression fits as a reference. As can be gleaned from these plots, piecewise regression models did not substantially improve the fit in either the North America or Israel study (gain in R2 ≤ .015).

172 

 Birdsong and Vanhove

North America study

Israel study

linear function (R 2 = 0.65) non-linear function (R 2 = 0.66)

180 160 140 120 100

200 Ultimate Attainment

Ultimate Attainment

200

linear function (R 2 = 0.63) non-linear function (R 2 = 0.63)

180 160 140 120 100

0

20

40

60

Age of Onset of Acquisition

0

20

40

60

Age of Onset of Acquisition

Fig. 9.3: Graphical summary of the reanalysis of DeKeyser et al.’s (2010) North America and Israel studies by Vanhove (2013). The dotted lines show the fitted values derived from a simple linear regression model; the solid lines were derived from piecewise regression models featuring a breakpoint at age 18. (The dotted line is not absent from the right-hand panel but is hardly visible due to its near-complete overlap with the solid line.) From “The Critical Period Hypothesis in Second Language Acquisition: A Statistical Critique and a Reanalysis,” by J. Vanhove, 2013, PLoS ONE, 8, e69172, Figures 5 and 6. Copyright 2013 by J. Vanhove. Adapted with permission.

It may be argued that the cutoff AoA of 18 years was decided on for partly arbitrary reasons (see DeKeyser et al., 2010) and that a cutoff at the true (but undefined) closure of the CP would have yielded different results. Vanhove (2013), therefore, computed a series of piecewise regression models for which the cut-off AoAs varied from 5 to 19 years. For the North America study, the best-fitting model had a cutoff at 16 years, which marginally improved the fit over a linear model (gain in R2 = .019, p = .05);3 for the Israel study, the best-fitting model had an inflection point at 6 years, which did not appreciably improve the fit (gain in R2 = .002). In conclusion, then, the DeKeyser et al. (2010) data do not provide cross-linguistic evidence in support of a critical period account but instead point to a continual linear decrease in ultimate L2 attainment as a function of the age of onset of L2 acquisition. Despite some earlier precedents (e.g., Birdsong & Molis, 2001; Hakuta et al., 2003), reservations about the use of piecewise regression to address CPH/L2A predictions have been expressed in the SLA literature (Long, 2013; Stevens, 2004). Unfortunately, these cannot be considered here due to spatial restrictions.

Granena and Long (2013) Granena and Long (2013) addressed the discontinuity prediction in three linguistic domains: morphosyntax, pronunciation, and lexis. The sample consisted of 65 speak3 The reported p values were not corrected for multiple testing. Looping through the data in search of the ideal cut-off age inflates the Type-I error rate. We estimate that for the North America study, the p value would have to be recalibrated to about p = .10 to take this Type I error inflation into account. For details concerning this estimation, see http://dx.doi.org/10.6084/m9.figshare.1206291.



Age of Second-Language Acquisition 

 173

ers of Chinese who had moved to Spain. In contrast to the studies reviewed thus far, these speakers were screened for their advanced L2 proficiency (i.e., the participants did not represent the full spectrum of L2 proficiencies). The participants’ AoA ranged from 3 to 29 years and their lengths of residence from 8 to 31 years. The participants completed a battery of linguistic tasks testing their pronunciation, lexical and col­ locational knowledge, and morphosyntactic attainment. Granena and Long (2013) reported global correlations between AoA and UA in line with those in the studies discussed above: r = -.81 for pronunciation, -.79 for lexical and collocational knowledge and -.73 for morphosyntax. Additionally, they fitted piecewise regression models with two inflection points: one at AoA 6 years and one at AoA 15 years.4 The increase in variance explained compared to simple linear models varied between 4 and 5 percentage points. In a further analysis, Granena and Long pooled the pronunciation, lexis and morphosyntax data into one dataset, which was then analyzed using a repeated-measures analysis of variance (ANOVA). Of specific interest was the question whether linguistic domain interacted with AoA-defined subgroup status (i.e., AoA < 6, AoA 7–15, AoA > 15). The authors found a significant interaction between linguistic domain and AoA-subgroup such that the earliest arrivals had a higher mean score on the pronunciation measures than on the lexical and morphosyntax measures and the later arrivals had a higher mean score on the morphosyntax measures. Granena and Long (2013) interpreted these findings as evidence of critical periods with different timings according to the linguistic domain affected: a closure for pronunciation in early childhood, followed by lexical and collocational knowledge and lastly a closure of a critical period for morphosyntax around age 15. However, the repeated-measures ANOVA that was carried out to bolster this claim made the fundamental assumption that the dependent variable consists of repeated measures of the same variable. The three types of outcomes observed in Granena and Long’s study are not the same variable measured under three different conditions (i.e., a pronunciation test condition, a lexical test condition, and a morphosyntactic test condition); rather, they represent three different, if correlated, qualities. Neither the fact that these outcomes can all be scaled to percentages nor that they could be construed as tapping into the same latent variable (i.e., L2 proficiency) changes this: The dependent variable has to be the same across all conditions. An inspection of the scatterplots presented 4 “A restricted one-slope model with AO [age of onset of acquisition] as a single predictor was compared against a full model that included interaction terms between the predictor and dummy-coded AO group variables” (Granena & Long, 2013, p. 16). Although this is the method for fitting piecewise regression models recommended by Baayen (2008; also cited in Long’s, 2013, discussion of this study) and used by Vanhove (2013) in his reanalysis of DeKeyser et al. (2010), it only yields sensible regression fits when modeling one inflection point. Fitting a piecewise regression model with multiple inflection points, though possible by other means, results in disjointed segments rather than a single connected regression fit.

174 

 Birdsong and Vanhove

in Figure 5 in Granena and Long (p. 328) reveals that the spread in the morphosyntax data was visibly lower than that in the pronunciation and lexis-collocation data. As a result, the difference in the mean scores between early arrivals and late arrivals was smaller for the morphosyntax data, resulting in—or at least contributing to— a significant interaction term. However, this interaction can be explained without having to invoke multiple critical periods by simply recognizing that differences in the variability of scores on different tests are also caused by differences in the design and rating of the tests. In brief, we maintain that the results that Granena and Long marshalled in support of a multiple CPs account may be artifacts of an unwarranted analysis.

Summary: The AoA–UA Function Controversy The studies discussed above sketch a picture of how the discontinuity prediction has been assessed. As is obvious from this selective overview, the issue is highly controversial with researchers on either side of the debate arguing forcefully that evidence garnered on the other side of the fence rests on faulty interpretations and/or design (e.g., Bialystok, 2002; DeKeyser, 2006; DeKeyser & Larson-Hall, 2005; Long, 2005, 2007; Stevens, 2004). But taking a bird’s-eye view, what general conclusions can be drawn from the current state of the art? First, a recurrent finding is that ultimate L2 attainment levels in the domains of pronunciation, morphosyntax, and lexical and collocational knowledge are negatively associated with the learners’ ages of onset of acquisition over a wide AoA range (see also, among others, DeKeyser, 2000; Flege, Yeni-Komshian, & Liu, 1999). This finding is widely accepted and is not a source of controversy. Second, researchers subscribing to the CPH/L2A have argued that the AoA–UA function shows a distinct flattening that marks the closure of the (or a) critical period, typically in the mid-teens. Although these conclusions typically rest on generous interpretations of group mean and correlation coefficient comparisons (Vanhove, 2013), a handful of studies have indeed produced evidence to such discontinuities, notably Johnson and Newport (1989) according to Birdsong and Molis’s (2001) reanalysis and, more recently, Granena and Long (2013; but see the caveat in Note 4). Other results contradict a critical period account, however. On the one hand, Birdsong and Molis found a steeper rather than a flattened slope in later arrivals. On the other hand, comparable linear decreases were found before and after the closure of the critical period under consideration (e.g., Flege et al., 1999). Such a continued linear decrease was observed even in Vanhove’s (2013) reanalysis of data from two studies that had been taken as evidence confirming the discontinuity prediction. We have previously proposed that these conflicting findings could be accounted for by assuming that UA levels decrease linearly across a wide AoA range (Birdsong, 2009; Vanhove, 2013), that is, without positing a temporally delineated period of heightened sensitivity to linguistic input or critical period. Depending on the specifics of the study (including



Age of Second-Language Acquisition 

 175

floor and ceiling effects) as well as random variability, “L” or “7” shaped deviations from linearity can arise. A general linear trend could also explain why incidences of L2 learners performing within a native control sample’s range decrease with advancing AoA (see Birdsong, 2005a, Figure 6.8). Interestingly, the suggestion that conflicting findings in CPH/L2A-related research can be reconciled by assuming a continual linear decline was also put forward by Hyltenstam and Abrahamsson (2003), who proposed a maturation-based account of L2 attainment without temporally constrained “windows of opportunity.” Last, we hope that our discussion has illustrated that seemingly minor technical details (e.g., the difference between correlation and regression and the specification of statistical models) can dramatically affect the inferences drawn from the data.

Second-Language Outcomes in the Social Context In this section, we consider the issue of age of L2 acquisition in its societal context. In the form of brief précis, we highlight three popular manifestations of the notion that (late) L2 acquisition is inferior relative to L1 acquisition: the idea that “earlier is better” in foreign language education, the emphasis on linguistic deficits among bilingual speakers as compared with monolingual speakers, and socially ingrained prejudices against non-nativelikeness.

The CPH/L2A and the Foreign Language Classroom Questions about the optimal age for beginning L2 classroom learning are at the center of ongoing debates among educators and parents about the timing, amount, and usefulness of classroom teaching of foreign languages (e.g., Gürsoy, 2011). CPH/L2A research has figured prominently in these debates. However, most studies relating to the CPH/ L2A involve ultimate attainment data gathered from bilinguals who have acquired, and who use, the L2 in naturalistic contexts. The application of findings under the CPH/ L2A is thus of limited utility in the classroom setting, which concerns neither ultimate attainment nor immersion. Still, the notion that earlier is better that is associated with the CPH/L2A is embraced by the general public and by stakeholders in elementary education. This notion was recently called into question by Huang (2015), who reviewed 40 empirical studies of the age of beginning foreign language instruction. Huang found that although there are nonlinguistic benefits of early foreign language instruction (e.g., in terms of cognitive development, attitudes toward languages other than the L1, and general academic achievement), there is no decisive evidence that earlier is better with respect to indices of either short-term or long-term L2 attainment in the classroom setting.

176 

 Birdsong and Vanhove

Deficits and Capacities The literature on age and L2 acquisition, with its emphasis on non-nativelikeness,5 often obscures the high levels of L2 ultimate attainment and communicative competence enjoyed by millions of users of multiple languages of varying AoA. Commenting on a 2014 special issue of Applied Linguistics “Complexities and Interactions of Age in Second Language Learning: Broadening the Research Agenda,” Yates and Kozar (2015) suggested that researchers’ preoccupation with L2 deficits as a function of AoA “is socially out of kilter with trends toward the re-evaluation of age-related expectations” (p. 2) which are redirecting modern societal attention to human cognitive and physical capacities over the lifespan. Similarly, although comparisons of L2 users with a monolingual standard have yielded important insights into L2 development over varying AoA, this methodology is insensitive to changes in the L1 among bilingual speakers over the lifespan, which cannot be ascribed to deficient language learning mechanisms. Given that more than half the world’s population is bilingual (Grosjean, 2010), it is important that societies understand that monolingual-likeness in both languages of speakers is an unrealistic expectation.

Societal Evaluation of L2 Outcomes The non-nativelike pronunciation that is typical of immigrants is known to trigger hostile reactions among native speakers. For example, human rights’ cases have examined employment-related discrimination and harassment based on speech accent (Munro, 2003). In contrast, positive attitudes are exhibited toward speech that is deemed standard by a majority of members of a speech community (Fuertes, Gottdiener, Martin, Gilbert, & Giles, 2012). Gluszek and Dovidio (2010) looked at social impressions of both nonnative accents and regional native accents. They found that nonnative accents, more so than regional accents, cue negative assessments of speakers’ overall linguistic competence. The researchers argued that stigmatization of nonnative accents extends beyond the immediate speech context to negatively impact future conversational interactions. On a more positive note, Weyant (2007) found that negative stereotyping of nonnative speakers can be reduced in experimental settings by perspective-taking activities (Galinsky & Moskowitz, 2000). 5 As mentioned earlier, many factors in addition to AoA have been examined in studies of (non-) nativelike attainment. These factors include degree of entrenchment of the L1, years of schooling, linguistic training, literacy, length of residence, L2 exposure/input/use, goals of learning, socialpsychological integration and identity, and motivation to become indistinguishable from natives. As Flege et al. (1999) demonstrated, different linguistic features of the L2 (e.g., morphosyntax, lexical knowledge, pronunciation) are not equally affected by these factors.



Age of Second-Language Acquisition 

 177

Conclusion Scientific theories should make predictions that, in principle, permit a falsification of the theory (Popper, 1959). The CPH/L2A is associated with two key predictions: nonnativelikeness and discontinuity. As we have seen, however, defining what constitutes sufficient evidence for nativelikeness is difficult if not impossible on epistemological grounds: the absence of evidence for non-nativelikeness does not equal evidence for nativelikeness. The discontinuity prediction can more readily be assessed, but in assessing the data available, several caveats have to be kept in mind, some of which we have discussed. We tentatively conclude that there is no convincing evidence for a CPH/L2A-compatible “bend” in the AoA–UA function. Such evidence may, however, still come in the form of highly-powered, carefully analyzed and cross-linguistically replicated studies. The scientific study of age and L2 attainment does not take place in a vacuum, disconnected from societal concerns. For example, it was pointed out that what learners achieve in classrooms may not reflect an advantage for early foreign language learning, and thus is inconsistent with results of ultimate attainment studies in immersion contexts. In addition, we underscored the risks of compar­ ing L2 users with mono­lingual natives, and suggested that it is in the interest of bilingual societies to recognize that monolingual–bilingual differences are normal, not deficient, outcomes in both the L1 and the L2. Finally, we noted that prejudices against the non-nativelikeness associated with late L2 learning are widespread, come in many forms, and can be corrosive cultural forces. In these various respects, the complete picture of age and L2 acquisition both incorporates and transcends theory.

References Abrahamsson, N., & Hyltenstam, K. (2008). The robustness of aptitude effects in near-native second language acquisition. Studies in Second Language Acquisition, 30, 481–509. http://dx.doi.org/ 10.1017/S027226310808073X Abrahamsson, N., & Hyltenstam, K. (2009). Age of onset and nativelikeness in a second language: Listener perception versus linguistic scrutiny. Language Learning, 59, 249–306. http://dx.doi.org/ 10.1111/j.1467-9922.2009.00507.x Andringa, S. (2014). The use of native speaker norms in critical period research. Studies in Second Language Acquisition, 36, 565–596. http://dx.doi.org/10.1017/S0272263113000600 Baayen, R. H. (2008). Analyzing linguistic data: A practical introduction to statistics using R. Cambridge, England: Cambridge University Press. http://dx.doi.org/10.1017/ CBO9780511801686 Bialystok, E. (2002). On the reliability of robustness. Studies in Second Language Acquisition, 24, 481–488. http://dx.doi.org/10.1017/S0272263102003054

178 

 Birdsong and Vanhove

Bialystok, E., & Hakuta, K. (1999). Confounded age: Linguistic and cognitive factors in age differences for second language acquisition. In D. Birdsong (Ed.), Second language acquisition and the critical period hypothesis (pp. 161–181). Mahwah, NJ: Erlbaum. Bialystok, E., & Miller, B. (1999). The problem of age in second-language acquisition: Influences from language, structure, and task. Bilingualism: Language and Cognition, 2, 127–145. http:// dx.doi.org/10.1017/S1366728999000231 Birdsong, D. (2005a). Interpreting age effects in second language acquisition. In J. F. Kroll & A. M. B. de Groot (Eds.), Handbook of bilingualism: Psycholinguistic approaches (pp. 109–127). Oxford, England: Oxford University Press. Birdsong, D. (2005b). Nativelikeness and non-nativelikeness in L2A research. International Review of Applied Linguistics in Language Teaching, 56, 9–49. Birdsong, D. (2006). Age and second language acquisition and processing: A selective overview. Language Learning, 56, 9–49. http://dx.doi.org/10.1111/j.1467-9922.2006.00353.x Birdsong, D. (2009). Age and the end state of second language acquisition. In W. C. Ritchie & T. K. Bhatia (Eds.), The new handbook of second language acquisition (2nd ed., pp. 401–424). Bingley, England: Emerald. Birdsong, D., & Gertken, L. (2013). In faint praise of folly: A critical review of native/non-native speaker comparisons, with examples from native and bilingual processing of French complex syntax. Language, Interaction and Acquisition, 4, 107–133. http://dx.doi.org/ 10.1075/lia.4.2.01bir Birdsong, D., & Molis, M. (2001). On the evidence for maturational constraints in second-language acquisition. Journal of Memory and Language, 44, 235–249. http://dx.doi.org/10.1006/ jmla.2000.2750 Bongaerts, T. (1999). Ultimate attainment in L2 pronunciation: The case of very advanced late L2 learners. In D. Birdsong (Ed.), Second language acquisition and the Critical Period Hypothesis (pp. 133–159). Mahwah, NJ: Erlbaum. Chiswick, B. R., & Miller, P. W. (2008). A test of the critical period hypothesis for language learning. Journal of Multilingual and Multicultural Development, 29, 16–29. http://dx.doi.org/10.2167/ jmmd555.0 Cook, V. (1999). Going beyond the native speaker in language teaching. TESOL Quarterly, 33, 185–209. http://dx.doi.org/10.2307/3587717 Davies, A. (2004). The native speaker in applied linguistics. In A. Davies & C. Elder (Eds.), The handbook of applied linguistics (pp. 431–450). Malden, MA: Blackwell. http://dx.doi.org/ 10.1002/9780470757000.ch17 DeKeyser, R. (2000). The robustness of critical period effects in second language acquisition. Studies in Second Language Acquisition, 22, 499–533. DeKeyser, R. (2006). A critique of recent arguments against the critical period hypothesis. In C. Abello-Contesse, R. Chacón-Beltrán, M. D. López-Jiménez, & M. M. Torreblanca-López (Eds.), Age in L2 acquisition and teaching (pp. 49–58). Bern, Switzerland: Peter Lang. DeKeyser, R. (2012). Age effects in second language learning. In S. M. Gass & A. Mackey (Eds.), The Routledge handbook of second language acquisition (pp. 442–460). London, England: Routledge. DeKeyser, R. (2013). Age effects in second language learning: Stepping stones toward better understanding. Language Learning, 63(Suppl. 1), 52–67. http://dx.doi.org/10.1111/ j.1467-9922.2012.00737.x



Age of Second-Language Acquisition 

 179

DeKeyser, R., Alfi-Shabtay, I., & Ravid, D. (2010). Cross-linguistic evidence for the nature of age effects in second language acquisition. Applied Psycholinguistics, 31, 413–438. http://dx.doi.org/ 10.1017/S0142716410000056 DeKeyser, R., & Larson-Hall, J. (2005). What does the critical period really mean? In J. F. Kroll & A. M. B. de Groot (Eds.), Handbook of bilingualism: Psycholinguistic approaches (pp. 88–108). New York, NY: Oxford University Press. Dijkstra, A., & Van Heuven, W. J. B. (2002). The architecture of the bilingual word recognition system: From identification to decision. Bilingualism: Language and Cognition, 5, 175–197. http:// dx.doi.org/10.1017/S1366728902003012 Flege, J. E., Yeni-Komshian, G. H., & Liu, S. (1999). Age constraints on second-language acquisition. Journal of Memory and Language, 41, 78–104. http://dx.doi.org/10.1006/jmla.1999.2638 Fowler, C. A., Sramko, V., Ostry, D. J., Rowland, S. A., & Hallé, P. (2008). Cross language phonetic influences on the speech of French–English bilinguals. Journal of Phonetics, 36, 649–663. http://dx.doi.org/10.1016/j.wocn.2008.04.001 Fuertes, J. N., Gottdiener, W., Martin, H., Gilbert, T. C., & Giles, H. (2012). A meta-analysis of the effects of speakers’ accents on interpersonal evaluations. European Journal of Social Psychology, 42, 120–133. http://dx.doi.org/10.1002/ejsp.862 Galinsky, A. D., & Moskowitz, G. B. (2000). Perspective-taking: Decreasing stereotype expression, stereotype accessibility, and in-group favoritism. Journal of Personality and Social Psychology, 78, 708–724. http://dx.doi.org/10.1037/0022-3514.78.4.708 Gluszek, A., & Dovidio, J. F. (2010). The way they speak: A social psychological perspective on the stigma of nonnative accents in communication. Personality and Social Psychology Review, 14, 214–237. http://dx.doi.org/10.1177/1088868309359288 Granena, G., & Long, M. H. (2013). Age of onset, length of residence, language aptitude, and ultimate L2 attainment in three linguistic domains. Second Language Research, 29, 311–343. http://dx.doi.org/10.1177/0267658312461497 Grosjean, F. (1989). Neurolinguists, beware! The bilingual is not two monolinguals in one person. Brain and Language, 36, 3–15. http://dx.doi.org/10.1016/0093-934X(89)90048-5 Grosjean, F. (2010). Bilingual. Cambridge, MA: Harvard University Press. http://dx.doi.org/10.4159/ 9780674056459 Gürsoy, E. (2011). The Critical Period Hypothesis revisited: The implications for current foreign language teaching to young learners. Journal of Language Teaching and Research, 2, 757–762. http://dx.doi.org/10.4304/jltr.2.4.757-762 Hakuta, K., Bialystok, E., & Wiley, E. (2003). Critical evidence: A test of the critical-period hypothesis for second-language acquisition. Psychological Science, 14, 31–38. http://dx.doi.org/10.1111/ 1467-9280.01415 Huang, B. H. (2015). A synthesis of empirical research on the linguistic outcomes of early foreign language instruction. International Journal of Multilingualism. Advance online publication. http://dx.doi.org/10.1080/14790718.2015.1066792 Hyltenstam, K., & Abrahamsson, N. (2003). Maturational constraints in SLA. In C. J. Doughty & M. H. Long (Eds.), The handbook of second language acquisition (pp. 538–588). Malden, MA: Blackwell. http://dx.doi.org/10.1002/9780470756492.ch17 Johnson, J. S., & Newport, E. L. (1989). Critical period effects in second language learning: The influence of maturational state on the acquisition of English as a second language. Cognitive Psychology, 21, 60–99. http://dx.doi.org/10.1016/0010-0285(89)90003-0

180 

 Birdsong and Vanhove

Lambelet, A., & Berthele, R. (2015). Age and foreign language learning in school. Basingstoke, England: Palgrave Macmillan. http://dx.doi.org/10.1057/9781137525901 Lenneberg, E. H. (1967). Biological foundations of language. New York, NY: Wiley. Long, M. H. (1990). Maturational constraints on language development. Studies in Second Language Acquisition, 12, 251–285. http://dx.doi.org/10.1017/S0272263100009165 Long, M. H. (2005). Problems with supposed counter-evidence to the critical period hypothesis. International Review of Applied Linguistics in Language Teaching, 43, 287–317. http://dx.doi.org/ 10.1515/iral.2005.43.4.287 Long, M. H. (2007). Problems in SLA. Mahwah, NJ: Erlbaum. Long, M. H. (2013). Maturational constraints on child and adult SLA. In G. Granena & M. H. Long (Eds.), Sensitive periods, language aptitude, and ultimate L2 attainment (pp. 3–42). Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1075/lllt.35.01lon MacLeod, A. N., & Stoel-Gammon, C. (2005). Are bilinguals different? What VOT tells us about simultaneous bilinguals. Journal of Multilingual Communication Disorders, 3, 118–127. http://dx.doi.org/ 10.1080/14769670500066313 Muñoz, C., & Singleton, D. (2011). A critical review of age-related research on L2 ultimate attainment. Language Teaching, 44, 1–35. http://dx.doi.org/10.1017/S0261444810000327 Munro, M. J. (2003). A primer on accent discrimination in the Canadian context. TESL Canada Journal/Revue TESL du Canada, 20, 38–51. Ortega, L. (2009). Understanding second language acquisition. London, England: Hodder. Penfield, W., & Roberts, L. (1959). Speech and brain mechanisms. Princeton, NJ: Princeton University Press. Popper, K. (1959). The logic of scientific discovery. London, England: Hutchinson. Rothman, J. (2008). Why all counter-evidence to the critical period hypothesis in second language acquisition is not equal or problematic. Language and Linguistics Compass, 2, 1063–1088. http://dx.doi.org/10.1111/j.1749-818X.2008.00098.x Schwartz, A. I., & Kroll, J. F. (2006). Bilingual lexical activation in sentence context. Journal of Memory and Language, 55, 197–212. http://dx.doi.org/10.1016/j.jml.2006.03.004 Scovel, T. (2006). Age, acquisition, and accent. In C. Abello-Contesse, R. Chacón-Beltrán, M. D. LópezJiménez, & M. M. Torreblanca-López (Eds.), Age in L2 acquisition and teaching (pp. 31–48). Bern, Switzerland: Peter Lang. Singleton, D. (2005). The critical period hypothesis: A coat of many colours. International Review of Applied Linguistics in Language Teaching, 43, 269–285. http://dx.doi.org/10.1515/ iral.2005.43.4.269 Stevens, G. (1999). Age of immigration and second language proficiency among foreign-born adults. Language in Society, 28, 555–578. http://dx.doi.org/10.1017/S0047404599004030 Stevens, G. (2004). Using census data to test the critical-period hypothesis for second-language acquisition. Psychological Science, 15, 215–216. http://dx.doi.org/10.1111/j.0956-7976.2004. 01503012.x Sundara, M., Polka, L., & Baum, S. (2006). Production of coronal stops by simultaneous bilingual adults. Bilingualism: Language and Cognition, 9, 97–114. http://dx.doi.org/10.1017/ S1366728905002403 Vanhove, J. (2013). The critical period hypothesis in second language acquisition: A statistical critique and a reanalysis. PLoS ONE, 8, e69172. http://dx.doi.org/10.1371/journal. pone.0069172



Age of Second-Language Acquisition 

 181

Weyant, J. M. (2007). Perspective taking as a means of reducing negative stereotyping of individuals who speak English as a second language. Journal of Applied Social Psychology, 37, 703–716. http://dx.doi.org/10.1111/j.1559-1816.2007.00181.x White, L., & Genesee, F. (1996). How native is near-native? The issue of ultimate attainment in adult second language acquisition. Second Language Research, 12, 233–265. http://dx.doi.org/ 10.1177/026765839601200301 Wiley, E., Bialystok, E., & Hakuta, K. (2005). New approaches to using census data to test the Critical Period Hypothesis for second language acquisition. Psychological Science, 16, 341–343. http:// dx.doi.org/10.1111/j.0956-7976.2005.01537.x Yates, L., & Kozar, O. (2015). Expanding the horizons of age-related research: A response to the Special Issue “Complexities and Interactions of Age in Second Language Learning: Broadening the Research Agenda.” Applied Linguistics. Advance online publication. Retrieved from http:// applij.oxfordjournals.org/content/early/2015/07/22/applin.amv031.short Yeni-Komshian, G. H., Flege, J. E., & Liu, S. (2000). Pronunciation proficiency in the first and second languages of Korean–English bilinguals. Bilingualism: Language and Cognition, 3, 131–149. http://dx.doi.org/10.1017/S1366728900000225

Jeff MacSwan

10 Code-Switching in Adulthood Code-switching is a speech style in which fluent bilingual speakers switch languages between or within sentences. For instance, Spanish–English bilingual speakers might say, This morning mi hermano y yo fuimos a comprar some milk (“This morning my brother and I went to buy some milk”), where the sentence begins in English, switches to Spanish, and then moves back to English again. Code-switching within a single sentence like this is called intrasentential code-switching; code-switching between sentences is known as intersentential code-switching. In the past, educators and others often assumed that code-switching was indicative of a language disability of some kind or perhaps was indicative of lazy speech. It was often alleged that bilingual speakers used code-switching as a coping strategy for incomplete mastery of both languages. However, recent research into the social and linguistic characteristics of code-switching indicates that it is a marker of bilingual identity and has its own internal grammatical structure, just like monolingual language. The term code-switching emerged in the 1950s and 1960s, and code-switching became a well-established, independent field of research by the 1970s. At the Thirteenth Annual Round Table Meeting on Linguistics and Languages at Georgetown University, held in 1962, Einar Haugen claimed origination of the term; however, the word first appeared in print in H. Vogt’s (1954) review of Uriel Weinreich’s Languages in Contact and 2 years later by Einar Haugen, according to Benson (2001). Despite these early interests, an actual code-switching research literature did not emerge until the late 1960s and early 1970s, when work focusing on social and grammatical aspects of language mixing began steadily appearing with scholarly engagement of previously published research. Early work on grammatical dimensions of code-switching aimed to illuminate language differences represented in the Latino community in the United States. A conventional perspective on educational and economic disparities represented among minority groups was that the poor did not do well in school because of inherent cultural, intellectual, and linguistic limitations (Raz, 2013). Bilingualism was among the many culprits of poor school achievement for deficit-oriented policymakers. Thus, educational researchers concerned with academic underachievement among bilingual students began to address important questions about the language these children brought to school and how it may factor into our understanding of student achievement. These concerns brought considerable attention among researchers to the linguistic study of code-switching and provided the momentum underlying the earliest work in the field. Guadalupe Valdés was among the first to emphasize the importance of a sound understanding of code-switching for classroom teachers. She observed that many http://dx.doi.org/10.1037/14939-011 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

184 

 Jeff MacSwan

teachers hold the view that “children who code-switch really speak neither English nor Spanish” (Valdés-Fallis, 1978, p. 2). This observation was echoed by Commins and Miramontes (1989), who observed that “a popular belief is that children who codeswitch . . . do so because they do not command enough pieces in either language to form a complete code; thus, they are considered semilingual” (p. 445). Indeed, the notion that code-switching might be evidence of semilingualism was made explicit in Cummins’s (1979) early formulation of the concept: Several investigators have drawn attention to the fact that some bilingual children who have been exposed to both languages in an unsystematic way prior to school come to school with less than native-like command of the vocabulary and syntactic structures of both L1 and L2 . . . Gonzalez (1977) suggests that under these conditions children may switch codes because they do not know the label for a particular concept in the language they are speaking but have it readily available in the other language. (p. 238)

Efforts have been made to evaluate the notion of semilingualism empirically, and these have consistently found no basis for the notion. Paulston (1983), for example, who conducted a review of empirical research on the topic, concluded her report this way: “Semilingualism does not exist, or put in a way which is non-refutable, has never been empirically demonstrated” (p. 54). Similarly, MacSwan (2000) reviewed four sources of reputed evidence for semilingualism—studies of language variation, linguistic structure, school performance, and language loss—and could find no empirical grounds for the idea. Similarly, studies of commercially available Spanish language tests, frequently used in school to assess Spanish language-background children as non- or limited speakers of their native language, also revealed that these tests lack validity. For instance, MacSwan and Rolstad (2006) found that although the Language Assessment Scales-Oral Español and the Idea Proficiency Test I-Oral Spanish identified 74% and 90% (respectively) of Spanish-speaking English language learners as limited speakers of their first language, a natural language measure found only 2% of the participants to have unexpectedly high morphological error rates (n = 145). Children with higher error rates were conjectured to be primarily speakers of English with limited exposure to their heritage language. See also MacSwan, Rolstad, and Glass (2002), MacSwan and Mahoney (2008), and MacSwan and Rolstad (2010). The way teachers, researchers, and others view children’s language ability is important because it will affect their views of what the children know, of their families and communities, and of the treatment children receive in school and other service contexts. The beliefs teachers hold about parents and communities may become tacit tracking mechanisms, serving to restrict the educational opportunities of bilingual students. Oakes’s (1985) classic work on tracking noted that ability grouping for children correlated highly with children’s self-perceptions of ability and career promise. Oakes noted that children who had been treated as “good students,” destined for success and achievement, wound up in higher tracks, whereas other children found themselves in lower tracks. Numerous school practices conspire to subtly put tracking



Code-Switching in Adulthood 

 185

structures in place, either overtly or covertly—attitudes and beliefs about children’s diverse language practices, and how these affect school success, among them. More specifically with regard to bilingual children, if teachers believe that codeswitching relates to an inherent disability that might be remedied by instruction, they will convey to the children that their “natural abilities” are severely limited, impacting on their success in school. In a study involving 278 elementary school teachers from 14 elementary schools in South Texas, Nava (2009) reported that a large majority of teachers viewed code-switching negatively and discouraged their students from using it. Teachers expressed the view that code-switching reflects limited proficiency in both languages and interferes with academic and cognitive development. Although teachers from more diverse school settings were more likely to have positive views about codeswitching, many teachers in the study significantly viewed code-switching as having a negative effect on learning (84% at less diverse schools, 17% at more diverse schools). Ramirez and Milk (1986) similarly found that teachers differentiated “standard American English” from three marked varieties, with “Hispanicized English” rated more favorably than grammatically ill-formed (descriptively speaking) English constructions and code-switching. Of the four varieties of language differentiated in the study, codeswitching was consistently ranked “least acceptable” by teachers. At the core of the concern about code-switching is the worry that bilingual speakers who code-switch may be motivated by inadequate linguistic competence of some kind. However, as in the case of language prejudice generally—indeed, as in the case of any species of deficit psychology—the burden of proof for the deficiency is on those who assert it, who must show that a specific variety reflects specified deficiencies, not just differences, and that these deficiencies have the sort of negative cognitive effects attributed to them. Although no such evidence has been developed, advocates of what Newmeyer (1986) called linguistic equality have taken to asserting the adequacy of stigmatized language varieties in much the same way as Franz Boas (1911) showed that non-Western languages were as linguistically sophisticated as their European counterparts, or as William Labov (1970) and many others showed that African American English is as rich and complex as the English of the educated classes—by presenting linguistic analyses that reveal its structure and richness. Research on code-switching has followed suit, falling into two major research areas: the study of code-switching as language use (discourse, pragmatics, and conversation) and the study of code-switching as a grammatical system. We will review each of these areas in turn.

Code-Switching as Language Use Myers-Scotton (1993b) credits Blom and Gumperz (1972) with sparking interest in social aspects of code-switching. Blom and Gumperz studied code-switching between dialects of Norwegian in Hemnesberget, a Norwegian fishing village. Although the topic

186 

 Jeff MacSwan

was actually introduced in previous work (Gumperz & Hernandez-Chavez, 1970), the chapter by Blom and Gumperz received considerably more exposure because it was included in Gumperz and Hymes’s (1972) edited collection, which became a standard textbook in the many new sociolinguistics courses created in the 1970s. In early work, Gumperz analyzed code-switching as situational or metaphorical, adding conversational code-switching in Gumperz (1982). A situational switch involves a change in participants and/or strategies, whereas metaphorical code-switching involves a change in topical emphasis (Gumperz & Hymes, 1972). These switches appear to be motivated by speaker-external factors, according to Gumperz. Gumperz (1982) proposed six major functions for conversational code-switching: (a) quotation, (b) addressee specification, (c) interjection, (d) reiteration, (e) message qualification, and (f) personification vs. objectification. Valdés (1981) posited two additional ones: (g) mitigating the illocutionary effects of speech acts and (h) aggravating the illocutionary effects of speech acts. Gumperz’s (1982) model focused on microlevel sociolinguistic analysis, using naturally occurring data from small-group interactions as the core data source. Whereas Labov and other sociolinguists tied language use to sociological variables, Gumperz considered language use to be a function of the dynamics of interaction. Thus, for Gumperz, language choice conveys intentional meaning of a sociopragmatic sort: “Detailed observation of verbal strategies revealed that an individual’s choice of speech style has symbolic value and interpretive consequences that cannot be explained simply by correlating the incidence of linguistic variants with independently determined social and contextual categories” (Gumperz, 1982, p. vii). Whereas Gumperz’s approach sought to distill symbolic and interpretive value from the analysis of code-switching, Myers-Scotton’s (1993a) markedness model sought to provide a predictive theory of language choice in a bilingual exchange; according to the markedness model, “speakers have a sense of markedness regarding available linguistic codes for any interaction, but choose their codes based on the persona and/ or relation with others which they wish to have in place” (Myers-Scotton, 1993b, p. 75). Myers-Scotton (1993b) posited a negotiation principle that underlies all code-switching events: (1) The Negotiation Principle: Choose the form of your conversation contribution such that it indexes the set of rights and obligations (the RO set) which you wish to be in force between speaker and addressee for the current exchange. (p. 113) The RO set is indexed by features. Speaking English in Nairobi, for instance, “may be indexical of any of a set of attributes, including most prominently ‘plus high educational level/socio-economic status’, ‘plus authority’, ‘plus formality’, and ‘plus official’” (Myers-Scotton, 1993b, p. 86). In other words, people make assumptions about others based on how they speak, and the choices they make in conversation may signal particular social relationships which they seek to establish. In relation



Code-Switching in Adulthood 

 187

to such taxonomies, Bailey (2002) critically observed, “The ease with which such cate­gories can be created—and discrepancies between the code switching taxonomies at which researchers have arrived—hint at the epistemological problems of such taxonomies” (p. 77). Indeed, the markedness model has been critically evaluated by some codeswitching scholars as well (Cashman, 2008; Meeuwis & Blommaert, 1994; Wei, 1994, 2002), who view it as a static perspective on social behavior in which speakers are described according to existing norms. As an alternative approach, Auer (1995) and Wei (2002) have advocated the sequential approach, which derives from work in conversational analysis (CA). Rather than provide “motivational speculation” regarding the code-switching of interlocutors, the CA approach is focused on a turn-by-turn analysis of language choice and how the meaning of code-switching is constructed (Wei, 2002). Hence, CA researchers are concerned more with how code-switching is used to make meaning and create identities rather than the underlying motivation of language switching, which dominated Gumperz’s (1982) original analysis as well as the markedness model (Myers-Scotton, 1993b). Research on language mixing with a focus on social identity formation has also emerged in the context of recent contributions to translanguaging as a topic of study. According to Lewis, Jones, and Baker (2012), the term translanguaging was originally used by Williams (1994) to denote natural language mixing in pedagogical settings before it was popularized by García (2009). Similar recent terms include heteroglossia (Bakhtin, 1975), hybrid language practices (Gutiérrez, Baquedano-López, & Tejeda, 1999), polylanguaging and polylingual languaging (Jørgensen, 2008; Jørgensen, Karrebæk, Madsen, & Møller, 2011), metrolingualism (Otsuji & Pennycook, 2011), code­ meshing (Canagarajah, 2011), translingual practice (Canagarajah, 2011), and multi­ languaging (Nguyen, 2012). Building on prior work in social network analysis of language choice in code-switching, Wei (2011) defined translanguaging as “the full range of linguistic performances of multilingual language users for purposes that transcend the combination of structures, the alternation between systems, the transmission of information and the representation of values, identities and relationships” (p. 1223). Translanguaging is built on the concept of languaging, which is an analysis of language used to gain knowledge and to make sense (Lado, 1979; Smagorinsky, 1998; Swain, 2006). As Chomsky (1959) noted in his classic review of behaviorism, efforts to develop a set of explanatory principles that can reliably predict when and how people will say what seems beyond reach, as Lipski (1978) echoed years ago in relation to our theory of code-switching: Indeed, since the role of individual idiosyncratic factors seems to be an important aspect of codeswitching, in that among groups of approximately equal bilingual abilities, some code-switch more than others, a complete determination of the sufficient conditions for code-switching probably lies beyond the reach of the behavioral sciences. With regard to the linguistic constraints, however, the path toward an eventual model seems more clearly indicated. (p. 261)

188 

 Jeff MacSwan

Although the markedness model sought to develop a predictive framework for understanding language choice among interlocutors, the sequence approach and subsequent work conducted through the analytic lens of translanguaging has sought to develop a descriptive framework permitting the analysis of code-switching as a conversational strategy for communication and knowledge construction. And as Lipski’s remarks further foreshadowed, research on linguistic constraints, which seek to develop a grammar of code-switching governing permissible mixed-language sequences, has proven to be a fruitful and highly successful undertaking, too.

Grammatical Studies of Code-Switching Like monolingual language, bilingual code-switching is highly structured and rulegoverned. Although linguists debate the specific details of the underlying rule system, that it is structured and rule-governed is not in doubt. One of the most important early contributions to code-switching research was Sankoff and Poplack’s (1981) implementation of the equivalence constraint. Several researchers had converged on the notion that language switching is controlled by some kind of syntactic equivalence requirement (Lipski, 1978; Pfaff, 1979; Poplack, 1978, 1981), but the ideas were best developed in Poplack’s (1981) two complementary constraints: (2) The equivalence constraint: Codes will tend to be switched at points where the surface structures of the languages map on each other. (3) The free morpheme constraint: A switch may occur at any point in the discourse at which it is possible to make a surface constituent cut and still retain a free morpheme. As a variationist, Poplack believed that linguistic rules correlate with social structure and should be stated in terms of statistical frequencies, hence (2) is expressed as a tendency. The general idea is nonetheless clear: code-switching is allowed within constituents so long as the word order requirements of both languages are met at surface structure. The constraint in (3) defines a restriction on morphology in code-switching contexts, also noted in Wentz and McClure (1977) and Pfaff (1979). To illustrate, (2) correctly predicts that the switch in (4) is disallowed, because the surface word order of English and Spanish differ with respect to object pronoun (clitic) placement; (3) correctly disallows (5), where an English stem is used with a Spanish bound morpheme without the phonological integration of the stem. (4) *told le, le told, him dije, dije him told to-him, to-him I-told, him I-told, I-told him “(I) told him” (Poplack, 1981, p. 176) (5) *estoy eat-iendo I-am eat-ing (Poplack, 1980, p. 586)



Code-Switching in Adulthood 

 189

Keep in mind that these restrictions are intended as descriptions, not prescriptions; as in linguistics generally, the intention is to capture the underlying psychological intuitions of native (or nativelike) speakers from a specific speech community— in this case, a bilingual speech community. Similarly, English speakers intuitively regard Martin painted the barn red as well-formed, and Martin built the barn red as ill-formed. Although Sankoff and Poplack (1981) expressed a strong preference for avoiding code-switching-specific mechanisms to mediate between the two languages in contact, they nonetheless concluded that such a mechanism is necessary on empirical grounds. Otherwise, the authors argued, the free union of Spanish and English phrase structure grammars would yield ill-formed results. For instance, whereas English requires pre-nominal adjectives (NP → Det Adj N), Spanish requires post-nominal adjective placement (NP → Det N Adj). A speaker is free to select the Spanish rule and lexically insert an English determiner, Spanish noun and English adjective (*the casa white) or even insert English lexical items for all categories (*the house white). Therefore, to constrain the grammars so that they do not generate violations of (2), Sankoff and Poplack introduced a superscripting mechanism (sometimes called bilingual tagging or language tagging) that restricted lexical insertion rules so that the grammar contributing the phrase structure rule would also be the grammar from which lexical insertion rules were drawn. Hence, under conditions of code-switching, the Spanish phrase structure rule would be annotated as in (6a), generating (6b). The superscripting conventions followed from heritability conditions, according to the authors, which essentially allowed phrase structure rules to look ahead and restrict the application of lexical insertion rules. (6a) NP → Det Nsp:n Adjsp:adj (6b) the casa blanca Sankoff and Poplack (1981) did not make explicit the mechanisms for superscript insertion; rather, they indicated that phrase structure rules are so superscripted when they are selected in the generation of a code-switched utterance, and are subsequently used to trigger language-specific lexical insertion rules (N → casa, for instance, in the case of Nsp:n). No account was presented as to how the superscript insertion mechanism is able to annotate the appropriate categories correctly—for instance, N and Adj in (6a), but not Det, where either language may be inserted without negative consequences. See MacSwan (2013, 2014) for further discussion. Despite limitations, Sankoff and Poplack’s work made important contributions to our understanding of the formal properties of code-switching and helped to clearly establish code-switching as a field of linguistic inquiry. Contemporary approaches have recast these and many other observations about code-switching in terms of lexically focused grammars that build structures from the bottom up rather than the top down, as in the syntactic theories available at the time of Sankoff’s original work. In a bilingual grammar, lexical items are drawn from

190 

 Jeff MacSwan

the lexicon of either language to introduce features into the lexical array, or the small subset of items that will make up the bilingual expression. According to Chametzky (2003), the lexical entry driven approach to syntax was part of the general effort underlying X’ reduction, with significant contributions from Stowell (1981) and Speas (1990), among others. With a return to its derivational roots, minimalist syntax reduced generation to the simplest possible form—free Merge, building structures from the ground (the lexical string) up (the hierarchical phrase structure) based on the specification of lexically encoded features. In the minimalist program there are two components of grammar: CHL, a computational system for human language, believed to be invariant across languages; and a lexicon, to which the idiosyncratic differences observed across languages are attributed. An operation called Select picks lexical items from the lexicon and introduces them into a Numeration or Lexical Array (LA), a finite subset of the lexicon used to construct a derivation. Merge takes items from the LA and forms new, hierarchically arranged syntactic objects. Movement operations (Internal Merge) apply to syntactic objects formed by Merge to rearrange elements within a tree (Chomsky, 1995, 2000). Phrase structure trees are thus built derivationally by the application of the operations Select and Merge, constrained by the condition that lexically encoded features match in the course of a derivation. Movements are driven by feature valuation and may be of two types. A head, or terminal in a phrase structure tree, may undergo movement and adjoin to another head. Similarly, a maximal projection may move to the specifier position of a head. In either case, the element moves for the purpose of valuing morphological features of case and f (number, person, and gender). In addition, its movement may be overt or covert. Overt movements are driven by strong features and are visible at PF (phonetic form; where they are pronounced) and LF (logical form; where they are interpreted). Covert movements, driven by weak features, are visible only at LF. At some point in the derivation, an operation Spell-Out applies to strip away from the derivation those elements relevant only to PF, prepared for articulation; what remains is mapped to LF by a subsystem of CHL called the covert component, preparing it for semantic interpretation. The elements relevant only to PF are mapped to PF by operations unlike the covert component, operations that comprise the phonological component. The phonological component is also regarded as a subsystem of CHL. The subsystem of CHL that maps the lexicon to Spell-Out is the overt component. Note that the various components (overt, covert, phonological) are all part of CHL, the computational system for human language. Applied to bilingual code-switching, the minimalist approach to syntax may be illustrated as in Figure 10.1. The leading aim of the minimalist program is the elimination of all mechanisms that are not necessary and essential on conceptual grounds alone; thus, only the minimal theoretical assumptions may be made to account for linguistic data, privileging more simplistic and elegant accounts over complex and cumbersome ones. These assumptions would naturally favor accounts of code-switching that make use of independently

Code-Switching in Adulthood 

 191

Cycle Again at Each Phase



Fig. 10.1: A minimalist model of bilingual code-switching.

motivated principles of grammar over those that posit rules, principles, or other constructs specific to it. Thus, although free union of phrase structure grammars led to the complications noted by Sankoff and Poplack (1981) for reasons associated with late lexical insertion, free union of lexically encoded grammars does not. We see that we can capture the range of data covered by the free morpheme constraint, and extend it further to other data, by examining whether other sorts of word merger may similarly be disallowed in code-switching. Consider, for example, the code-switching examples in (7) from Spanish–Nahuatl bilingual speakers (MacSwan, 1999). (7a) *No nitekititoc no ni-tekiti-toc not 1S-work-DUR “I’m not working” (7b) Amo estoy trabajando amo estoy trabaja-ndo not be/PRES/1Ss work-DUR “I’m not working”

192 

 Jeff MacSwan

Although Spanish and Nahuatl have the same basic word order requirements with respect to negation, Spanish negation does not permit a Nahuatl verb in its complement position in (7a), but Nahuatl negation followed by a Spanish verb is well formed in (7b). Consider some properties of Spanish negation. Zagona (1988) argued that Spanish no is a syntactic clitic and forms part of the Spanish verbal complex as a result of head movement. To make a case for this analysis, Zagona pointed out that Spanish no must be fronted with the verb in (8), unlike the adverbs in (9). (8)  ¿Qué no dijo Juan? what not say/1Ss/PAST Juan “What didn’t Juan say?” (9a) *¿Qué sólo leyó Juan? what only read/1Ss/PAST Juan “What did Juan only read?” (9b) *¿Qué meramente leyó Juan? what merely read/1Ss/PAST Juan “What did Juan merely read?” Zagona (1988) further pointed out that Spanish no cannot be contrastively stressed in (10a) as its English counterpart in (10b) can be, owing to the fact that clitics are inherently unstressable. The example in (10b) shows that in English, in contrast to Spanish, the negative element is not a syntactic clitic. (10a) *Juan no ha no hecho la tarea Juan not has not done the task “Juan hasn’t not done the task” (10b) Juan hasn’t not done the task These facts suggest that in Spanish, the verb is a host for negation. Nahuatl, on the other hand, behaves differently from French and Spanish with regard to negation. A test similar to the one Zagona uses (1988) in (10) shows that Nahuatl patterns with English: (11) Amo nio amo niktati nowelti amo ni-o amo ni-k-tati no-welti not 1S-go amo 1S-3Os-see my-sister “I’m not going to not see my sister” Because clitics are inherently unstressable, we may conclude from (11) that amo is not a clitic in Nahuatl.



Code-Switching in Adulthood 

 193

The facts suggest the possibility of a ban on code-switching in head movement contexts in which two word-level syntactic units merge at a structural level. As corroborating evidence, consider code-switching in restructuring contexts, a well-studied and classic example of head movement. According to Rizzi’s (1982) analysis, Italian modals, aspectuals, and motion verbs comprise the class of restructuring verbs, which behave differently from other verbs, as illustrated in the Italian examples in (12) and (13). (12a) Finalmente si comincerà a costruire le nuove case popolari Finally si begin/fut to build the new houses people/gen “Finally we’ll begin to build the new houses for the poor.” (12b) Finalmente le nuove case popolari si cominceranno a costruire (Same meaning as (12a).) (13a) Finalmente si otterrà di costruire le nuove case popolari Finally si get.permission/fut to build the new houses people/gen “Finally we’ll get permission to build the new houses for the poor.” (13b) *Finalmente le nuove case popolari si otterranno di costruire (Same meaning as (13a).) In Rizzi’s analysis, comincerà “will begin,” but not otterrà “will get permission,” triggers an optional reanalysis of the form Vx (P) V2 ⇒ V, where Vx is a verb of the restructuring class, (P) an optional intervening preposition, and V2 is the verb of the embedded sentence. This restructuring process may be seen as a type of compounding by way of head movement, resulting in [V Vx V2]. In (12) a reanalysis of the constituents allows the object of the embedded clause in an impersonal si construction to move to the subject position of the matrix clause; in (13) this promotion is barred because reanalysis cannot apply for otterrà. Importantly, reanalysis is optional in Italian; it has applied in (13b), allowing the promotion of the embedded object to subject position, but it has not applied in (13a) where the object of the embedded clause remains in situ. Aspectual essere is used with a past participle in Italian passive impersonal si constructions. In constructions such as (14a), essere too may be viewed as a restructuring verb, allowing promotion of the embedded object to subject position, shown in (14b). (14a) Si è dato un regalo si essere given a gift “A gift is given.” (14b) Un regalo si è dato a gift si essere given “A gift is given.”

194 

 Jeff MacSwan

Within Rizzi’s (1982) system, restructuring has applied to (14b) but not to (14a), forcing the promotion of [NP un regalo] in the former case. To test further whether code-switching is prohibited in head movement contexts, we may now examine cases of language switching in this context, as in (15). (15a) Si è donné un cadeau si essere given a gift (15b) *Un cadeau si è donné a gift si essere given The movement of [NP un cadeau] indicates that reanalysis has occurred in (15b), just as it did in (14b). The verbal complexes are identical in (15a) and (15b): A mixture of the Italian aspectual auxiliary è immediately adjacent to the French past participle donné. Once again, the facts appear to indicate that code-switching in restructuring configurations, an instance of head movement, is prohibited, leading us to conclude that a general ban on code-switching in head movement contexts is at play. These facts extend the observation of the free morpheme constraint discussed earlier. The prohibition against merging word-like units from different languages in code-switching appears to be related to the nature of phonological processing, which does not tolerate systemic alteration. Descriptively, these facts support Poplack’s (1981) free morpheme constraint, which has been attested in numerous corpora (Bentahila & Davies, 1983; BerkSeligson, 1986; Clyne, 1987; MacSwan, 1999). However, some researchers disagree about the generalization, and have reported counterexamples (Bokamba, 1989; Chan, 1999; Halmari, 1997; Hlavac, 2003; Myers-Scotton, 1993a; Nartey, 1982). For further discussion, see MacSwan (2005a) and MacSwan and Colina (2014). The approach just sketched within the minimalist framework is a constraintfree (CF) implementation of code-switching, and contrasts with constraint-based (CB) approaches, illustrated by Sankoff and Poplack’s (1981) approach. The guiding research principle underlying the CF approach is (16) (MacSwan, 1999, 2014). (16) Nothing constrains code-switching apart from the requirements of the mixed grammars. This principle implies that any theory purporting to provide a grammatical theory of language mixing may not make reference to code-switching–specific mechanisms, like (2) and (3), or posit solutions which make reference to the (geopolitical) identities of specific languages. It should be pointed out that code-switching as a field remains divided along these lines, with the matrix language frame (MLF) model (Myers-Scotton, 1993a) representing the current leading CB approach. The MLF model differentiates the languages involved in code-switching as the matrix language (ML) and embedded language (EL).



Code-Switching in Adulthood 

 195

According to the MLF Model, the matrix language defines the surface structure positions for content words and functional elements using two basic principles: the morpheme order principle, which requires that morphemes within a bilingual constituent follow the order prescribed by the ML, and the system morpheme principle, which states that all system morphemes–defined as morphemes that have grammatical relations with other constituents outside their maximal projections—come from the ML in any code-switched utterance. The MLF model is a CB approach because it makes explicit reference to language mixing in its principles, rather than deriving the grammaticality facts associated with mixed-language expressions by reference to independently motivated principles of grammar. For further discussion of the debate between these approaches, see Hebblethwaite (2010); Herring, Deuchar, Couto, and Quintanilla (2010); Jake, Myers-Scotton, and Gross (2002, 2005); and MacSwan (2005a, 2005b, 2014), among others.

Conclusion Evidence uncontroversially shows that bilingual code-switching is rule-governed and structured, and that bilingual speakers who code-switch exhibit profound sensitivity to extremely subtle properties of language which lie deep within the subconscious. Thus, not only is there no evidence that code-switching reflects a linguistic deficiency of some sort, but, significantly, evidence suggests that code-switchers have the very same kind of sensitivity to linguistic structure as monolinguals do. An Italian–French bilingual, for instance, seeks to move an embedded Italian subject in (15b) only to recognize that the construction crashes: Raising the embedded subject causes restructuring of the verbal complex in (15b), forcing language switching during phonological parsing—prohibited by the nature of the phonological rule systems. Similar restrictions emerge when a Spanish–Nahuatl speaker code-switches between Spanish negation and a Nahuatl verb, but not between a Nahuatl verb and Spanish negation, as shown in (7), because the bilingual speaker implicitly knows that Spanish negation syntactically merges with its verb, whereas Nahuatl negation does not. The evidence in (5) indicates that Spanish–English bilingual speakers know these restrictions too, as they apply to presyntactic affixation, and (4) shows that Spanish–English bilingual speakers know that pronominal clitics, like cliticized negation, do not tolerate codeswitching as this would entail language switching during phonological parsing. All this knowledge is subconscious, of course, just as it is for the monolingual speakers of the languages involved. Contrary to casual assertions, the linguistic study of bilingual code-switching has revealed that simultaneous bilingual speakers, just like monolingual speakers, are sensitive to extremely subtle requirements of their linguistic systems, and use their languages creatively to satisfy a variety of social purposes, and to achieve a sense of identity as part of a bilingual or multilingual community. Furthermore, research

196 

 Jeff MacSwan

suggests that code-switching is an intrinsic identity marker—bilingual communities mix languages in part to assert their bilingual identities, which distinguish them from the monolingual identity associated with either of the languages they speak. As such, researchers and practitioners should respond to bilingual speakers as bilingual speakers, treating language mixing as a highly skilled, intricate practice that shows profound linguistic competence and asserts a specific identity.

References Auer, P. (1995). The pragmatics of code-switching: A sequential approach. In L. Milroy & P. Muysken (Eds.), One speaker, two languages: Cross-disciplinary perspectives (pp. 115–135). Cambridge, England: Cambridge University Press. http://dx.doi.org/10.1017/CBO9780511620867.006 Bailey, B. (2002). Language, race, and negotiation of identity: A study of Dominican Americans. New York, NY: LFB Scholarly Publishing. Bakhtin, M. (1975). Heteroglossia in the novel. In S. Dentith (Ed.), Bakhtinian thought: An introductory reader (195–224). London, England: Routledge. Benson, E. J. (2001). The neglected early history of codeswitching research in the United States. Language & Communication, 21, 23–36. http://dx.doi.org/10.1016/S0271-5309(00)00012-4 Bentahila, A., & Davies, E. E. (1983). The syntax of Arabic–French code-switching. Lingua, 59, 301–330. http://dx.doi.org/10.1016/0024-3841(83)90007-4 Berk-Seligson, S. (1986). Linguistic constraints on intrasentential code-switching: A study of Spanish/Hebrew bilingualism. Language in Society, 15, 313–348. http://dx.doi.org/10.1017/ S0047404500011799 Blom, J. P., & Gumperz, J. J. (1972). Social meaning and structure: Code-switching in Norway. In J. Gumperz & D. Hymes (Eds.), Directions in sociolinguistics (pp. 407–434). New York, NY: Holt, Reinhart & Winston. Boas, F. (1911). Handbook of American Indian languages. Washington, DC: Smithsonian Institution, Bureau of American Ethnology. Bokamba, E. G. (1989). Are there syntactic constraints on code-mixing? World Englishes, 8, 277–292. http://dx.doi.org/10.1111/j.1467-971X.1989.tb00669.x Canagarajah, S. (2011). Codemeshing in academic writing: Identifying teachable strategies of translanguaging. Modern Language Journal, 95, 401–417. http://dx.doi.org/10.1111/ j.1540-4781.2011.01207.x Cashman, H. R. (2008). Conversation and Interactional Analysis. In L. Wei & M. G. Moyer (Eds.), The Blackwell guide to research methods in bilingualism and multilingualism (pp. 275–295). Oxford, England: Blackwell. Chametzky, R. A. (2003). Phrase structure. In R. Hendrick (Ed.), Minimalist syntax (pp. 192–225). Oxford, England: Blackwell. http://dx.doi.org/10.1002/9780470758342.ch5 Chan, B. H.-S. (1999). Aspects of the syntax, production and pragmatics of code-switching with special reference to Cantonese-English (Unpublished doctoral dissertation). University College London, London, England. Chomsky, N. (1959). Review of B. F. Skinner’s Verbal Behavior. Language, 35, 26–58. http://dx.doi.org/ 10.2307/411334



Code-Switching in Adulthood 

 197

Chomsky, N. (1995). The minimalist program. Cambridge, MA: MIT Press. Chomsky, N. (2000). Minimalist inquiries: The framework. In R. Martin, D. Michaels, & J. Uriagereka (Eds.), Step by step: Essays on minimalist syntax in honor of Howard Lasnik (pp. 89–155). Cambridge, MA: MIT Press. Clyne, M. (1987). Constraints on code switching: How universal are they? Linguistics, 25, 739–764. http://dx.doi.org/10.1515/ling.1987.25.4.739 Commins, N. L., & Miramontes, O. B. (1989). Perceived and actual linguistic competence: A descriptive study of four low-achieving Hispanic bilingual students. American Educational Research Journal, 26, 443–472. http://dx.doi.org/10.3102/00028312026004443 Cummins, J. (1979). Linguistic interdependence and the educational development of bilingual children. Review of Educational Research, 49, 222–251. http://dx.doi.org/10.3102/ 00346543049002222 García, O. (2009). Bilingual education in the 21st century: A global perspective. Oxford, England: Wiley-Blackwell. Gumperz, J. J. (1982). Discourse strategies. New York, NY: Cambridge University Press. http:// dx.doi.org/10.1017/CBO9780511611834 Gumperz, J. J., & Hernandez-Chavez, E. (1970). Cognitive aspects of bilingual communication. In W. H. Whitely (Ed.), Language and social change (pp. 111–125). Oxford, England: Oxford University Press. Gumperz, J. J., & Hymes, D. (Eds.). (1972). Directions in sociolinguistics. New York, NY: Holt, Reinhart & Winston. Gutiérrez, K., Baquedano-López, P., & Tejeda, C. (1999). Rethinking diversity: Hybridity and hybrid language practices in the third space. Mind, Culture and Activity, 6, 286–303. http://dx.doi.org/ 10.1080/10749039909524733 Halmari, H. (1997). Government and code switching: Explaining American Finnish. New York, NY: John Benjamins. http://dx.doi.org/10.1075/sibil.12 Hebblethwaite, B. (2010). Adverb code-switching among Miami’s Haitian Creole–English second generation. Bilingualism: Language and Cognition, 13, 409–428. http://dx.doi.org/10.1017/ S1366728909990563 Herring, J. R., Deuchar, M., Couto, M. C., & Quintanilla, M. M. (2010). ‘I saw the madre’: Evaluating predictions about codeswitched determiner-noun sequences using Spanish–English and Welsh–English data. International Journal of Bilingual Education and Bilingualism, 13, 553–573. http://dx.doi.org/10.1080/13670050.2010.488286 Hlavac, J. (2003). Second-generation speech: Lexicon, code-switching, and morpho-syntax of Croatian–English bilinguals. New York, NY: Peter Lang. Jake, J., Myers-Scotton, C., & Gross, S. (2002). Making a minimalist approach to codeswitching work: Adding the matrix language. Bilingualism: Language and Cognition, 5, 69–91. http://dx.doi.org/ 10.1017/S1366728902000147 Jake, J., Myers-Scotton, C., & Gross, S. (2005). A response to MacSwan (2005): Keeping the matrix language. Bilingualism: Language and Cognition, 8, 271–276. http://dx.doi.org/10.1017/ S1366728905002300 Jørgensen, J. N. (2008). Polylingual languaging around and among children and adolescents. International Journal of Multilingualism, 5, 161–176. http://dx.doi.org/10.1080/ 14790710802387562 Jørgensen, J. N., Karrebæk, M. S., Madsen, L. M., & Møller, J. S. (2011). Polylanguaging in super­ diversity. Diversities, 13(2), 23–38.

198 

 Jeff MacSwan

Labov, W. (1970). The logic of non-standard English. In F. Williams (Ed.), Language and poverty (pp. 153–189). Chicago, IL: Rand McNally. http://dx.doi.org/10.1016/ B978-0-12-754850-0.50014-3 Lado, R. (1979). Thinking and “languaging”: A psycholinguistic model of performance and learning. Sophia Linguistica, 12, 3–24. Lewis, G., Jones, B., & Baker, C. (2012). Translanguaging: Origins and development from school to street and beyond. Educational Research and Evaluation, 18, 641–654. http://dx.doi.org/ 10.1080/13803611.2012.718488 Lipski, J. (1978). Code-switching and the problem of bilingual competence. In J. Paradis (Ed.), Aspects of bilingualism (pp. 250–264). Columbia, SC: Hornbeam Press. MacSwan, J. (1999). A minimalist approach to intrasentential code switching. New York, NY: Garland Press. MacSwan, J. (2000). The threshold hypothesis, semilingualism, and other contributions to a deficit view of linguistic minorities. Hispanic Journal of Behavioral Sciences, 22(1), 3–45. http://dx.doi.org/ 10.1177/0739986300221001 MacSwan, J. (2005a). Codeswitching and generative grammar: A critique of the MLF model and some remarks on “modified minimalism.” Bilingualism: Language and Cognition, 8, 1–22. http:// dx.doi.org/10.1017/S1366728904002068 MacSwan, J. (2005b). Remarks on Jake, Myers-Scotton and Gross’s response: There is no “matrix language.” Bilingualism: Language and Cognition, 8, 277–284. http://dx.doi.org/10.1017/ S1366728905002312 MacSwan, J. (2013). Code switching and linguistic theory. In T. K. Bhatia & W. Ritchie (Eds.), Handbook of bilingualism and multilingualism (2nd ed., pp. 223–350). Oxford, England: Blackwell. MacSwan, J. (2014). Programs and proposals in codeswitching research: Unconstraining theories of bilingual language mixing. In J. MacSwan (Ed.), Grammatical theory and bilingual codeswitching (pp. 1–33). Cambridge, MA: MIT Press. MacSwan, J., & Colina, S. (2014). Some consequences of language design: Codeswitching at the PF interface. In J. MacSwan (Ed.), Grammatical theory and bilingual codeswitching (pp. 185–210). Cambridge, MA: MIT Press. MacSwan, J., & Mahoney, K. (2008). Academic bias in language testing: A construct validity critique of the IPT I Oral Grades K-6 Spanish Second Edition. Journal of Educational Research and Policy Studies, 8, 85–100. MacSwan, J., & Rolstad, K. (2006). How language proficiency tests mislead us about ability: Implications for English language learner placement in special education. Teachers College Record, 108, 2304–2328. http://dx.doi.org/10.1111/j.1467-9620.2006.00783.x MacSwan, J., & Rolstad, K. (2010). The role of language in theories of academic failure for linguistic minorities. In J. Petrovic (Ed.), International perspectives on bilingual education: Policy, practice, and controversy (pp. 173–195). Charlotte, NC: Information Age. MacSwan, J., Rolstad, K., & Glass, G. V. (2002). Do some school-age children have no language? Some problems of construct validity in the Pre-LAS Español. Bilingual Research Journal, 26, 395–420. http://dx.doi.org/10.1080/15235882.2002.10668718 Meeuwis, M., & Blommaert, J. (1994). The “markedness model” and the absence of society: Remarks on codeswitching. Multilingua, 13, 387–423. Myers-Scotton, C. (1993a). Dueling languages: Grammatical structure in codeswitching. Oxford, England: Clarendon Press.



Code-Switching in Adulthood 

 199

Myers-Scotton, C. (1993b). Social motivations for codeswitching: Evidence from Africa. Oxford, England: Clarendon Press. Nartey, J. S. (1982). Code-switching, interference or faddism? Language use among educated Ghanaians. Anthropological Linguistics, 24, 183–192. Nava, N. (2009). Elementary teachers’ attitudes and beliefs towards their students’ use of codeswitching in South Texas. Lenguaje, 37, 135–158. Newmeyer, F. J. (1986). The politics of linguistics. Chicago, IL: The University of Chicago Press. Nguyen, H. H. (2012). The multilanguaging of a Vietnamese American in South Philadelphia. Working Papers in Educational Linguistics, 27, 65–85. Oakes, J. (1985). Keeping track: How schools structure inequality. New Haven, CT: Yale University Press. Otsuji, E., & Pennycook, A. (2011). Social inclusion and metrolingual practices. International Journal of Bilingual Education and Bilingualism, 14, 413–426. http://dx.doi.org/10.1080/ 13670050.2011.573065 Paulston, C. B. (1983). Swedish research and debate about bilingualism. Stockholm, Sweden: National Swedish Board of Education. Pfaff, C. (1979). Constraints on language mixing: Intrasentential code-switching and borrowing in Spanish/English. Language, 55, 291–318. http://dx.doi.org/10.2307/412586 Poplack, S. (1978). Syntactic structure and social function of code-switching. In R. Duran (Ed.), Latino discourse and communicative behavior (pp. 169–184). New York, NY: Ablex. Poplack, S. (1980). “Sometimes I’ll start a sentence in Spanish y termino en Español”: Toward a typology of code-switching. Linguistics, 18, 581–618. http://dx.doi.org/10.1515/ling.1980.18.7-8.581 Poplack, S. (1981). The syntactic structure and social function of code-switching. In R. Durán (Ed.), Latino language and communicative behavior (pp. 169–184). Norwood, NJ: Ablex. Ramirez, A. G., & Milk, R. D. (1986). Notions of grammaticality among teachers of bilingual pupils. TESOL Quarterly, 20, 495–513. http://dx.doi.org/10.2307/3586296 Raz, M. (2013). What’s wrong with the poor? Psychiatry, race, and the war on poverty. Chapel Hill, NC: The University of North Carolina Press. Rizzi, L. (1982). Issues in Italian syntax. Dordrecht, the Netherlands: Foris. http://dx.doi.org/10.1515/ 9783110883718 Sankoff, D., & Poplack, S. (1981). A formal grammar for code-switching. Papers in Linguistics (Edmonton), 14, 3–45. http://dx.doi.org/10.1080/08351818109370523 Smagorinsky, P. (1998). Thinking and speech and protocol analysis. Mind, Culture, and Activity, 5, 157–177. http://dx.doi.org/10.1207/s15327884mca0503_2 Speas, M. (1990). Phrase structure in natural language. Dordrecht, the Netherlands: Kluwer. Stowell, T. (1981). Origins of phrase structure (Unpublished doctoral dissertation). Massachusetts Institute of Technology, Cambridge, MA. Swain, M. (2006). Languaging, agency and collaboration in advanced second language learning. In H. Byrnes (Ed.), Advanced language learning: The contributions of Halliday and Vygotsky (pp. 95–108). London, England: Continuum. Valdés, G. (1981). Code switching as deliberate verbal strategy: A microanalysis of direct and indirect requests among bilingual Chicano speakers. In R. Durán (Ed.), Latino language and communicative behavior (pp. 95–107). Norwood, NJ: Ablex. Valdés-Fallis, G. (1978). Code switching and the classroom teacher. Language in education: Theory and practice 4. Washington, DC: Center for Applied Linguistics. Vogt, H. (1954). Review of Languages in Contact. Word, 10, 79–82.

200 

 Jeff MacSwan

Wei, L. (1994). Three generations, two languages, one family. Clevedon, England: Multilingual Matters. Wei, L. (2002). ‘What do you want me to say?’ On the conversation analysis approach to bilingual interaction. Language in Society, 31, 159–180. http://dx.doi.org/10.1017/S0047404501020140 Wei, L. (2011). Moment analysis and translanguaging space: Discursive construction of identities by multilingual Chinese youth in Britain. Journal of Pragmatics, 43, 1222–1235. Wentz, J., & McClure, E. (1977). Aspects of the syntax of the code-switched discourse of bilingual children. In F. J. Ingemann (Ed.), 1975 Mid-American Linguistics Conference papers (pp. 351–370). Lawrence: University of Kansas. Williams, C. (1994). Arfarniad o ddulliau dysgu ac addysgu yng nghyd-destun addysg uwchradd ddwyieithog [An evaluation of teaching and learning methods in the context of bilingual secondary education] (Unpublished doctoral dissertation). University of Wales, Bangor, England. Zagona, K. (1988). Verb phrase syntax: A parametric study of English and Spanish. Boston, MA: Kluwer Academic. http://dx.doi.org/10.1007/978-94-009-2717-9

III Academic Achievement and Literacy in Bilinguals

Kathryn Lindholm-Leary

11 Bilingualism and Academic Achievement in Children in Dual Language Programs The past two decades have witnessed a proliferation of dual language (DL) programs in the United States, whose stated goals are bilingualism, biliteracy, and gradelevel achievement. This chapter examines research on the student outcomes of bilin­gualism and academic achievement, largely in the context of DL programs. In addition, outcomes for different demographic profiles of student participants in DL programs from preschool through secondary levels are presented. More specifically, this chapter addresses (a) the extent to which children in DL programs develop bilingual proficiency and academic achievement in two languages compared with children in English mainstream programs, (b) school and program factors that may impact student outcomes in DL programs, (c) student demographic characteristics that influence outcomes of bilingualism and achievement, and (d) the influence of language proficiency on academic achievement. Before addressing these points, it is important to provide a brief definition and overview of DL programs, how they are implemented, and to provide some information about the characteristics of students who attend these programs.

Dual Language Programs There are two major types of language education programs for students in the United States. These programs differ largely in terms of whether students receive instruction only in English or in English and an additional language. Currently, most programs that use both English and another language for instruction are referred to as dual language programs. In all programs, students are required to learn English, be taught to the same district and state standards, and be assessed using the same state standards. These programs are briefly described below (for more information about program types, see Hamayan, Genesee, & Cloud, 2013). •

English mainstream: This is the typical classroom in the United States, in which students receive instruction in English, and there are usually students of all English proficiency levels in the same classroom—from native English speakers to students who speak little or no English.

http://dx.doi.org/10.1037/14939-012 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

204 

•

 Kathryn Lindholm-Leary

Dual language (DL) programs: Students receive instruction in English and an additional language. There are three major types of DL programs that promote bilingualism: –– Developmental bilingual education: Also sometimes referred to as maintenance or late exit bilingual programs; this is an enrichment form of DL education, specifically designed for English learners (ELs), that uses ELs’ home language and English for literacy and academic instruction beginning in kindergarten and continuing through as many grades as the school district will support. –– One-way immersion education: Usually just called immersion, these DL programs are enrichment forms of education specifically designed for native English speakers (NESs). Instruction is provided in English and a second language. –– Two-way immersion education: These programs integrate students who enter school speaking a language other than English from a common native language background (e.g., Spanish, Mandarin) and native English-speaking students in the same classroom for academic instruction through both languages.

In DL programs, the partner language (e.g., Spanish, Mandarin, Korean) is used for a significant portion (from 50% to either 90% in two-way or 100% in one-way) of the students’ instructional day. In the 90:10 program, students spend 90% of their day being instructed in the partner language, learning content (math, social studies) and literacy in that language, and they spend 10% of their day in English, developing oral academic language. However, despite the name, it is only in kindergarten and first grade that 90% of instruction occurs in the partner language. Each grade after that more English is added until the children are spending half their day in English and half their day in the partner language by about Grade 4. In 50:50 programs, students spend half their instructional day in English and half in the partner language across all grade levels. Although programs are most common at the K–5 grade levels, in many areas, students can continue to study through the partner language in middle and high school.

Student Demographic Characteristics This chapter focuses on results that include native speakers of both the partner language and English who participate in a DL program. Furthermore, research will concentrate more specifically on children who participate in two-way DL programs, also drawing on research from one-way immersion for NES and developmental programs for ELs with which to compare or contrast research results in two-way programs.



Bilingualism and Academic Achievement in Children 

 205

Native Other Languages Speakers Native other language speakers (NOLSs) are typically referred to as ELs or as limited English proficient in the research literature and in educational and policy communities. Sometimes these children are referred to as second-language learners, or even bilingual speakers, though they may possess minimal, if any, proficiency in a second language. In fact, most research on bilingualism of school-age children focuses on this group of students, particularly ELs who are native Spanish speakers. These PreK–12 students have no or weak proficiency in English when they begin schooling in the United States. According to U.S. Census Bureau statistics, in 2011, about 22% of schoolage children spoke a language other than English at home (Ryan, 2013), and this is the fastest growing population in the United States (Cosentino de Cohen & Clewell, 2007). Hispanic children are the largest group of children who are more likely to speak English with difficulty (2.1 million), and children who are poor are more likely to speak English with difficulty than nonpoor children (10% vs. 3%). There is very little information about ELs in DL programs other than Spanish; yet there are DL programs in languages such as Korean, Japanese, Mandarin, Cantonese, Russian, Hmong, Italian, German, and French that include EL students who are native speakers of those languages. As this brief introduction to EL children indicates, this group is not homogenous and students may differ in terms of bilingual proficiency and socioeconomic status (SES), among other things. These are the two student demographic characteristics that will be considered here. •

Bilingual proficiency: There is a range of language proficiency in ELs, with some students essentially monolingual in the home language, whereas others may have parents who speak the home language but these children are more dominant in English when they begin schooling in the United States because of childcare arrangements or other exposure to English.

•

Socioeconomic status: Differences in SES may result from low to high levels of formal parental education (from no formal education to advanced college degrees) and low to high levels of income (from poverty to great wealth). These SES differences can impact the kinds of schools that children attend, the kinds of educational and other experiences in which the children participate, and the levels of language and literacy enrichment that students receive at home. For example, the great majority of ELs (70%) are concentrated in 10% of U.S. elementary schools, which tend to be highly segregated with ELs comprising from half to most of the student population (Cosentino de Cohen & Clewell, 2007).

Native English Speakers (NESs) There is little demographic information about native speakers of English in DL programs other than the students’ ethnicity, SES, or other specific characteristics;

206 

 Kathryn Lindholm-Leary

typically, research has not much addressed bilingualism in this group unless the research is focused on students in foreign or DL education or on immigration/ acculturation research in second or later-generation children of a particular ethnic background.

Academic Achievement and Bilingualism Outcomes in Dual Language Programs Considerable research has examined the reading and math achievement of students in DL programs (for reviews see Bikle, Hakuta, & Billings, 2004; Lindholm-Leary, 2001; Lindholm-Leary & Borsato, 2006; Lindholm-Leary & Genesee, 2010; LindholmLeary & Howard, 2008). This research base consistently demonstrates that students in DL programs achieve at or above the performance of their peers who are not in DL programs. Although most of this research looks at reading achievement, a few studies have examined math achievement and even fewer, achievement in other content areas; these results hold whether we examine outcomes in reading or other content areas. Furthermore, the result that DL students score as well as or higher than their peers in non-DL programs has been reported in Spanish, Italian, Chinese, and Korean DL programs (e.g., Bae, 2007; Chiappe, Glaeser, & Ferko, 2007; Genesee & Lindholm-Leary, 2013; Lindholm-Leary, 2001, 2011; Lindholm-Leary & Howard, 2008; Montanari, 2013). In addition, studies show that DL programs are successful at the secondary level. Compared with their peers in English mainstream programs, DL middle and high school students are (a) as or more likely to be enrolled in higher level math courses (Lindholm-Leary & Borsato, 2005), (b) as or more likely to pass the high school exit exam (Lindholm-Leary & Genesee, 2010), (c) less likely to drop out of school (Thomas & Collier, 2002), and (d) more likely to close the achievement gap with NES peers at least by the end of high school (Lindholm-Leary, 2014; Lindholm-Leary & Borsato, 2006; Thomas & Collier, 2002). Turning to the development of bilingual proficiency, we know that second-language learning is a challenging and lengthy process especially when it calls for the acquisition of academic language and not simply conversational language skills. Yet, there is little research on the linguistic development of either English or a second (non-English) language of students in DL programs in the United States (Lindholm-Leary & Howard, 2008; Saunders & O’Brien, 2006). Most of the research is concentrated on native speakers of Spanish and their proficiency in English, but not on the development of language skills in the second language (or primary language for NOLSs; for information on bilingual development throughout the lifespan, see Chapter 7, this volume). Further, the majority of research is focused on language development or proficiency derived



Bilingualism and Academic Achievement in Children 

 207

from various assessments (e.g., Language Assessment Scales, California English Language Development Test) rather than systematic studies of students’ language skills or language use. Nevertheless, this body of research shows that students in DL programs develop bilingual proficiency in DL programs; that is, they have developed oral and literate proficiencies in both languages that enable them to engage in schoolwork at the appropriate grade level. So far, this research has reported on the average outcomes for students in DL programs. However, it is far more interesting to disaggregate the results and better understand the bilingualism and academic achievement outcomes according to other factors. First, we examine various school and program factors that can impact student outcomes in DL programs.

Bilingualism and Achievement Outcomes by School and Program Variables School Demographics ELs, especially Spanish speakers, often experience segregated or isolated schooling experiences or schools with high percentages of ELs, minority populations, and poverty (Aud, Fox, & KewalRamani, 2010; Cosentino de Cohen & Clewell, 2007; LindholmLeary & Block, 2010), factors that are often associated with educational underachievement. Many ethnic minority, especially low-income, children experience segregated schooling as well. Some of these schools have implemented DL programs because of the strong research base demonstrating their success in promoting higher levels of language proficiency and achievement. Others have implemented DL programs in an attempt to desegregate the school. What evidence is there that DL students who attend such schools can achieve at similar or higher levels than their non-DL peers in English mainstream programs? Of the few studies that have examined students’ language proficiency and academic achievement in these segregated low-income schools, results show that students in DL programs score as well as or better than similar background students who are in mainstream (non-DL) programs (Lindholm-Leary, 2001; Lindholm-Leary & Block, 2010). More specifically, Lindholm-Leary (2001) examined the performance of native English speakers and native Spanish speakers in Hispanic segregated, lowincome vs. nonsegregated, high-income schools. On standardized tests that measured reading achievement in English, though achievement scores at both school types were at or above grade level by fifth grade, students scored higher at schools with less segregated (lower Hispanic) populations compared with more segregated schools (with high Hispanic populations). On the other hand, in looking at oral proficiency and

208 

 Kathryn Lindholm-Leary

reading achievement in Spanish, students scored much higher in the more segregated schools with a higher Hispanic population. Although it appeared that DL students at more segregated schools were at a disadvantage in their development of English literacy relative to their peers in less segregated schools, it was not clear whether being in DL programs augmented or diminished this tendency. In a study designed to further elucidate this issue with comparison groups, Lindholm-Leary and Block (2010) examined Hispanic English-speaking and Spanish-speaking EL students in predominantly Hispanic, low-income schools. They reported that Hispanic students—both English speakers and Spanish speakers—participating in DL programs in predominantly Hispanic, low-income schools achieved at similar or higher levels compared with their peers in English mainstream classes in tests of English. Furthermore, results showed that the DL students achieved above grade level in assessments conducted in Spanish as well. Thus, whereas more segregated (predominantly Hispanic, low income) schools typically place students at risk for underachievement, DL programs within these schools enable the students to achieve at similar or higher levels than their non-DL peers. In addition, the DL students at these schools demonstrate higher levels of bilingualism than their peers in less segregated schools. It appears that these schools may provide more opportunities for students to use Spanish and thus enhance their bilingual proficiency.

Program Type—Level of Exposure to English and a Partner Language Many educators, policymakers, and parents assume that the time-on-task principle applies to language proficiency; that is, more time spent studying through English leads to higher proficiency in English. Because of high-pressure accountability requiring schools to demonstrate English proficiency and literacy skills at early grade levels, there is considerable pressure for DL programs to add more instruction time in English, especially at earlier grades. However, research shows that this time-on-task principle applies to the development of the partner or nonsocietal language but not to English, the societal language (e.g., Gathercole & Thomas, 2009; Genesee, 2004). Various researchers have empirically demonstrated or reviewed the research to determine whether more English in the instructional day is associated with higher levels of proficiency and reading in English or whether more partner language in the instructional day is associated with higher levels of proficiency and reading achievement in the partner language (for reviews, see August & Shanahan, 2006; Genesee & Lindholm-Leary, 2013; Lindholm-Leary & Borsato, 2006; Lindholm-Leary & Genesee, 2010; Lindholm-Leary & Howard, 2008). Briefly, this research has compared programs



Bilingualism and Academic Achievement in Children 

 209

with more or less English in the instructional day, for example, English mainstream (more English) vs. DL (less English) programs or 50:50 (more English) vs. 90:10 (less English) DL programs. This research is consistent in showing that students who spend less time in English in DL programs tend to score at similar levels as their peers who receive more English; this is true for level of English language proficiency (listening, speaking, reading, writing), reclassification rates from EL to fluent English proficient (FEP), and reading achievement measured in English. Further, these findings are observed as early as preschool (Barnett, Yarosz, Thomas, Jung, & Blanco, 2007; Lindholm-Leary, 2014; Páez, Tabors, & LÓpez, 2007). Furthermore, differences between DL and non-DL students that appear to favor non-DL peers tend to disappear by later elementary grades, and some studies show that children in DL programs may outperform their peers in non-DL English mainstream programs in English (LindholmLeary & Borsato, 2006; Lindholm-Leary & Genesee, 2010; Lindholm-Leary & Howard, 2008; MacSwan & Pray, 2005). Thus, these findings corroborate previous reviews of research in DL (particularly bilingual education) and in immersion education for NESs showing that greater amounts of instruction through English are not necessarily associated with higher levels of proficiency in English or higher reading or math achievement in English (for reviews, see Genesee, 2004; Genesee & Lindholm-Leary, 2013; Lindholm-Leary & Borsato, 2006; Lindholm-Leary & Genesee, 2010; LindholmLeary & Howard, 2008). However, with respect to proficiency in the partner language, comparative studies show that students demonstrate higher levels of partner language proficiency when they participate in programs with higher levels of the partner language, that is, in 90:10 compared with 50:50 programs or DL vs. English mainstream programs (Lindholm-Leary, 2001, 2007; Lindholm-Leary & Borsato, 2006; Lindholm-Leary & Howard, 2008). In addition, students rate their proficiency in Spanish and level of bilingualism higher in 90:10 than 50:50 programs (Lindholm-Leary, 2007). Finally, reading achievement measured in Spanish is higher in 90:10 than 50:50 programs, especially for EL students (Lindholm-Leary, 2013; Lindholm-Leary & Howard, 2008). This research is consistent with research in Canada and elsewhere, showing that students in total immersion programs have higher levels of proficiency in the partner language than students in 50:50 immersion programs (Genesee, 2004; Genesee & Lindholm-Leary, 2013). In summary, these program and school factors clearly impact student outcomes of bilingualism and achievement, but not necessarily in expected ways. First, findings demonstrate that students in segregated/low-income schools can develop grade-level language proficiency and academic achievement as measured in both languages, and they can achieve at similar or higher levels than their peers in English mainstream. Second, programs with a higher amount of instruction through the partner language can lead to stronger proficiency in the partner language with no sacrifice to English proficiency. Thus, bilingualism and biliteracy may be enhanced to a greater degree when children receive higher levels of instruction in the partner language.

210 

 Kathryn Lindholm-Leary

Student Outcomes According to Student Background Characteristics So far, student outcomes in DL programs have shown that student participants develop bilingual proficiencies and reading and math achievement in both languages. Also, we have seen that there are differences in student outcomes according to school demographics. Now we turn to look at the extent to which student outcomes may vary according to student background characteristics such as ethnicity and SES. Research on the student outcomes—language proficiency and academic achievement—in DL programs indicates that there are important differences according to these background characteristics (for reviews, see Genesee, Lindholm-Leary, et al., 2006; Lindholm-Leary, 2001; Lindholm-Leary & Howard, 2008).

Comparisons Across Ethnic Backgrounds Students from different ethnic or racial groups develop oral and literate proficiency in two languages and achieve proficiency in English to the same or higher levels as their peers in English mainstream programs at least by the end of elementary school. These results have been obtained in studies with students from European American, Hispanic, African American, and Asian American backgrounds (Haj-Broussard, 2005; Howard, 2003; Lindholm-Leary, 2001, 2011; Lindholm-Leary & Block, 2010; Lindholm-Leary & Borsato, 2006; Lindholm-Leary & Hernández, 2011; Lindholm-Leary & Howard, 2008).

Comparisons According to Socioeconomic Status Although research with most populations clearly shows that SES and parent education impact student achievement, little research has examined the influence of SES, and especially parent education, on the achievement of DL students (Genesee, LindholmLeary, et al., 2006). Most of this research includes Hispanic students from low-income families. Thus, it is difficult to discern the true effect of SES because of the limited variation in SES among the samples examined. Notwithstanding this caveat, studies have demonstrated that Spanish-speakers tend to have very low SES in general, and that SES has a powerful impact in explaining the lower achievement and language proficiency of Spanish-speaking children (e.g., Lindholm-Leary & Borsato, 2006; Lindholm-Leary & Genesee, 2010; Roberts, Mohammed, & Vaughn, 2010). However, the few research studies within DL programs that have used parental education as a measure of SES have reported either no effect for parental education, or an effect only for English but not Spanish achievement (Genesee, Lindholm-Leary, et al., 2006; Lindholm-Leary, 2001). Perhaps parental education is not reported as a significant



Bilingualism and Academic Achievement in Children 

 211

effect in the few studies of EL students because there is very little range of parental education in the predominantly Hispanic, low-income families studied. Research shows that young Spanish-speaking children from low-income homes enter preschool on average with very low levels of proficiency in their primary language of Spanish and in English (Espinosa, 2009; Hammer, Miccio, & Wagstaff, 2003; Lindholm-Leary, 2014; Páez et al., 2007; Roberts et al., 2010). Furthermore, even for these primarily low-income children, parent education has a significant impact on entering preschool children’s English, but not Spanish, oral language proficiency (Lindholm-Leary, 2014). Parent education has a highly significant effect on language proficiency and reading achievement assessments conducted in both Spanish and English for school-age English and Spanish speakers (Lindholm-Leary, 2013; Lindholm-Leary & Hernández, 2011). Students who were still classified as ELs later in elementary school or into middle school (current ELs) differed significantly from students who were reclassified as FEPs in level of parent education overall. The more successful FEP students tended to have parents with higher levels of education, but even FEPs who had parents with comparable levels of education to current ELs scored higher than the current ELs. Perhaps parents of FEPs, who have more advanced levels of education, positively influence the use of complex communicative approaches and literacy events in the home that may affect the students’ proficiencies in both languages.

Student Outcomes According to Student Language Background Research has examined whether there are differences in language proficiency outcomes according to whether the students are native speakers of English or of the partner language. Results from this body of research comparing NESs to native speakers of the partner language indicate that, not surprisingly, there is a strong native speaker effect in the partner language. That is, (a) native Spanish speakers tend to develop higher levels of Spanish than native English speakers, and this holds true in measures of writing development as well as speaking (e.g., Lindholm-Leary & Howard, 2008); and (b) native Chinese speakers generally perform higher than NES in terms of both oral and written Chinese language proficiency (Lindholm-Leary, 2011). However, in measures of English, whereas NES tend to score higher in English than native speakers of Spanish, this gap diminishes considerably by fifth grade. Furthermore, there seem to be slightly different patterns for NES and native Spanish speakers, with NESs always showing a clear dominance in and preference for English, and native Spanish speakers demonstrating more balanced bilingualism (Howard, 2003; Howard, Christian, & Genesee, 2004; Lindholm-Leary, 2001). Sometimes the native Spanish speakers perform slightly higher in their native language, and other times

212 

 Kathryn Lindholm-Leary

slightly higher in English, but in general, their performance on language and literacy measures across the two languages is much more similar than that of their NES peers (Lindholm-Leary & Howard, 2008).

Native Other Language Speakers Initially, young Spanish-speaking children enter preschool and kindergarten with low levels of phonological awareness, letter identification, vocabulary, and emergent literacy skills in English, and they perform below monolingual children of the same age in both languages (Castro, Garcia, & Markos, 2013; Espinosa, 2013; Lindholm-Leary, 2014; Páez et al., 2007). Lindholm-Leary (2014) reported that at preschool entry, 30% of Spanish-speaking children had no proficiency in either language on the Language Assessment Scale (Duncan & De Avila, 1998) and only 1% were rated as proficient in both languages. This is not to say that these 30% of children had “no language,” but rather that their proficiency, as assessed by a standardized language test, was at a low level, which may place these children at risk for developing language and literacy skills below grade level. There is little research about these children unless they are later classified for special education. Thus, it is important to better understand children who enter school with differing levels of proficiency and who fall at different places along a continuum of bilingualism, a point that will be addressed later. With respect to the development of English for ELs, MacSwan and Pray (2005) pointed out that school-age ELs, unlike their native-speaking peers, “have developed only partial knowledge of the structure of their target language [English], and exhibit substantial errors associated with tense, case, grammatical agreement, word order, pronunciation, and other aspects of language structure” (p. 656). However, in a review of research on the oral language development of English, Saunders and O’Brien (2006) concluded that with increased oral proficiency in English, ELs can display a wider range of language skills, including skills associated with academic uses of language, such as question forms that involve higher level thinking (e.g., What makes you think the argument was valid?) and definitional skills. Correspondingly, Gathercole (2002a) showed that although second-grade ELs in DL programs may lag behind their mono­ lingual English-speaking peers in developing certain grammatical structures such as mass and count distinctions, there is no longer a gap by fifth grade and there is no difference between the bilingual English learners and their monolingual English-speaking peers. That these EL students eventually develop proficiency in English has been corroborated in reviews of research on ELs in two-way programs, which have shown that almost all ELs in two-way programs were rated as orally proficient in English, particularly by the upper elementary or middle school grade levels (Howard & Sugarman, 2007; LindholmLeary, 2013; Lindholm-Leary & Hernández, 2011; Lindholm-Leary & Howard, 2008). These results were on the basis of several large-scale studies of various two-way pro-



Bilingualism and Academic Achievement in Children 

 213

grams in different regions of the United States. Moreover, these results were obtained whether they were on the basis of cross-sectional or longitudinal studies and regardless of the language measures used (e.g., Language Assessment Scales, Student Oral Proficiency Assessment). However, most of these measures were limited to oral proficiency (speaking, listening) and not literacy. As discussed subsequently, though, even assessments that include oral and literate proficiencies demonstrate that most EL students in DL programs develop full proficiency in English (e.g., Lindholm-Leary, 2013; LindholmLeary & Hernández, 2011). There is very little research on EL students’ proficiency in their native language; the scant research that has been conducted on ELs’ native-language proficiency is mostly limited to teacher ratings and results on standardized language tests. Nonetheless, the available research shows that DL programs promote relatively high levels of Spanish proficiency (e.g., Gathercole, 2002a, 2002b; Howard et al., 2004; Howard & Sugarman, 2007; Lindholm-Leary, 2001, 2014; Lindholm-Leary & Howard, 2008). Finally, Gathercole (2002b) pointed out that by fifth grade, ELs in two-way programs show significantly greater proficiency in certain Spanish grammatical structures (e.g., gender distinctions) over ELs in English mainstream programs, even when those ELs come from homes in which Spanish is spoken. This research is important in demonstrating the significant impact that DL programs can serve in promoting both English and Spanish language proficiency. In the few studies of reading and writing development of ELs, students in DL programs make good progress in acquiring literacy skills in English and the partner language, often showing statistically significant gains in both languages in reading and writing (Bae, 2007; Chiappe, Glaeser, & Ferko, 2007; Howard et al., 2004; LindholmLeary, 2001; Lindholm-Leary & Howard, 2008; Montanari, 2013). Although native Spanish speakers tend to be stronger in Spanish than in English on the Cloze measures of reading comprehension (Howard et al., 2004), they tend to score the same in English and Spanish in their writing development from third through fifth grade. Further, Serrano and Howard (2003) reported more influence of English in their Spanish writing samples than vice versa. Similar findings were reported by Montanari (2013), though there were very few ELs in the Italian DL program; from first to third grade, students showed significant growth in both English and Italian reading fluency and accuracy, and the gap between these literacy skills decreased across the grade levels. In an examination of the English reading and writing progress of native Korean speaking first and second graders in Korean DL programs, studies showed that these students showed reading and writing development in English that was comparable or superior to their English speaking peers in English mainstream programs (Bae, 2007; Chiappe et al., 2007). Further, Lindholm-Leary (2011) demonstrated that native Chinese speakers possessed fairly high levels of oral and literate skills in Chinese, as measured by the program’s teacher and district-developed language assessments.

214 

 Kathryn Lindholm-Leary

Native English Speakers There are even fewer studies in the United States that examine English or second language development on the part of native speakers of English, unless these students are identified for special education. Most of the knowledge base of NES students’ development comes from 40 years of research on English speakers enrolled in French immersion programs in Canada (for a review, see Genesee, 2004). However, as mentioned previously, research on the NES enrolled in DL programs in the United States compare favorably with findings of immersion students in Canada (Genesee, 2004; Genesee & Lindholm-Leary, 2013; Lindholm-Leary & Howard, 2008). In the few studies that have been reported with respect to English language development of NES students, most to all of the native speakers of English are rated highly on oral proficiency tests in both 90:10 and 50:50 programs at program entry and again in the later elementary grades (Howard et al., 2004; Lindholm-Leary, 2001; LindholmLeary & Howard, 2008). With respect to language proficiency in the partner language, studies show that a range of half to most all NES students were scored at fairly high levels of oral language proficiency by fourth to sixth grade, regardless of the type of oral proficiency measure (Howard & Sugarman, 2007; Lindholm-Leary, 2001; Lindholm-Leary & Howard, 2008). Similar results were obtained in a study of NES students in Chinese DL programs (Lindholm-Leary, 2011). In a variety of studies representing different partner languages (Italian, Korean, Mandarin, Spanish), native English speaking students made great strides in reading and writing development across the grade levels (Bae, 2007; Chiappe et al., 2007; Howard et al., 2004; Lindholm-Leary, 2001; Lindholm-Leary & Howard, 2008; Montanari, 2013). However, despite several years of instruction through the partner language, native English speakers tended to be stronger in English than in the partner language in both reading and writing. However, as Montanari (2013) pointed out, native English speakers learned to read first in Italian; thus, initially, their reading skills were more advanced in Italian than in English. In addition, Montanari (2013) also showed that the Italian reading skills of the native English speakers were similar to those of native Italian children instructed in Italy whereas the English reading skills were similar to or higher than those of native English speakers in the United States. Similarly, Bae (2007) and Chiappe et al. (2007) investigated the English reading and writing progress of native English speaking first and second graders in Korean DL programs and reported that these students showed reading and writing development in English that was comparable or superior to their English speaking peers in English mainstream programs.

Students’ Ratings of their Bilingual Proficiency In the previous section, students’ language proficiency was assessed through teacher ratings, oral or reading development measures, or other standardized measures of



Bilingualism and Academic Achievement in Children 

 215

language proficiency. In this section, students in Grades 4 through 8 rated their own proficiency in the partner language; these students were participating in either a Spanish–English or a Chinese–English program. In these studies, students—both NOLS/EL and NES—who have participated in DL programs for 6 to 8 years think their skills in the partner language are highly functional and, in particular, that they have the skills they need to participate in a variety of classroom and social exchanges at their grade level (Lindholm-Leary, 2003, 2007, 2011; Lindholm-Leary & Ferrante, 2005). In one measure of Spanish or Chinese proficiency, students indicated their extent of agreement on a four-point Likert scale (from strongly agree to strongly disagree) about whether they have a variety of language skills in the partner language. Overall, the great majority (86%–93%) of students believe they possess this array of skills in the partner language, including: have conversations with their peers, provide and obtain information, express feelings and emotions, express opinions about various topics, understand and interpret written and spoken Spanish or Chinese on a variety of topics, and present information, concepts, and ideas to an audience on a variety of topics. Almost all students believe that they can read Spanish or Chinese and English well for their grade level (87%, 92%) and they can translate from one language to the other (94%). However, in both Spanish and Chinese programs, there are about 10% of native speakers and English speakers who rate their Chinese or Spanish skills at a fairly low level. When students were asked whether they believe they are “very,” “somewhat,” or “not very” bilingual, 47% reported “somewhat” and 51% responded “very,” though students in Spanish– English programs rated their bilingualism significantly higher than students in Chinese–English programs. Further, teacher ratings were highly correlated with students’ ratings of their bilingualism (Lindholm-Leary, 2007).

Student Outcomes for Heritage Language Students Lindholm-Leary (2003, 2011) and her colleagues (Lindholm-Leary & Ferrante, 2005) compared DL middle school and high school students according to whether they were native speakers of English or of the partner language, but they also looked separately at students who were heritage language learners (i.e., they were NES and did not speak the partner language, but they were ethnically similar to the native speakers of the partner language and therefore shared the language as a heritage language). Sometimes, but not consistently, the heritage language students (Chinese English speakers, Hispanic English speakers) show higher levels of self-rated proficiency in the partner language than nonheritage English speakers (Lindholm-Leary, 2003; Lindholm-Leary & Ferrante, 2005). In looking at students’ language use according to their ethnic background, Hispanic students—both native Spanish speakers and native English speakers—found more opportunities to use Spanish, were more likely to use Spanish at home, and were more comfortable speaking Spanish in public compared with European American English speakers (Lindholm-Leary & Ferrante, 2005).

216 

 Kathryn Lindholm-Leary

Language Proficiency and Its Relation to Academic Achievement There is little research on how oral language or bilingual proficiency is related to students’ overall academic success, although we know that it is. It is clear that oral English proficiency of an academic nature correlates positively with English reading achievement (Genesee et al., 2005; Genesee, Lindholm-Leary, et al., 2006; LindholmLeary, 2001, 2013; Lindholm-Leary & Howard, 2008), that oral Spanish proficiency is associated with Spanish reading achievement (Genesee et al., 2005; Genesee, Lindholm-Leary, et al., 2006; Lindholm-Leary, 2001, 2013; Lindholm-Leary & Hernández, 2011; Lindholm-Leary & Howard, 2008), and that bilingual proficiency is associated with academic achievement (Lindholm & Aclan, 1991; Lindholm-Leary, 2013; Lindholm-Leary & Hernández, 2011; Lindholm-Leary & Howard, 2008). So far, we have seen differences in achievement according to language background but not how language background may impact student success across the grade levels. This next section further disaggregates outcomes according to students’ level of proficiency in English and other student background variables to demonstrate different trajectories of growth over time.

Long-Term Student Outcomes According to Bilingual Language Proficiency and Other Student Background Variables Few studies have examined the level of language proficiency or level of bilingualism and its relation to long-term reading or math achievement in DL programs, particularly with the goal of understanding why some children close the achievement gap with English speakers, whereas others do not. One of the problems in this research is that studies have disaggregated the EL population to look at differences in language proficiency and achievement, and the background factors that may account for these differences. Lindholm-Leary and Hernández (2011) reported on a longitudinal study of three groups of fourth through eighth grade Hispanic students in DL programs who differed in English language proficiency: (a) native English speakers, (b) previous EL but current English proficient students, and (c) current ELs. They found that the three groups varied significantly in parent education, language proficiency in Spanish, and achievement as measured in Spanish and English. They also found that FEP/previous ELs were the most Spanish proficient and the most bilingual, achieved at higher levels in English and Spanish, and closed the achievement gap with native English speakers in English mainstream programs. Drawing on the same sample of fourth- through eighth-grade Hispanic students, Lindholm-Leary (2013) further differentiated the ELs into four language-proficiency categories based on their current English language pro-



Bilingualism and Academic Achievement in Children 

 217

ficiency score: (a) beginner/early intermediate, (b) intermediate, (c) early advanced/ advanced, or (d) reclassified as FEP. A retrospective analysis was used in which the students in each of the four language proficiency levels were followed back across the grades to determine whether there were differences in their scores at kindergarten or first grade and how students scored in language proficiency and academic achievement across the grade levels. Results showed that in each of the three outcome measures (English language proficiency, English reading achievement, Spanish reading achievement), language proficiency group had a significant impact on the outcome measure, with FEP students outscoring early advanced/advanced, who outscored intermediate, who outscored beginner/early intermediate students. However, students also varied significantly in their background characteristics. That is, students who were still at beginner/early intermediate levels were most likely to be economically disadvantaged, to have parents who had a high school diploma or less, to have special education services and least likely to have a parent who was a college graduate. In addition, students within each increasing proficiency level had greater economic and parent education advantages and less likelihood of being identified for special education. These outcome and student background differences suggest that these students are not at all homogeneous, but are quite distinct, though they were all identified as primarily low SES, Spanish-speaking ELs when they started school. To better understand these same groups of students, Lindholm-Leary (2013) looked at scores at or near program entry, in kindergarten or first grade. Although there was clearly a range of scores from lower to higher in each of the four proficiency groups at program entry (e.g., 16% of FEP and 24% of early advanced/advanced students began kindergarten at the beginning level in English language proficiency), overall the starting scale score varied significantly by language proficiency group. The FEP students scored much higher, and the beginner/early intermediate much lower, than the other groups, and this result was just as true for reading achievement in English and Spanish. Furthermore, whereas one might expect that the beginner/early intermediate group scored lower in English but was stronger in Spanish, and that the FEP group was the strongest in English but perhaps weaker in Spanish, that was not the case. In fact, the highest achievers in English were also the highest achievers in Spanish, and the lowest achievers in Spanish were the lowest achievers in English. Thus, these findings again lend credence to the research showing the strong relationship in reading achievement across the two languages for students instructed in both languages (August & Shanahan, 2006; Genesee, Geva, Dressler, & Kamil, 2006; Lindholm & Aclan, 1991; Lindholm-Leary & Genesee, 2010). In these two studies, FEP students scored the highest in all measures. Though these FEP students entered school as ELs, they continued to develop their Spanish to levels that exceeded those of the current EL students, both in oral language proficiency and in literacy. Furthermore, their English oral language and literacy skills were strong as assessed on the state-developed measures for language proficiency and reading achievement. In addition, FEP students outperformed their Hispanic

218 

 Kathryn Lindholm-Leary

English-speaking peers and also English-speaking students in English mainstream classes. Similar findings have been reported in other research on DL programs and in other educational contexts, and show that these previous EL and current bilingual Hispanics demonstrate higher levels of achievement and educational expectations than their monolingual English-speaking or Spanish-speaking Hispanic peers (e.g., Genesee, Lindholm-Leary, et al., 2006; Lindholm & Aclan, 1991; LindholmLeary, 2001). Some have attributed this educational success to the positive impact of bilingualism on cognition and academic achievement (e.g., Lindholm & Aclan, 1991; Lindholm-Leary, 2001). This high achievement of FEP students in English and Spanish might be associated with bilingualism or a higher level of parental formal education or other background variables that affect their language proficiency development. This is clearly an area requiring further research on the interplay between bilingual language proficiency, reading achievement, and various background variables.

Conclusion In this chapter, we have examined the bilingualism and academic achievement of students in one- and two-way DL programs. The findings have demonstrated that, overall, (a) students make remarkable progress toward full bilingualism, (b) students achieve at or above grade level in English and the partner language, and (c) they perform at comparable or higher levels compared with non-DL peers in English mainstream programs. We also saw that school and program factors can have an impact on student outcomes of bilingualism and achievement. Research indicated that students in more segregated schools can develop grade-level proficiency in language and literacy as measured in both languages, and they perform at higher levels than their peers in English mainstream. Next we saw that programs with a higher amount of instruction through the partner language can lead to stronger proficiency in the partner language with no sacrifice to English proficiency. Far more research needs to be conducted to look at different schooling contexts that promote or hinder bilingualism and academic achievement in two languages. Although Lindholm-Leary and Block (2010) noted that students scored higher in Spanish and were more bilingual in predominantly Hispanic low-income schools compared with their DL peers in less segregated schools, it may simply be that the use of Spanish is more accepted in these segregated schools and that the greater exposure to and use of Spanish yields higher proficiency in Spanish. More research on the use of the partner language within DL programs and outside the classrooms would also be helpful, particularly at later elementary and secondary levels. Student bilingualism and academic achievement outcomes are also highly influenced by students’ background factors and language proficiency. Overall, SES and parental formal education have an effect on language proficiency and academic



Bilingualism and Academic Achievement in Children 

 219

achievement; however, parent education still has an impact on reading achievement in English and Spanish even when holding SES constant within low-income Hispanic families. These data reveal that even low SES children achieve as well as or higher than their peers in English mainstream classes, showing that the DL programs do not work only for children from more advantaged family backgrounds. It would be helpful to include research with a greater representation of families with different levels of education, especially Hispanic families, and to look more carefully at the impact of parent education level on non-Hispanic NES as well. There is a great need for investigative work that examines familial and other background factors that may impact achievement with DL programs or other programs that can potentially improve bilingualism. Finally, we looked at the influence of language background on students’ bilingualism and reading achievement. First, and most important, both NESs and native speakers of the partner language develop bilingual proficiency and literacy skills in both languages, and they are able to achieve in English at levels that meet or exceed their non-DL peers. Second, the data demonstrate a native speaker effect, such that native speakers generally perform higher than second language speakers in terms of both oral and written language proficiency. Another important finding is that sometimes, but not consistently, heritage language students (e.g., Chinese or Hispanic English speakers) have higher levels of self-rated proficiency in the heritage language than nonheritage English speakers. This is another area in need of further research, to better understand how student access to languages within their home and community can influence their bilingualism. This research also points to challenges in developing high levels of bilingual proficiency across the grade spans within DL programs. Potowski’s (2007) observations of fifth- and eighth-grade students in a Spanish–English DL program showed that although students did develop bilingual skills, they did not develop highly proficient or balanced bilingual skills, because they were dominant, and felt more comfortable, speaking in English than in Spanish. In addition, de Jong and Bearse (2011) showed that secondary DL students feel that they do not receive sufficient opportunity or support to develop high levels of Spanish within the DL classroom. Difficulties in developing fully proficient bilingual speakers in the United States may also reflect other challenges in the United States. For example, accountability is usually associated with demonstrating language proficiency and academic success as measured only in English; language proficiency and literacy skills in the partner language may be considered an added benefit, but they are not seen as critical in demonstrating student achievement according to federal, state, and local requirements. Accountability challenges may also explain the paucity of research on the bilingual language skills that are necessary for promoting high levels of literacy and school success. Furthermore, there is little, if any, discussion about bilingualism at the school or societal level. Immigrant parents are still advised not to speak their native language to their child, resulting in language loss in the primary language and communication barriers. In addition, schools at all levels—preschool through college—do not support

220 

 Kathryn Lindholm-Leary

the development of full bilingual proficiency. Rather, preschools provide instruction largely through English to young children who speak only or mostly a native language other than English, which results in language loss and communication barriers with parents, rather than yielding early levels of bilingual proficiency. Additionally, middle and high school programs that receive bilingual children from elementary one- and two-way DL programs are often reluctant, or even refuse, to develop a secondary program that offers more than standard foreign language instruction. Thus, there is insufficient opportunity for interested students in continuing their DL program, and thus for promoting high levels of proficiency in the partner language. The result of these misguided approaches is continuation of a policy that attempts to promote high levels of achievement, but ignores the mounting evidence on the positive impact of bilingualism on children’s language, cognitive and sociocultural development, along with their academic achievement.

References Aud, S., Fox, M. A., & KewalRamani, A. (2010). Status and trends in the education of racial and ethnic groups (NCES 2010-015). U.S. Department of Education, National Center for Education Statistics. Washington, DC: U.S. Government Printing Office. August, D., & Shanahan, T. (Eds.). (2006). Developing literacy in second-language learners: Report of the national literacy panel on language-minority children and youth. Mahwah, NJ: Erlbaum. Bae, J. (2007). Development of English skills need not suffer as a result of immersion: Grades 1 and 2 writing assessment in a Korean/English two-way immersion program. Language Learning, 57, 299–332. http://dx.doi.org/10.1111/j.1467-9922.2007.00410.x Barnett, W. S., Yarosz, D. J., Thomas, J. H., Jung, K., & Blanco, D. (2007). Two-way monolingual English immersion in preschool education: An experimental comparison. Early Childhood Research Quarterly, 22, 277–293. http://dx.doi.org/10.1016/j.ecresq.2007.03.003 Bikle, K., Hakuta, K., & Billings, E. S. (2004). Trends in two-way immersion research. In J. A. Banks & C. A. McGee Banks (Eds.), Handbook of research on multicultural education (2nd ed., pp. 589–604). New York, NY: Macmillan. Castro, D. C., Garcia, E. E., & Markos, A. M. (2013). Dual language learners: Research informing policy. Chapel Hill: The University of North Carolina. Chiappe, P., Glaeser, B., & Ferko, D. (2007). Speech perception, vocabulary, and the development of reading skills in English among Korean- and English-speaking children. Journal of Educational Psychology, 99, 154–166. http://dx.doi.org/10.1037/0022-0663.99.1.154 Cosentino de Cohen, C., & Clewell, B. C. (2007). Putting English language learners on the educational map: The No Child Left Behind Act implemented. Washington, DC: The Urban Institute. http://www.urban.org/publications/311468.html de Jong, E. J., & Bearse, C. (2011). The same outcomes for all? High school students reflect on their two-way immersion program experiences. In D. Christian, D. Tedick, & T. Fortune (Eds.), Immersion education: Pathways to bilingualism and beyond (pp. 104–122). Clevedon, England: Multilingual Matters.



Bilingualism and Academic Achievement in Children 

 221

Duncan, S. E., & De Avila, E. (1998). Pre-Language Assessment Scale 2000. Monterey, CA: CTB McGraw-Hill. Espinosa, L. (2009). Getting it RIGHT for young children from diverse backgrounds: Applying research to improve practice. Englewood Cliffs, NJ: Prentice Hall. Espinosa, L. (2013). Early education for dual language learners: Promoting school readiness and early school success. Washington, DC: Migration Policy Institute. Gathercole, V. C. M. (2002a). Command of the mass/count distinction in bilingual and monolingual children: An English morphosyntactic distinction. In D. K. Oller & R. E. Eilers (Eds.), Language and literacy in bilingual children (pp. 175–206). Avon, England: Multi­lingual Matters. Gathercole, V. C. M. (2002b). Grammatical gender in bilingual and monolingual children: A Spanish morphosyntactic distinction. In D. K. Oller & R. E. Eilers (Eds.), Language and literacy in bilingual children (pp. 207–219). Avon, England: Multilingual Matters. Gathercole, V. C. M., & Thomas, E. M. (2009). Bilingual first-language development: Dominant language takeover, threatened minority language take-up. Bilingualism: Language and Cognition, 12, 213–237. http://dx.doi.org/10.1017/S1366728909004015 Genesee, F. (2004). What do we know about bilingual education for majority language students? In T. K. Bhatia & W. Ritchie (Eds.), Handbook of bilingualism and multiculturalism (pp. 547–576). Malden, MA: Blackwell. Genesee, F., Geva, E., Dressler, C., & Kamil, M. (2006). Synthesis: Cross-linguistic relationships. In D. August & T. Shanahan (Eds.), Developing literacy in second language learners: Report of the National Literacy Panel on Minority-Language Children and Youth (pp. 153–174). Mahwah, NJ: Erlbaum. Genesee, F., & Lindholm-Leary, K. (2013). Two case studies of content-based language education. Journal of Immersion and Content-Based Language Education, 1, 3–33. http://dx.doi.org/ 10.1075/jicb.1.1.02gen Genesee, F., Lindholm-Leary, K. J., Saunders, W., & Christian, D. (2005). English language learners in U.S. schools: An overview of research findings. Journal of Education for Students Placed at Risk, 10, 363–385. http://dx.doi.org/10.1207/s15327671espr1004_2 Genesee, F., Lindholm-Leary, K. J., Saunders, W., & Christian, D. (2006). Educating English language learners. New York, NY: Cambridge University Press. http://dx.doi.org/10.1017/ CBO9780511499913 Haj-Broussard, M. (2005). Comparison contexts: African-American students, immersion and achievement. The ACIE Newsletter, 8(3). Hamayan, E., Genesee, F., & Cloud, N. (2013). Dual language instruction from A to Z. Portsmouth, NH: Heinemann. Hammer, C. S., Miccio, A. W., & Wagstaff, D. A. (2003). Home literacy experiences and their relationship to bilingual preschoolers’ developing English literacy abilities: An initial investigation. Language, Speech, and Hearing Services in Schools, 34, 20–30. http://dx.doi.org/10.1044/ 0161-1461(2003/003) Howard, E. R. (2003). Biliteracy development in two-way immersion education programs: A multilevel analysis of the effects of native language and home language use on the development of narrative writing ability in English and Spanish (Unpublished doctoral dissertation). Harvard University, Cambridge, MA. Howard, E. R., Christian, D., & Genesee, F. (2004). The development of bilingualism and biliteracy from grades 3 to 5: A summary of findings from the CAL/CREDE study of two-way immersion education.

222 

 Kathryn Lindholm-Leary

Santa Cruz, CA: Center for Research on Education, Diversity & Excellence and Center for Applied Linguistics. Howard, E. R., & Sugarman, J. (2007). Realizing the vision of two-way immersion: Fostering effective programs and classrooms. Washington, DC: Delta Systems and ERIC Clearinghouse on Languages and Linguistics. Lindholm, K. J., & Aclan, Z. (1991). Bilingual proficiency as a bridge to academic achievement: Results from bilingual/immersion programs. Journal of Education, 173, 99–113. Lindholm-Leary, K. (2007, March). Got bilingualism? Spanish and Chinese immersion students’ perceptions of bilingualism and biculturalism. Paper presented at the California Association for Bilingual Education, Long Beach, California. Lindholm-Leary, K. (2013, April). Understanding ELLs at different English proficiency levels in dual language programs. Paper presented at American Educational Research Association, San Francisco, CA. Lindholm-Leary, K. (2014). Bilingual and biliteracy skills in young Spanish-speaking low-SES children: Impact of instructional language and primary language proficiency. Inter­national Journal of Bilingual Education and Bilingualism, 17, 144–159. http://dx.doi.org/10.1080/ 13670050.2013.866625 Lindholm-Leary, K., & Genesee, F. (2010). Alternative educational programs for English language learners. In Research on English language learners (pp. 323–382). Sacramento, CA: California Department of Education Press. Lindholm-Leary, K., & Hernández, A. (2011). Achievement and language proficiency of Latino students in dual language programs: Native English speakers, fluent English/previous ELLs, and current ELLs. Journal of Multilingual and Multicultural Development, 32, 531–545. http://dx.doi.org/ 10.1080/01434632.2011.611596 Lindholm-Leary, K. J. (2001). Dual language education. Clevedon, England: Multilingual Matters. Lindholm-Leary, K. J. (2003). Dual language achievement, proficiency, and attitudes among current high school graduates of two-way programs. NABE Journal, 26, 20–25. Lindholm-Leary, K. J. (2011). Student outcomes in Chinese two-way immersion programs: Language proficiency, academic achievement, and student attitudes. In D. J. Tedick, D. Christian, & T. W. Fortune (Eds.), Immersion education: Practices, policies, possibilities (pp. 81–103). Bristol, England: Multilingual Matters. Lindholm-Leary, K. J., & Block, N. (2010). Achievement in predominantly low-SES Hispanic dual language schools. International Journal of Bilingual Education and Bilingualism, 13, 43–60. http:// dx.doi.org/10.1080/13670050902777546 Lindholm-Leary, K. J., & Borsato, G. (2005). Hispanic high schoolers and mathematics: Follow-up of students who had participated in two-way bilingual elementary programs. Bilingual Research Journal, 29, 641–652. http://dx.doi.org/10.1080/15235882.2005.10162856 Lindholm-Leary, K. J., & Borsato, G. (2006). Academic achievement. In F. Genesee, K. Lindholm-Leary, W. Saunders, & D. Christian (Eds.), Educating English language learners (pp. 176–222). New York, NY: Cambridge University Press. Lindholm-Leary, K. J., & Ferrante, A. (2005). Follow-up study of middle school two-way students: Language proficiency, achievement and attitudes. In R. Hoosain & F. Salili (Eds.), Language in multicultural education (pp. 157–179). Greenwich, CT: Information Age. Lindholm-Leary, K. J., & Howard, E. (2008). Language and academic achievement in two-way immersion programs. In T. Fortune & D. Tedick (Eds.), Pathways to bilingualism: Evolving perspectives on immersion education (pp. 177–200). Clevedon, England: Multilingual Matters.



Bilingualism and Academic Achievement in Children 

 223

MacSwan, J., & Pray, L. (2005). Learning English bilingually: Age of onset of exposure and rate of acquisition among English language learners in a bilingual education program. Bilingual Research Journal, 29, 653–678. http://dx.doi.org/10.1080/15235882.2005.10162857 Montanari, S. (2013). A case study of bi-literacy development among children enrolled in an Italian– English dual language program in Southern California. International Journal of Bilingual Education and Bilingualism, 17, 509–525. http://dx.doi.org/10.1080/13670050.2013.833892. Páez, M. M., Tabors, P. O., & López, L. M. (2007). Dual language and literacy development of Spanishspeaking preschool children. Journal of Applied Developmental Psychology, 28(2), 85–102. http://dx.doi.org/10.1016/j.appdev.2006.12.007 Potowski, K. (2007). Language and identity in a dual immersion school. Clevedon, England: Multilingual Matters. Roberts, G., Mohammed, S. S., & Vaughn, S. (2010). Reading achievement across three language groups: Growth estimates for overall reading and reading sub-skills obtained with the early childhood longitudinal survey. Journal of Educational Psychology, 102, 668–686. http://dx.doi.org/ 10.1037/a0018983 Ryan, C. (2013). Language use in the United States: 2011. American Community Survey Reports, ACS-22. Washington, DC: U.S. Census Bureau. Saunders, W., & O’Brien, G. (2006). Oral language. In F. Genesee, K. Lindholm-Leary, W. Saunders, & D. Christian (Eds.), Educating English language learners: A synthesis of research evidence (pp. 14–63). New York, NY: Cambridge University Press. Serrano, R., & Howard, E. R. (2003). Maintaining Spanish proficiency in the United States: The influence of English on the Spanish writing of native Spanish speakers in two-way immersion programs. In L. Sayahi (Ed.), Selected proceedings of the First Workshop on Spanish Socio­linguistics (pp. 77–88). Somerville, MA: Cascadilla Proceedings Project. Thomas, W., & Collier, V. (2002). A national study of school effectiveness for language minority students’ long-term academic achievement. Santa Cruz, CA: Center for Research on Education, Diversity and Excellence.

Judith F. Kroll, Jason Gullifer, and Megan Zirnstein

12 Literacy in Adulthood: Reading in Two Languages

Bilingual adults who are proficient in a second language (L2) come to their bilingualism in different ways. Early bilingual speakers who have acquired two languages from childhood may read proficiently in each of their two languages or only in one, depending on the circumstances in which they use each language. Late bilingual speakers who acquire the L2 past early childhood typically come to the task of L2 learning with a high level of literacy in their native or first language (L1). For many late L2 learners, reading is the first encounter with the L2 in the context of classroom instruction, although some adults may have initial exposure to a spoken L2 after being immersed in an L2 environment by reasons of travel or immigration, and only later learn to read the L2. The range of bilingual experience suggests that there may be different profiles of literacy for adult readers of an L2. As we will see in this chapter, although there are many ways that individuals come to read in an L2 that shape the nature of their literacy, research in the last 2 decades has shown that there is a high level of permeability across bilingual speakers’ two languages (e.g., Dijkstra, 2005; Kroll, Bobb, & Hoshino, 2014; Kroll, Dussias, Bogulski, & Kroff, 2012). Cross-language interactions can be seen at every level of language processing, from the lexicon to the phonology and grammar, for the most proficient bilingual speakers and for L2 learners. The discovery that has changed our understanding of bilingual reading is that both languages are active even when bilingual readers read in one language alone (e.g., see Kroll & Dussias, 2013; Kroll, Gullifer, & Rossi, 2013). The parallel activation of the two languages produces competition that needs to be resolved. That competition takes different forms but has the consequence of creating interactions in which the L1 affects the L2, the typical direction of cross-language transfer from the native language to the weaker L2 (e.g., Kroll & Stewart, 1994; MacWhinney, 2005), and the L2 affects the L1 (e.g., Ameel, Storms, Malt, & Sloman, 2005; Dussias & Sagarra, 2007). It was once thought that these effects of the L2 on the L1 were restricted to highly proficient bilingual speakers for whom the L2 is associated with a high level of skill and automaticity (e.g., Van Hell & Dijkstra, 2002). But new studies suggest changes in the L1 at even very early stages of L2 learning that reflect the high degree of permeability across the two languages (e.g., Chang, 2013; Mei et al., 2014). The writing of this chapter was supported in part by NIH Grant HD053146, NSF Grants BCS-0955090, BCS-1251896, SMA-1409973 and OISE-0968369, and a Guggenheim Fellowship. http://dx.doi.org/10.1037/14939-013 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

226 

 Kroll, Gullifer, and Zirnstein

In this chapter, we review the highlights of the recent research on bilingual reading, focusing on bilingual adults who are highly proficient in the two languages but who may have acquired the L2 beyond early childhood, and for readers who use two languages in which there may be different scripts or only one language that is written. Critically, an implication of the new research on bilingualism is that there is a high degree of plasticity, even for late acquirers of an L2. As we will see in the review that follows, the current research no longer assumes that the acquisition and use of an L2 past early childhood is limited (see also Chapter 9, this volume). Even late bilingual learners can become highly proficient in the L2. In addition, the active use of two or more languages appears to confer a set of cognitive advantages that extend beyond language use to enable bilingual speakers to regulate the use of each language (e.g., Bialystok, Craik, & Luk, 2012). These observations hold profound consequences for pedagogy and for social policy that affects late L2 learners. Whereas, in the past, bilingualism may have been viewed as a circumstance to be dealt with, it is now understood as goal to be achieved. Not only does bilingualism present the opportunity to read in two languages, providing enhanced access to information across languages and cultures, but it also changes the mind and brain in ways that create greater cognitive flexibility and resilience.

Cross-Language Lexical Activation Much of the research that demonstrates that both languages are active when reading in one language alone has been conducted at the level of the lexicon, asking whether word recognition is affected by knowledge of the language not in use. Evidence for parallel activation (or language nonselective access) is most apparent for words that share overlap across the two languages. As noted above, parallel activation of the unintended language does not occur solely as the result of low L2 proficiency. Although lower proficiency L2 speakers appear to rely on coactivation of the L1 to process the L2 under some circumstances (e.g., Kroll & Stewart, 1994; Kroll, Van Hell, Tokowicz, & Green, 2010), and sometimes show less evidence of L2 activation during L1 processing, highly proficient L2 speakers experience coactivation of both languages, such that each language becomes activated when processing the other. Cross-language activation is not dependent on the type of task or the types of items used in experiments. Parallel activation is observed in a variety of word recognition tasks, using behavioral (e.g., Dijkstra, Van Jaarsveld, & Ten Brinke, 1998) and electrophysiological measurements (e.g., Midgley, Holcomb, & Grainger, 2009), and even when words share no explicit overlap across the two languages (e.g., Morford, Wilkinson, Villwock, Piñar, & Kroll, 2011; Thierry & Wu, 2007). A subject of great interest is how bilingual readers reduce the activation of the unintended language. In speaking, inhibitory mechanisms suppress unintended alter-



Literacy in Adulthood 

 227

natives (e.g., Abutalebi & Green, 2007; Green, 1998; Misra, Guo, Bobb, & Kroll, 2012). A similar mechanism appears to function in visual word recognition (e.g., Blumenfeld & Marian, 2011; Martín, Macizo, & Bajo, 2010). Because reading involves the processing of a written input, one possibility is that bilingual readers recognize features of the input that cue the intended language. When a Chinese–English bilingual speaker reads English, it is obvious that the text is not Chinese. Counterintuitively, many studies show that the presence of language-specific cues, such as different written script, is not sufficient to reduce the activation of the unintended language (e.g., Thierry & Wu, 2007). The primary evidence for parallel activation in reading comes from studies that examine the processing of words that contain cross-language overlap such as cognates and homographs. Cognates are words that contain form and meaning overlap across two languages. For example the word bus refers to the same object, shares complete orthographic overlap, and is pronounced similarly in English and Spanish. Inter­ lingual homographs, or “false friends,” share lexical form but differ in meaning across two languages. For example, the word red in Spanish refers not to the color red, as in English, but to a net; however, it is written and pronounced similarly in English and Spanish. Bilingual speakers process words with cross-language overlap differently compared with nonambiguous words that are matched on their lexical properties and monolingual speakers show no evidence of processing differences. Dijkstra et al. (1998) reported that Dutch–English bilingual readers activated Dutch during an English lexical decision task when required to respond “yes” if a presented letter string was a word in English. They showed that these bilingual readers were faster to respond to cognates than to matched control words. In contrast to cognates, Dijkstra et al. (1998) showed that Dutch–English homographs exhibited no processing differences unless the task was modified to more strongly involve Dutch as the unintended language. When Dutch distractor words (that required a “no” response) were added to the set of nonword materials in English lexical decision to boost the relevance of Dutch, homographs were slower than control words. When the task was changed into a generalized lexical decision task such that bilingual speakers responded to a word in either English or Dutch, they were faster to process homographs than control words, illustrating sensitivity to form overlap but without the need to resolve a semantic conflict that might otherwise lead to erroneous performance. Cognate and homograph effects indicate that bilingual speakers are unable to function as monolingual speakers in each language. However, the dependence of the homograph effect on the context of the task indicates that not all words with cross language overlap index language coactivation in the same way, and suggests that some homograph words in particular may be sensitive to contextual features of the environment. However, the stability of the cognate effect even when the unintended language is not present in the set of materials indicates that bilingual speakers coactivate the unintended language. If bilingual speakers could selectively access the intended language, then the presence of cross-language overlap should have no bearing on processing.

228 

 Kroll, Gullifer, and Zirnstein

In some ways, parallel activation of the L1 during L2 reading is not surprising; a core assertion in theories of L2 learning (e.g., Kroll & Stewart, 1994; MacWhinney, 2005) is that the stronger language will influence the weaker language during learning through reliance on transfer from or translation into the L1. However, parallel activation is not limited to low-proficiency L2 users reading in their L2. Highly proficient L2 speakers (e.g., those in Dijkstra et al., 1998) continue to activate the L1 during L2 reading. Furthermore, many studies now also support the idea that L1 reading is influenced by coactivation of the L2 (e.g., Gullifer, Kroll, & Dussias, 2013; Titone, Libben, Mercier, Whitford, & Pivneva, 2011; Van Assche, Drieghe, Duyck, Welvaert, & Hartsuiker, 2011; Van Hell & Dijkstra, 2002). Parallel activation of the unintended language during L1 reading is evident for bilingual readers who are highly proficient speakers of both languages and can be considered to have higher baseline activation levels for alternatives in both languages, though emerging electrophysiological and fMRI evidence suggests that even early learners are sensitive to cross-language overlap when they read words in their dominant L1 (e.g., Bice & Kroll, 2015; Mei et al., 2014). Although the evidence for parallel activation of bilingual speakers’ two languages is compelling, there have also been criticisms of these studies. First, it is not clear from the study of cognate and homograph processing alone that nonselective access extends to processing of all words in the lexicon. Second, one can argue that the processing differences observed for words with cross-language overlap are simply the result of increased frequency of word usage. For example, Spanish–English bilingual readers will experience the cognate word bus twice as often as their monolingual counterparts, and this increased experience may lead to a processing advantage for that word in bilingual readers’ lexicons (e.g., see evidence on the weaker-links or frequency lag hypothesis; Gollan, Montoya, Cera, & Sandoval, 2008; Gollan et al., 2011). Although frequency may play a role in changing lexical processing for bilingual readers, it does not rule out parallel activation of languages. Cognate and homograph effects depend on the degree of orthographic overlap (e.g., which can be calculated via the van Orden or Levenshtein distance methods; van Orden, 1987) and phonological overlap (that can be elicited from participants making auditory judgments on sound overlap) of a word between the two. Smaller cross-language effects are observed for language-ambiguous words with a less orthographic or phonological overlap, and this relationship is linear (e.g., Dijkstra, Grainger, & Van Heuven, 1999; Schwartz, Kroll, & Diaz, 2007; Van Assche et al., 2011). A purely frequency-dependent hypothesis would not predict sensitivity to cross-language overlap within languageambiguous words. Further evidence in favor of the parallel activation hypothesis is the observation of cross-language effects for stimuli that share no overt similarities (Morford et al., 2011; Thierry & Wu, 2007; Wu, Cristino, Leek, & Thierry, 2013). To illustrate, when proficient Chinese–English bilingual speakers and monolingual speakers of each language make semantic relatedness judgments on English words, both groups of speakers show semantic priming via a reduction of the N400 response for related



Literacy in Adulthood 

 229

words. The N400 is a negative-deflecting event-related potential (ERP) effect, peaking around 400 milliseconds following stimulus onset. ERP effects are measures of the brain’s electrophysiological response to cognitive events, such as the presentation and subsequent processing of a word. The N400, in particular, is typically associated with ease of semantic processing (Federmeier & Kutas, 1999; Kutas & Hillyard, 1984; for a review, see Kutas & Federmeier, 2011; Swaab, Ledoux, Camblin, & Boudewyn, 2011). Critically, only bilingual readers in these studies showed an additional modulation of the N400 when the Chinese translation of the English words shared characters and phonology, features that were not present in the experiment (Thierry & Wu, 2007; Wu & Thierry, 2010).

Language Selection How do bilingual readers overcome coactivation to select the intended language? Research on production suggests that inhibition of the more dominant language enables selection of the weaker language. Inhibition is seen in tasks that require language switching. Bilingual readers exhibit a cost to switch from one language to another. The magnitude of the switch cost is related to the dominance of the language being switched. A switch into the weaker language following the dominant language results in a smaller cost compared to a switch into the dominant language following the weaker language. It becomes counterintuitively harder to speak the dominant language in a mixed-language situation. If both languages are of equal dominance, a switch cost of equal magnitude is observed (see Bobb & Wodniecka, 2013). Asymmetric switch costs are most easily explained by an inhibitory mechanism of language selection in which the dominant language requires stronger inhibition to suppress it compared to a weaker language. When the suppressed language later becomes the target language, the inhibition must be overcome, resulting in a switch cost proportional to the amount of inhibition applied to that language (and hence proportional to the dominance of the suppressed language). A mechanism of selection is particularly important during language production because there is often only one output channel: the mouth (unless the bilingual reader in question knows one spoken and one signed language). Because only one word can be spoken at any given moment, only one candidate can become the target output. During language comprehension there is no physical limit that necessitates language selection, and there may be no penalty to having multiple coactivated words particularly if those words share meaning overlap. Yet recent studies suggest a similar inhibitory mechanism during language comprehension, with some important differences. Language switch costs are observed during visual word recognition (Thomas & Allport, 2000; von Studnitz & Green, 2002), indicative of an inhibitory mechanism. Yet the comprehension-based switch costs tend to be symmetric, even when participants are more dominant in one language. Other comprehension studies

230 

 Kroll, Gullifer, and Zirnstein

have found that switch costs may be reduced altogether in certain situations. For example, Ibáñez, Macizo, and Bajo (2010) failed to find switch costs during self-paced reading when participants were required to switch languages following the end of a sentence. Further work from Martín et al. (2010) suggests that inhibition of activated alternatives when bilingual readers read words for meaning decays over the course of 500–750 ms, a marked difference to language production where switch costs persist even when bilingual readers are given time on the order of seconds to prepare for a language switch (e.g., Misra et al., 2012). In sum, it seems that an inhibitory mechanism is recruited during reading, but that inhibition decays relatively quickly compared to inhibition recruited during language production.

Cues to Language Selection Reading is the process of taking written input and discerning meaning. Because bilingual readers are supplied with an input that is ultimately indicative of the intended language, it is possible that bilingual readers can recognize and use features of the input that may serve as pointers to the intended language (i.e., a language cue) to selectively access the intended language. Overall, there has been little support for the notion that language cues in the input restrict activation to the intended language, though only a few potential cues have been studied in the extant body of literature. Parallel activation in early literature was commonly measured in isolated word recognition tasks. However, outside the laboratory, words are rarely processed without surrounding context. One language cue that has since been thoroughly investigated is the presence of a sentence context in one of the two languages. However, the mere presence of a unilingual sentence context is not sufficient to restrict coactivation to that language. When words with cross-language overlap are embedded inside of a single-language sentence context, bilingual readers still show bidirectional (L1–L2 and L2–L1) sensitivity to the overlap with the unintended language, particularly with cognate words (e.g., Duyck, Van Assche, Drieghe, & Hartsuiker, 2007; Gullifer et al., 2013; Schwartz & Kroll, 2006; Van Assche, Duyck, Hartsuiker, & Diependaele, 2009; Van Hell & De Groot, 2008). Although the mere presence of a sentence context cannot restrict parallel activation, if it is strongly biased toward a single interpretation (i.e., the sentence is semantically constrained), effects indicative of cross-language activation are reduced or eliminated. For example, when bilingual readers read a sentence such as “The boy ate the juicy red apple” in which the word apple (a cognate word between many languages) is highly predictable given the surrounding context, cross-language activation is reduced indicating that the semantics of a sentence can function as a cue to language (Libben & Titone, 2009; Van Hell & De Groot, 2008; Titone et al., 2011; but see Van Assche et al., 2011). However, it should be noted that initial lexical access becomes selective relatively late in the time course of word-recognition following a period at which both languages are initially activated. For example, Libben and



Literacy in Adulthood 

 231

Titone (2009) showed that when French–English bilingual readers read semantically constrained sentences in English, cognates were facilitated and homographs were inhibited relative to control words for measures of eye movements that are thought to track initial lexical access. These effects were eliminated in later measures of eye movements that involved processes of sentential integration. In contrast, when upcoming words were not predictable given the context, cognate and homograph effects are present in early and late measures of eye movements. The finding that the semantics of a sentence reduces lexical coactivation indicates that it is possible for bilingual readers to eventually select the appropriate language in comprehension and that the semantics of a sentence can function as a cue. It also may be the case that readers may require multiple convergent cues (e.g., a unilingual sentence context and semantic constraints) to achieve language-selective recognition. Features of the environment in which a language is processed may also function as language cues. Bilingual readers coactivate both languages even when they are immersed in an environment that requires the use of a single language alone. In the strongest case, parallel activation of the L2 on the L1 is observable when participants are recruited and tested in a purely monolingual setting (e.g., Van Hell & Dijkstra, 2002). However, when cross-language effects are compared across different settings, there is demonstrable sensitivity to the global context in the magnitude of the cross-language effects. With homographs as an index of coactivation, readers have been shown to zoom-in to the intended language as they progress through an experiment requiring the use of one language. For example, Elston-Güttler and colleagues (Elston-Güttler & Gunter, 2009; Elston-Güttler, Gunter, & Kotz, 2005; Paulmann, Elston-Güttler, Gunter, & Kotz, 2006) have shown that behavioral and electrophysiological homograph effects are reduced during the second half of experiments relative to the first half. Furthermore, when participants watch a film in the unintended language before beginning a task or if they hear background words in the unintended language during the task, the adjustment to the intended language is slowed. Taken together these results indicate that the global context of language usage can influence language nonselectivity during reading. Global context seems to have a weaker effect when cognates are used as the index of lexical co-activation may only be revealed using a highly sensitive measure such as ERPs, which provide information not only about the time course of an effect, but also what underlying cognitive processes may be involved. Gullifer et al. (2013) tested the global context effect on cross-language activation in a behavioral task. They asked two groups of Spanish–English bilingual readers to read sentences in Spanish and English, and sentences contained cognates between the two languages or lexically matched noncognates. One group of participants read the sentences in blocks of each language (i.e., a single-language context) and another group read the same sentences in mixed-language blocks (i.e., a mixed-language context). If the global context influences coactivation of the unintended language, then the magnitude of the cognate effect should be greater in mixed-language blocks relative to single language blocks. Yet, the magnitude of the cognate effect is strikingly similar

232 

 Kroll, Gullifer, and Zirnstein

regardless of the global context, hinting that this may serve a limited role in reducing coactivation when the target words strongly coactivate both languages. However, ERP studies have found a reduction in the N400 cognate effect for isolated word recognition in the nonimmersed language, indicating that there is sensitivity to global context (Midgley, Holcomb, & Grainger, 2011). Overall, global context that requires the use of one language can reduce, but not eliminate, coactivation of the unintended language.

Cross-Language Processes Within Sentence Context A critical issue when considering the nonselectivity of bilingual readers’ two languages is what happens when reading is situated in context. As discussed previously, when bilingual readers are reading in one language alone, there is still the potential for interference and support from the nontarget language. Bilingual readers are faced with a situation where the two languages often compete for selection at the level of the lexicon and grammar. The evidence suggests that when the two languages converge, syntactic priming can be observed from one language to the other (e.g., Hartsuiker, Pickering, & Veltkamp, 2004). When the two languages differ, they may do so because one language has a unique syntactic structure, or because one language makes different commitments in terms of parsing preferences (see Dussias, Marful, Gerfen, & Bajo Molina, 2010). This has repercussions for how the mechanisms underlying sentence and discourse processing unfold in real time and whether successful comprehension is ultimately attained (e.g., Kroll & Dussias, 2013). For example, syntactic parsing preferences in the L1 have been shown to be mediated by L2 exposure and proficiency (Dussias & Sagarra, 2007). The more L2 experience nonnative speakers have (e.g., the longer they have been immersed in an L2 context), the more likely it is that the L2 structure will interfere with processing in the native L1. This permeability between the frequency and processing of syntactic structures in the L1 and L2 reveals a complicated relationship between bilingual readers’ experience in both languages and how the L1 and L2 typically interact during natural reading. However, the work previously discussed focused on the processing of syntactically ambiguous sentences that are not frequently encountered in natural language use. In what way, then, might the coactivation of bilingual readers’ two languages be taken advantage of during normal comprehension? Recent research on code-switching in written language contexts may begin to provide an answer to this question.

Reading Code-Switched Text A code-switch occurs when a speaker switches from one language to another, sometimes within the same phrase or sentence, without changing the intended meaning of the utterance (see Example 1 from Altarriba, Kroll, Sholl, & Rayner, 1996).



Literacy in Adulthood 

 233

(1) He wanted to place all of his dinero at the credit union. (2) He wanted to place all of his money at the credit union. In cases like (1) a switch occurs within a noun phrase but is comparable in meaning to the translation equivalent in (2), where no switch occurs. Although these types of code-switches are typically studied in bilingual speech production, investigating the processing of code-switches within a written sentence context (i.e., intrasentential code-switches) can help reveal repercussions for language nonselectivity in the comprehension domain. The frequency with which bilingual speakers code-switch in natural conversation is thought to reflect the coactivation of bilingual speakers’ two languages. Altarriba et al. (1996) asked how bilingual speakers process sentences like (1) and contrasted them to single-language sentences like (2), and found that switching from one language to another produced a switch cost in longer reading times on the switched word (e.g., dinero). With both languages active, it is easy to see how interference or competition might occur. However, especially in dense code-switched speech, bilingual speakers appear to select the word that is most readily available in either language. Recent electrophysiological work sheds some light on this discrepancy. Moreno, Federmeier, and Kutas (2002) investigated the brain responses of bilingual readers when they encountered code-switches. In this task, bilingual participants read sentences that contained language switches, as in (3), within-language lexical switches (4), or no switch (5). (3) Each night the campers built a . . . fuego. (code-switch) (4) Each night the campers built a . . . blaze. (lexical switch) (5) Each night the campers built a . . . fire. (expected) Moreno et al. (2002) also included a condition in which target words were embedded in sentences that were highly semantically constrained (i.e., idioms). They hypoth­ esized that switch costs in reading may reflect difficulty in either accessing the meaning of the code-switched word in the nontarget language or in encountering an unexpected word. Either process should produce different results in ERP responses to the code-switches. If code-switching costs are because of difficulty in lexical access, then they should elicit an N400 response, related to ease of lexicosemantic integration (see Swaab et al., 2011). If code-switched words are more difficult because of their unexpectedness, then they should elicit a late positive ERP response, which is commonly found for words that are unexpected (Coulson, King, & Kutas, 1998; McCallum, Farmer, & Pocock, 1984), that require or trigger reanalysis (Friederici, 1995; Kolk & Chwilla, 2007), or are simply more difficult to integrate in sentence contexts (Kaan, Harris, Gibson, & Holcomb, 2000). Results indicated that lexical switches consistently generated N400s, whereas code-switches only did so for highly constrained, idiomatic sentences. Code-switches also produced late positive responses, suggesting

234 

 Kroll, Gullifer, and Zirnstein

that costs are due not only to difficulty with lexicosemantic integration but also to the unexpectedness of a code-switched word (cf. Proverbio, Leoni, & Zani, 2004; see van der Meij, Cuetos, Carreiras, & Barber, 2011). If code-switches encountered in natural reading are interpreted as unexpected, and therefore more difficult to process, in what way are code-switches in natural speech more advantageous to the listener or conversational partner? Moreno et al. (2002) found that bilingual speakers with higher L2 proficiency, tended to produce a late positivity that was earlier in onset than that produced by those with low L2 proficiency. This earlier shift in the late positive component may reflect earlier awareness of the code-switch for more proficient bilingual speakers. The pattern across these recent studies indicates that code-switches may be costly in unexpected situations, but that this cost can be modulated by coactivation of bilingual speakers’ L1 and L2. Language nonselectivity in bilingual reading can interfere with comprehension, but can also provide benefits if similarities between the languages are fully exploited.

The Costs of Prediction in L2 Sentence Context In the past, it was suggested that bilingual readers reading in the L2 have constraints on their cognitive resources (see Clahsen & Felser, 2006; McDonald, 2006), possibly as a repercussion of lower proficiency and/or later age of acquisition. Within the scope of this difficulty, to what extent does reading in the L2 draw on a bilingual reader’s cognitive and neural resources? Furthermore, is there any evidence to suggest that bilingual readers, even those who acquired their L2 later in life, are able to read in a nativelike fashion? A recent debate in the reading comprehension literature has focused on readers’ ability to engage in a task that is seen as highly resource demanding: predicting the features of upcoming words. These features could be orthographic (Laszlo & Federmeier, 2009), semantic (Federmeier, McLennan, De Ochoa, & Kutas, 2002; Federmeier, Wlotko, De Ochoa-Dewald, & Kutas, 2007), thematic (Kamide, Altmann, & Haywood, 2003), morphosyntactic (Otten & Van Berkum, 2009; Van Berkum, Brown, Zwitserlood, Kooijman, & Hagoort, 2005), or pragmatic (Van Berkum, 2008); however, of primary concern is how readers are able to generate these predictions and whether they are similarly capable of adapting their expectations to more closely match their reading experience. Prediction was thought to be too demanding for most readers to engage in, as the potential benefits of generating a prediction for a specific word may not outweigh the costs of encountering a different word entirely. However, in the last several years, it has been widely shown that young adult readers not only do tend to engage in prediction in their native language (see Federmeier, 2007; Van Berkum, 2008, 2012) but will also do so even if they encounter situations in which predictions are less helpful (Federmeier et al., 2007). In a recent study on prediction effects in L2 processing, Martin et al. (2013) asked whether proficient bilingual readers can generate predictions for the meaning of



Literacy in Adulthood 

 235

upcoming words in the L2. Bilingual and monolingual participants read sentences that were highly constrained but differed in the expectedness of the sentence final word. In addition, the indefinite article preceding the final word either agreed with the expected word (6) or did not (7). (6) I asked you a question because I need . . . an answer. (expected) (7) I asked you a question because I need . . . a response. (unexpected) The prediction effects in this study were therefore dependent on readers’ sensitivity to phonological agreement between the target noun and its preceding indefinite article. Their results suggested that the bilingual readers were not able to predict in the L2, whereas the monolingual readers were. In the unexpected condition, monolingual participants produced a late frontal positivity, an ERP response that has previously been shown in response to cases where predictions are disconfirmed (Federmeier et al., 2007). However, bilingual readers reading in their L2 did not produce this response. The authors took this to suggest that the bilingual readers in their study were not predicting, and that being able to form predictions in the L2 may require early age of acquisition and possibly a degree of proficiency on par with that of native speakers. Indeed, in a later study, Foucart, Martin, Moreno, and Costa (2014) found that earlier age of acquisition may play an important role in determining whether bilingual readers have the resources necessary to generate predictions in the L2. There are two possible issues with respect to this interpretation. One is that this level of proficiency may be more difficult to reach the later in life an individual starts to learn a second language. Late L2 learners may have more difficulty generating predictions than monolingual readers. This could be because of an inability to engage in language processing in a nativelike fashion (e.g., Clahsen & Felser, 2006) or to an inability to engage sufficient cognitive resources to process the L2 in real time (e.g., McDonald, 2006). In addition, it is possible that the late frontal positivity found in response to disconfirmed predictions is not solely a repercussion of predicting and having said prediction disconfirmed. Instead, it may reflect difficulty with inhibiting a previously formed prediction or mediating conflict between a prediction that has been generated and the unexpected word. If this process does rely on inhibition skills, then there is reason to expect that bilingual readers may be quite successful at resolving this type of conflict, as bilingual readers have often been shown to out­ perform matched monolingual readers on cognitive control tasks measuring inhibition skill (Bialystok, Craik, Klein, & Viswanathan, 2004; Gold, Kim, Johnson, Kryscio, & Smith, 2013). Zirnstein, Van Hell, and Kroll (2014a, 2014b) tested this by presenting bilingual readers with L2 sentences and measured their ERP responses to target words while they read for comprehension. The sentences included those like (8) that were highly semantically constraining and included a target word that was highly expected or

236 

 Kroll, Gullifer, and Zirnstein

predictable, and sentences like (9) that were also highly constraining, but had a target word that was plausible, but unexpected. (8) They paid for their meals, but forgot to leave a . . . tip for the waitress. (9) They paid for their meals, but forgot to leave a . . . ten for the waitress. In addition to the reading task, participants completed a cognitive control measure (i.e., the AX-CPT; Cohen, Barch, Carter, & Servan-Schreiber, 1999) and measures of verbal production fluency in the L1 and L2 (e.g., Luo, Luk, & Bialystok, 2010). The results from the reading task showed that bilingual readers reading in the L2 did produce a late frontal positivity in response in the unexpected condition, and that the magnitude of this effect was predicted by performance on the cognitive control task. Readers with better performance in the control task showed no costs in their ERPs during the unexpected condition. Only those with poorer performance on the control task generated the positivity, suggesting that this ERP effect is related to cognitive control ability and difficulty with inhibiting a previously formed prediction. In addition, this effect of cognitive control interacted with verbal fluency. Bilingual readers with lower control and lower verbal fluency in the L1 and L2 did not generate a frontal positive ERP response, possibly as a consequence of not having the resources necessary to generate a prediction in the L2. The findings from these ERP studies suggest that experience in the L2, and in mediating conflict between languages, has a direct influence on what strategies bilingual readers can adopt and how L2 reading unfolds over time. A recent issue in bilingual reading comprehension is to what extent late L2 learners are constrained in their ability to extract information from the L2, and whether these constraints apply to all late learners. The bilingual participants in Martín et al. (2010) and Zirnstein et al. (2014a, 2014b) were relatively late acquirers of their L2 (i.e., age of acquisition was after the age of 5). However, a consistent result across both studies was a lack of frontal positivity for some bilingual readers. Zirnstein and colleagues interpreted this effect as indicating successful recovery when a prediction had been disconfirmed. Recent work by Morgan-Short and colleagues (Morgan-Short, Finger, Grey, & Ullman, 2012; Morgan-Short, Steinhauer, Sanz, & Ullman, 2012) also indicates that even late bilingual readers can achieve and maintain nativelike processing. The evidence on L2 prediction and bilingual reading comprehension shows that nativelike processing is attainable, and that bilingual readers have the potential to overcome the demands of processing in an L2.

The Consequences of Bilingualism for Reading The research we have reviewed on reading in and out of sentence context suggests a high degree of interaction between bilingual readers’ two languages. On one hand, there is cross-language activation and competition that must be resolved for one lan-



Literacy in Adulthood 

 237

guage to be selected. On the other, there is evidence that bilingual readers are able to exploit that activation when they are required to code-switch from one language to the other. The ability of bilingual readers to control or regulate the activation of the language not in use has been hypothesized to confer a set of benefits to bilingual readers more generally (e.g., Bialystok et al., 2012; see also Chapter 16, this volume). These benefits or cognitive advantages have been observed in tasks that tap into executive function and that do not necessarily engage language explicitly, so that bilingual readers are often reported to outperform their monolingual counterparts when asked to ignore irrelevant information, resolve conflict, and switch between conditions. Although some have argued that the reported benefits are more likely to be because of experience in producing rather than comprehending language (e.g., Emmorey, Luk, Pyers, & Bialystok, 2008), as we noted earlier in this chapter, there is evidence for momentary inhibition of the nontarget language in reading text as well as in preparing speech (e.g., Martín et al., 2010). What we do not yet understand is how the moment-to-moment processes that engage the cognitive system as individuals read may accumulate over the life experience of bilingual readers to produce domain general changes to cognition. Those changes may be important, not only for issues of cognitive control, but also for how cognitive control processes come to shape reading. One area of research on reading new words in a new language has examined the consequences of bilingualism for the act of word learning. Bilingual readers are better able to acquire new vocabulary in a novel language than monolingual readers (e.g., Bogulski, Bice, & Kroll, 2015; Kaushanskaya & Marian, 2009). The benefit for bilingual readers in word learning may be a reflection of enhanced executive function that affects learning more generally or specific to those learning contexts in which bilingual readers have particular experience. Bogulski et al. (2015) compared these two alternatives by asking whether the bilingual advantage in new word learning would be found only for bilingual readers reading in their L1 and learning to link the new words (functionally an L3) to the L1, or also for bilingual readers reading in their L2 and linking the L3 to the L2. If the vocabulary learning benefit is due to domain general cognitive consequences of bilingualism, then all types of bilingual readers should reveal the effect. Bogulski et al. found the advantage only for bilingual readers reading in their L1 and argued that it is the L1 with which they have had the most inhibitory experience. It is the stronger of the two languages, typically the L1, that requires regulation to resolve the sort of cross-language competition described earlier in our review. The result that suggested this explanation was the observation that bilingual readers learning in L1 were slower to name the L1 translations of the new L3 words than monolingual readers. Critically, the same bilingual readers were not slower than monolingual readers in performing tasks outside of the learning context. Bogulski et al. argued that desirable difficulties in learning may produce better memory when the strategy at the time of learning enhances conceptual elaboration and meaningful feedback (e.g., Bjork, Dunlosky, & Kornell, 2013; see Bjork & Kroll, 2015).

238 

 Kroll, Gullifer, and Zirnstein

Our review has focused on reading in the ordinary sense, when readers process text for its meaning with the goal of comprehension. Much of the research on L2 reading has been addressed to the question of whether bilingual adults, and particularly late bilingual adults who acquired the L2 after early childhood, are able to read fluently in the L2, appreciating the nuances of meaning, the implications of L2-specific syntax, and generating predictions in the way that native speakers do. But in addition to processing code-switched text, bilingual readers also perform some other extra­ ordinary reading tasks, as when they translate from one language to the other. Monolingual readers can paraphrase text as they read, but a comparison of paraphrasing and translating for bilingual readers suggests that translating is the more natural of the two tasks (e.g., Christoffels & De Groot, 2004). Most bilingual readers are not trained as professional translators or simultaneous interpreters. Although the work of professional interpreters is a truly impressive feat with evidence for exceptional cognitive and memory abilities (e.g., Bajo, Padilla, & Padilla, 2000; Christoffels & De Groot, 2005), even ordinary bilingual readers are able to translate from one language to the other with relative ease. Although it seems clear that bilingual readers do not typically read text with the intention of translating it into the other language, a series of studies have asked how reading changes when the goal of reading is pure comprehension and when the goal is to translate. Macizo and Bajo (2006) examined reading performance when bilingual readers were told to read expecting to translate or to be able to repeat the text that they had read. They found evidence for enhanced sensitivity to the other language and increased processing load when bilingual readers anticipated producing the translation of the text they were reading. These findings suggest that comprehension is goal directed, with different cognitive control processes engaged as a function of the task.

Conclusion The research we have reviewed in this chapter suggests that bilingual readers’ two languages are fundamentally open to interactions that influence reading in each of their languages. That openness creates a level of competition that requires resolution but also enables bilingual readers do to more with language than their monolingual peers. What are the implications of these findings about cross-language interaction during reading? The costs associated with the requirement to regulate the language not in use appear to confer a set of positive consequences for language processing and for cognition. That topic is itself a focus of ongoing research across the lifespan to understand how the use of two or more languages changes cognition and the neural networks that support it (e.g., Green & Abutalebi, 2013; Kroll & Bialystok, 2013). But perhaps the most striking implication of the new research findings on bilingual reading is that the model of the native language and monolingual reader is no longer taken to characterize successful literacy. Reading in two languages is a



Literacy in Adulthood 

 239

natural process that comes to change both languages, not only the L2. The skilled bilingual reader differs from a monolingual reader, even when reading in the L1. We are just beginning to appreciate the scope of these differences but a full account will eventually inform everything from the evaluation of reading skill to approaches to reading instruction. The prevalence of bilingualism suggests that this new model of literacy that accommodates the presence of two or more languages will eventually become the norm.

References Abutalebi, J., & Green, D. (2007). Bilingual language production: The neurocognition of language representation and control. Journal of Neurolinguistics, 20, 242–275. http://dx.doi.org/ 10.1016/j.jneuroling.2006.10.003 Altarriba, J., Kroll, J. F., Sholl, A., & Rayner, K. (1996). The influence of lexical and conceptual constraints on reading mixed-language sentences: Evidence from eye fixations and naming times. Memory & Cognition, 24, 477–492. http://dx.doi.org/10.3758/BF03200936

Ameel, E., Storms, G., Malt, B. C., & Sloman, S. A. (2005). How bilinguals solve the naming problem. Journal of Memory and Language, 53, 60–80. http://dx.doi.org/10.1016/j.jml.2005.02.004

Bajo, M. T., Padilla, F., & Padilla, P. (2000). Comprehension processes in simultaneous interpreting.

In A. Chesterman, N. Gallardo, & Y. Gambier (Eds.), Translation in context (pp. 127–142). Amsterdam, the Netherlands: John Benjamins. http://dx.doi.org/10.1075/btl.39.15baj

Bialystok, E., Craik, F. I. M., Klein, R., & Viswanathan, M. (2004). Bilingualism, aging, and cognitive

control: Evidence from the Simon task. Psychology and Aging, 19, 290–303. http://dx.doi.org/ 10.1037/0882-7974.19.2.290

Bialystok, E., Craik, F. I. M. & Luk, G. (2012). Bilingualism: Consequences for mind and brain. Trends in Cognitive Sciences, 16, 240–250. http://dx.doi.org/10.1016/j.tics.2012.03.001

Bice, K., & Kroll, J. F. (2015). Native language change during early stages of second language learning. NeuroReport, 26, 966–971.

Bjork, R. A., Dunlosky, J., & Kornell, N. (2013). Self-regulated learning: Beliefs, techniques, and illusions. Annual Review of Psychology, 64, 417–444. http://dx.doi.org/10.1146/ annurev-psych-113011-143823

Bjork, R. A., & Kroll, J. F. (2015). Desirable difficulties in vocabulary learning. The American Journal of Psychology, 128, 241–252. http://dx.doi.org/10.5406/amerjpsyc.128.2.0241

Blumenfeld, H. K., & Marian, V. (2011). Bilingualism influences inhibitory control in auditory comprehension. Cognition, 118, 245–257. http://dx.doi.org/10.1016/j.cognition.2010.10.012

Bobb, S. C., & Wodniecka, Z. (2013). Language switching in picture naming: What asymmetric switch

costs (do not) tell us about inhibition in bilingual speech planning. Journal of Cognitive Psychology, 25, 568–585. http://dx.doi.org/10.1080/20445911.2013.792822

Bogulski, C. A., Bice, K., & Kroll, J. F. (2015). A bilingual advantage in vocabulary acquisition depends

on learning via the dominant language (Unpublished manuscript). Pennsylvania State University, University Park.

Chang, C. B. (2013). A novelty effect in phonetic drift of the native language. Journal of Phonetics, 41, 520–533. http://dx.doi.org/10.1016/j.wocn.2013.09.006

240 

 Kroll, Gullifer, and Zirnstein

Christoffels, I. K., & De Groot, A. (2004). Components of simultaneous interpreting: Comparing interpreting with shadowing and paraphrasing. Bilingualism: Language and Cognition, 7, 227–240. http://dx.doi.org/10.1017/S1366728904001609 Christoffels, I. K., & De Groot, A. M. B. (2005). Simultaneous interpreting: A cognitive perspective. In J. F. Kroll & A. M. B. de Groot (Eds.), Handbook of bilingualism: Psycholinguistic approaches (pp. 454–479). New York, NY: Oxford University Press. Clahsen, H., & Felser, C. (2006). Grammatical processing in language learners. Applied Psycholinguistics, 27, 3–42. http://dx.doi.org/10.1017/S0142716406060024 Cohen, J. D., Barch, D. M., Carter, C., & Servan-Schreiber, D. (1999). Context-processing deficits in schizophrenia: Converging evidence from three theoretically motivated cognitive tasks. Journal of Abnormal Psychology, 108, 120–133. http://dx.doi.org/10.1037/0021-843X.108.1.120 Coulson, S., King, J. W., & Kutas, M. (1998). Expect the unexpected: Event-related brain response to morphosyntactic violations. Language and Cognitive Processes, 13, 21–58. http://dx.doi.org/ 10.1080/016909698386582 Dijkstra, T. (2005). Bilingual word recognition and lexical access. In J. F. Kroll & A. M. B. de Groot (Eds.), Handbook of bilingualism: Psycholinguistic approaches (pp. 179–201). New York, NY: Oxford University Press. Dijkstra, T., Grainger, J., & Van Heuven, W. J. (1999). Recognition of cognates and interlingual homographs: The neglected role of phonology. Journal of Memory and Language, 41, 496–518. http://dx.doi.org/10.1006/jmla.1999.2654 Dijkstra, T., Van Jaarsveld, H., & Ten Brinke, S. (1998). Interlingual homograph recognition: Effects of task demands and language intermixing. Bilingualism: Language and Cognition, 1, 51–66. http://dx.doi.org/10.1017/S1366728998000121 Dussias, P. E., Marful, A., Gerfen, C., & Bajo Molina, M. T. (2010). Usage frequencies of complementtaking verbs in Spanish and English: Data from Spanish monolinguals and Spanish-English bilinguals. Behavior Research Methods, 42, 1004–1011. http://dx.doi.org/10.3758/ BRM.42.4.1004 Dussias, P. E., & Sagarra, N. (2007). The effect of exposure on syntactic parsing in Spanish–English bilinguals. Bilingualism: Language and Cognition, 10, 101–116. http://dx.doi.org/10.1017/ S1366728906002847 Duyck, W., Van Assche, E., Drieghe, D., & Hartsuiker, R. J. (2007). Visual word recognition by bilinguals in a sentence context: Evidence for nonselective lexical access. Journal of Experimental Psychology: Learning, Memory, and Cognition, 33, 663–679. http://dx.doi.org/10.1037/ 0278-7393.33.4.663 Elston-Güttler, K. E., & Gunter, T. C. (2009). Fine-tuned: Phonology and semantics affect first- to second-language zooming in. Journal of Cognitive Neuroscience, 21, 180–196. http://dx.doi.org/ 10.1162/jocn.2009.21015 Elston-Güttler, K. E., Gunter, T. C., & Kotz, S. A. (2005). Zooming into L2: Global language context and adjustment affect processing of interlingual homographs in sentences. Cognitive Brain Research, 25, 57–70. http://dx.doi.org/10.1016/j.cogbrainres.2005.04.007 Emmorey, K., Luk, G., Pyers, J. E., & Bialystok, E. (2008). The source of enhanced cognitive control in bilinguals: Evidence from bimodal bilinguals. Psychological Science, 19, 1201–1206. http:// dx.doi.org/10.1111/j.1467-9280.2008.02224.x Federmeier, K. D. (2007). Thinking ahead: The role and roots of prediction in language comprehension. Psychophysiology, 44, 491–505. http://dx.doi.org/10.1111/j.1469-8986.2007.00531.x



Literacy in Adulthood 

 241

Federmeier, K. D., & Kutas, M. (1999). A rose by any other name: Long-term memory structure and sentence processing. Journal of Memory and Language, 41, 469–495. http://dx.doi.org/ 10.1006/jmla.1999.2660 Federmeier, K. D., McLennan, D. B., De Ochoa, E., & Kutas, M. (2002). The impact of semantic memory organization and sentence context information on spoken language processing by younger and older adults: An ERP study. Psychophysiology, 39, 133–146. http://dx.doi.org/ 10.1111/1469-8986.3920133 Federmeier, K. D., Wlotko, E. W., De Ochoa-Dewald, E., & Kutas, M. (2007). Multiple effects of sentential constraint on word processing. Brain Research, 1146, 75–84. http://dx.doi.org/10.1016/ j.brainres.2006.06.101 Foucart, A., Martin, C. D., Moreno, E., & Costa, A. (2014). Can bilinguals guess what’s coming? Word anticipation in L2 sentence reading. Journal of Experimental Psychology: Learning, Memory, and Cognition, 40, 1461–1469. Friederici, A. D. (1995). The time course of syntactic activation during language processing: A model based on neuropsychological and neurophysiological data. Brain and Language, 50, 259–281. http://dx.doi.org/10.1006/brln.1995.1048 Gold, B. T., Kim, C., Johnson, N. F., Kryscio, R. J., & Smith, C. D. (2013). Lifelong bilingualism maintains neural efficiency for cognitive control in aging. The Journal of Neuroscience, 33, 387–396. http://dx.doi.org/10.1523/JNEUROSCI.3837-12.2013 Gollan, T. H., Montoya, R. I., Cera, C., & Sandoval, T. C. (2008). More use almost always means a smaller frequency effect: Aging, bilingualism, and the weaker links hypothesis. Journal of Memory and Language, 58, 787–814. http://dx.doi.org/10.1016/j.jml.2007.07.001 Gollan, T. H., Slattery, T. J., Goldenberg, D., Van Assche, E., Duyck, W., & Rayner, K. (2011). Frequency drives lexical access in reading but not in speaking: The frequency-lag hypothesis. Journal of Experimental Psychology: General, 140, 186–209. http://dx.doi.org/10.1037/a0022256 Green, D. W. (1998). Mental control of the bilingual lexico-semantic system. Bilingualism: Language and Cognition, 1, 67–81. http://dx.doi.org/10.1017/S1366728998000133 Green, D. W., & Abutalebi, J. (2013). Language control in bilinguals: The adaptive control hypothesis. Journal of Cognitive Psychology, 25, 515–530. http://dx.doi.org/10.1080/ 20445911.2013.796377 Gullifer, J. W., Kroll, J. F., & Dussias, P. E. (2013). When language switching has no apparent cost: Lexical access in sentence context. Frontiers in Psychology, 4, 278. http://dx.doi.org/10.3389/ fpsyg.2013.00278 Hartsuiker, R. J., Pickering, M. J., & Veltkamp, E. (2004). Is syntax separate or shared between languages? Cross-linguistic syntactic priming in Spanish-English bilinguals. Psychological Science, 15, 409–414. http://dx.doi.org/10.1111/j.0956-7976.2004.00693.x Ibáñez, A. J., Macizo, P., & Bajo, M. T. (2010). Language access and language selection in professional translators. Acta Psychologica, 135, 257–266. http://dx.doi.org/10.1016/ j.actpsy.2010.07.009 Kaan, E., Harris, A., Gibson, E., & Holcomb, P. J. (2000). The P600 as an index of syntactic integration difficulty. Language and Cognitive Processes, 15, 159–201. http://dx.doi.org/10.1080/ 016909600386084 Kamide, Y., Altmann, G. T. M., & Haywood, S. L. (2003). The time-course of prediction in incremental sentence processing: Evidence from anticipatory eye movements. Journal of Memory and Language, 49, 133–156. http://dx.doi.org/10.1016/S0749-596X(03)00023-8

242 

 Kroll, Gullifer, and Zirnstein

Kaushanskaya, M., & Marian, V. (2009). Bilingualism reduces native-language interference during novel-word learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35, 829–835. http://dx.doi.org/10.1037/a0015275 Kolk, H., & Chwilla, D. (2007). Late positivities in unusual situations. Brain and Language, 100, 257–261. http://dx.doi.org/10.1016/j.bandl.2006.07.006 Kroll, J. F., & Bialystok, E. (2013). Understanding the consequences of bilingualism for language processing and cognition. Journal of Cognitive Psychology, 25, 497–514. http://dx.doi.org/ 10.1080/20445911.2013.799170 Kroll, J. F., Bobb, S. C., & Hoshino, N. (2014). Two languages in mind: Bilingualism as a tool to investigate language, cognition, and the brain. Current Directions in Psychological Science, 23, 159–163. http://dx.doi.org/10.1177/0963721414528511 Kroll, J. F., & Dussias, P. E. (2013). The comprehension of words and sentences in two languages. In T. Bhatia & W. Ritchie (Eds.), The handbook of bilingualism and multilingualism (2nd ed., pp. 216–243). Malden, MA: Wiley-Blackwell. Kroll, J. F., Dussias, P. E., Bogulski, C. A., & Kroff, J. R. V. (2012). Juggling two languages in one mind: What bilinguals tell us about language processing and its consequences for cognition. In B. Ross (Ed.), The psychology of learning and motivation (Vol. 56, pp. 229–262). San Diego, CA: Academic Press. http://dx.doi.org/10.1016/B978-0-12-394393-4.00007-8 Kroll, J. F., Gullifer, J., & Rossi, E. (2013). The multilingual lexicon: The cognitive and neural basis of lexical comprehension and production in two languages. Annual Review of Applied Linguistics, 33, 102–127. http://dx.doi.org/10.1017/S0267190513000111 Kroll, J. F., & Stewart, E. (1994). Category interference in translation and picture naming: Evidence for asymmetric connections between bilingual memory representations. Journal of Memory and Language, 33, 149–174. http://dx.doi.org/10.1006/jmla.1994.1008 Kroll, J. F., van Hell, J. G., Tokowicz, N., & Green, D. W. (2010). The Revised Hierarchical Model: A critical review and assessment. Bilingualism: Language and Cognition, 13, 373–381. http://dx.doi.org/ 10.1017/S136672891000009X Kutas, M., & Federmeier, K. D. (2011). Thirty years and counting: Finding meaning in the N400 component of the event-related brain potential (ERP). Annual Review of Psychology, 62, 621–647. http://dx.doi.org/10.1146/annurev.psych.093008.131123 Kutas, M., & Hillyard, S. A. (1984). Brain potentials during reading reflect word expectancy and semantic association. Nature, 307, 161–163. http://dx.doi.org/10.1038/307161a0 Laszlo, S., & Federmeier, K. D. (2009). A beautiful day in the neighborhood: An event-related potential study of lexical relationships and prediction in context. Journal of Memory and Language, 61, 326–338. http://dx.doi.org/10.1016/j.jml.2009.06.004 Libben, M. R., & Titone, D. A. (2009). Bilingual lexical access in context: Evidence from eye movements during reading. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35, 381–390. http://dx.doi.org/10.1037/a0014875 Luo, L., Luk, G., & Bialystok, E. (2010). Effect of language proficiency and executive control on verbal fluency performance in bilinguals. Cognition, 114, 29–41. http://dx.doi.org/10.1016/ j.cognition.2009.08.014 Macizo, P., & Bajo, M. T. (2006). Reading for repetition and reading for translation: Do they involve the same processes? Cognition, 99, 1–34. http://dx.doi.org/10.1016/j.cognition.2004.09.012 MacWhinney, B. (2005). A unified model of language acquisition. In J. F. Kroll & A. M. B. de Groot (Eds.), Handbook of bilingualism: Psycholinguistic approaches (pp. 49–67). New York, NY: Oxford University Press.



Literacy in Adulthood 

 243

Martin, C. D., Thierry, G., Kuipers, J.-R., Boutonnet, B., Foucart, A., & Costa, A. (2013). Bilinguals reading in their second language do not predict upcoming words as native readers do. Journal of Memory and Language, 69, 574–588. http://dx.doi.org/10.1016/j.jml.2013.08.001 Martín, M. C., Macizo, P., & Bajo, T. (2010). Time course of inhibitory processes in bilingual language processing. British Journal of Psychology, 101, 679–693. http://dx.doi.org/10.1348/ 000712609X480571 McCallum, W. C., Farmer, S. F., & Pocock, P. V. (1984). The effects of physical and semantic incongruities on auditory event-related potentials. Electroencephalography and Clinical Neurophysiology: Evoked Potentials, 59, 477–488. http://dx.doi.org/10.1016/0168-5597(84)90006-6 McDonald, J. L. (2006). Beyond the critical period: Processing-based explanations for poor grammaticality judgment performance by late second language learners. Journal of Memory and Language, 55, 381–401. http://dx.doi.org/10.1016/j.jml.2006.06.006 Mei, L., Xue, G., Lu, Z. L., Chen, C., Zhang, M., He, Q., . . . Dong, Q. (2014). Learning to read words in a new language shapes the neural organization of the prior languages. Neuropsychologia, 65, 156–168. http://dx.doi.org/10.1016/j.neuropsychologia.2014.10.019 Midgley, K. J., Holcomb, P. J., & Grainger, J. (2009). Language effects in second language learners and proficient bilinguals investigated with event-related potentials. Journal of Neurolinguistics, 22, 281–300. http://dx.doi.org/10.1016/j.jneuroling.2008.08.001 Midgley, K. J., Holcomb, P. J., & Grainger, J. (2011). Effects of cognate status on word comprehension in second language learners: An ERP investigation. Journal of Cognitive Neuroscience, 23, 1634–1647. http://dx.doi.org/10.1162/jocn.2010.21463 Misra, M., Guo, T., Bobb, S. C., & Kroll, J. F. (2012). When bilinguals choose a single word to speak: Electrophysiological evidence for inhibition of the native language. Journal of Memory and Language, 67, 224–237. http://dx.doi.org/10.1016/j.jml.2012.05.001 Moreno, E. M., Federmeier, K. D., & Kutas, M. (2002). Switching languages, switching palabras (words): An electrophysiological study of code switching. Brain and Language, 80, 188–207. http://dx.doi.org/10.1006/brln.2001.2588 Morford, J. P., Wilkinson, E., Villwock, A., Piñar, P., & Kroll, J. F. (2011). When deaf signers read English: Do written words activate their sign translations? Cognition, 118, 286–292. http:// dx.doi.org/10.1016/j.cognition.2010.11.006 Morgan-Short, K., Finger, I., Grey, S., & Ullman, M. T. (2012). Second language processing shows increased native-like neural responses after months of no exposure. PLoS ONE, 7, e32974. http://dx.doi.org/10.1371/journal.pone.0032974 Morgan-Short, K., Steinhauer, K., Sanz, C., & Ullman, M. T. (2012). Explicit and implicit second language training differentially affect the achievement of native-like brain activation patterns. Journal of Cognitive Neuroscience, 24, 933–947. http://dx.doi.org/10.1162/jocn_a_00119 Otten, M., & Van Berkum, J. J. A. (2009). Does working memory capacity affect the ability to predict upcoming words in discourse? Brain Research, 1291, 92–101. http://dx.doi.org/10.1016/ j.brainres.2009.07.042 Paulmann, S., Elston-Güttler, K. E., Gunter, T. C., & Kotz, S. A. (2006). Is bilingual lexical access influenced by language context? NeuroReport, 17, 727–731. http://dx.doi.org/10.1097/ 01.wnr.0000214400.88845.fa Proverbio, A. M., Leoni, G., & Zani, A. (2004). Language switching mechanisms in simul­taneous interpreters: An ERP study. Neuropsychologia, 42, 1636–1656. http://dx.doi.org/10.1016/ j.neuropsychologia.2004.04.013

244 

 Kroll, Gullifer, and Zirnstein

Schwartz, A. I., & Kroll, J. F. (2006). Bilingual lexical activation in sentence context. Journal of Memory and Language, 55, 197–212. http://dx.doi.org/10.1016/j.jml.2006.03.004 Schwartz, A. I., Kroll, J. F., & Diaz, M. (2007). Reading words in Spanish and English: Mapping orthography to phonology in two languages. Language and Cognitive Processes, 22, 106–129. http://dx.doi.org/10.1080/01690960500463920 Swaab, T. Y., Ledoux, K., Camblin, C. C., & Boudewyn, M. A. (2011). Language-related ERP components. In S. J. Luck & E. S. Kappenman (Eds.), The Oxford handbook of event-related potential components (pp. 397–440). Oxford, England: Oxford University Press. Thierry, G., & Wu, Y. J. (2007). Brain potentials reveal unconscious translation during foreign-language comprehension. Proceedings of the National Academy of Sciences of the United States of America, 104, 12530–12535. http://dx.doi.org/10.1073/pnas.0609927104 Thomas, M. S., & Allport, A. (2000). Language switching costs in bilingual visual word recognition. Journal of Memory and Language, 43, 44–66. http://dx.doi.org/10.1006/jmla.1999.2700 Titone, D., Libben, M., Mercier, J., Whitford, V., & Pivneva, I. (2011). Bilingual lexical access during L1 sentence reading: The effects of L2 knowledge, semantic constraint, and L1–L2 intermixing. Journal of Experimental Psychology: Learning, Memory, and Cognition, 37, 1412–1431. http:// dx.doi.org/10.1037/a0024492 Van Assche, E., Drieghe, D., Duyck, W., Welvaert, M., & Hartsuiker, R. J. (2011). The influence of semantic constraints on bilingual word recognition during sentence reading. Journal of Memory and Language, 64, 88–107. http://dx.doi.org/10.1016/j.jml.2010.08.006 Van Assche, E., Duyck, W., Hartsuiker, R. J., & Diependaele, K. (2009). Does bilingualism change native-language reading? Cognate effects in a sentence context. Psychological Science, 20, 923–927. http://dx.doi.org/10.1111/j.1467-9280.2009.02389.x Van Berkum, J. J. A. (2008). Understanding sentences in context: What brain waves can tell us. Current Directions in Psychological Science, 17, 376–380. http://dx.doi.org/10.1111/ j.1467-8721.2008.00609.x Van Berkum, J. J. A. (2012). The electrophysiology of discourse and conversation. In M. Spivey, K. McRae, & M. Joanisse (Eds.), The Cambridge handbook of psycholinguistics (pp. 589–612). Cambridge, England: Cambridge University Press. http://dx.doi.org/10.1017/ CBO9781139029377.040 Van Berkum, J. J. A., Brown, C. M., Zwitserlood, P., Kooijman, V., & Hagoort, P. (2005). Anticipating upcoming words in discourse: Evidence from ERPs and reading times. Journal of Experimental Psychology: Learning, Memory, and Cognition, 31, 443–467. http://dx.doi.org/10.1037/ 0278-7393.31.3.443 van der Meij, M., Cuetos, F., Carreiras, M., & Barber, H. A. (2011). Electrophysiological correlates of language switching in second language learners. Psychophysiology, 48, 44–54. http:// dx.doi.org/10.1111/j.1469-8986.2010.01039.x Van Hell, J. G., & De Groot, A. M. (2008). Sentence context modulates visual word recognition and translation in bilinguals. Acta Psychologica, 128, 431–451. http://dx.doi.org/10.1016/ j.actpsy.2008.03.010 Van Hell, J. G., & Dijkstra, T. (2002). Foreign language knowledge can influence native language performance in exclusively native contexts. Psychonomic Bulletin & Review, 9, 780–789. http:// dx.doi.org/10.3758/BF03196335 van Orden, G. C. (1987). A ROWS is a ROSE: Spelling, sound, and reading. Memory & Cognition, 15, 181–198. http://dx.doi.org/10.3758/BF03197716



Literacy in Adulthood 

 245

von Studnitz, R. E., & Green, D. W. (2002). The cost of switching language in a semantic categorization task. Bilingualism: Language and Cognition, 5, 241–251. http://dx.doi.org/10.1017/ S1366728902003036 Wu, Y. J., Cristino, F., Leek, C., & Thierry, G. (2013). Non-selective lexical access in bilinguals is spontaneous and independent of input monitoring: Evidence from eye tracking. Cognition, 129, 418–425. http://dx.doi.org/10.1016/j.cognition.2013.08.005 Wu, Y. J., & Thierry, G. (2010). Chinese–English bilinguals reading English hear Chinese. The Journal of Neuroscience, 30, 7646–7651. http://dx.doi.org/10.1523/JNEUROSCI.1602-10.2010 Zirnstein, M., Van Hell, J. G., & Kroll, J. F. (2014a). Cognitive control ability mediates prediction costs in monolinguals and bilinguals. Manuscript submitted for publication. Zirnstein, M., Van Hell, J. G., & Kroll, J. F. (2014b, August). Inhibition plays a role when readers predict in a first or second language. Poster presented at the 21st Annual Meeting of the Cognitive Neuroscience Society, Boston, MA.

IV Cognitive Effects of Bilingualism

Ágnes Melinda Kovács

13 Cognitive Effects of Bilingualism in Infancy Exposure to multiple languages is a very common phenomenon even during early childhood. Although learning just one language is a major accomplishment in itself, the challenge for infants born in multilingual environments must be still greater. In contemporary societies many children grow up in bilingual families and have to learn to cope with different languages. However, a single language milieu is still the standard model for investigating language acquisition even though a great proportion of children are raised with more than one language. As bilingual children presumably have to learn twice as much as their monolingual peers, their language learning could be expected to be somewhat delayed. Yet, infants who acquire two languages simultaneously pass language production milestones at the same age as monolingual infants (see Chapter 4, this volume), and display only minor differences in language processing (see Chapter 3). Thus, the big puzzle becomes uncovering what mechanisms infants exposed to two languages from birth (crib bilinguals) use to efficiently deal with a linguistic signal coming from different languages.

Milestones in Bilingual Language Acquisition While examining bilingual language acquisition one is faced with a bilingual language acquisition paradox (Petitto et al., 2001). This refers, on the one hand, to the amazement of parents and scientists observing how effortlessly children acquire two or more languages. On the other hand, it captures the worry that exposing children to two languages might result in language delays and confusion. Two main theories have been formulated regarding how young children may deal with bilingual input (see Chapter 2, this volume, for an extensive account of this issue). According to the unitary language system account, in the early phases of language learning, children form a single language system for both languages (Leopold, 1978; Volterra & Taeschner, 1978). This account relies on findings showing that bilingual speakers have few translation equivalents across their two languages in the one-word stage (Volterra & Taeschner, 1978), and they frequently mix languages in their word combinations (Vihman, Macken, Miller, Simmons, & Miller, 1985). Of course, bilingual language acquisition is different from monolingual language acquisition in specific ways; for instance, bilingual speakers tend to have smaller http://dx.doi.org/10.1037/14939-014 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

250 

 Ágnes Melinda Kovács

vocabularies than monolingual speakers if only one language is considered (Chapter 4, this volume). However, surprisingly, bilingual speakers seem to reach the basic milestones in acquiring both of their languages (first-word stage, first 50 words, and twoword combinations) at the same age as monolingual speakers (see Table 13.1). This is the case even for children who learn two languages that belong to different modalities (e.g., Sign Language and French; Petitto et al., 2001). In addition, in contrast to the findings that bilingual children produce few translation equivalents (Volterra & Taeschner, 1978), more recent research has found that bilingual speakers’ early vocabularies consist on average of 30% of such “doublets” already in the one-word stage (Pearson, Fernández, & Oller, 1995). Furthermore, bilingual children who receive approximately equal input in both languages mix them following their parents’ language mixing patterns (Genesee, Nicoladis, & Paradis, 1995). Such results support a differentiated language system hypothesis according to which young bilingual speakers construct two distinct representational systems for the two languages early on. In sum, crib bilingualism does not appear to significantly alter the course of language development. However, the mechanisms that allow language differentiation are still unclear and it is possible that specific changes in cognitive processes occur as a result of being exposed to and having to learn multiple languages. Mechanisms outside the domain of language such as attention, inhibition and selection might be used to a greater extent when dealing simultaneously with two languages. Before discussing the possible cognitive changes I will first address processes closely related to language learning that may provide the foundation for a fast and efficient acquisition of the language(s) infants hear in their environment.

Tab. 13.1: Linguistic Milestones in Monolingual and Bilingual Children

Study

Participants

Vihman & McCune,  1994 Petitto, 1987

English monolingual  speakers English monolingual  speakers English monolingual  speakers Spanish–English  bilingual speakers French–English  bilingual speakers Sign Language–  French bilingual  speakers

Pearson, Fernández,  & Oller, 1993

Petitto et al., 2001

First-word stage

First 50 words

First two-word combination

—

— 1.7 years

1 year

1.6 years (range:  1.5–2.2 years) 1.6 years

1.1 year

1.7 years

1.8 years

1.1 years

1.6 years

1.7 years

10 months

1.6 years

1.5 years

1 year (range:  9–14 months) —

1.7 years



 251

Cognitive Effects of Bilingualism in Infancy 

Cognitive Processes in the Service of Language Learning: Infant Language Discrimination Monolingual and bilingual infants have to process speech signals to acquire language. However, only bilingual infants are exposed to utterances from two languages. If they were unable to sort utterances into the different source languages, bilingual children would present considerable learning difficulties and display delays. However, such delays and confusions are rare or inexistent (Genesee et al., 1995; Petitto et al., 2001). Likely, infants are equipped with abilities to monitor and segregate the linguistic input into distinct categories from very early on. Specific processes, such as an automatic rhythmic clustering of the languages, may allow language differentiation even when both are unknown to the listener (Ramus & Mehler, 1999; see also Chapter 3, this volume, for a review of this issue). Already a few days after birth, infants possess impressive language discrimination abilities, distinguishing different languages on the basis of their prosodic properties (Nazzi, Bertoncini, & Mehler, 1998; Ramus, Hauser, Miller, Morris, & Mehler, 2000). However, two languages that share prosodic similarities (e.g., English and Dutch) are difficult to discriminate by young learners. French newborns and two-monthold English learning infants fail indeed to show such differentiation (Christophe & Morton, 1998; Nazzi et al., 1998; see Table 13.2). By their fourth month, however, monolingual and bilingual infants can tell apart two languages that have similar rhythmic characteristics (e.g., Catalan and Spanish; Bosch & Sebastián-Gallés, 1997,

Tab. 13.2: Early Discrimination Abilities in Monolingual and Bilingual Infants Study

Participants

Language pairs

Nazzi, Bertoncini, &  Mehler, 1998

French newborns

Ramus et al., 2000

French newborns

Low-pass filtered Japanese–English Dutch–English English + Dutch vs.  Spanish + Italian Natural Dutch–Japanese  resynthesized Dutch–Japanese Japanese–English Dutch–English French–Japanese Dutch–Japanese Catalan–Spanish

Christophe & Morton,  1998

2-month-old English  monolingual infants

Bosch & Sebastián Gallés, 2001

4-month-old Spanish  monolingual infants 4-month-old Catalan–Spanish  bilingual infants

Catalan–Spanish

Discrimination √ X √ √ √ √ X X √(X) √ √

252 

 Ágnes Melinda Kovács

2001), suggesting that, after a certain amount of exposure to their native languages, infants are able to use other cues for discrimination besides the rhythmic properties of their language(s). Languages also differ in properties other than prosodic (e.g., phonetic repertoire, phonotactic constraints) and young bilingual speakers can make use of these properties to make fine-grained discriminations between rhythmically similar languages (Sebastián-Gallés & Bosch, 2002). These powerful and early abilities, besides other nonlanguage specific processes, may allow crib bilinguals to differentiate their languages prelexically, to avoid delays and confusion, and to reach linguistic milestones at the same age as monolingual peers.

Processing a Bimodal Linguistic Input Enhances Executive Functions in Infancy How does the infant’s developing cognitive system manage to deal with utterances belonging to two different languages? What is the impact of receiving such a complex linguistic input on the development of diverse cognitive abilities? Processing two languages simultaneously may result in specific changes in the cognitive system as well as in structural reorganization at the neuronal level (García-Pentón, Pérez Fernández, Iturria-Medina, Gillon-Dowens, & Carreiras, 2014; Mechelli et al., 2004). Neuroimaging data suggest that bilingual adults indeed have greater gray matter density than monolingual adults in certain brain areas, such as the left inferior parietal cortex (Mechelli et al., 2004). Such reorganization is more pronounced in early bilingual speakers. Behavioral studies further suggest that mastering two languages from an early age influences certain domains of cognitive functioning; bilingual adults and preschool-age children indeed display enhanced cognitive control abilities (executive functions) because of practice in suppressing one language while speaking the other (Bialystok, 1999; Costa, Hernández, & Sebastián-Gallés, 2008). However, how early such a bilingual advantage may start has been investigated little. Fast learning of new regularities by neglecting or overwriting the old ones is crucial for adjusting behavior to the changing requirements of the environment and for performing daily activities in different domains (Burgess, Veitch, de Lacy Costello, & Shallice, 2000). Such abilities are usually termed executive functions (EF), which is an umbrella term for inhibition, monitoring, and attention switching. Because of the continuous monitoring of dual language input, bilingualism could boost executive control before children start producing words. First, to learn two different languages, bilingual speakers have to sort the speech utterances according to the source language. Then, to learn the distinct patterns and regularities that belong to each language and to build different language representations, young learners might recruit complex control and monitoring processes to keep the two representational sets sepa-



Cognitive Effects of Bilingualism in Infancy 

 253

rate and avoid conflict and interference between the two language systems. Furthermore, bilingual language learning might also involve a continuous switch of attention between the two language systems. An early and extensive use of EF in young bilingual speakers might thus lead to an accelerated development of the involved abilities. Before describing studies that aimed to test this possibility, I will first discuss data from children and adults that point to advantages of bilingual speakers in EF.

Sharpening Executive Functions Through Language Switching The daily language switching in bilingual language production could result in specific changes in the cognitive system. Bialystok (1999) found that the experience gained during bilingual language selection and inhibition leads to advantages in performing tasks that require inhibitory functions in preschool-age children. Studies with adults have documented similar advantages for bilingual speakers (Bialystok, Craik, et al., 2005). These studies were motivated by the conjecture that bilingual speakers may involve the same mechanisms for language switching that they recruit for solving different executive function tasks, such as the Stroop task, the Simon task, or the attentional network task. To communicate efficiently, bilingual speakers have to control which of their languages they currently use and possibly inhibit the lexical items of one language when switching to the other (Green, 1998). The practice bilingual speakers have in managing the two languages seems to lead to more efficient executive control in general. Bilingual adults outperform monolingual adults on diverse EF tasks; for instance, they show reduced interference effect in a Simon task (Bialystok, Craik, et al., 2005) where there is a spatial stimulus-response incompatibility (e.g., responding to a colored stimulus appearing on the right side of the screen with the left hand, compared to when it appears on the left side). Likewise, bilingual speakers show better attentional control and reduced task switching costs in the Attentional Network Task (Costa et al., 2008). In this task, participants are exposed to five arrows and asked to indicate whether the central arrow points to the right or left. Responses tend to be slower when the central arrow is presented along with distractor arrows pointing to the opposite direction (incongruent trials) rather than the same direction (congruent trials). The bilingual advantage in executive control tasks seems to persist from toddlerhood and preschool age (Poulin-Dubois, Blaye, Coutya, & Bialystok, 2011; see also Chapter 14, this volume), through young adulthood to elderly ages (Bialystok, Martin, & Viswanathan, 2005; Chapter 16, this volume), although such advantages are not observed in some cases (Duñabeitia et al., 2014). Yet, little is known about whether a comparable enhancement could result from the mere exposure to different languages before language production begins. Processing utterances belonging to two different languages may already be sufficient for the enhancement of such abilities. Hence, an

254 

 Ágnes Melinda Kovács

EF advantage might exist even in bilingually exposed infants who are not yet able to produce words.

Monitoring Two Languages Boosts Executive Control in Infancy The ability to select between competing stimuli and behavioral responses and override them seems to have a slow development. These abilities reach adult levels only toward puberty (Casey et al., 1997), with important improvements around the end of the first year (Diamond, 1985) and around age 4 (Gerstadt, Hong, & Diamond, 1994). Behavioral studies suggests that inhibitory abilities are still little developed in 7-month-olds, as they perform poorly on the A not B tasks that require the inhibition of a previously rewarded response (e.g., searching for a toy in location B after a delay period and after having previously found it in location A for several times, Diamond, 1985). There can be special circumstances that boost inhibitory and control abilities in young infants, possibly through accelerating developmental changes in specific brain areas. Such evidence comes from a study by Matthews, Ellis, and Nelson (1996), who compared preterm and full-term infants of the same conception age on a non-reaching type of A not B task. In the A not B tasks infants are exposed to situations in which first an object is repeatedly hidden in location A, and then it is hidden in location B in the full view of the child. Typical errors (searching in location A as a prepotent response) are observed when the delay between the hiding and the search is lengthened (Diamond, 1985). Preterm infants, who had more experience with the events of the surrounding world, tolerated greater delays compared with full-term infants (Matthews et al., 1996). This suggests that they had better-developed inhibition. The results of this study are in favor of the proposal that development of the brain structures that mediate performance in the A not B task may be strongly influenced by postnatal experience. A further enhancing factor may be the rich environment of an infant exposed to two languages from birth. Possibly, the continuous monitoring of a bimodal linguistic input results in an early boost of attentional control and inhibition already in infancy. In a series of studies we asked whether such enhancements might arise in crib bilinguals prior to language production (Kovács & Mehler, 2009a, 2009b). We tested this possibility by measuring the performance of monolingual and bilingual infants with an eye tracker on tasks that require executive control (Kovács & Mehler, 2009a). If monitoring bilingual input boosts executive control early on, bilingual infants should outperform monolingual infants on such tasks. Additionally, we coupled this investigation with the question of how general such an advantage may be. Crib bilingualism may initially lead to improved control processes only in the language domain because bilingual infants have experience in dealing with conflicting input in language. Alternatively, crib bilingualism may result in a domain general advantage in executive control abilities. We tested monolingual and bilingual 7-month-old infants on a response-switching task where a previously valid and repeated response (an eye movement to location A



Cognitive Effects of Bilingualism in Infancy 

 255

after a cue) had to be inhibited to perform a new response (an eye movement to location B). We hypothesized that both groups would learn the first response equally well (an eye movement to location A) because no control abilities were involved in this phase. However, if bilingual infants had better executive control functions, they would outperform monolingual infants when learning the second response (an eye movement to location B) for which they had to inhibit the earlier learned response (an eye movement to location A). The participants were infants with parents who addressed them in their respective native languages and with daily exposure to two languages (minimum 35% to each language). The majority of bilingual infants heard Italian and Slovenian from their parents, whereas the rest heard Italian and either Spanish, English, Arabic, Danish, French, or Russian. All monolingual infants heard only Italian in their families. Infants came from upper-middle-class families and were matched for their parents’ socioeconomic status and family size (number of siblings). Participants were recruited from Trieste (Italy), where bilingualism has historic roots and has been present for generations. The study consisted of a preswitch and a postswitch phase (see Figure 13.1A). In the preswitch phase, infants were presented with nine trials where a trisyllabic nonce word was followed by a visual reward always appearing on the same side of the screen one second after the cue. The word-cue was composed from three different syllables in Experiment 1 and from syllables with an AAB or ABB structure in Experiment 2 (e.g., le-le-mo or le-mo-mo, where As and Bs stand for a syllable; Kovács & Mehler, 2009a). Thus, infants had to learn that the words predicted the appearance of the rewards in a certain location. In the postswitch phase, infants were exposed to an additional nine trials with the words now indicating that the rewards would appear on the other side of the screen. To see the reward object, infants thus had to learn to look to the opposite side of the screen. We measured learning by recording the infants’ anticipatory looks for the visual reward with an eye tracker. In Experiment 3 we used visual cue sequences instead of linguistic stimuli. Data from Experiment 1 and 2 suggest that bilingual infants are more efficient than monolingual infants in inhibiting a previously learned regularity that involves a contingency between a structured or a random linguistic stimulus and the location of a visual stimulus (e.g., look right after a word; see Figure 1B after Kovács & Mehler, 2009a). Thus, they succeed in quickly learning a second regularity that involves a new pairing. Although both groups of infants learned to correctly anticipate the reward in the preswitch phase, only bilingual infants showed learning in the postswitch phase. Next, we asked how general this advantage in executive control might be and whether it would apply to stimuli from domains other than language by using visual stimuli. If a boost of executive functions extends beyond the language domain from a very early age, bilingual infants should also perform better than monolingual infants on a switching task that involves solely visual stimuli. In contrast, if the advantage is restricted to situations where language is involved, the two groups should perform similarly on the nonlinguistic task. Experiment 3 was structurally similar to the one

Fig. 13.1: Inhibiting a previously learned response in 7-month-old monolingual and bilingual infants. Trial structure (A) and results of experiments 2 and 3 (B and C) are shown. Proportion of infants with correct anticipatory looks. From “Cognitive Gains in 7-Month-Old Bilingual Infants,” by Á. M. Kovács and J. Mehler, 2009, Proceedings of the National Academy of Sciences of the United States of America, 106, pp. 6557–6558. Copyright 2009 by Á. M. Kovács and J. Mehler. Adapted with permission.



Cognitive Effects of Bilingualism in Infancy 

 257

described earlier, except that we used visual sequences as cues. These sequences followed the same regularities as the linguistic stimuli; that is, they had identical geometrical shapes at the beginning of the sequence, such as in AAB, or at the end of the sequence, such as in ABB, where As and Bs stand for different shapes. The findings were similar to the previous results, even when using solely visual stimuli: only bilingual infants showed successful learning over the trials of the postswitch phase, whereas both groups learned in the preswitch phase. These data show that monolingual infants have difficulty overcoming a welllearned response, a finding that fits well with previous results showing that 7-montholds display difficulties in inhibiting previously rewarded responses because of their immature EF (Diamond, 1985). In our study, however, bilingual infants significantly decreased their perseverative responses and increased anticipations to the new location. This suggests that a multilingual environment improves aspects of EF even in preverbal infants. Whereas monolingual and bilingual infants learned equally well that a speech or visual cue predicted the position of a visual reward in the preswitch phase of each experiment, we observed a significant behavioral difference between the two groups in the respective postswitch phases. Bilingual infants readily suppressed the previously learned response and updated their predictions according to the changing requirements of the task, whereas monolingual infants did not learn to correctly modify their responses during the trials of the postswitch phase. The bilingual infants’ enhanced performance cannot be attributed to a systematic difference in general information processing abilities because the performance of the two groups was comparable during the preswitch phases of the experiments. Taken together, the results suggest that perceiving and processing utterances from two languages during the first months of life improves domain-general components of EF well before language production begins. Hence, although suppression of one language when speaking the other is well attested, it is not necessarily required for an EF improvement. Solely processing two languages and having to deal with the representations of each of them seems sufficient for enhancing executive control in 7-month-old infants. Their well-developed EF abilities may help bilingual infants to successfully monitor and keep separate the linguistic representations of the two languages, possibly allowing them to efficiently acquire each language.

Bilingualism Leads to Flexibility in Learning Two Structural Regularities in Infancy How young children manage to find the crucial regularities in the vast linguistic signal they hear in their environment is still unclear. This problem becomes particularly salient for infants born into bilingual families. In a set of studies (Kovács & Mehler, 2009b), we aimed to explore how monolingual and bilingual infants learn

258 

 Ágnes Melinda Kovács

and generalize repetition-based regularities implemented in speechlike stimuli when they are exposed to two kinds of structures simultaneously. Such simultaneous learning might involve executive control and attentional abilities to a great extent, because infants have to concomitantly monitor and construct representations for each of the two regularities. Thus, if bilingual infants develop better executive functions, they should also outperform monolingual infants in this task. In two experiments we investigated how 12-month-old monolingual and bilingual infants extract the underlying structure from an ambiguous speech input using an eye-tracker. Previous research suggested that infants are able to generalize repetitionbased regularities when trained with speechlike stimuli following a single pattern (e.g., AAB; Marcus et al., 1999). In contrast, we familiarized infants with two structures simultaneously. To facilitate learning in this interleaved task, we used two structures that are easily distinguishable. For instance, in two experiments we used adjacent repetition-based structures (AAB) and nonadjacent repetition-based ones (ABA; Kovács & Mehler, 2009b), whereas in other experiments we used adjacent repetitionbased structures (AA) and diversity-based patterns (AB patterns that contain different syllables; Kovács, 2014a). In a typical trial of the experiments of Kovács and Mehler (2009b), the infant was presented with a nonce word conforming to a specific structure (e.g., AAB, such as zo-zo-mo) followed by a visual reward (e.g., a colorful toy) in a specific location (e.g., left). In a different trial, we presented another nonce word conforming to the other structure (e.g., ABA, such as zo-mo-zo) followed by a visual reward on the opposite side (e.g., right). The familiarization consisted of 36 such ABA and AAB intermixed trials, followed by eight test trials, where infants heard new words composed of new syllables and no reward appeared (Figure 13.2 A & B). Thus, after repeated trials infants could learn that the structure of the word predicted the location where the reward would appear, and possibly they could also develop expectations about new exemplars of linguistic stimuli that followed the same structure. We asked whether bilingual infants would outperform their monolingual peers in learning two repetition-based regularities simultaneously by measuring their anticipatory looks in the test trials. Our principal measure to assess learning was infants’ first look after hearing a new speech item in the test. If they learned the structures, they should first search for the toy where it used to appear for that specific structure. Second, we measured infants’ overall accuracy in looking to the correct side. That is, trials were scored as correct if the infant looked longer to the correct side within the two seconds after hearing a new item and before the start of the next trial. We selected the participants and determined their language status as in Kovács and Mehler (2009a). Results are shown in Figure 13.2C. Bilingual 12-month-old infants looked more often to the correct side for both structures during the test trials as shown by the first look analysis and by the overall accuracy analysis (Figure 13.2C). Monolingual infants, in contrast, looked more often to the correct side when the speech items had

Cognitive Effects of Bilingualism in Infancy 

Bilingual Infants

(Exp1)

Monolingual Infants

(Exp1)

Monolingual Infants

(Exp 2)

Difference Score

Difference Score



Bilingual Infants

(Exp1)

Monolingual Infants

(Exp1)

 259

Monolingual Infants

(Exp 2)

Fig. 13.2: Learning multiple rules simultaneously. Familiarization phase of experiment 1 (A); test phase of experiment 1 (B). On the right, two scan paths of an infant are depicted on two trials. Measures of learning for the two structures or the two voice cues (C). On the left are difference scores for first looks ([number of correct looks - number of incorrect looks]/[number of correct looks + number of incorrect looks]) related to the chance level of 0 for ABA and AAB structures (experiment 1); and for male and female voices (experiment 2). On the right are difference scores for overall accuracy. Error bars represent SE. From “Flexible Learning of Multiple Speech Structures in Bilingual Infants,” by A. M. Kovács and J. Mehler, 2009, Science, 325, p. 611. Copyright 2009 by A. M. Kovács and J. Mehler. Reprinted with permission.

the structure AAB, but not when they had the structure ABA. When faced with two regularities, both consisting of well-defined structures (AAB and ABA), monolingual 12-months-olds generalized only the AAB structure. However, they failed to learn the nonadjacent repetitions. Keeping in mind two regularities simultaneously might be too difficult for infants and they might focus on only one regularity and disregard the other. These results also suggest that close and distant identity relations may involve different processing demands. Seemingly, adjacent repetitions are easier or more salient than nonadjacent repetitions. The asymmetry we found is in line with recent neuroimaging studies showing that even newborns can detect adjacent repetitions, but they fail with nonadjacent ones (Gervain et al., 2008). In contrast, preverbal bilingual 12-month-olds learned both structures and seemed to be more efficient in learning two regularities simultaneously than their monolingual peers.

260 

 Ágnes Melinda Kovács

Next, we asked whether exposure to rhythmically similar or dissimilar languages leads to a different performance in simultaneously learning two regularities. When two languages share rhythmic properties, bilingual infants might have a more difficult task because they must find other cues to discriminate them. Thus, we analyzed the bilingual infants’ performance as a function of the similarity of the languages they learned at home, that is, whether they were exposed to two languages from different rhythmic classes (e.g., Slovenian and Italian), or two languages belonging to the same rhythmic class (e.g., Italian and Spanish). However, we found no main effect of language similarity and no other effects or interactions. Further studies involving larger samples are needed to explore how language similarity may influence learning in bilingual infants. To control for the possibility that the bilingual infants’ advantage reflects just a better learning of contingencies between sounds and locations, rather than their ability to learn multiple structural regularities, we ran an additional experiment involving only monolingual infants. In Experiment 2, a new group of monolingual 12-month-olds heard speech items that differed not only in their structure, but also in their pitch (e.g., female voice for ABA and male for AAB). Infants successfully learned to predict the toy locations based on the voices as shown by the first look analysis and the overall accuracy analysis (Figure 13.2C). Together, the results of these experiments show that, in contrast to bilingual infants, monolingual infants cannot extract two structures, although they can associate two speakers to different locations. Thus, these data suggest that preverbal 12-month-old bilingual infants are more efficient in learning two regularities simultaneously than their monolingual peers. In a situation where infants had the opportunity to learn two mutually inconsistent regularities, bilingual infants learned both, whereas monolingual infants learned only one. This advantage is likely related to a precocious development of control and selection abilities, which we have discussed in the previous parts (Kovács & Mehler, 2009a), and which has also been documented in bilingual preschoolers and adults (Bialystok, Craik, et al., 2005; see also Chapters 14–16, this volume). Such abilities may allow bilingual speakers to deal more efficiently with two conflicting representations in the domain of language. In the studies previously described, we investigated how different domainspecific and domain-general processes interact in the service of language acqui­ sition, and how these are shaped by early bilingual exposure. According to the findings, bilingual language learning seems to involve domain-general EF abilities, even at an age when infants do not yet produce words. The practice of using such abilities during language acquisition results in their enhancement even outside the domain of language. Improved EF will, in turn, be used to deal more efficiently with conflicting linguistic representations and possibly also with nonlinguistic ones. In addition to revealing differences in EF between monolingual and bilingual infants, these results also speak to the long-standing debate whether bilingual infants



Cognitive Effects of Bilingualism in Infancy 

 261

start out constructing only one language system comprising both of the languages they acquire, or whether they represent their two languages separately from the start. Our results seem to support the differentiated language system hypothesis. We conjecture that the EF enhancement observed in bilingual 7-month-olds (Kovács & Mehler, 2009a) is possible only if preverbal infants process the two languages distinctively. Only if infants can represent the two languages differently, will they be able to selectively attend to them. Yet, one might argue that an EF enhancement could arise even if infants do not separate the two languages at the age of 7 months; if so, these results would be consistent with the unitary language system hypothesis. For example, learning from a more complex linguistic input might require more attentional resources as the input is more variable. Bilingual infants would thus have to deploy greater attentional control during language acquisition. Because attentional processes are part of the EF system, such practice might also enhance the development of EF abilities, even if bilingual infants have not yet succeeded to separate the two languages. However, it is unlikely that bilingual infants are unable to separate the two languages they are exposed to by their seventh month. Previous evidence shows that infants discriminate rhythmically different languages at birth, and prosodically similar languages by their fourth month (Bosch & Sebastián-Gallés, 1997; Nazzi et al., 1998). Thus, infants might use different cues early on (e.g., prosodic, phonetic, or phonotactic differences) to separate the languages; this might allow them to selectively attend to the two languages and construct different systems. Also, the finding that bilingual infants could simultaneously extract two regularities (Kovács & Mehler, 2009b) is not consistent with the predictions of a unitary language system hypothesis either. This hypothesis would assume that bilingual infants should construct a single system also when confronted with artificial speechlike stimuli that contain two regularities. However, the 12-month-old bilingual infants in our study extracted and generalized simultaneously two structural regularities. In contrast, monolingual infants systematically learned only one regularity from the artificial stream, irrespective of whether the signal contained two structures or a structure and a random pattern (Kovács, 2014a; Kovács & Mehler, 2009b). In other words, monolingual infants managed to extract only one structure and presumably considered contrary evidence (e.g., the other structure) as noise. Such findings open the field for a further conjecture. Infants exposed to a monolingual language input, which contains a well-defined system of regularities, might be tuned to search for a single consistent pattern in speechlike stimuli. This might explain why they might fail to simultaneously learn multiple patterns. Thus, early on, monolingual infants might expect that the speech input they hear has a single coherent system of regularities. This expectation, of course, will be modified by experience. As described earlier, infants exposed to two languages demonstrated the ability to learn two conflicting regularities simultaneously. Thus, bilingual infants presumably allow multiple conflicting sets of regularities in speechlike input.

262 

 Ágnes Melinda Kovács

It is possible that infants start the task of language acquisition with some expectations about the consistency of the linguistic signal. This expectancy will undergo an exposure-dependent specialization, in the sense that it will allow a single set of regularities or multiple ones depending on whether infants are exposed to one language or two. Developing such an expectancy would help rapid language acquisition. It would also diminish the amount of cognitive resources allocated to language learning, because it would permit considering as noise any evidence that does not exactly fit with the current “conjectures” of the infant learner. Such a possibility remains hypothetical until further studies provide support for the framework that infants have specific expectations about the coherence of the linguistic input.

Further Cognitive Effects of Early Bilingualism Although the studies previously discussed suggest that there are specific enhancements in how bilingual infants deal with learning new regularities after having already learned one and in extracting two rules simultaneously, recent studies suggest that there might be other changes in the cognitive system that seem to be triggered by bilingual input from very early on in childhood. Such changes might, for instance, result in a boost in visual language discrimination (Sebastián-Gallés, Albareda-Castellot, Weikum, & Werker, 2012), in different word learning strategies (Byers-Heinlein & Werker, 2009, 2013) and habituation patterns (Singh et al., 2015), or in enhanced memory flexibility (Brito & Barr, 2012). Research by Sebastián-Gallés et al. (2012) suggests, for instance, that bilingual 8-month-olds can visually discriminate two languages they are not familiar with, whereas monolingual infants cannot. This study investigated whether bilingual exposure enhances the attentional system because of the need to pay attention to the distinctive perceptual information in the language input. More specifically, two alternative hypotheses were tested. Earlier studies found that whereas monolingual 8-month-olds could not discriminate the visual features of their native language and a new language, bilingual infants of the same age succeeded in the discrimination of their two native languages (Weikum et al., 2007). Sebastián-Gallés et al. asked whether such a bilingual advantage is because of perceptual narrowing to the exact languages of exposure or to increased perceptual attentiveness to visual languagerelated cues even for unfamiliar languages. Bilingual infants could discriminate visual speech even for two foreign languages, a finding that supports the greater perceptual attentiveness hypothesis (see also Chapter 3, this volume). However, it is still unclear how processes of perceptual narrowing and perceptual attentiveness interact in bilingual language acquisition. Further changes in cognitive processing in bilingual infants are also observed in the domain of word learning. Early work targeting monolingual infants suggests that they use specific disambiguation heuristics to identify referents for words. For



Cognitive Effects of Bilingualism in Infancy 

 263

instance, if they encounter a new label (e.g., dax) together with a familiar object for which they already have a lexical referent (e.g., shoe) and a new object, they will infer that the new word’s referent must be the new object. This is also referred to as the mutual exclusivity principle, according to which infants assume that one object has only one label (Markman & Wachtel, 1988). Although monolingual infants can successfully use mutual exclusivity to learn new labels for new referents, infants exposed to bilingual or multilingual input are often exposed to lexical equivalents in two languages. Thus, in their case, the one object–one label strategy does not seem to be easily applicable. Experimental data suggest that multilingual children seem to apply the principle of mutual exclusivity less than monolingual children (Byers-Heinlein & Werker, 2009; Houston-Price, Caloghiris, & Raviglione, 2010; see also Chapter 4, this volume). However, it is not clear whether this is because of the fact that they accept that an object can have multiple verbal referents. Recent studies provide evidence in this direction, suggesting that infants exposed to multiple languages seem to accept that one object can have multiple referents more frequently than monolingual infants in a synonym task (Kovács, 2014b). Furthermore, Byers-Heinlein and Werker (2013) investigated whether bilingual infants’ use of mutual exclusivity as a disambiguation strategy is related to the extent to which their lexicons had a one-to-one versus a many-to-one mapping structure, as operationalized by their knowledge of translation equivalents in their two languages. The data show that bilingual infants who understood translation equivalents for more than half the words in their vocabularies applied the principle of mutual exclusivity less compared to those bilingual children who knew fewer translation equivalents. It is not very surprising that multilingual infants do not apply the one object–one label heuristic as they usually encounter two labels for the same object—one in each language. However, as pointed out by Costa and Sebastián-Gallés (2014), it is unclear how bilingual infants compensate for the lack of this principle during word learning, what alternative strategies they use, and whether early vocabulary development is compromised in some way by its absence. A further study investigating other possible advantages of bilingualism in infancy found that bilingual but not monolingual 18-month-old infants showed generalization in a deferred imitation task, pointing to a cognitive advantage in memory generalization for bilingual infants (Brito & Barr, 2012). Bilingual infants showed better memory flexibility, generalizing successfully when a demonstration was performed on object A, while the test involved object B. Although the mechanisms through which a better generalization ability emerges from learning simultaneously two languages are not clear, the study found a relation between the degree of how balanced bilingual children were in their two languages and their generalization performance. Regarding other possible advantages, Singh et al. (2015) found differences in how bilingual and monolingual 6-month-olds habituated to visual stimuli in a visual discrimination task. Similar to the studies mentioned previously, a relation was found here as well between habituation performance and the amount of second language

264 

 Ágnes Melinda Kovács

exposure. Although more and more studies point to specific advantages of bilingual infants in various cognitive domains, it is unlikely that a generalized bilingual advantage will be uncovered by future studies. It is unclear whether the observed advantages persist throughout childhood and adulthood providing overall learning benefits, whether they reach a plateau at some point (e.g., when the two languages were acquired), and why no general learning benefits are usually observed in bilingual children and adults in previous studies.

New Perspectives: Effects of Bilingualism as Cognitive Adaptations One possible way to conceptualize the changes induced by the bilingual environment is to think of them similarly to other adaptations in the cognitive system, for instance, visual adaptations. Wearing prism glasses that lead to an inversion of the left and right or the up and down dimensions certainly lead to adaptation costs. Performance initially decreases followed by an improvement that signals that the visual system adapted to the new input. Analogously, one would expect that when exposed to two languages the developing brain would quickly adapt to processing two mutually exclusive language systems (Kovács, 2015). Behavioral observations, according to which bilingual language acquisition follows the same milestones as the monolingual one despite the more complex input, point to such plasticity. However, the analogy between visual adaptation and a possible bilingual adaptation becomes problematic when one realizes that in contrast to a situation where the prism glasses are removed at the end of a study, bilingual individuals will likely continue using “the prism glass of bilingualism” for the rest of their lives. Importantly, besides having to extract patterns from a complex and variable dual language input, bilingual infants likely need yet another, possibly separate adaptation that allows them to flexibly switch attention between two languages systems, to be able to distinctively acquire both of them. To continue with the prism glasses comparison, one would need to ask participants in a visual adaptation study to put on and off the prism glasses many times a day, and consequently to constantly switch between upright and upside down visual input. Likely, the cognitive system will optimize such switches, and minimize adaptation times. This should also be valid for cases in which the language input contains frequent between-language switches. Bilingual adults change from one language to another rather smoothly and even young bilingual children must be able to successfully deal with these frequent switches. In the earlier parts of this chapter, I discussed different studies that could be reframed as possible adaptations involved in dealing with a multilingual input. Various results pointing to advantages or disadvantages in specific domains might all be seen as cognitive adaptations to a bilingual environment (see Kovács, 2015).



Cognitive Effects of Bilingualism in Infancy 

 265

For instance, better attention switching and simultaneous learning of two regularities (Kovács & Mehler, 2009a, 2009b) and better memory generalization (Brito & Barr, 2012) may allow bilingual speakers to successfully cope with the rapidly changing bilingual input. Additionally, more fine-tuned visual language discrimination abilities observed in bilingual 8-month-olds even for languages they are not familiar with (Sebastián-Gallés et al., 2012) might be an adaptation that stems from greater attention to nonfamiliar languages and leads to more efficient language learning in the case of scarcer input. Furthermore, even studies that highlight “costs” of bilingualism may point to possible adaptations. For instance, in a study by Fennell, Byers-Heinlein, and Werker (2007), bilingual infants only learned word–object associations with nonwords that were minimal pairs (e.g., bih/dih) at 20 months, lagging behind monolingual infants by about 3 months. One could argue that a later emerging sensitivity to minimal pairs in learning word–object associations might be explained by a flexibility of bilingual infants in forming broader phonological categories. In line with an earlier proposal, these results may reflect an adaptive strategy to learn two languages (SebastiánGallés, 2010; see also Chapters 3 and 4, this volume). In a similar vein, bilingual speakers’ smaller vocabulary when only one language is measured may be explained with a possible adaptation that ensures a “fair” division of the possibly limited cognitive resources between the two languages. Interestingly, however, when bilingual speakers’ vocabulary is measured taking together both languages, their cumulative scores are equal to or higher than those of monolingual speakers, pointing to a possibly more general memory enhancement (as documented in some studies; Brito & Barr, 2012). In sum, early exposure to more than one language seems to lead to specific adaptations in the cognitive system, which will influence, in turn, how language is acquired, and might result in changes in other domains as well. However, these presumably do not imply that bilingualism leads to radical representational changes in the human mind. Instead, they indicate that the cognitive system of a young child is ready to successfully deal with the challenge coming from multiple languages.

References Bialystok, E. (1999). Cognitive complexity and attentional control in the bilingual mind. Child Development, 70, 636–644. http://dx.doi.org/10.1111/1467-8624.00046 Bialystok, E., Craik, F. I. M., Grady, C., Chau, W., Ishii, R., Gunji, A., & Pantev, C. (2005). Effect of bilingualism on cognitive control in the Simon task: Evidence from MEG. NeuroImage, 24, 40–49. http://dx.doi.org/10.1016/j.neuroimage.2004.09.044 Bialystok, E., Martin, M. M., & Viswanathan, M. (2005). Bilingualism across the lifespan: The rise and fall of inhibitory control. The International Journal of Bilingualism, 9, 103–119. http://dx.doi.org/ 10.1177/13670069050090010701

266 

 Ágnes Melinda Kovács

Bosch, L., & Sebastián-Gallés, N. (1997). Native-language recognition abilities in 4-month-old infants from monolingual and bilingual environments. Cognition, 65, 33–69. http://dx.doi.org/ 10.1016/S0010-0277(97)00040-1 Bosch, L., & Sebastián-Gallés, N. (2001). Evidence of early language discrimination abilities in infants from bilingual environments. Infancy, 2, 29–49. http://dx.doi.org/10.1207/S15327078IN0201_3 Brito, N., & Barr, R. (2012). Influence of bilingualism on memory generalization during infancy. Developmental Science, 15, 812–816. http://dx.doi.org/10.1111/j.1467-7687.2012.1184.x Burgess, P. W., Veitch, E., de Lacy Costello, A., & Shallice, T. (2000). The cognitive and neuroanatomical correlates of multitasking. Neuropsychologia, 38, 848–863. http://dx.doi.org/10.1016/ S0028-3932(99)00134-7 Byers-Heinlein, K., & Werker, J. F. (2009). Monolingual, bilingual, trilingual: Infants’ language experience influences the development of a word-learning heuristic. Developmental Science, 12, 815–823. http://dx.doi.org/10.1111/j.1467-7687.2009.00902.x Byers-Heinlein, K., & Werker, J. F. (2013). Lexicon structure and the disambiguation of novel words: Evidence from bilingual infants. Cognition, 128, 407–416. http://dx.doi.org/10.1016/ j.cognition.2013.05.010 Casey, B. J., Trainor, R. J., Orendi, J. L., Schubert, A. B., Nystrom, L. E., Giedd, J. N., . . . Rapoport, J. L. (1997). A developmental functional MRI study of prefrontal activation during performance of a go–no-go task. Journal of Cognitive Neuroscience, 9, 835–847. http://dx.doi.org/10.1162/ jocn.1997.9.6.835 Christophe, A., & Morton, J. (1998). Is Dutch native English? Linguistic analysis by 2-month-olds. Developmental Science, 1, 215–219. http://dx.doi.org/10.1111/1467-7687.00033 Costa, A., Hernández, M., & Sebastián-Gallés, N. (2008). Bilingualism aids conflict resolution: Evidence from the ANT task. Cognition, 106, 59–86. http://dx.doi.org/10.1016/j.cognition. 2006.12.013 Costa, A., & Sebastián-Gallés, N. (2014). How does the bilingual experience sculpt the brain? Nature Reviews Neuroscience, 15, 336–345. http://dx.doi.org/10.1038/nrn3709 Diamond, A. (1985). Development of the ability to use recall to guide action, as indicated by infants’ performance on AB. Child Development, 56, 868–883. http://dx.doi.org/10.2307/1130099 Duñabeitia, J. A., Hernández, J. A., Antón, E., Macizo, P., Estévez, A., Fuentes, L. J., & Carreiras, M. (2014). The inhibitory advantage in bilingual children revisited: Myth or reality? Experimental Psychology, 61, 234–251. http://dx.doi.org/10.1027/1618-3169/a000243 Fennell, C. T., Byers-Heinlein, K., & Werker, J. F. (2007). Using speech sounds to guide word learning: The case of bilingual infants. Child Development, 78, 1510–1525. http://dx.doi.org/10.1111/ j.1467-8624.2007.01080.x García-Pentón, L., Pérez Fernández, A., Iturria-Medina, Y., Gillon-Dowens, M., & Carreiras, M. (2014). Anatomical connectivity changes in the bilingual brain. NeuroImage, 84, 495–504. http:// dx.doi.org/10.1016/j.neuroimage.2013.08.064 Genesee, F., Nicoladis, E., & Paradis, J. (1995). Language differentiation in early bilingual development. Journal of Child Language, 22, 611–631. http://dx.doi.org/10.1017/S0305000900009971 Gerstadt, C. L., Hong, Y. J., & Diamond, A. (1994). The relationship between cognition and action: Performance of children 3 1/2–7 years old on a Stroop-like day–night test. Cognition, 53, 129–153. http://dx.doi.org/10.1016/0010-0277(94)90068-X Gervain, J., Macagno, F., Cogoi, S., Peña, M., & Mehler, J. (2008). The neonate brain detects speech structure. Proceedings of the National Academy of Sciences of the United States of America, 105, 14222–14227. http://dx.doi.org/10.1073/pnas.0806530105



Cognitive Effects of Bilingualism in Infancy 

 267

Green, D. W. (1998). Mental control of the bilingual lexico-semantic system. Bilingualism: Language and Cognition, 1, 67–81. http://dx.doi.org/10.1017/S1366728998000133 Houston-Price, C., Caloghiris, Z., & Raviglione, E. (2010). Language experience shapes the development of the mutual exclusivity bias. Infancy, 15, 125–150. http://dx.doi.org/10.1111/ j.1532-7078.2009.00009.x Kovács, Á. M. (2014a). Extracting regularities from noise: Do infants encode patterns based on same and different relations? Language Learning, 64, 65–85. http://dx.doi.org/10.1111/lang.12056 Kovács, Á. M. (2014b, July). Flexible learning of words and rules in bilingual and monolingual infants. Paper presented at the International Language Production Workshop, Geneva, Switzerland. Kovács, Á. M. (2015). Cognitive adaptations induced by a multi-language input in early development. Current Opinion in Neurobiology, 35, 80–86. http://dx.doi.org/10.1016/j.conb.2015.07.003 Kovács, Á. M., & Mehler, J. (2009a). Cognitive gains in 7-month-old bilingual infants. Proceedings of the National Academy of Sciences of the United States of America, 106, 6556–6560. http://dx.doi.org/ 10.1073/pnas.0811323106 Kovács, Á. M., & Mehler, J. (2009b). Flexible learning of multiple speech structures in bilingual infants. Science, 325, 611–612. http://dx.doi.org/10.1126/science.1173947 Leopold, W. (1978). A child’s learning of two languages. In E. Hatch (Ed.), Second language acquisition (pp. 23–32). Rowley, MA: Newbury House. Marcus, G. F., Vijayan, S., Bandi Rao, S., & Vishton, P. M. (1999). Rule learning by seven-monthold infants. Science, 283, 77–80. http://dx.doi.org/10.1126/science.283.5398.77 Markman, E. M., & Wachtel, G. F. (1988). Children’s use of mutual exclusivity to constrain the meanings of words. Cognitive Psychology, 20, 121–157. http://dx.doi.org/10.1016/ 0010-0285(88)90017-5 Matthews, A., Ellis, A. E., & Nelson, C. A. (1996). Development of preterm and full-term infant ability on AB, recall memory, transparent barrier detour, and means-end tasks. Child Development, 67, 2658–2676. http://dx.doi.org/10.2307/1131745 Mechelli, A., Crinion, J. T., Noppeney, U., O’Doherty, J., Ashburner, J., Frackowiak, R. S., & Price, C. J. (2004). Neurolinguistics: Structural plasticity in the bilingual brain. Nature, 431, 757. http:// dx.doi.org/10.1038/431757a Nazzi, T., Bertoncini, J., & Mehler, J. (1998). Language discrimination by newborns: Toward an understanding of the role of rhythm. Journal of Experimental Psychology: Human Perception and Performance, 24, 756–766. http://dx.doi.org/10.1037/0096-1523.24.3.756 Pearson, B. Z., Fernández, S. C., & Oller, D. K. (1993). Lexical development in bilingual infants and toddlers, comparison to monolingual norms. Language Learning, 43, 93–120. http://dx.doi.org/ 10.1111/j.1467-1770.1993.tb00174.x Pearson, B. Z., Fernández, S., & Oller, D. K. (1995). Cross-language synonyms in the lexicons of bilingual infants: One language or two? Journal of Child Language, 22, 345–368. http://dx.doi.org/10.1017/ S030500090000982X Petitto, L. A. (1987). On the autonomy of language and gesture: Evidence from the acquisition of personal pronouns in American Sign Language. Cognition, 27, 1–52. http://dx.doi.org/10.1016/ 0010-0277(87)90034-5 Petitto, L. A., Katerelos, M., Levy, B. G., Gauna, K., Tétreault, K., & Ferraro, V. (2001). Bilingual signed and spoken language acquisition from birth: Implications for the mechanisms underlying early bilingual language acquisition. Journal of Child Language, 28, 453–496. http://dx.doi.org/ 10.1017/S0305000901004718

268 

 Ágnes Melinda Kovács

Poulin-Dubois, D., Blaye, A., Coutya, J., & Bialystok, E. (2011). The effects of bilingualism on toddlers’ executive functioning. Journal of Experimental Child Psychology, 108, 567–579. http://dx.doi.org/ 10.1016/j.jecp.2010.10.009 Ramus, F., Hauser, M. D., Miller, C., Morris, D., & Mehler, J. (2000). Language discrimination by human newborns and by cotton-top tamarin monkeys. Science, 288, 349–351. http://dx.doi.org/10.1126/ science.288.5464.349 Ramus, F., & Mehler, J. (1999). Language identification with suprasegmental cues: A study based on speech resynthesis. Journal of the Acoustical Society of America, 105, 512–521. http://dx.doi.org/ 10.1121/1.424522 Sebastián-Gallés, N. (2010). Bilingual language acquisition: Where does the difference lie? Human Development, 53, 245–255. http://dx.doi.org/10.1159/000321282 Sebastián-Gallés, N., Albareda-Castellot, B., Weikum, W. M., & Werker, J. F. (2012). A bilingual advantage in visual language discrimination in infancy. Psychological Science, 23, 994–999. http:// dx.doi.org/10.1177/0956797612436817 Sebastián-Gallés, N., & Bosch, L. (2002). Building phonotactic knowledge in bilinguals: Role of early exposure. Journal of Experimental Psychology: Human Perception and Performance, 28, 974–989. http://dx.doi.org/10.1037/0096-1523.28.4.974 Singh, L., Fu, C. S., Rahman, A. A., Hameed, W. B., Sanmugam, S., Agarwal, P., . . . & the GUSTO Research Team. (2015). Back to basics: A bilingual advantage in infant visual habituation. Child Development, 86, 294–302. http://dx.doi.org/10.1111/cdev.12271 Vihman, M. M., Macken, M. A., Miller, R., Simmons, H., & Miller, J. (1985). From babbling to speech. Language, 61, 397–443. http://dx.doi.org/10.2307/414151 Vihman, M. M., & McCune, L. (1994). When is a word a word? Journal of Child Language, 21, 517–542. Volterra, V., & Taeschner, T. (1978). The acquisition and development of language by bilingual children. Journal of Child Language, 5, 311–326. http://dx.doi.org/10.1017/S0305000900007492 Weikum, W. M., Vouloumanos, A., Navarra, J., Soto-Faraco, S., Sebastián-Gallés, N., & Werker, J. F. (2007). Visual language discrimination in infancy. Science, 316, 1159. http://dx.doi.org/ 10.1126/science.1137686

Elena Nicoladis

14 Bilingual Speakers’ Cognitive Development in Childhood Children go through enormous changes in how they think and speak from infancy to adolescence. From not speaking at all, they develop a concept of self (Lewis & Ramsay, 2004), learn how to take the perspective of others (Sodian & Kristen, 2010), how to count and compare quantities (Lecorre, Van de Walle, Brannon, & Carey, 2006), among many other abilities. Within the preschool years, children become able to use complex grammatical constructions (Guasti, Stavrakaki, & Arosio, 2012), tell lies (Wilson, Smith, & Ross, 2003), and tell stories about their own lives (Nelson & Fivush, 2004) as well as fictional stories (Li & Tan, 2015). Some of these developments start earlier than childhood. Even infants, for example, can demonstrate some ability to remember and selectively attend to figures in their environment (Fivush, 2011; see also Chapter 13, this volume). During childhood, memory abilities continue to develop in capacity, flexibility, and speed (Kail & Ferrer, 2007). Similarly, the ability to selectively attend to the problem at hand and inhibit attention to distractors continues to develop through childhood (Zelazo et al., 2003). Researchers have argued that bilingual children might have some experiences that impact their cognitive development. For example, the experience of knowing two languages could lead to differences in thinking relative to knowing just one language. Every language, through semantics, grammar, morphology, phonology, and usage, encodes a way of thinking (see Lucy, 2011). To speak a particular language, a speaker must highlight some aspects of events and downplay others, a phenomenon Slobin (1996) called thinking for speaking. Bilingual speakers have access to two different ways of thinking, through each of their languages. They may be more flexible in their thinking (Hakuta & Diaz, 1985). For example, bilingual children might be more creative than monolingual children in generating ideas of what to do with an object (Ben-Zeev, 1977; Peal & Lambert, 1962). Bilingual children might therefore be more flexible in their thinking and creative than their monolingual counterparts. Another possible ramification from being able to think in two languages is that bilingual children might have to store information in such a way that they could speak about it in either language. They might therefore remember information more in terms of the visuospatial characteristics than monolingual children. Another experience that bilingual children have is choosing the appropriate language for the context (see Chapter 2, this volume). Bilingual children have to learn that not all people speak the same language and that two different words can have the same referent (e.g., perro and dog both refer to the same animal). These experiences http://dx.doi.org/10.1037/14939-015 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

270 

 Elena Nicoladis

could have several impacts on bilingual children’s cognitive development. Switching between languages could lead bilingual children to earlier development of executive functions, particularly selective attention to what is relevant to a particular context (Barac, Bialystok, Castro, & Sanchez, 2014; see also Chapter 13, this volume). Attending to the social context could lead bilingual children to earlier development of an understanding of other people’s minds or a better ability to attend to nonverbal communication. In this chapter, I review the evidence for bilingual children’s cognitive develop­ment differing from that of monolingual children in terms of cognitive flexibility, visuospatial abilities, executive functions, and social cognition. I will show that researchers’ predictions of bilingual advantages in cognitive development sometimes but not always hold true. I will discuss some possible reasons for why research has had such varied results. Before turning to the research on cognitive development, it is important to review the assumptions on which differences between bilingual speakers and monolingual speakers are based. Researchers have generally assumed that bilingual speakers often have to choose different languages for different contexts, leading to differences in executive functions and social cognition. There is now extensive evidence that bilingual children can choose the appropriate language for the context from early in development (Chapter 2, this volume). Another assumption among researchers has been that bilingual children have access to two different ways of thinking, corresponding to each language. This difference could lead to enhanced cognitive flexibility and visuospatial encoding among bilingual speakers relative to monolingual speakers. I first review the evidence that bilingual children think differently when speaking each of their languages.

Do Bilingual Children Think Differently When Speaking Each of Their Languages? Using different languages for different functions could imply that bilingual speakers think differently to speak each language. Bilingual children often use their two languages in different contexts to accomplish different goals (Chapter 2, this volume). There is some evidence that bilingual children may be able to link particular languages to particular functions from early in development. Vihman (1998) reported that two Estonian–English bilingual children used English when engaging in dramatic play with each other (e.g., in playing store) and Estonian to negotiate how to engage in that play. For these children who were growing up in the United States, Estonian was the language spoken at home and their stronger language. The children associated each of their languages with different social functions, related to how they heard the two languages, such as English in stores and Estonian at home to negotiate the next activity. Bilingual speakers sometimes speak two languages that differ in how one conceptualizes an event. When talking about motion (like a man rolling down a hill), different



Bilingual Speakers’ Cognitive Development in Childhood 

 271

languages encode different aspects of the motion. When describing a moving figure, in some languages, like English or Chinese, the main verb is often how the movement took place (as in rolling) and the path or direction of the movement is in another part of the phrase (as in down; Slobin, 1996). In Turkish, Spanish, or French, the main verb often encodes path (as in descending) whereas manner is only expressed optionally. Bilingual adults who speak languages that differ on this typological dimension tend to show differences in how they describe motion events in their two languages. Hohenstein, Eisenberg, and Naigles (2006) found that in describing moving figures Spanish–English bilingual adults used more manner verbs in English (like rolling) and more path verbs in Spanish (like salir [to go out]). Bilingual children show the same tendency to describe motion events differently in each of their languages. Evidence comes from a study with French–English bilingual children between 5 and 10 years of age (Miller, Furman, & Nicoladis, 2015). The children watched a cartoon and told the story back. When describing how the character moved in English, the bilingual children used more different manner verbs (e.g., swim, roll, fly) than in French and more different path verbs in French (e.g., descendre [to go down], suivre [to follow], retourner [to return]) than in English. This result suggests that bilingual children, like adults, highlight different aspects of events depending on which language they are speaking. These results are consistent with the argument that bilingual children think differently when speaking each of their languages. Another study showed similar results for a more subtle language preference. In forming possessive constructions, monolingual English speakers avoid periphrastic constructions for proper nouns (like the hat of Julie), preferring constructions with’s (like Julie’s hat; Skarabela & Serratrice, 2009). In contrast, periphrastic possessive constructions with common nouns (like the shoes of the dog) are acceptable to monolingual English speakers. Nicoladis (2012) elicited possessive constructions from French–English bilingual preschoolers, as well as age-matched French and English monolingual children. In French, there is only one way to form a possessive with both proper and common nouns, a periphrastic construction like les chaussures du chien [the shoes of the dog]. In English, the bilingual children showed the same tendency as the monolingual children to avoid periphrastic constructions with proper nouns (Nicoladis, 2012). In French, the bilingual children did not show any differences between proper and common nouns in how they formed possessive constructions. This result is suggestive that when speaking English, both bilingual and monolingual children conceptualize proper nouns differently from common nouns, at least in the context of speaking about their possessions. This distinction does not apply in French and bilingual (and monolingual) children do not highlight the distinction when thinking for speaking in French. Although the studies discussed so far are consistent with the argument that bilingual speakers think differently for each of their languages, there is another possible explanation. The bilingual children might have learned language-appropriate ways of speaking, without necessarily thinking differently. Recall that there is extensive

272 

 Elena Nicoladis

evidence that bilingual children can differentiate their languages from early in development (Chapter 2, this volume). More convincing evidence that bilingual speakers truly think differently for each of their languages would come from studies that included other cognitive measures, such as memory. One such study was with French– English bilingual preschool children describing moving figures, like a cat that was running (Nicoladis, Rose, & Foursha-Stevenson, 2010). In labeling moving figures, there are differences in preferred constructions in English and French. In English, a construction like a dancing cow that highlights the movement is preferred. In French, a construction like une vache qui danse [a cow that dances] is preferred, with the qui danse [that dances] being optional. The French construction highlights the figure itself, rather than the movement. When speaking English, French–English bilingual preschoolers tended to mention the movement more often than they did in French and equally often as English monolingual preschoolers (once vocabulary differences were controlled for). When speaking French, the bilingual children tended to mention the figure itself and omit the movement. Nicoladis et al. (2010) went on to show that French–English bilingual speakers could equally well remember and name the movement performed by the figures, regardless of what language they were tested in. The children simply chose to omit mentioning the movement more often when speaking French than when speaking English. This result shows that bilingual children can remember aspects of the movement that they do not choose to encode when speaking. In other words, they think differently when speaking each of their languages. Thus far, the discussion has focused on how bilingual children think for speaking where their two languages show different tendencies or preferences. What if one language encodes a dimension that the other language does not? There is some evidence that bilingual children’s way of thinking transfers from the language that encodes that dimension to the other language. Consider the case of grammatical gender. Some languages mark every noun for grammatical gender (like French and German), whereas other languages (like English) have no systematic markers of grammatical gender. Grammatical gender is often related to natural gender (so, words for biological females are often grammatically feminine), but often show some arbitrary assignment (e.g., the word for key is masculine in German but feminine in Spanish). By the time monolingual speakers of many gendered languages reach middle childhood (about 8–9 years of age), they show evidence of thinking about even arbitrarily assigned genders as related to natural gender (Sera et al., 2002). For example, when asked to choose a male or female voice for objects that would appear in animated cartoons, Spanish-speaking children prefer a female voice for a spoon (the word for spoon in Spanish is feminine; Sera et al., 2002). As for bilingual speakers, one study showed that French–English bilingual children between 8 and 10 years of age also showed effects of conceptualizing French grammatical gender even when speaking English (Nicoladis & Foursha-Stevenson, 2012). They were asked in English to classify objects (e.g., a spoon or a toy zebra) as either “boys” or “girls.” Their classifications were related to the grammatical gender of those objects in French. This result sug-



Bilingual Speakers’ Cognitive Development in Childhood 

 273

gests that even when speaking English, bilingual children were conceptualizing in ways derived from their French experience (as do adults; see Boroditsky, Schmidt, & Phillips, 2003). In sum, the studies to date suggest that bilingual children generally think differently to speak each of their languages. This is not to say that bilingual speakers never make errors or never show influences of the style of thinking associated with the other language. For example, where one language does not encode a particular structure (like grammatical gender), bilingual children’s conceptualization in that language can be influenced by the other language. Furthermore, there are many reported instances of bilingual children using the words of one language but the grammatical construction of the other. For example, in talking about the gun that Santa Claus had given him, a Cantonese–English bilingual child asked in English, “Where’s the Santa Claus give me the gun?” (Yip & Matthews, 2000). The word order reflects appropriate word order in Cantonese. This cross-linguistic influence almost never represents the majority of what bilingual children say. Nicoladis and Gavrila (2015) showed that Welsh–English bilingual children used the word order for adjective placement from their other language less than 15% of the time, in both languages. As such, crosslinguistic influence may simply be a kind of speech error (Nicoladis, 2006). Most of the time and even from an early age, bilingual children show evidence of thinking for speaking differently in each of their languages.

Do Bilingual Children Show Cognitive Development Differences From Monolingual Children? As shown in the previous section, bilingual children think differently for speaking each of their languages; do they then show any differences in cognitive development from monolingual children? In this section, four domains of cognition are considered: cognitive flexibility, visuospatial abilities, executive functions, and social cognition.

Cognitive Flexibility Because bilingual children have access to two different ways of thinking, they may come to an early understanding that there are multiple ways of conceptualizing objects and events (Ben-Zeev, 1977; Peal & Lambert, 1962). One study showed bilingual advantages in young children’s flexibility in producing drawings of nonexistent things (Adi-Japha, Berberich-Artzi, & Libnawi, 2010). Four- and 5-year-old Hebrew– English and Hebrew–Arabic bilingual children as well as Hebrew monolingual children were asked to draw things (e.g., a flower) that do not exist. The bilingual children were significantly more likely than the monolingual children to add an element to the

274 

 Elena Nicoladis

picture from another semantic category (e.g., a giraffe with the flower). The monolingual children were significantly more likely to simply omit an element (i.e., petal) from the drawing (so they produced a flower with few petals). The authors interpreted these results as showing that bilingual speakers have greater flexibility in their thinking than monolingual speakers. Bilingual children are more likely than monolingual children to accept that words can have more than one meaning (Bialystok, Barac, Blaye, & Poulin-Dubois, 2010; Davidson, Jergovic, Imami, & Theodos, 1997). Bilingual children start showing this flexibility in understanding words as soon as they start to have translation equivalents in their vocabulary (Byers-Heinlein & Werker, 2013). That is, once they understand that a single referent could have more than one name, bilingual children no longer assume that a single word has only one referent. This understanding could lead bilingual children to be more flexible than monolingual children in interpreting novel words (Marinova-Todd, 2012). In the Marinova-Todd (2012) study, French– English bilingual children and English monolingual children in the United States were asked to interpret a nonsense words (like corplum, with the target meaning of a stick) presented in sentences starting with ambiguous contexts (e.g., “A corplum may be used for support”) and becoming more and more concrete (e.g., “The painter used a corplum to mix his paints”). The bilingual children were more willing to guess at the possible meanings of the nonsense words and required fewer concrete presentations to correctly guess the meanings. These results suggest that bilingual children may be willing to interpret words with greater flexibility than monolingual children. This flexibility could help bilingual children in being successful communicators even when they have difficulty accessing particular words. In a recent study, French–English bilingual children as well as English and French monolingual children between 8 and 10 years of age watched a Pink Panther cartoon and told the story back (Hoang, Nicoladis, Smithson, & Furman, in press). The bilingual children did this twice, once in French and once in English. The analyses focused on particular words that were crucial to conveying the important parts of the story. The researchers found that the bilingual children were less likely to produce target words (e.g., panther to refer to the Pink Panther) than monolingual children in both of their languages. Nevertheless, the bilingual children tried to convey more of the important concepts in speech than the monolingual children and chose ways of expressing those concepts that would likely make the story clear for the listener (e.g., calling the panther lion or big cat). These results suggest that bilingual speakers can use their cognitive flexibility in a communicative setting to make their points clear to their listeners. In sum, bilingual speakers do seem to display a greater degree of cognitive flexibility than monolingual speakers. There are as yet not enough studies to know whether these differences generalize to all bilingual speakers, because much of the previous work has been done in Canada with French–English bilingual speakers.



Bilingual Speakers’ Cognitive Development in Childhood 

 275

Visuospatial Tasks Bilingual speakers might be more likely than monolingual speakers to store their memories in visuospatial images, rather than verbally or propositionally (Paivio, Clark, & Lambert, 1988). Indeed, among adults, bilingual speakers can outperform monolingual speakers on some visuospatial tasks, including the time to complete a maze (Chuneya, 2008) and understanding how static pictures of knots were likely to come to a particular configuration (McLeay, 2003). There have been fewer studies on bilingual children. Some researchers have pointed out that bilingual children’s ability to transfer skills and concepts from one language to another could be related to storing information in visuospatial modalities (Castilla, Restrepo, & Perez-Leroux, 2009). Francis (2000) reported that Spanish–Náhuatl bilingual children were able to transfer knowledge about how to construct a good story from the school language to their other language. There is also some more direct evidence that bilingual children might outperform monolingual children on visuospatial tasks. One such study focused on Spanish– English and Vietnamese–English bilingual children’s ability to perform a block design task relative to same-aged English monolingual children (Gorrell, Bregman, McAllister, & Lipscomb, 1982). In the block design task, children are shown a pattern on blocks and asked to reproduce it. Both bilingual groups outperformed the monolingual group. However, not all studies have shown bilingual advantages on visuospatial tasks. Windsor, Kohnert, Loxtercamp, and Kan (2008) found that bilingual speakers were significantly slower than monolingual speakers on solving some visuospatial tasks. In the study by Gorrell et al. (1982), children’s performance on a spatial role-taking ability was also assessed. For this task, children were asked to predict what an array of objects would look like from the perspective of another person. The Spanish–English bilingual children were more accurate than the monolingual children on this task whereas the Vietnamese–English bilingual children were less accurate. One possible reason for the varied findings is that visuospatial abilities might interact with other cognitive abilities, such as executive functions. For visuospatial tasks involving an executive function component, such as identifying both interpretations of an ambiguous figure or reproducing the hand positions of another person, bilingual advantages have been found. Bialystok and Shapero (2005) found that bilingual 6-year-olds were more successful at identifying both interpretations of ambiguous figures (e.g., a vase-face) than same-aged monolingual children. Furthermore, children’s success in doing this task was correlated with their performance on selective attention task, suggesting that their performance was related to executive functions. Similarly, Garratt and Kelly (2008) found that Spanish–English bilingual children outperformed English monolingual children on a number of executive function tasks as well as the ability to imitate the hand positions of the experimenter. Imitating hand positions could require both visuospatial abilities (to remember the position of the

276 

 Elena Nicoladis

hands) and executive functions, that is, cognitive abilities related to managing information and behaviour (to switch perspectives to perform the same hand position as the experimenter). In sum, there is some intriguing evidence that bilingual children might out­ perform monolingual children on some visuospatial tasks. Because not all studies have found a bilingual advantage, it is possible that bilingual speakers outperform monolingual speakers on visuospatial tasks that also involve executive functions.

Executive Functions Executive functions refer to a variety of cognitive processes related to managing information, including selective attention, working memory, and planning (Zelazo et al., 2003; see also Chapters 13 and 16, this volume). In many cases, bilingual speakers have experience switching between languages in different contexts (Chapter 2). This experience could lead bilingual speakers to enhanced executive functioning relative to same-age monolingual speakers, possibly selective attention in particular (Tare & Gelman, 2010). Selective attention refers to the ability to attend to the variables that are important to achieve the goal at hand. Many studies have shown a bilingual advantage on selective attention tasks, both those involving language and those that do not. One way to test selective attention is to focus on children’s ability to judge whether semantically anomalous sentences are grammatical (e.g., giraffes grapple with green grapes) or not (e.g., grapple grapes with giraffes). Bilingual children show a greater ability to ignore the meaning of semantically anomalous sentences to judge the sentences simply on grammaticality (Bialystok, 2001; Cromdal, 1999). Other studies have focused on bilingual children’s performance on nonlinguistic attentional tasks, such as dimensional change card sorting. In these tasks, children are first asked to sort cards with pictures according to one dimension (e.g., the color, so cards with blue shapes might go in one pile and cards with red shapes in another). They are then asked to change the dimension by which they sort the very same cards (so cards with circles, both blue and red, now go in one pile and cards with squares, both blue and red, now go in another pile). Bilingual children can outperform monolingual children on these tasks (Bialystok, 2011), starting as young as 3 years of age (Bialystok et al., 2010). Not all studies have shown bilingual advantages in selective attention tasks and related executive functioning tasks (Duñabeitia et al., 2014; Gathercole et al., 2014; Morton & Harper, 2007). It is not yet clear whether the variability in the results is related to the tasks used to test the bilingual advantage (Duñabeitia et al., 2014) or something about the bilingual populations under study (Gathercole et al., 2014). Morton and Harper (2007) pointed out that many studies reporting bilingual advantages have focused on immigrants to Canada; yet, Canadian immigration policy specifically targets families from high socioeconomic backgrounds. When controlling for



Bilingual Speakers’ Cognitive Development in Childhood 

 277

socioeconomic status, Morton and Harper (2007) found no differences between Canadian bilingual and monolingual children on executive functioning. In sum, some studies have shown bilingual advantages over monolingual children in selective attention, whereas other studies have shown no such bilingual advantages on the very same tasks. This same variability has been reported in research with bilingual adults, leading researchers to point out that many different factors are involved in both executive functions and bilingualism (Titone, Pivneva, Sheikh, Webb, & Whitford, 2015). For example, as noted in the previous section, executive functions may interact with other cognitive abilities, such as visuospatial abilities. Future research will undoubtedly focus on identifying the multitude of factors related to bilingual speakers’ (and monolingual speakers’) performance on executive function tasks.

Social Cognition Social cognition refers to the thinking used in interacting with other people. Researchers have argued that the experience of using two languages with different people could lead to effects on bilingual children’s development of social cognition. In particular, two areas of social cognition have been explored: theory of mind and the use of nonverbal cues. Bilingualism could also affect how children attribute beliefs to other people. A number of studies have tested the hypothesis that bilingual children might develop a theory of mind earlier than monolingual children. Theory of mind refers to the belief that other people have beliefs (Bartsch & Wellman, 1995). A crucial test of children’s theory of mind is whether children understand that other people have beliefs that differ from their own, or a false belief task (Perner, Leekam, & Wimmer, 1987). Goetz (2003) showed that more Mandarin–English bilingual children passed a false belief task than same-aged monolingual peers. However, other studies have shown no difference between bilingual and monolingual children on theory-of-mind tasks (e.g., Jean-Louis, 1999) or a bilingual advantage when differences on vocabulary scores have been statistically controlled (e.g., Bialystok & Senman, 2004). One possible reason for varying results from different studies is that it may not be so much bilingualism that leads to theory-of-mind advantages, but that the languages index which cultures influence children’s development. Sabbagh, Xu, Carlson, Moses, and Lee (2006) showed that Chinese monolingual children performed better than English (American) monolingual children at tasks linked with theory-of-mind development. So, bilingual children may sometimes show advantages in theory of mind over monolingual children, but only if their bilingualism corresponds with participation in at least one culture that favors early development. Another aspect of social cognition development that could be affected by bilingualism is the use of nonverbal cues in social contexts. Bilingual speakers might have to monitor social interactions for nonverbal cues about what speakers mean more so

278 

 Elena Nicoladis

than monolingual speakers. Yow and Markman (2011) showed that bilingual preschool children were more successful in following the gaze of an adult to locate a hidden toy relative to monolingual children. Yow and Markman (2015) further showed that bilingual preschoolers are not simply better at interpreting nonverbal cues, they can also integrate the nonverbal cues with a spoken message better than same-aged monolingual children. In that study, children were asked to either find a toy or look at a toy, with or without nonverbal cues about what they were being asked to do. A similar conclusion comes from a study in which children were asked to interpret sentences with ambiguous pronouns (like he that could refer to more than one masculine referent). Yow (2015) showed that bilingual preschoolers could use gestures more successfully than their monolingual peers to figure out the meaning of the ambiguous pronouns. In sum, some studies, but not all, have shown bilingual advantages in theoryof-mind development. It is possible that bilingual advantages in theory of mind are more likely to emerge when bilingual children are growing up in a cultural context that favors early emergence of theory of mind. Studies of children’s ability to interpret nonverbal cues have thus far shown more consistent bilingual advantages. These results suggest that bilingual children might learn to monitor social interactions for non-verbal cues for appropriate behavior.

Summary and Conclusion Bilingual children develop their cognitive abilities while having some experiences that differ from monolingual children. Previous research has shown that bilingual children can switch between languages for different contexts from early in development (Chapter 2, this volume). In this chapter, I have shown that bilingual children can also think differently to speak each of their two languages. In other words, bilingual children have access to two different ways of conceptualizing the world. These different experiences could lead to differences in cognitive development between bilingual and monolingual children. Having two different sets of conceptualizations could lead to greater cognitive flexibility and a propensity to store information in visuospatial images for bilingual speakers relative to monolingual speakers. The experience that bilingual speakers have switching languages for different contexts could lead to earlier development of executive functions and social cognition for bilingual children relative to monolingual children. In this chapter, I have shown that, indeed, some studies (but not all) have supported the predictions for bilingual advantages in terms of cognitive flexibility, visuospatial abilities, executive functions, and social cognition. It is clear that researchers do not yet fully understand what it is about bilingualism that can lead to cognitive differences from monolingual speakers. One possibility is that certain levels of proficiency (see Chapter 7, this volume) in both languages are required for cognitive advantages (Cummins, 1976), although some studies have shown that proficiency in



Bilingual Speakers’ Cognitive Development in Childhood 

 279

the language of the school may be a more important factor (Lauchlan, Parisi, & Fadda, 2013). Another possibility is that the cognitive abilities that I have treated as separate abilities interact and bilingual advantages emerge in the interactions of these abilities. For example, bilingual speakers might tend to show advantages on executive function tasks or social cognitive abilities that rely heavily on visuospatial abilities. Future research will also test the possibility that culture and socioeconomic status may be more important than bilingualism in leading to cognitive advantages. In this chapter, I have often referred to effects of bilingualism as if simply knowing two languages will have the same effects regardless of the language pair spoken or the community in which the children are growing up. An important direction for future research is to test these assumptions. It is possible that a Spanish–Portuguese bilingual speaker might not necessarily develop exactly the same way as a Chinese– Finnish bilingual speaker. The degree of similarity between the two languages, either structurally or in usage, could influence development. Bosch and Ramon-Casas (2014) showed that Spanish–Catalan bilingual children’s vocabulary size was enhanced by the phonological similarities of their two languages. The community in which bilingual children are growing up could also influence how language and cognitive development unfold. Bilingual children’s outcomes might differ in communities that support bilingualism vs. communities that do not (see Smithson, Paradis, & Nicoladis, 2014). How community attitudes and beliefs come to influence how bilingual children develop is not yet well understood. We know that adult attitudes can affect children’s language choice and code-mixing within the family (Lanza, 1992; see also Chapter 2, this volume). In bilingual communities in which both languages are actively maintained, bilingual children’s language use comes to resemble that of their community within the preschool years (Baker & van den Bogaerde, 2008; Nicoladis & Genesee, 1997; Paradis & Nicoladis, 2007). Even if subsequent research shows that there are few, if any, cognitive advantages linked solely to bilingualism, it is important to keep in mind that bilingualism is an advantage in and of itself. The ability to speak two (or more) languages allows children to communicate with more people. Each language embodies a way of thinking and each way of thinking differs at least a little bit from that associated with any other language. Bilingual children have access to two different ways of thinking and being, two windows on the world.

References Adi-Japha, E., Berberich-Artzi, J., & Libnawi, A. (2010). Cognitive flexibility in drawings of bilingual children. Child Development, 81, 1356–1366. http://dx.doi.org/10.1111/ j.1467-8624.2010.01477.x

280 

 Elena Nicoladis

Baker, A., & van den Bogaerde, B. (2008). Codemixing in signs and words in the input to and output from children. In C. Plaza-Pust & E. Morales Lopéz (Eds.), Sign bilingualism: Language development, interaction, and maintenance in sign language contact situations (pp. 1–27). Amsterdam, the Netherlands: Benjamins. Barac, R., Bialystok, E., Castro, D. C., & Sanchez, M. (2014). The cognitive development of young dual language learners: A critical review. Early Childhood Research Quarterly, 29, 699–714. http:// dx.doi.org/10.1016/j.ecresq.2014.02.003 Bartsch, K., & Wellman, H. M. (1995). Children talk about the mind. New York, NY: Oxford University Press. Ben-Zeev, S. (1977). Mechanisms by which childhood bilingualism affects understanding of language and cognitive structures. In P. A. Hornby (Ed.), Bilingualism: Psychological, social, and educational implications (pp. 29–55). New York, NY: Academic Press. Bialystok, E. (2001). Bilingualism in development: Language, literacy, and cognition. New York, NY: Cambridge University Press. http://dx.doi.org/10.1017/CBO9780511605963 Bialystok, E. (2011). Coordination of executive functions in monolingual and bilingual children. Journal of Experimental Child Psychology, 110, 461–468. http://dx.doi.org/10.1016/ j.jecp.2011.05.005 Bialystok, E., Barac, R., Blaye, A., & Poulin-Dubois, D. (2010). Word mapping and executive functioning in young monolingual and bilingual children. Journal of Cognition and Development, 11, 485–508. http://dx.doi.org/10.1080/15248372.2010.516420 Bialystok, E., & Senman, L. (2004). Executive processes in appearance-reality tasks: The role of inhibition of attention and symbolic representation. Child Development, 75, 562–579. http://dx.doi.org/ 10.1111/j.1467-8624.2004.00693.x Bialystok, E., & Shapero, D. (2005). Ambiguous benefits: The effect of bilingualism on reversing ambiguous figures. Developmental Science, 8, 595–604. http://dx.doi.org/10.1111/ j.1467-7687.2005.00451.x Boroditsky, L., Schmidt, L. A., & Phillips, W. (2003). Sex, syntax, and semantics. In S. Gentner & S. Goldin-Meadow (Eds.), Language in mind: Advances in the study of language and thought (pp. 61–79). Cambridge, MA: MIT Press. Bosch, L., & Ramon-Casas, M. (2014). First translation equivalents in bilingual toddlers’ expressive vocabulary: Does form similarity matter? International Journal of Behavioral Development, 38, 317–322. http://dx.doi.org/10.1177/0165025414532559 Byers-Heinlein, K., & Werker, J. F. (2013). Lexicon structure and the disambiguation of novel words: Evidence from bilingual infants. Cognition, 128, 407–416. http://dx.doi.org/10.1016/ j.cognition.2013.05.010 Castilla, A. P., Restrepo, M. A., & Perez-Leroux, A. T. (2009). Individual differences and language interdependence: A study of sequential bilingual development in Spanish–English preschool children. International Journal of Bilingual Education and Bilingualism, 12, 565–580. http://dx.doi.org/ 10.1080/13670050802357795 Chuneya, A. (2008). Assessing cognitive functions in Russian–English bilinguals and English monolinguals using visual spatial tasks. (Unpublished doctoral dissertation). Carlos Albizu University, Miami, FL. Cromdal, J. (1999). Childhood bilingualism and metalinguistic skills: Analysis and control in young Swedish–English bilinguals. Applied Psycholinguistics, 20, 1–20. http://dx.doi.org/10.1017/ S0142716499001010



Bilingual Speakers’ Cognitive Development in Childhood 

 281

Cummins, J. (1976). The influence of bilingualism on cognitive growth: A synthesis of research findings and explanatory hypotheses. Working Papers on Bilingualism, 9, 1–43. Davidson, D., Jergovic, D., Imami, Z., & Theodos, V. (1997). Monolingual and bilingual children’s use of the mutual exclusivity constraint. Journal of Child Language, 24, 3–24. http://dx.doi.org/ 10.1017/S0305000996002917 Duñabeitia, J. A., Hernández, J. A., Antón, E., Macizo, P., Estévez, A., Fuentes, L. J., & Carreiras, M. (2014). The inhibitory advantage in bilingual children revisited: Myth or reality? Experimental Psychology, 61, 234–251. http://dx.doi.org/10.1027/1618-3169/a000243 Fivush, R. (2011). The development of autobiographical memory. Annual Review of Psychology, 62, 559–582. http://dx.doi.org/10.1146/annurev.psych.121208.131702 Francis, N. (2000). The shared conceptual system and language processing in bilingual children: Findings from literacy assessment in Spanish and Náhuatl. Applied Linguistics, 21, 170–204. http://dx.doi.org/10.1093/applin/21.2.170 Garratt, L. C., & Kelly, T. P. (2008). To what extent does bilingualism affect children’s performance on the NEPSY? Child Neuropsychology, 14, 71–81. http://dx.doi.org/10.1080/09297040701218405 Gathercole, V. C. M., Thomas, E. M., Kennedy, I., Prys, C., Young, N., Viñas Guasch, N., . . . Jones, L. (2014). Does language dominance affect cognitive performance in bilinguals? Lifespan evidence from preschoolers through older adults on card sorting, Simon, and metalinguistic tasks. Frontiers in Psychology, 5, 11. http://dx.doi.org/10.3389/fpsyg.2014.00011 Goetz, P. J. (2003). The effects of bilingualism on theory of mind development. Bilingualism: Language and Cognition, 6, 1–15. http://dx.doi.org/10.1017/S1366728903001007 Gorrell, J. J., Bregman, N. J., McAllister, H. A., & Lipscomb, T. J. (1982). A comparison of spatial role taking in the monolingual and bilingual child. The Journal of Genetic Psychology, 140, 3–10. http://dx.doi.org/10.1080/00221325.1982.10534170 Guasti, M. T., Stavrakaki, S., & Arosio, F. (2012). Cross-linguistic differences and similarities in the acquisition of relative clauses: Evidence from Greek and Italian. Lingua, 122, 700–713. http:// dx.doi.org/10.1016/j.lingua.2012.02.001 Hakuta, K., & Diaz, R. (1985). The relationship between degree of bilingualism and cognitive ability: A critical discussion and some new longitudinal data. In K. E. Nelson (Ed.), Children’s language (Vol. 5, pp. 319–344). Hillsdale, NJ: Erlbaum. Hoang, H., Nicoladis, E., Smithson, L., & Furman, R. (in press). French–English bilingual children’s tense use and shift in narration. International Journal of Bilingualism. http://dx.doi.org/ 10.1017/1367006915613161 Hohenstein, J., Eisenberg, A., & Naigles, L. (2006). Is he floating across or crossing afloat? Crossinfluence of L1 and L2 in Spanish–English bilingual adults. Bilingualism: Language and Cognition, 9, 249–261. http://dx.doi.org/10.1017/S1366728906002616 Jean-Louis, B. (1999). Does childhood bilingualism facilitate the development of theory of mind? (Unpublished doctoral dissertation). Yale University, New Haven, CT. Kail, R. V., & Ferrer, E. (2007). Processing speed in childhood and adolescence: Longitudinal models for examining developmental change. Child Development, 78, 1760–1770. http://dx.doi.org/ 10.1111/j.1467-8624.2007.01088.x Lanza, E. (1992). Can bilingual two-year-olds code-switch? Journal of Child Language, 19, 633–658. http://dx.doi.org/10.1017/S0305000900011600 Lauchlan, F., Parisi, M., & Fadda, R. (2013). Bilingualism in Sardinia and Scotland: Exploring the cognitive benefits of speaking a “minority” language. The International Journal of Bilingualism, 17, 43–56. http://dx.doi.org/10.1177/1367006911429622

282 

 Elena Nicoladis

Lecorre, M., Van de Walle, G., Brannon, E. M., & Carey, S. (2006). Re-visiting the competence/ performance debate in the acquisition of the counting principles. Cognitive Psychology, 52, 130–169. http://dx.doi.org/10.1016/j.cogpsych.2005.07.002 Lewis, M., & Ramsay, D. (2004). Development of self-recognition, personal pronoun use, and pretend play during the 2nd year. Child Development, 75, 1821–1831. http://dx.doi.org/10.1111/ j.1467-8624.2004.00819.x Li, L., & Tan, C. L. (2015). Home literacy environment and its influence on Singaporean children’s Chinese oral and written language abilities. Early Childhood Education Journal. Advance online publication. Retrieved from http://link.springer.com/article/10.1007/s10643-015-0723-4 Lucy, J. (2011). Language and cognition: The view from anthropology. In V. Cook & B. Bassetti (Eds.), Language and bilingual cognition (pp. 43–68). New York, NY: Psychology Press. Marinova-Todd, S. H. (2012). “Corplum is a core from a plum”: The advantage of bilingual children in the analysis of word meaning. Bilingualism: Language and Cognition, 15, 117–127. http:// dx.doi.org/10.1017/S136672891000043X McLeay, H. (2003). The relationship between bilingualism and the performance of spatial tasks. International Journal of Bilingual Education and Bilingualism, 6, 423–438. http://dx.doi.org/ 10.1080/13670050308667795 Miller, N., Furman, R., & Nicoladis, E. (2015). French–English bilingual children’s motion event communication shows cross-linguistic influence in speech but not gesture. Manuscript submitted for publication. Morton, J. B., & Harper, S. N. (2007). What did Simon say? Revisiting the bilingual advantage. Developmental Science, 10, 719–726. http://dx.doi.org/10.1111/j.1467-7687.2007.00623.x Nelson, K., & Fivush, R. (2004). The emergence of autobiographical memory: A social cultural developmental theory. Psychological Review, 111, 486–511. http://dx.doi.org/10.1037/ 0033-295X.111.2.486 Nicoladis, E. (2006). Cross-linguistic transfer in adjective–noun strings by preschool bilingual children. Bilingualism: Language and Cognition, 9, 15–32. http://dx.doi.org/10.1017/ S136672890500235X Nicoladis, E. (2012). Cross-linguistic influence in French–English bilingual children’s possessive constructions. Bilingualism: Language and Cognition, 15, 320–328. http://dx.doi.org/10.1017/ S1366728911000101 Nicoladis, E., & Foursha-Stevenson, C. (2012). Language and culture effects on gender classification of objects. Journal of Cross-Cultural Psychology, 43, 1095–1109. http://dx.doi.org/10.1177/ 0022022111420144 Nicoladis, E., & Gavrila, A. (2015). Cross-linguistic influence in Welsh–English bilingual children’s adjectival constructions. Journal of Child Language, 42, 903–916. http://dx.doi.org/10.1177/ 0022022111420144 Nicoladis, E., & Genesee, F. (1997). The role of parental input and language dominance in bilingual children’s code-mixing. In E. Hughes, M. Hughes, & A. Greenhill (Eds.), Proceedings of the 21st Annual Boston University Conference on Language Development (pp. 422–432). Somerville, MA: Cascadilla Press. Nicoladis, E., Rose, A., & Foursha-Stevenson, C. (2010). Thinking for speaking and cross-linguistic transfer in preschool bilingual children. International Journal of Bilingual Education and Bilingualism, 13, 345–370. http://dx.doi.org/10.1080/13670050903243043



Bilingual Speakers’ Cognitive Development in Childhood 

 283

Paivio, A., Clark, J. M., & Lambert, W. E. (1988). Bilingual dual-coding theory and semantic repetition effects on recall. Journal of Experimental Psychology: Learning, Memory, and Cognition, 14, 163–172. http://dx.doi.org/10.1037/0278-7393.14.1.163 Paradis, J., & Nicoladis, E. (2007). The influence of dominance and sociolinguistic context on bilingual preschoolers’ language choice. International Journal of Bilingual Education and Bilingualism, 10, 277–297. http://dx.doi.org/10.2167/beb444.0 Peal, E., & Lambert, W. (1962). The relation of bilingualism to intelligence. Psychological Monographs: General and Applied, 76, 1–23. http://dx.doi.org/10.1037/h0093840 Perner, J., Leekam, S. R., & Wimmer, H. (1987). Three-year-olds’ difficulty with false belief: The case for a conceptual deficit. British Journal of Developmental Psychology, 5, 125–137. http://dx.doi.org/ 10.1111/j.2044-835X.1987.tb01048.x Sabbagh, M. A., Xu, F., Carlson, S. M., Moses, L. J., & Lee, K. (2006). The development of executive functioning and theory of mind: A comparison of Chinese and U.S. preschoolers. Psychological Science, 17, 74–81. http://dx.doi.org/10.1111/j.1467-9280.2005.01667.x Sera, M. D., Elieff, C., Forbes, J., Burch, M. C., Rodríguez, W., & Dubois, D. P. (2002). When language affects cognition and when it does not: An analysis of grammatical gender and classification. Journal of Experimental Psychology: General, 131, 377–397. http://dx.doi.org/10.1037/ 0096-3445.131.3.377 Skarabela, B., & Serratrice, L. (2009). “The doctor’s mother” or “the mother of the doctor”? Syntactic priming of possessive noun phrases in English preschoolers. Online Supplement of the Proceedings of the Boston University Conference on Language Development. Retrieved from http://www.bu.edu/bucld/proceedings/supplement/vol33/ Slobin, D. I. (1996). Two ways to travel: Verbs of motion in English and Spanish. In M. Shibatani & S. A. Thompson (Eds.), Grammatical constructions: Their form and meaning (pp. 195–219). Oxford, England: Oxford University Press. Smithson, L., Paradis, J., & Nicoladis, E. (2014). Bilingualism and receptive vocabulary achievement: Could sociocultural context make a difference? Bilingualism: Language and Cognition, 17, 810–821. http://dx.doi.org/10.1017/S1366728913000813 Sodian, B., & Kristen, S. (2010). Theory of mind. In B. M. Glatzeder, V. Goel, & A. von Müller (Eds.), Towards a theory of thinking: Building blocks for a conceptual framework (pp. 189–201). Berlin, Germany: Springer-Verlag. http://dx.doi.org/10.1007/978-3-642-03129-8_13 Tare, M., & Gelman, S. A. (2010). Can you say it another way? Cognitive factors in bilingual children’s pragmatic language skills. Journal of Cognition and Development, 11, 137–158. http://dx.doi.org/ 10.1080/15248371003699951 Titone, D., Pivneva, I., Sheikh, N. A., Webb, N., & Whitford, V. M. (2015). Doubling down on multifactorial approaches to the study of bilingualism & executive control. Bilingualism: Language and Cognition, 18, 43–44. http://dx.doi.org/10.1017/S1366728914000595 Vihman, M. M. (1998). A developmental perspective on codeswitching: Conversations between a pair of bilingual siblings. The International Journal of Bilingualism, 2, 45–84. Wilson, A. E., Smith, M. D., & Ross, H. S. (2003). The nature and effects of young children’s lies. Social Development, 12, 21–45. http://dx.doi.org/10.1111/1467-9507.00220 Windsor, J., Kohnert, K., Loxtercamp, A. L., & Kan, P.-F. (2008). Performance on nonlinguistic visual tasks by children with language impairment. Applied Psycholinguistics, 29, 237–268. http:// dx.doi.org/10.1017/S0142716407080113

284 

 Elena Nicoladis

Yip, V., & Matthews, S. (2000). Syntactic transfer in a Cantonese–English bilingual child. Bilingualism: Language and Cognition, 3, 193–208. http://dx.doi.org/10.1017/ S136672890000033X Yow, W. Q. (2015). Monolingual and bilingual preschoolers’ use of gestures to interpret ambiguous pronouns. Journal of Child Language, 42, 1394–1407. http://dx.doi.org/10.1017/ S0305000914000737 Yow, W. Q., & Markman, E. M. (2011). Young bilingual children’s heightened sensitivity to referential cues. Journal of Cognition and Development, 12, 12–31. http://dx.doi.org/10.1080/ 15248372.2011.539524 Yow, W. Q., & Markman, E. M. (2015). A bilingual advantage in how children integrate multiple cues to understand a speaker’s referential intent. Bilingualism: Language and Cognition, 18, 391–399. http://dx.doi.org/10.1017/S1366728914000133 Zelazo, P. D., Müller, U., Frye, D., Marcovitch, S., Argitis, G., Boseovski, J., . . . Carlson, S. M. (2003). The development of executive function in early childhood. Monographs of the Society for Research in Child Development, 68, vii–137. Retrieved from http://www.jstor.org/stable/1166202

Max R. Freeman, Anthony Shook, and Viorica Marian

15 Cognitive and Emotional Effects of Bilingualism in Adulthood The limits of my language mean the limits of my world. —Ludwig Wittgenstein

Language is a powerful human ability that shapes our thoughts and our perceptions of everyday experiences. At a social level, many aspects of human communication are dependent on language; at a cognitive level, language helps us to understand events and experiences in our environment, and ascribe meaning to these events within our minds. For bilingual speakers, the ability to communicate, think, and understand the environment may be different from monolingual speakers due to the influence of managing multiple languages on the cognitive system. As a result, the ability to communicate in multiple languages may serve as a springboard for transforming higher order cognitive processes, ultimately impacting how we think about and interpret the world. Consider how language influences two highly interactive aspects of the cognitive system: cognitive control and emotion. These aspects of cognition are important to communication, as they affect our ability to understand and process language. When engaged in conversation with a friend in a crowded restaurant, we use cognitive control (e.g., inhibitory control, or the ability to suppress a response or irrelevant information) to ignore the other conversations and noise from the surrounding tables. Additionally, when asked to recall and describe an event (e.g., your fifth birthday party), the language we use to describe that memory (and the memory itself) is tightly bound to the emotions associated with the experience. These examples illustrate how cognitive control and emotion are connected, and how these aspects of cognition are filtered through and by language. Because language serves as a cornerstone for cognition, impacting human abilities like cognitive control, memory, and emotion, experience with two or more languages may change how we attend to the environment. In addition, how we think and feel about our experiences may be shaped by the additional linguistic channels through which we perceive daily events. Neurological research suggests a link between cognitive control and emotional processes in the brain. The regions involved in cognitive control (e.g., attention, selfregulation; prefrontal cortex [PFC] and anterior cingulate cortex [ACC]) and emotion We thank the Northwestern University Bilingualism and Psycholinguistics Research Group for valuable feedback. Preparation of this chapter was supported in part by grant NICHD 1R01HD059858 to the third author. http://dx.doi.org/10.1037/14939-016 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

286 

 Freeman, Shook, and Marian

(e.g., amygdala) become activated in concert when situations demand emotion regulation or control of emotional thoughts, such as in response to an aversive event (e.g., pain; see Ochsner & Gross, 2005, for a review). Interestingly, bilingual speakers simultaneously activate areas associated with cognitive control and emotion on language switching tasks, even when processing emotionally neutral words (e.g., Hernandez, 2009), suggesting that cognitive control and emotion are highly interactive. To further understand the interactivity of cognitive control and emotion in bilingual speakers, we discuss research on the impact of bilingual experience on these processes. Research on bilingualism has recently begun to explore the underlying mechanisms involved in the experience of managing two (or more) languages, as well as the cognitive changes incurred as a result of this experience. The evidence suggests that the bilingual mind may be organized differently than “single-language minds” (Kroll & Bialystok, 2013, pp. 497–498). Experience with multiple languages may also result in an enriched cultural identity that incorporates elements from both cultures, which may transform bilingual speakers’ mental conceptual structure and the way they process their environment. These rich language and cultural experiences further impact cognition, resulting in enhanced cognitive control abilities, and shaping how emotional events are processed and retrieved from memory. In this chapter, we present evidence that highlights how and why bilingual experience impacts the human mind. Though the effects of bilingual experience on cognition are wide ranging, we focus primarily on two aspects of cognition: cognitive control and emotion, as well as how these aspects interact. We draw on studies that examine different bilingual groups (e.g., simultaneous bilingual speakers, who learn both languages at the same time; sequential bilingual speakers, who learn one language prior to learning another; and adult second-language [L2] learners), and discuss how bilingual and monolingual speakers differ on cognitive control abilities and in emotional processing. Throughout this chapter, we explore how bilingual experience changes cognitive control and emotion, the underlying sources of change within each, and the links between them.

Bilingualism and Cognitive Control Cognitive control can be thought of as the way individuals manage distraction or conflicting information in their daily lives, and is comprised of a constellation of skills that help individuals navigate their environment. One aspect of cognitive control is a process known as conflict monitoring, which is the ability to keep track of conflict or potential conflict in the environment (as measured by the Simon Task, see Figure 15.1A). Once aware of a potential conflict, individuals rely on another important feature of cognitive control known as competition resolution, or the ability to manage and resolve distractions while trying to focus on relevant information (as measured by the Stroop Task, see Figure 15.1B). Finally, cognitive control also includes skills like interference suppression



Cognitive and Emotional Effects of Bilingualism 

 287

Fig. 15.1: (A) Simon task: Congruent trials represent the coinciding location of the target stimulus on the screen and the location of the response on the keyboard. Incongruent trials represent a conflict with the location of the target stimulus on the screen and the location of the response on the keyboard. (B) Nonlinguistic Stroop task: Congruent trials represent no conflict between the direction of the arrow and its location on the screen. Incongruent trials represent a conflict between the direction of the arrow and its location on the screen. (C) Flanker task: Congruent trials contain a target chevron embedded in a group of competitor chevrons pointing in the same direction of the target. Incongruent trials contain a target chevron embedded in a group of competitor chevrons pointing in the opposite direction of the target.

and response inhibition, which involve ignoring distracting information or viable alternative responses when responding to a direction or question (as measured by the Flanker task, see Figure 15.1C). Of central importance is the idea that cognitive control is domain general. In other words, cognitive control is not limited to a single mental process, like language or vision, but extends into every aspect of perception and cognition.

Effects of Bilingualism on Inhibitory Control and Aging The effects of bilingualism on cognitive control have been widely studied over the past 30 years and suggest that experience with multiple languages has a positive effect on cognitive control (e.g., Bialystok, Craik, & Luk, 2008; Kroll & Bialystok, 2013; Luk,

288 

 Freeman, Shook, and Marian

De Sa, & Bialystok, 2011; Ridderinkhof, Ullsperger, Crone, & Nieuwenhuis, 2004; but see Hilchey & Klein, 2011; Paap & Greenberg, 2013). For example, bilingual speakers have shown advantages over monolingual speakers across various aspects of cognitive control, including conflict monitoring (Bialystok, Craik, Klein, & Viswanathan, 2004; Bialystok, Martin, & Viswanathan, 2005; Bialystok et al., 2008; Martin-Rhee & Bialystok, 2008), competition resolution (Bialystok et al., 2008; Costa, Hernández, Costa-Faidella, & Sebastián-Gallés, 2009; Costa, Hernández, & Sebastián-Gallés, 2008; Hernández, Costa, Fuentes, Vivas, & Sebastián-Gallés, 2010; Luk, De Sa, & Bialystok, 2011), and interference suppression and response inhibition (Costa et al., 2008; Luk, Anderson, Craik, Grady, & Bialystok, 2010). These bilingual advantages in cognitive control also translate to real-world benefits. For example, lifelong bilingualism increases cognitive reserve, a protective mechanism against dementia that arises as a result of sustained mental activity (e.g., Schweizer, Ware, Fischer, Craik, & Bialystok, 2012). Active bilingualism throughout the lifespan is associated with a delayed onset of symptoms of Alzheimer’s disease by 4 to 5 years (Alladi et al., 2013; Bialystok, Craik, & Freedman, 2007; Craik, Bialystok, & Freedman, 2010). Also within bilingual speakers, behavioral symptoms associated with Alzheimer’s are less severe than what would be expected from the amount of disease-related neurological breakdown (Schweizer et al., 2012). That is, when monolingual speakers and bilingual speakers are matched on neurological decline associated with Alzheimer’s disease, because bilingualism serves as a protective measure against age-related cognitive decline, bilingual speakers demonstrate fewer behavioral symptoms associated with Alzheimer’s. When Alzheimer’s is not present in older bilingual adults, active bilingualism throughout the lifespan can lead to an advantage in the ability to remember events (Schroeder & Marian, 2012). Also in healthy older lifelong bilingual adults, certain neurological tissues are better preserved in the brain relative to monolingual peers, such as white matter (which is responsible for coordinating communication between brain areas; Luk, Bialystok, Craik, & Grady, 2011) and gray matter (which involves specific processes such as executive control and muscle movement; Abutalebi et al., 2014). Thus, as a result of sustained experience with managing two languages, the general cognitive system is enhanced, demonstrating cognitive control benefits as well as decreased cognitive and neurological decline relative to monolingual speakers. The lifelong effects of bilingualism are discussed in further detail in Chapter 16 of this volume.

Effects of Bilingualism on Language Activation and Interaction What aspects of bilingual experience give rise to cognitive enhancement? When engaged in a conversation, whether speaking or listening, bilingual speakers constantly activate both of their languages simultaneously. This parallel activation occurs across different linguistic contexts for bilingual speakers (see Kroll & Bialystok,



Cognitive and Emotional Effects of Bilingualism 

 289

2013, for a review), including in language production and comprehension (Bartolotti & Marian, 2012; Blumenfeld & Marian, 2007, 2011, 2013; Green, 1998; Kroll, Bobb, Misra, & Guo, 2008; Linck, Hoshino, & Kroll, 2008; Marian & Spivey, 2003a, 2003b; Martín, Macizo, & Bajo, 2010). Activating two languages in parallel increases bilingual speakers’ overall cognitive load, as they have to manage the activation of an irrelevant language while trying to comprehend or produce speech in the target language. For example, as monolingual speakers hear a word unfold over time, they activate multiple competing candidate words (e.g., the target plug activates phonological competitors like plum; e.g., Blumenfeld & Marian, 2011). As bilingual speakers hear the same word, they experience the same within-language activation, in addition to between-language competitor activation (e.g., plug activates plancha/iron for a Spanish–English bilingual speaker; e.g., Blumenfeld & Marian, 2013). To manage this increased competition, bilingual speakers rely on inhibitory control (e.g., Green, 1998). Because similar inhibitory mechanisms underlie both linguistic (e.g., managing simultaneous activation of two languages) and nonlinguistic control (e.g., competition resolution), an interdependent relationship likely exists between language and cognition in bilingual speakers (e.g., Blumenfeld & Marian, 2013). Thus, being a “mental juggler” of two languages (Kroll, 2009) increases cognitive control efficiency and transforms the cognitive system, allowing it to accommodate and manage multiple languages. Evidence from language switching studies may also offer an explanation as to why inhibitory control is more efficient in bilingual speakers. Bilingual speakers who are less proficient in their L2 must inhibit their first language (L1) more strongly to avoid interference while using their L2 (e.g., Gollan & Ferreira, 2009; Meuter & Allport, 1999; Misra, Guo, Bobb, & Kroll, 2012). Functional magnetic resonance imaging (fMRI) studies have suggested that the same cortical regions required to overcome L1 inhibition are also recruited in nonlinguistic cognitive control tasks (e.g., interference suppression as measured by the Flanker task). Specifically, the dorsolateral prefrontal cortex (DLPFC) and the left inferior parietal gyrus were recruited when bilingual speakers switched from L2 blocked naming to L1 blocked naming (Guo, Liu, Misra, & Kroll, 2011). Studies exploring how listeners manage competition in language-free contexts (i.e., nonlinguistic cognitive control) have demonstrated that the DLPFC is recruited during the Flanker task (e.g., Bunge, Dudukovic, Thomason, Vaidya, & Gabrieli, 2002; see Figure 15.1C for an example) and attentional control in bilingual speakers (Hernandez, Martinez, & Kohnert, 2000). Together, these findings suggest that the DLPFC is involved in both linguistic and nonlinguistic cognitive control. Results from research on the simultaneous activation of languages and on language switching in bilingual speakers are consistent with the inhibitory control model (Green, 1998), which predicts that differences in switch costs between languages, which are asymmetrical depending on the direction of the language switch (i.e., L1 to L2 vs. L2 to L1), are a function of language dominance. For instance, research has found that as L2 proficiency increases, the asymmetry in cost between switching from

290 

 Freeman, Shook, and Marian

L1 to L2 vs. L2 to L1 diminishes (Costa & Santesteban, 2004). During language switching tasks, both languages in a bilingual speaker are activated in parallel and compete for selection. Successful selection requires the inhibition of nontarget language competitors. Green (1998) noted that overcoming suppression of the previously inhibited language is modulated by how strongly it was suppressed; for an unbalanced bilingual speaker, it will take longer to switch into the dominant, more-suppressed language. For a balanced bilingual speaker, both languages are activated and inhibited to a similar degree and the direction of the language switch does not affect the magnitude of the cost. That is, as proficiency with an L2 increases over time, so does the efficiency with which bilingual speakers manage parallel language activation and language switching. Thus, along with experience with parallel activation, language switching may serve as a potential mechanism for cognitive enhancement in bilingual speakers.

Effects of Bilingualism on the Interaction Between Linguistic and Nonlinguistic Cognitive Control Parallel activation and language switching in bilingual speakers require the use of linguistic cognitive control. More specifically, the link between language and cognition in bilingual speakers has led to a growing line of research exploring the effects of parallel language activation and linguistic and nonlinguistic inhibitory control. Eyetracking methodology has been used to index the relationship between lexical activation and inhibitory control. For instance, Blumenfeld and Marian (2011) used eyetracking to examine the time course (i.e., duration) of within-language competition (e.g., English target plum, competitor plug) in bilingual and monolingual speakers. To measure how listeners used inhibitory control in managing lexical competition, each within-language competition trial was followed by a nonlinguistic priming trial that measured listeners’ residual (continued) inhibition of targets and competitors (see Figure 15.2). Although bilingual and monolingual speakers showed similar within-language competition, bilingual speakers demonstrated less competitor inhibition on the subsequent priming probe trials (i.e., fewer looks to the competitor asterisk), suggesting that bilingual speakers overcame linguistic interference more quickly and efficiently than monolingual speakers. In addition, Blumenfeld and Marian (2011) tested participants’ performance on a nonlinguistic Stroop task, where they indicated, via button press, the direction of an arrow, which was sometimes incongruent with the arrow’s location on a screen (e.g., rightward pointing arrow on the left side of a screen; see Figure 15.1B for an example). The authors found that better performance on this task was related to better resolution of linguistic competition in bilingual speakers but not in monolingual speakers; bilingual speakers with greater cognitive control abilities in a nonlinguistic task were better able to inhibit linguistic competition. These find-



Cognitive and Emotional Effects of Bilingualism 

 291

Fig. 15.2: An illustration of the eye-tracking/negative priming paradigm. Participants heard English words and identified corresponding pictures in the presence of English phonological competitor pictures while their eye movements were tracked (top panel). Word Recognition trials were followed by priming probe trials (bottom panel) that probed inhibition of competitor words (light asterisks in locations previously occupied by competitor pictures, column A), relative to control items (light asterisks in locations previously occupied by control pictures, column B), and target items (light asterisks in locations previously occupied by target pictures, column C). Participants responded by pressing one of four keys, arranged in a square corresponding to the location of items on the stimulus display. From “Bilingualism Influences Inhibitory Control in Auditory Comprehension,” by H. K. Blumenfeld & V. Marian, 2011, Cognition, 118, p. 247. Copyright 2011 by Elsevier. Adapted with permission.

ings demonstrate a relationship between nonlinguistic cognitive control and management of linguistic competition in bilingual speakers, as well as the impact of language experience on cognitive control. To further explore the relationship between language activation and cognitive control, in a subsequent study, Blumenfeld and Marian (2013) measured betweenlanguage competition and cognitive control abilities in bilingual speakers. Monolingual speakers and both high- and low-proficiency bilingual speakers were presented with four objects and asked to find the target object (e.g., comb) amongst a betweenlanguage competitor (e.g., conejo/rabbit) and two filler items. The high-proficiency bilingual speakers demonstrated increased early parallel activation (i.e., looks to the rabbit) and reduced later parallel activation relative to low-proficiency bilingual speakers. Stronger early parallel activation followed by reduced later parallel activation was associated with smaller Stroop effects (incongruent trials minus neutral

292 

 Freeman, Shook, and Marian

trials, see Figure 15.1B), indicating more efficient non-linguistic inhibition in only the high-proficiency bilingual group. These findings are consistent with research demonstrating that asymmetrical language switch costs diminish with increased L2 skills (e.g., Costa & Santesteban, 2004), suggesting that increased abilities in the L2 lead to more efficient management of cross-linguistic interference. Similarly, brain imaging evidence indicates that bilingual speakers recruit cognitive control networks more extensively during language processing (e.g., Guo et al., 2011) relative to monolingual speakers (e.g., Abutalebi et al., 2007, 2008), which results in more efficient performance on linguistic and nonlinguistic tasks. Thus, we see a positive correlation between L2 proficiency and efficiency of cognitive control in bilingual speakers. Research exploring the effect of bilingualism on language processing and cognitive control suggests that bilingual speakers’ reliance on domain-general cognitive control mechanisms during language processing results in cognitive benefits. The cognitive benefits may arise from the need to manage switching between two languages that are activated in parallel. Importantly, the studies discussed in this section demonstrate that bilingual speakers can be considered experts at managing linguistic competition. This research paints a picture of a highly-plastic cognitive and neurological system that is sensitive to experience with multiple languages. The cognitive system adaptively changes in ways that are beneficial not only to bilingual language processing but also to cognition in general.

Bilingualism and Emotion Effects of Bilingualism on Emotion Processing Along with enhancements to cognitive control, bilingualism also affects another higher order aspect of the cognitive system: emotion. When we think about emotional experiences, we often activate the language associated with those experiences. It has been suggested that across the world’s languages, some concepts, words, and emotions may not transfer (Pavlenko, 2008; but see Odlin, 2005), and within bilingual speakers, we may see a “lost in translation” effect when attempting to express or process an emotion for which there is no label in one of their languages. Perhaps, then, it is no surprise that research on emotion in bilingual speakers has revealed mixed evidence on how bilingual speakers process emotions across both of their languages. The emotional content of words appears to be processed differently in the L1 and the L2, especially with late L2 learners (see Pavlenko, 2012, for a review). For example, processing emotion-laden words in the L1 is thought to be more automatic than in the L2, resulting in stronger reactions to emotion-laden words (e.g., kiss) in a bilingual speaker’s native language. For instance, emotion-laden words in a late bilingual speaker’s L1 result in increased arousal (i.e., electrodermal reactivity: the skin’s electrostatic response), relative to emotion-laden words in the L2 (e.g., Harris, Ayçiçegi, &



Cognitive and Emotional Effects of Bilingualism 

 293

Gleason, 2003). A possible explanation for the difference in L1 and L2 emotional word processing is that adult L2 learners process L2 emotional words at a more semantic rather than affective level (Pavlenko, 2012). In other words, although late bilingual speakers understand the meaning of L2 emotional words, they do not process emotional words in the L2 as deeply as in the L1. It is important to consider the role of L2 proficiency when thinking about the factors that might modulate how emotional words are processed across the L1 and L2. Theoretical evidence from bilingual language models, such as the revised hierarchical model (Kroll & Stewart, 1994), may offer insight into how proficiency modulates processing of the emotional content of words across the L1 and the L2. For unbalanced bilingual speakers, an L2 word requires a mediator—the L1 translation equivalent— to access its conceptual representation; thus, the emotional word in the L2 is less directly linked to its conceptual representation than the L1 translation, suggesting that there may be an asymmetry in how each language accesses the deeper, affective meaning of the emotional word. However, as L2 proficiency increases and direct L2-to-concept links are strengthened, so too are the links between the concept and its corresponding emotional state, suggesting that balanced bilingual speakers may not show the same bias toward deeper emotional processing in their L1. In contrast, there is also evidence suggesting that emotional words may be processed similarly across the L1 and L2 regardless of proficiency. Ferré, García, Fraga, Sánchez-Casas, and Molero (2010) tested memory of emotional words across three bilingual groups (early Spanish–Catalan bilingual speakers who were Catalan dominant, early Spanish–Catalan bilingual speakers who were Spanish dominant, and late Spanish–English bilingual speakers). All three bilingual groups recalled emotional words equally well in both languages, and factors such as age of acquisition, dominance, and similarity between language pairs (Spanish–English vs. SpanishCatalan) did not affect bilingual speakers’ recall ability. Ferré et al. also suggested that, for proficient bilingual speakers, emotional words in the memory task had the same intensity or impact across the L1 and L2, supporting the notion that as L2 proficiency increases, so does the processing equivalence of emotional words across languages. Further support for the relative equivalence in emotional word processing across languages comes from an investigation using the emotional Stroop task with bilingual speakers (Sutton, Altarriba, Gianico, & Basnight-Brown, 2007). The emotional Stroop task requires participants to name the color of ink in which a word is printed (e.g., red, blue, or black ink). Unlike the traditional Stroop task, there is no incongruent condition (e.g., left arrow pointing to the right, or naming blue when it is presented in red ink), and thus no incongruence effect. Instead, participants view both emotional and neutral words (e.g., they may see the word exam written in green ink, and would respond by saying green); emotional words typically result in slower response latencies than neutral words, suggesting that they interfere with the naming of the color ink by virtue of being processed more deeply. Sutton et al. (2007) found that balanced Spanish–English

294 

 Freeman, Shook, and Marian

bilingual speakers’ reaction times to emotional words were slower than reaction times to neutral words regardless of the language of presentation. In other words, the bilingual speakers in Sutton et al. processed emotional words equally across both L1 and L2. These findings are consistent with the Revised Hierarchical Model (Kroll & Stewart, 1994), which would predict that for balanced bilingual speakers, L1 and L2 translation equivalents are strongly connected to a shared conceptual representation, and balanced bilingual speakers can thus access the affective meaning of each word equally. The findings from behavioral research on how bilingual speakers process emotions across both of their languages demonstrate the complexity of emotion processing in bilingual speakers, which is also highlighted within recent electrophysiological research. Conrad, Recio, and Jacobs (2011) examined the emotional valence (i.e., the emotional response an individual may have to a word: positive, negative, or neutral) of German and Spanish words in a visual lexical decision task with two groups of proficient bilingual speakers (native German speakers of Spanish and native Spanish speakers of German) whose skills across their two languages were not equivalent (i.e., unbalanced bilingual speakers). The authors found early posterior negativity (EPN) and late positive complex (LPC) components in both groups, which are believed to index emotional valence (e.g., Kissler, Herbert, Peyk, & Junghofer, 2007; Schacht & Sommer, 2009). However, whereas the EPN and LPC components were present when the bilingual speakers processed both the L1 and the L2, the onset of the EPN component was delayed (~50 ms) when bilingual speakers performed the task in their L2. This response was attributed to the delay in speed of retrieving the word in the L2 and not to qualitative differences in emotion representation in the L1 and L2. An alternative explanation provided by Opitz and Degner (2012) is that not only does the delay in the EPN component result from delayed lexical access to the L2, but also that access to emotionality in the L2 is less automatic and requires additional processing demands. Therefore, we see a diminished effect of a word’s emotionality in the L2. However, supported by the evidence discussed on inhibitory control, as proficiency increases, cognitive control abilities may increase as well, making balanced bilingual speakers more efficient at accessing emotional content in the L2. One way to reconcile conflicting findings of emotion processing across languages is to consider that although emotion processing in the L1 may be more automatic (i.e., occurs more quickly) than in the L2 (Opitz & Degner, 2012; Pavlenko, 2012), the quality of these emotional experiences (e.g., arousal, feelings toward a stimulus) may be similar across both languages (Conrad et al., 2011; Opitz & Degner, 2012). Language experience, specifically proficiency, also appears to play a role in the degree to which emotional words are processed similarly across the L1 and L2, with balanced bilingual speakers processing L1 and L2 emotional words more automatically and efficiently (e.g., Sutton et al., 2007) than unbalanced bilingual speakers (e.g., Opitz & Degner, 2012). The degree to which emotional words are processed equally across languages in bilingual speakers may also impact bilingual speakers’ emotional memory. That is, how deeply the emotional content of an event is encoded may depend on the lan-



Cognitive and Emotional Effects of Bilingualism 

 295

guage used at the time of the event, and this difference in encoding may influence bilingual speakers’ ability to subsequently recall emotional events.

Effects of Bilingualism on Emotional Memory Previous work on memory retrieval in bilingual speakers demonstrates that a memory is better expressed in the language in which it was encoded at the time the event took place (Marian & Fausey, 2006; Marian & Kaushanskaya, 2007; Marian & Neisser, 2000). Not only can the language of retrieval impact how well the event is expressed, but the language of retrieval can also affect the emotional content of the memory. In Marian and Kaushanskaya (2008), Russian–English bilingual participants were asked to describe (via narrative samples) their memories of immigration to the United States in either Russian or English to a Russian–English bilingual experimenter. The results demonstrated that the Russian–English bilingual speakers produced more emotional words in L2 narratives. One possible explanation for this finding could be that the L2 provides more distance from the emotional experience and is thought to be less emotionally laden (Bond & Lai, 1986; Costa et al., 2014; Gonzalez-Reigosa, 1976; Harris, Gleason, & Ayçiçegi, 2006; Keysar, Hayakawa, & An, 2012). As a result, to convey emotion, more emotion words may be needed in L2 (where there is more distance from emotional experience) than in L1 (where there is less distance from emotional experience). In other words, the bilingual speakers in this study may have used more emotion words in their nonnative language because the emotional content of the L2 words was less salient. In addition to differences in emotionality across languages, the results of Marian and Kaushanskaya (2008) were influenced by age of immigration. When speaking Russian, immigrants who came to the United States from Russia early in life produced longer narratives, more positive emotion words, and used more emotional words overall than immigrants who arrived later. That is, bilingual speakers who immigrated earlier reconstructed their memory of the event based off of current experiences with the L2 culture, which tended to be more positive. The early immigrants in this study may have been less connected to their L1 by virtue of having additional years of experience using their L2. More time in an English environment may have resulted in English becoming their preferred language, and the results from Marian and Kaushanskaya demonstrated that the language of preference had an effect on the emotional content in immigrants’ narratives. More broadly, age of immigration affects how language, emotion, and memory interact within the cognitive system.

Effects of Biculturalism on Emotion In addition to the interplay between language, memory, and emotion, language and culture are closely intertwined. When exploring how bilingual experience impacts

296 

 Freeman, Shook, and Marian

emotion, it is necessary to consider cultural influences, as culture imposes norms on how we interpret emotional events. Marian and Kaushanskaya (2004) explored the variability of the type of emotional words used to describe events across languages (and cultures) in Russian–English bilingual immigrants who came to the United States from the former Soviet Union. Participants were asked to describe two separate events (e.g., birthday, snowfall), one in each language. Pronoun usage was used to measure cultural influence: collectivism (i.e., emphasis on the group; former Soviet Union) was measured by the increased use of group pronouns, like we, and individualism (i.e., emphasis on the individual; United States) was measured by the increased use of individual pronouns, like I. The results demonstrated that, consistent with the language-dependent memory hypothesis in which memories are best recalled when the language of encoding and retrieval match (e.g., Marian & Neisser, 2000), bilingual speakers produced narratives that were rated as more emotionally intense when the language of encoding and retrieval matched. In addition, bilingual speakers shifted toward a collectivist nature (use of we, us, our pronouns) when speaking in Russian and shifted toward an individualistic nature (use of I, me, my pronouns) when speaking in English. Because of cultural distinctions (e.g., collectivism vs. individualism), immigrants may experience difficulty reorganizing their sense of self when immigrating to a foreign country in which cultural values are distinct from their native country. This, in turn, may force immigrants to adjust the way they view themselves, and society at large, which could affect not only their interactions with others, but also the way they organize and encode events and emotions. The results from Marian and Kaushanskaya (2004) suggested that the linguistic context and the culture with which one affiliates influence the organization of emotions and retrieval of memories within the cognitive system. Evidence from processing, memory, and cultural effects of bilingualism on emotion demonstrates that the accessibility and processing of emotion is highly complex and influenced by language experience. We suggest that language proficiency, age of L2 onset, match between language of encoding and language of retrieval, and cultural values likely affect bilingual emotional experiences. Because language, emotion, and memory are tightly linked within the cognitive system, we also suggest that language usage drives bilingual speakers to remember events on the basis of linguistic cues, and that memories themselves may store information or cues about the language, and by extension, the culture.

Effects of Bilingualism on the Intersection of Cognitive Control and Emotion Although bilingual experience impacts cognitive control and emotion independently within the cognitive system, experience with multiple languages can also shape how cognitive control and emotion interact with each other. In addition to research from



Cognitive and Emotional Effects of Bilingualism 

 297

Blumenfeld and colleagues (2007, 2011, 2013) suggesting that bilingual speakers activate their languages in parallel, the valence of an emotional word also affects bilingual speakers’ ability to activate or access their L1 translation equivalent when processing that emotional word in L2. Wu and Thierry (2012) found that bilingual speakers experienced restricted access to translation equivalents of emotional words with a negative connotation (e.g., violence), but emotional words with a positive or a neutral connotation (e.g., holiday, theory) resulted in simultaneous access (parallel activation) to translation equivalents. Interestingly, this effect occurred without the participants having to make an explicit judgment on the emotional valence of the words. One possible reason for this difference in emotional valence across the L1 and the L2 could be that when processing a word with negative valence in the L2, late L2 learners (as in Wu & Thierry’s, 2012, study) may inhibit access to the L1 translation equivalent. Words with negative valence may be particularly distressing and subsequently, the full semantic representation of the negative word may only become activated in the L2. More broadly, the results of this study suggest that when bilingual speakers process language, they deploy cognitive control mechanisms that limit cross-linguistic access on the basis of the emotional content of translation equivalents. Further evidence highlighting the interactivity between cognitive control and emotion comes from neural network research on humans and nonhuman animal models. Common to cognitive control and emotion is the ACC, which may be involved in error detection and correction, and is activated during tasks that involve cognitive control and tasks that involve emotion processing (Bush, Luu, & Posner, 2000). In rhesus monkeys, the ventromedial frontal region, known as the limbic area of the frontal cortex, includes the ACC, which receives input from the amygdala (Porrino, Crane, & Goldman-Rakic, 1981). Evidence from these nonhuman primates sheds light on the interactivity of the regions associated with cognitive control (frontal areas, ACC) and emotion (amygdala, ACC), and suggests that the ACC serves as a processor and filter for both cognitive control and emotion. These same brain activation patterns in cognitive control (PFC and ACC; e.g., Abutalebi et al., 2007) and in emotion (amygdala and ACC; e.g., Killgore & Yurgelun-Todd, 2004) have been observed in humans as well. In addition, there is evidence to suggest that the prefrontal cortex (PFC) is activated when individuals are required to control or regulate emotional responses (see Ochsner & Gross, 2005, for a review). Thus, the neural networks in human and nonhuman primates provide additional evidence that cognitive control and emotion are served by the same distributed neural and cognitive mechanisms. It is clear that the neural network subserving cognitive control and emotion is highly integrated in humans, but what remains unclear is how this network is impacted by experience with managing multiple languages. In a language switching task, bilingual speakers activated cognitive control regions, such as the DLPFC, and also activated regions central to emotional processing, such as the amygdala, despite the fact that the stimuli used reflected low or neutral emotional valence (e.g., fruits, furniture; Hernandez, 2009). Activation of the amygdala only occurred when the bilingual

298 

 Freeman, Shook, and Marian

participants were naming pictures in their L1 (Spanish), and not in their L2 (English), despite the participants’ dominant language being English. A possible explanation for this pattern of results, similar to the idea that bilingual speakers process emotional words differently across the L1 and L2, is that bilingual speakers attend to (neutral) words differently in their L1 and L2, and processing neutral words in the L1 may result in automatic activation of an emotional state by virtue of being processed in the L1. Therefore, whereas previous studies have shown the activation of cognitive control regions during emotional tasks and events, Hernandez (2009) demonstrated that bilingual speakers can activate brain areas associated with emotion during an emotionally neutral linguistic–cognitive control task. The underlying cause for activation of overlapping areas involved in cognitive control and emotion in bilingual speakers remains unclear and is subject to future neuroimaging research. However, the evidence discussed in this section clearly suggests that experience with multiple languages affects the interplay between cognitive control and emotion.

Conclusion Bilingual speakers are thought to be “mental jugglers” (Kroll, 2009), constantly having to navigate both of their languages, even under contexts in which only one language is required. The demands imposed by parallel activation and language switching appear to confer benefits to bilingual speakers in both linguistic and nonlinguistic contexts. As reviewed in this chapter, bilingual speakers demonstrate advantages in cognitive control (e.g., Bialystok et al., 2008; Kroll & Bialystok, 2013; Luk, De Sa, & Bialystok, 2011; but see Hilchey & Klein, 2011; Paap & Greenberg, 2013) and lifelong bilingual speakers experience delays the onset of the symptoms of Alzheimer’s disease (Alladi et al., 2013; Bialystok et al., 2007; Craik et al., 2010; Schweizer et al., 2012). Brain pathways and networks that are responsible for cognitive control and language control overlap in bilingual speakers, and the cognitive control network is enhanced and recruited more efficiently in bilingual speakers than in monolingual speakers during language processing (e.g., Abutalebi et al., 2008; Guo et al., 2011). Likewise, eye-tracking research demonstrates that linguistic and nonlinguistic cognitive control are correlated in bilingual speakers (Blumenfeld & Marian, 2011, 2013), suggesting a tight relationship between language and cognition. Research on bilingualism has also revealed a complex relationship between linguistic experience and emotional processing, where a bilingual speaker’s ability to access the emotional content of words or memories depends on the language being used at a given time. For instance, there is mixed evidence as to whether bilingual speakers process emotional content similarly across their languages. In unbalanced bilingual speakers, emotional word processing may be more automatic in the L1 than in the L2 (e.g., Pavlenko, 2012), but how emotion is represented in the brain may be similar across both languages (e.g., Conrad et al., 2011; Opitz & Degner, 2012). In bal-



Cognitive and Emotional Effects of Bilingualism 

 299

anced bilingual speakers, L1 and L2 emotional words may be processed equivalently (Sutton et al., 2007), as direct links to concepts are formed (Kroll & Stewart, 1994). In addition, evidence from bilingual speakers also demonstrates how emotion influences the encoding and retrieval of memories. Marian and Kaushanskaya (2004, 2008) have shown high interactivity between emotion and memory, and experience with multiple languages affects how bilingual speakers form, store, and retrieve memories. Our emotional perception of an event is influenced by the linguistic and cultural context in which the event takes place, as well as the language in which the emotional event is retrieved. We conclude that language, cognitive control, and emotion are intertwined. Indeed, cognitive control and emotion appear to be both filtered through and filtered by language. Our perception of emotional experiences relies on cognitive control to modulate emotional reactions, and may use language to encode and express these reactions. The research explored in this chapter demonstrates that bilingual experience is unique in that knowing multiple languages can have profound effects on the cognitive system, altering the interactivity between cognitive control and emotion, as well as affecting each of these components independently. Experience with multiple languages can change how we think about and interpret daily events in the environment, allowing for more than one channel through which we view the world.

References Abutalebi, J., Annoni, J. M., Zimine, I., Pegna, A. J., Seghier, M. L., Lee-Jahnke, H., . . . Khateb, A. (2008). Language control and lexical competition in bilinguals: An event-related fMRI study. Cerebral Cortex, 18, 1496–1505. http://dx.doi.org/10.1093/cercor/bhm182 Abutalebi, J., Brambati, S. M., Annoni, J. M., Moro, A., Cappa, S. F., & Perani, D. (2007). The neural cost of the auditory perception of language switches: An event-related functional magnetic resonance imaging study in bilinguals. The Journal of Neuroscience, 27, 13762–13769. http:// dx.doi.org/10.1523/JNEUROSCI.3294-07.2007 Abutalebi, J., Canini, M., Della Rosa, P. A., Sheung, L. P., Green, D. W., & Weekes, B. S. (2014). Bilingualism protects anterior temporal lobe integrity in aging. Neurobiology of Aging, 35, 2126–2133. http://dx.doi.org/10.1016/j.neurobiolaging.2014.03.010 Alladi, S., Bak, T. H., Duggirala, V., Surampudi, B., Shailaja, M., Shukla, A. K., . . . Kaul, S. (2013). Bilingualism delays age at onset of dementia, independent of education and immigration status. Neurology, 81, 1938–1944. http://dx.doi.org/10.1212/01.wnl.0000436620.33155.a4 Bartolotti, J., & Marian, V. (2012). Language learning and control in monolinguals and bilinguals. Cognitive Science, 36, 1129–1147. http://dx.doi.org/10.1111/j.1551-6709.2012.01243.x Bialystok, E., Craik, F., & Luk, G. (2008). Cognitive control and lexical access in younger and older bilinguals. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34, 859–873. http://dx.doi.org/10.1037/0278-7393.34.4.859 Bialystok, E., Craik, F. I. M., & Freedman, M. (2007). Bilingualism as a protection against the onset of symptoms of dementia. Neuropsychologia, 45, 459–464. http://dx.doi.org/10.1016/ j.neuropsychologia.2006.10.009

300 

 Freeman, Shook, and Marian

Bialystok, E., Craik, F. I. M., Klein, R., & Viswanathan, M. (2004). Bilingualism, aging, and cognitive control: Evidence from the Simon task. Psychology and Aging, 19, 290–303. Bialystok, E., Martin, M. M., & Viswanathan, M. (2005). Bilingualism across the lifespan: The rise and fall of inhibitory control. The International Journal of Bilingualism, 9, 103–119. http:// dx.doi.org/10.1177/13670069050090010701 Blumenfeld, H. K., & Marian, V. (2007). Constraints on parallel activation in bilingual spoken language processing: Examining proficiency and lexical status using eye-tracking. Language and Cognitive Processes, 22, 633–660. http://dx.doi.org/10.1080/01690960601000746 Blumenfeld, H. K., & Marian, V. (2011). Bilingualism influences inhibitory control in auditory comprehension. Cognition, 118, 245–257. http://dx.doi.org/10.1016/j.cognition.2010.10.012 Blumenfeld, H. K., & Marian, V. (2013). Parallel language activation and cognitive control during spoken word recognition in bilinguals. Journal of Cognitive Psychology, 25, 547–567. http:// dx.doi.org/10.1080/20445911.2013.812093 Bond, M. H., & Lai, T. M. (1986). Embarrassment and code-switching into a second language. The Journal of Social Psychology, 126, 179–186. Bunge, S. A., Dudukovic, N. M., Thomason, M. E., Vaidya, C. J., & Gabrieli, J. D. E. (2002). Immature frontal lobe contributions to cognitive control in children: Evidence from fMRI. Neuron, 33, 301–311. http://dx.doi.org/10.1016/S0896-6273(01)00583-9 Bush, G., Luu, P., & Posner, M. I. (2000). Cognitive and emotional influences in anterior cingulate cortex. Trends in Cognitive Sciences, 4, 215–222. http://dx.doi.org/10.1016/S1364-6613(00)01483-2 Conrad, M., Recio, G., & Jacobs, A. M. (2011). The time course of emotion effects in first and second language processing: A cross-cultural ERP study with German–Spanish bilinguals. Frontiers in Psychology, 2, 351. http://dx.doi.org/10.3389/fpsyg.2011.00351 Costa, A., Foucart, A., Hayakawa, S., Aparici, M., Apesteguia, J., Heafner, J., & Keysar, B. (2014). Your morals depend on language. PLoS ONE, 9, e94842. http://dx.doi.org/10.1371/ journal.pone.0094842 Costa, A., Hernández, M., Costa-Faidella, J., & Sebastián-Gallés, N. (2009). On the bilingual advantage in conflict processing: Now you see it, now you don’t. Cognition, 113, 135–149. http:// dx.doi.org/10.1016/j.cognition.2009.08.001 Costa, A., Hernández, M., & Sebastián-Gallés, N. (2008). Bilingualism aids conflict resolution: Evidence from the ANT task. Cognition, 106, 59–86. http://dx.doi.org/10.1016/ j.cognition.2006.12.013 Costa, A., & Santesteban, M. (2004). Lexical access in bilingual speech production: Evidence from language switching in highly proficient bilinguals and L2 learners. Journal of Memory and Language, 50, 491–511. http://dx.doi.org/10.1016/j.jml.2004.02.002 Craik, F. I. M., Bialystok, E., & Freedman, M. (2010). Delaying the onset of Alzheimer disease: Bilingualism as a form of cognitive reserve. Neurology, 75, 1726–1729. http://dx.doi.org/10.1212/ WNL.0b013e3181fc2a1c Ferré, P., García, T., Fraga, I., Sánchez-Casas, R., & Molero, M. (2010). Memory for emotional words in bilinguals: Do words have the same emotional intensity in the first and in the second language? Cognition and Emotion, 24, 760–785. http://dx.doi.org/10.1080/ 02699930902985779 Gollan, T. H., & Ferreira, V. S. (2009). Should I stay or should I switch? A cost-benefit analysis of voluntary language switching in young and aging bilinguals. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35, 640–665. http://dx.doi.org/10.1037/a0014981



Cognitive and Emotional Effects of Bilingualism 

 301

Gonzalez-Reigosa, F. (1976). The anxiety-arousing effect of taboo words in bilinguals. In C. D. Spielberger & R. Diaz-Guerrero (Eds.), Cross-cultural anxiety (pp. 89–105). Washington, DC: Hemisphere. Green, D. W. (1998). Mental control of the bilingual lexico-semantic system. Bilingualism: Language and Cognition, 1, 67–81. http://dx.doi.org/10.1017/S1366728998000133 Guo, T., Liu, H., Misra, M., & Kroll, J. F. (2011). Local and global inhibition in bilingual word production: fMRI evidence from Chinese–English bilinguals. NeuroImage, 56, 2300–2309. http:// dx.doi.org/10.1016/j.neuroimage.2011.03.049 Harris, C. L., Ayçiçegi, A., & Gleason, J. (2003). Taboo words and reprimands elicit greater autonomic reactivity in a first language than in a second language. Applied Psycholinguistics, 24, 561–571. http://dx.doi.org/10.1017/S0142716403000286 Harris, C. L., Gleason, J. B., & Ayçiçegi, A. (2006). When is a first language more emotional? Psychophysiological evidence from bilingual speakers. In A. Pavlenko (Ed.), Bilingual minds: Emotional experience, expression and representation (pp. 257–283). Clevedon, England: Multilingual Matters. Hernandez, A. E. (2009). Language switching in the bilingual brain: What’s next? Brain and Language, 109, 133–140. http://dx.doi.org/10.1016/j.bandl.2008.12.005 Hernandez, A. E., Martinez, A., & Kohnert, K. (2000). In search of the language switch: An fMRI study of picture naming in Spanish–English bilinguals. Brain and Language, 73, 421–431. http:// dx.doi.org/10.1006/brln.1999.2278 Hernández, M., Costa, A., Fuentes, L. J., Vivas, A. B., & Sebastián-Gallés, N. (2010). The impact of bilingualism on the executive control and orienting networks of attention. Bilingualism: Language and Cognition, 13, 315–325. http://dx.doi.org/10.1017/S1366728909990010 Hilchey, M. D., & Klein, R. M. (2011). Are there bilingual advantages on nonlinguistic interference tasks? Implications for the plasticity of executive control processes. Psychonomic Bulletin & Review, 18, 625–658. http://dx.doi.org/10.3758/s13423-011-0116-7 Keysar, B., Hayakawa, S. L., & An, S. G. (2012). The foreign-language effect: Thinking in a foreign tongue reduces decision biases. Psychological Science, 23, 661–668. http://dx.doi.org/10.1177/ 0956797611432178 Killgore, W. D., & Yurgelun-Todd, D. A. (2004). Activation of the amygdala and anterior cingulate during nonconscious processing of sad versus happy faces. NeuroImage, 21, 1215–1223. http://dx.doi.org/10.1016/j.neuroimage.2003.12.033 Kissler, J., Herbert, C., Peyk, P., & Junghofer, M. (2007). Buzzwords: Early cortical responses to emotional words during reading. Psychological Science, 18, 475–480. http://dx.doi.org/10.1111/ j.1467-9280.2007.01924.x Kroll, J. F. (2009). The consequences of bilingualism for the mind and the brain. Psychological Science in the Public Interest, 10, i–ii. http://dx.doi.org/10.1177/1529100610389314 Kroll, J. F., & Bialystok, E. (2013). Understanding the consequences of bilingualism for language processing and cognition. Journal of Cognitive Psychology, 25, 497–514. http://dx.doi.org/ 10.1080/20445911.2013.799170 Kroll, J. F., Bobb, S. C., Misra, M., & Guo, T. (2008). Language selection in bilingual speech: Evidence for inhibitory processes. Acta Psychologica, 128, 416–430. http://dx.doi.org/10.1016/ j.actpsy.2008.02.001 Kroll, J. F., & Stewart, E. (1994). Category interference in translation and picture naming: Evidence for asymmetric connections between bilingual memory representations. Journal of Memory and Language, 33, 149–174. http://dx.doi.org/10.1006/jmla.1994.1008

302 

 Freeman, Shook, and Marian

Linck, J. A., Hoshino, N., & Kroll, J. F. (2008). Cross-language lexical processes and inhibitory control. The Mental Lexicon, 3, 349–374. http://dx.doi.org/10.1075/ml.3.3.06lin Luk, G., Anderson, J. A. E., Craik, F. I. M., Grady, C., & Bialystok, E. (2010). Distinct neural correlates for two types of inhibition in bilinguals: Response inhibition versus interference suppression. Brain and Cognition, 74, 347–357. http://dx.doi.org/10.1016/j.bandc.2010.09.004 Luk, G., Bialystok, E., Craik, F. I. M., & Grady, C. L. (2011). Lifelong bilingualism maintains white matter integrity in older adults. The Journal of Neuroscience, 31, 16808–16813. http://dx.doi.org/ 10.1523/JNEUROSCI.4563-11.2011 Luk, G., De Sa, E., & Bialystok, E. (2011). Is there a relation between onset age of bilingualism and enhancement of cognitive control? Bilingualism: Language and Cognition, 14, 588–595. http:// dx.doi.org/10.1017/S1366728911000010 Marian, V., & Fausey, C. M. (2006). Language-dependent memory in bilingual learning. Applied Cognitive Psychology, 20, 1025–1047. http://dx.doi.org/10.1002/acp.1242 Marian, V., & Kaushanskaya, M. (2004). Self-construal and emotion in bicultural bilinguals. Journal of Memory and Language, 51, 190–201. http://dx.doi.org/10.1016/j.jml.2004.04.003 Marian, V., & Kaushanskaya, M. (2007). Language context guides memory content. Psychonomic Bulletin & Review, 14, 925–933. http://dx.doi.org/10.3758/BF03194123 Marian, V., & Kaushanskaya, M. (2008). Words, feelings, and bilingualism: Cross-linguistic differences in emotionality of autobiographical memories. The Mental Lexicon, 3, 72–90. http:// dx.doi.org/10.1075/ml.3.1.06mar Marian, V., & Neisser, U. (2000). Language-dependent recall of autobiographical memories. Journal of Experimental Psychology: General, 129, 361–368. http://dx.doi.org/10.1037/ 0096-3445.129.3.361 Marian, V., & Spivey, M. (2003a). Bilingual and monolingual processing of competing lexical items. Applied Psycholinguistics, 24, 173–193. http://dx.doi.org/10.1017/S0142716403000092 Marian, V., & Spivey, M. (2003b). Competing activation in bilingual language processing: Withinand between-language competition. Bilingualism: Language and Cognition, 6, 97–115. http:// dx.doi.org/10.1017/S1366728903001068 Martín, M. C., Macizo, P., & Bajo, T. (2010). Time course of inhibitory processes in bilingual language processing. British Journal of Psychology, 101, 679–693. http://dx.doi.org/10.1348/ 000712609X480571 Martin-Rhee, M. M., & Bialystok, E. (2008). The development of two types of inhibitory control in monolingual and bilingual children. Bilingualism: Language and Cognition, 11, 81–93. http:// dx.doi.org/10.1017/S1366728907003227 Meuter, R., & Allport, A. (1999). Bilingual language switching in naming: Asymmetrical costs of language selection. Journal of Memory and Language, 40, 25–40. http://dx.doi.org/10.1006/ jmla.1998.2602 Misra, M., Guo, T., Bobb, S. C., & Kroll, J. F. (2012). When bilinguals choose a single word to speak: Electrophysiological evidence for inhibition of the native language. Journal of Memory and Language, 67, 224–237. http://dx.doi.org/10.1016/j.jml.2012.05.001 Ochsner, K. N., & Gross, J. J. (2005). The cognitive control of emotion. Trends in Cognitive Sciences, 9, 242–249. http://dx.doi.org/10.1016/j.tics.2005.03.010 Odlin, T. (2005). Cross-linguistic influence and conceptual transfer: What are the concepts? Annual Review of Applied Linguistics, 25, 3–25. http://dx.doi.org/10.1017/S0267190505000012 Opitz, B., & Degner, J. (2012). Emotionality in a second language: It’s a matter of time. Neuropsychologia, 50, 1961–1967. http://dx.doi.org/10.1016/j.neuropsychologia.2012.04.021



Cognitive and Emotional Effects of Bilingualism 

 303

Paap, K. R., & Greenberg, Z. I. (2013). There is no coherent evidence for a bilingual advantage in executive processing. Cognitive Psychology, 66, 232–258. http://dx.doi.org/10.1016/ j.cogpsych.2012.12.002 Pavlenko, A. (2008). Emotion and emotion-laden words in the bilingual lexicon. Bilingualism: Language and Cognition, 11, 147–164. http://dx.doi.org/10.1017/S1366728908003283 Pavlenko, A. (2012). Affective processing in bilingual speakers: Disembodied cognition? International Journal of Psychology, 47, 405–428. http://dx.doi.org/10.1080/00207594.2012.743665 Porrino, L. J., Crane, A. M., & Goldman-Rakic, P. S. (1981). Direct and indirect pathways from the amygdala to the frontal lobe in rhesus monkeys. The Journal of Comparative Neurology, 198, 121–136. http://dx.doi.org/10.1002/cne.901980111 Ridderinkhof, K. R., Ullsperger, M., Crone, E. A., & Nieuwenhuis, S. (2004). The role of the medial frontal cortex in cognitive control. Science, 306, 443–447. http://dx.doi.org/10.1126/ science.1100301 Schacht, A., & Sommer, W. (2009). Emotions in word and face processing: Early and late cortical responses. Brain and Cognition, 69, 538–550. http://dx.doi.org/10.1016/j.bandc.2008.11.005 Schroeder, S. R., & Marian, V. (2012). A bilingual advantage for episodic memory in older adults. Journal of Cognitive Psychology, 24, 591–601. http://dx.doi.org/10.1080/20445911.2012.669367 Schweizer, T. A., Ware, J., Fischer, C. E., Craik, F. I. M., & Bialystok, E. (2012). Bilingualism as a contributor to cognitive reserve: Evidence from brain atrophy in Alzheimer’s disease. Cortex, 48, 991–996. http://dx.doi.org/10.1016/j.cortex.2011.04.009 Sutton, T. M., Altarriba, J., Gianico, J. L., & Basnight-Brown, D. M. (2007). The automatic access of emotion: Emotional Stroop effects in Spanish–English bilingual speakers. Cognition and Emotion, 21, 1077–1090. http://dx.doi.org/10.1080/02699930601054133 Wu, Y. J., & Thierry, G. (2012). How reading in a second language protects your heart. The Journal of Neuroscience, 32, 6485–6489. http://dx.doi.org/10.1523/JNEUROSCI.6119-11.2012

Hilary D. Duncan and Natalie A. Phillips

16 The Contribution of Bilingualism to Cognitive Reserve in Healthy Aging and Dementia This chapter reviews how speaking more than one language may mitigate the cognitive decline often seen in aging, and possibly delay the onset of dementia. In the first section, we review key concepts, including cognitive reserve and the bilingual benefit, and outline the cognitive decline commonly seen in aging in a discussion about how the bilingual benefit may contribute to cognitive reserve to help offset age-related cognitive decline. In the second section, which deals with bilingualism in healthy aging, we review studies examining the differences in language, cognition, and memory between healthy older monolingual and bilingual adults. We then outline the few studies that look at the potential neuroanatomical differences between older monolingual and bilingual adults, and large cohort studies of healthy older adults. In the final section of the chapter, we synthesize and analyze the research in a relatively new area of interest, namely, the study of the effects of bilingualism on the development of dementia. This chapter demonstrates that the study of cognitive and brain differences between monolingual and bilingual adults not only contributes to our understanding of the mechanisms of the bilingual brain but also adds to our growing understanding of the concept of cognitive reserve and how cognitive reserve might play a role in healthy and pathological aging. Additionally, as demonstrated in other chapters in this volume and by others (e.g., Luk & Bialystok, 2013), bilingualism is a multifaceted phenomenon that can be defined in many ways, and varies across individuals, groups, and cultures. As such, it is important to keep in mind when reviewing the following research that the studies vary in terms of the breadth and depth of their assessment of important language-related variables. This is a topic that resurfaces in each part of the chapter. We try to pay specific attention to how a number of variables are assessed and whether they impact the overall pattern of results. One set of these variables is related to bilingualism and language group assessment. This set includes factors like age of acquisition, proficiency, the time spent using the languages, and the amount of switching between two languages. Another set consists of variables that have been shown to contribute to cognitive reserve, like education. Finally, we consider variables that are related to bilingualism, but not language per se, such as immigration status or language context. This work was supported by a doctoral fellowship from the Alzheimer Society of Canada and by Grant 203751 from the Natural Sciences and Engineering Research Council of Canada. http://dx.doi.org/10.1037/14939-017 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

306 

 Duncan and Phillips

Key Concepts Cognitive Reserve The concept of cognitive reserve arose after the repeated finding that the degree of damage to the brain does not always correlate with functional and cognitive abilities. In other words, two people with similar levels of brain damage or pathology do not necessarily function at the same level (e.g., Stern, 2009). For example, in one study, about 45% of older adults were found on autopsy to have evidence of Alzheimer’s disease pathology in their brains, although they appeared to have had normal cognitive function while living (Bennett et al., 2006). Briefly, the theory of cognitive reserve states that individuals with more cognitive reserve (e.g., people with higher IQ, more years of education, or those who participate in mentally or physically stimulating activities) are able to maintain the same level of functional or cognitive performance compared with those with less cognitive reserve, despite having greater amounts of brain pathology or age-related brain changes (Barulli & Stern, 2013). A number of activities appear to protect the brain against the effects of aging, including evidence for the benefits of late life recreational activity (e.g., Brewster et al., 2014), higher levels of education (e.g., Bennett, Schneider, Wilson, Bienias, & Arnold, 2005), cognitively stimulating activities (e.g., Wilson et al., 2013), and social engagement (e.g., Engelhardt, Buber, Skirbekk, & Prskawetz, 2010). There are two hypothesized ways in which cognitive reserve mechanisms might function: neural reserve and neural compensation (Stern, 2009). Neural reserve refers to the differences between healthy, nonimpaired individuals in the strength or efficiency of the cognitive networks set up in their brain. Thus, activities contributing to cognitive reserve could train brain networks used for completing a task, making a network more efficient and effective (neural reserve). In contrast, neural compensation refers to how individuals use alternate brain networks to compensate when their brain is weakened by disease, such as Alzheimer’s disease. Thus, if one network sustains damage, then a network that has been strengthened through cognitive reserve could hypothetically be relied on, allowing a person to use compensatory mechanisms. Research has shown that when matched on Clinical Dementia Rating scores, older adults with higher socioeconomic status (SES) have reduced brain volume and accelerated brain atrophy than those with lower SES (Fotenos, Mintun, Snyder, Morris, & Buckner, 2008). Although these results may seem initially counterintuitive, they are in line with the cognitive reserve theory because they indicate that those persons with higher cognitive reserve (i.e., higher SES) are able to cope longer with brain pathology (i.e., greater brain atrophy) before they begin to show signs of cognitive deficit or succumb to dementia. The inverse of this hypothesis also appears to be true; in a group of older adults matched on brain atrophy, those with higher cognitive reserve will have better cognitive and functional performance. For example, when balanced on brain atrophy, older adults



Bilingualism in Healthy Aging and Dementia 

 307

with higher levels of education showed better memory performance, and older adults with higher occupational attainment showed better memory and reasoning performance (Staff, Murray, Deary, & Whalley, 2004).

The Theory of the Bilingual Benefit How might being bilingual relate to cognitive reserve? Speaking more than one language may be similar to other mentally stimulating activities, and therefore could be a contributor to cognitive reserve and protect against age-related decline and the onset of dementia. But what exactly is it about speaking two languages that might exercise the brain? As outlined in Chapter 15 of this volume, research suggests that a bilingual speaker’s two languages are active when completing word recognition or language production tasks, even when only one language is required. The simultaneous activation of languages means that bilingual speakers will hold two lexical representations in mind and will require inhibitory or control mechanisms to manage the competition between their languages (e.g., Green, 1998). This is believed to benefit other aspects of cognition, particularly components of executive functioning known as attention control and inhibitory control.

Cognitive Reserve and Bilingualism in Aging During normal aging, the brain is most vulnerable to atrophy of the prefrontal lobes (Good et al., 2001; Raz et al., 1997), which support executive functioning. Older adults perform more poorly than younger adults on many components of executive functioning, including tests of attention control and inhibition (e.g., Belleville, Rouleau, & Van der Linden, 2006; Sylvain-Roy & Belleville, 2015), planning (e.g., Sorel & Pennequin, 2008), and set-shifting (e.g., Crawford, Bryan, Luszcz, Obonsawin, & Stewart, 2000; Goffaux et al., 2008). This area is particularly important as executive functions predict functional living skills in cognitively healthy elderly adults and patients with dementia (Pereira, Yassuda, Oliveira, & Forlenza, 2008). Furthermore, good executive functioning can allow an older adult to remain independent even when suffering from other forms of cognitive loss (e.g., Cahn-Weiner, Malloy, Boyle, Marran, & Salloway, 2000). In relation to the cognitive reserve theory, this would mean that older bilingual adults, who have spent years exercising these hypothesized control mechanisms, would build up neural reserve and demonstrate more preserved executive functioning than older monolingual adults. Additionally, for older adults who develop dementia, their experience dealing with two languages could then be an example of neural compensation; that is, the brain areas and cognitive functions typically affected by dementia (i.e., hippocampus, temporal lobes, and memory functions) might be supported by their superior executive functioning capabilities.

308 

 Duncan and Phillips

The Effects of Bilingualism on Aging Attention and Executive Functioning Attention control, inhibition, and other aspects of executive functioning have been the most studied areas with respect to the effects of bilingualism on cognitive functioning. Numerous studies have shown a bilingual advantage on executive control tasks (Bialystok, 1999; Bialystok & Martin, 2004; Costa, Hernández, & SebastiánGallés, 2008) with the magnitude of the advantage increasing from adulthood into older age (Bialystok, Craik, Klein, & Viswanathan, 2004; Bialystok, Craik, & Luk, 2008; Bialystok, Craik, & Ruocco, 2006; Bialystok, Craik, & Ryan, 2006). Typically, executive control tasks examine reaction time and accuracy during tasks that contain congruent (low-conflict) trials and incongruent (high-conflict) trials. For example, during the Stroop task, participants are typically slower to name incongruent ink colors than congruent colors (i.e., the Stroop effect). Older bilingual adults showed a smaller Stroop effect relative to their monolingual counterparts, supporting a bilingual advantage in executive control of attention (Bialystok et al., 2008). However, this finding has not always been replicated (Kousaie & Phillips, 2012), and it has been suggested that the apparent bilingual advantage would be better termed a more general executive benefit, rather than a specific benefit on attention control or inhibition (see studies reviewed in Hilchey & Klein, 2011). A more recent study using multiple executive functioning tasks also found no reliable advantage for older bilingual speakers compared with older monolingual speakers, and suggested that the discrepancy between studies in the field in terms of finding a bilingual advantage may be related to language variables outside of the dichotomous monolingual versus bilingual classification (Kousaie, Sheppard, Lemieux, Monetta, & Taler, 2014). As outlined in the introduction to this chapter, and as noted by Kousaie and Phillips (2012), a number of issues remain in this field of research. For instance, it is not clear whether executive functioning advantages are confined to bilingual speakers with high levels of proficiency in both their languages, those with many years of speaking two languages, those who live in a context that requires them to switch between languages frequently, or those who speak specific languages that may increase the need for cognitive control. Furthermore, very few of the studies looking at cognitive control have addressed the many nonlanguage variables that could be related to being bilingual, such as being an immigrant, one’s level of social engagement, and participation in cultural life and education. Note that these later factors could contribute more directly to cognitive reserve, independent of their relationship with being bilingual.

Memory Surprisingly, given the significance of memory functions in aging and dementia, and its purported relationship with executive functioning (e.g., Troyer, Graves, & Cullum,



Bilingualism in Healthy Aging and Dementia 

 309

1994), there is a dearth of literature examining whether a bilingual advantage may be seen in episodic memory. Episodic memory is a type of long-term memory for past experiences that can be recalled as having occurred at a particular time and place. Impairment in this cognitive function is one of the hallmarks of Alzheimer’s disease. Schroeder and Marian (2012) tested a group of older bilingual adults from a variety of cultural backgrounds, and for whom the language of testing (English) was a second language. In a picture recall task, older bilingual adults recalled more pictures than did the older monolingual adults, and those with an earlier age of second language acquisition and more experience with a second language had better recall. The bilingual participants in this study spoke a variety of native languages (e.g., Bengali, French, German, Gujarati), with the majority (72%) being more proficient in their first language; however, the majority also indicated that English was currently their most frequently used language. This study could provide evidence toward the theory that bilingualism contributes to cognitive reserve in the domain of memory for those who must frequently speak their less proficient language. Unfortunately, this study does not give information on the immigration status of the bilingual adults, making it difficult to assess whether the results are a benefit of bilingualism, or, instead, are due to immigration-related factors. A cross-sectional study by Ljungberg, Hansson, Andrés, Josefsson, and Nilsson (2013) using only Swedish nonimmigrants found that bilingual adults had better verbal episodic memory than monolingual adults in middle adulthood and older age when all were tested in their first language. However, the difference between the two language groups did not increase in the older age groups. Unlike many other studies, the bilingual adults in this group were culturally similar to their monolingual counterparts (all were born in Sweden). However, they also differed from bilingual adults used in many other research studies in that they learned their second language through formal training, the majority spent only 0 to 2 hours a day on average using their second language, and most (64%) only spoke the second language when traveling (Ljungberg et al., 2013). Thus, the participants’ level of second language proficiency and their use of their second language might have been low. As such, although the study addressed the potentially confounding issue of immigration, the definition of bilingualism differs from the majority of studies assessing the bilingual benefit, and would be more aptly categorized as a study assessing the benefits of learning a second language. Finally, Wodniecka, Craik, Luo, and Bialystok (2010) conducted two studies comparing groups of older, mostly immigrant, bilingual adults with English as a second language, versus older, nonimmigrant, English-speaking monolingual adults on verbal and nonverbal memory tasks and found mixed results. In Study 1, they found moderate support for a bilingual advantage on recollection of faces but a monolingual advantage for recollection of verbal material. Study 2 found no difference between a different group of monolingual and bilingual adults for recollection of abstract objects, and a bilingual advantage for recollection of verbal material. Notably, the bilingual adults in Study 1 had significantly lower vocabulary scores than the monolingual adults, and showed a disadvantage on the verbal memory task,

310 

 Duncan and Phillips

whereas the bilingual adults in Study 2 had higher vocabulary scores and scored better than monolingual adults on the verbal memory task. As such, it is difficult to declare this study unequivocal evidence for a bilingual benefit in episodic memory, as it could be that those with higher vocabulary scores, regardless of their language group, demonstrate better verbal memory. With such mixed results, and so few studies to draw from, it is not yet possible to conclude whether bilingualism confers an advantage on memory processing. Additionally, as with studies assessing the effects of bilingualism on cognition, it is difficult to untangle which aspects of bilingualism may contribute to any advantage (i.e., immigration status, age of acquisition, time spent speaking a second language). However, given that episodic memory declines with age, and episodic memory impairment is the earliest and most significant change observed in early Alzheimer’s disease, this area is an important one for future study.

Language Bilingualism does not confer advantages in all cognitive abilities. Many studies of vocabulary knowledge report that bilingual speakers score lower in each of their languages than monolingual speakers of that language (e.g., Bialystok, Luk, Peets, & Yang, 2010). This deficit is found at all ages, including older adulthood (Bialystok & Luk, 2012). Older bilingual speakers, similar to younger bilingual speakers, are also disadvantaged on naming tasks (Bialystok et al., 2008). When it comes to lexical access and retrieval, the impact of bilingualism appears to differ with age. For example, younger bilingual speakers generate fewer words in phonemic and semantic fluency than do younger monolingual speakers (Gollan, Montoya, & Werner, 2002); however, older bilingual speakers generate fewer category words than older monolingual speakers but perform similarly on letter fluency (Rosselli et al., 2000). Additionally, older monolingual speakers have significantly lower phonemic and semantic fluency scores than younger monolingual speakers, whereas older bilingual speakers have only lower semantic scores compared with their younger counterparts (Gollan et al., 2002). Importantly, the degree of similarity between a bilingual speaker’s two languages could have an effect on lexical retrieval as phonologically similar words from the two languages interfere with each other. The majority of studies assessing verbal fluency or lexical access in bilingual speakers have been conducted with Spanish–English bilingual speakers (e.g., Gollan et al., 2002) or English monolingual speakers compared with mixed groups of second-language English bilingual speakers (e.g., Bialystok et al., 2008). A study examining level of bilingualism (as a ratio between native Marathi and second-language Hindi proficiency) indicated that a higher level of bilingualism was associated with better phonemic and semantic verbal fluency in Marathi (Kamat et al., 2012). The authors of this study suggested that the high level of cognates (words that have a common root or origin, like English hospital and French hôpital)



Bilingualism in Healthy Aging and Dementia 

 311

between the two languages may mediate the relationship between bilingualism and verbal fluency.

Neuroanatomical Another avenue for investigating the potential impact of bilingualism on cognitive decline in aging is to directly examine brain structure to assess whether speaking a second language shapes the brain. A large area of research is devoted to analyzing how bilingual speakers process language compared with monolinguals using brain imaging techniques such as fMRI (for reviews, see Luk, Green, Abutalebi, & Grady, 2011; Perani & Abutalebi, 2005); however, there is less research examining brain differences in relation to cognitive abilities, and even fewer studies that look at the effect in the aging brain. A study examining white matter in older monolingual and bilingual adults found that lifelong bilingual adults had higher white matter integrity than monolingual adults in the corpus callosum, which is an important structure for interhemispheric communication (Luk, Bialystok, Craik, & Grady, 2011). The authors posited that being bilingual contributes to cognitive reserve by helping maintain white matter integrity in aging. However, another group of researchers found the exact opposite finding—older bilingual adults had lower corpus callosum white matter integrity than a group of monolingual adults matched on age and cognitive functioning (Gold, Johnson, & Powell, 2013). The authors in this case posited that, in line with the cognitive reserve hypothesis, older bilingual adults were able to perform at the same level as older monolingual adults despite more age-related brain damage. There are multiple possibilities for the discrepancy between these two studies. For example, although both studies attempted to account for language variables such as proficiency (in fact, the studies used the same criteria to define bilingualism), only one of the studies gives information on immigration status. In the Luk, Bialystok, et al. (2011) study, over half of the bilingual group were immigrants, compared with 14% of the monolingual group, whereas this information is not given in the Gold et al. (2013) study. More important, however, the older adults in the Luk, Bialystok, et al. (2011) study were significantly older than those examined by Gold et al. (2013). According to the cognitive reserve hypothesis, it is possible that the “younger” older bilingual adults tested by Gold et al. (2013) were demonstrating evidence of the structural effects of cognitive reserve, prior to the onset of significant age-related brain atrophy (i.e., neural reserve), whereas the “older” older bilingual adults in the Luk, Bialystok, et al. (2011) study were demonstrating the protective effects of cognitive reserve (i.e., neural compensation), after the onset of significant age-related brain atrophy. Finally, Abutalebi et al. (2014) found that older bilingual adults had more gray matter volume than age-matched monolingual adults in the left anterior temporal pole, an area that is hypothesized to be involved in lexical retrieval and is activated by both of a bilingual speaker’s languages. Unfortunately, the monolingual adults in this study spoke

312 

 Duncan and Phillips

Italian and were from Italy, whereas the bilingual adults spoke English and Cantonese and were from Hong Kong, offering a number of other potential causes for the difference in gray matter (i.e., health behavior, diet). They did, however, find that within the bilingual group, gray matter volume was significantly associated with naming performance in the second language, suggesting that the proficiency in a second language may be related to brain structure.

Cohort Studies A cohort study is a type of longitudinal research that follows a large group of healthy people and uses correlational analyses to determine factors associated with particular outcomes. In the study of bilingualism, the benefit of a cohort study is that it often takes advantage of previously collected data being used for epidemiological purposes and allows one to examine a larger number of participants than would be possible with a typical experimental study. The difficulty with using cohort data is that, given the broad range of measures often collected, the measurement of any one variable may not be particularly in-depth (e.g., education, language proficiency, immigration status), and variables are measured retrospectively, such as obtaining language-related information via self or caregiver report rather than objective assessment. Nonetheless, the data from two large-scale cohort studies can shed light on whether being bilingual affects cognitive decline in older adults. Kavé, Eyal, Shorek, and Cohen-Mansfield (2008) examined 814 older adults from a representative sample of the Israeli Jewish population. They found that the number of languages spoken correlated positively with cognitive screening test scores beyond the effect of other demographic variables, such as age, gender, place of birth, age at immigration, or education. However, this study did not contain any monolingual adults; all participants spoke Hebrew and at least one other language; therefore, individuals who may have had particularly weak language abilities, having been unable to learn Hebrew in a Hebrew-dominant country, were excluded. Additionally, the authors were able to circumvent the potentially confounding effects of education by examining a subgroup of noneducated older adults, where they also found that the number of languages spoken predicted cognitive functioning. In this study, participants self-reported which languages they spoke, which ones they spoke at home, and which language they were most comfortable speaking; however, the amount of time speaking each language, age of acquisition, and proficiency were not assessed. A cohort study by Bak, Nissan, Allerhand, and Deary (2014) assessed later-life cognition in 853 older English native speakers, 262 of whom reported speaking a second language. All participants were born and raised in Edinburgh, Scotland, eliminating any potential effects of immigration or cultural variables. This study found that older bilingual adults outperformed age-matched monolingual adults on general intelligence and reading. This finding differs from previously reviewed research, which



Bilingualism in Healthy Aging and Dementia 

 313

has typically found any benefit to be in executive functioning and not general intelligence, or language-related tasks. The bilingual adults in this study, however, would not have met criteria for bilingualism in many of the previously reviewed studies. For example, the bilingual adults were English native speakers born and living in an officially English-speaking country. One quarter of them did not learn their second language until after the age of 18, and 65% were not using their second language in their everyday life. Similar to the Ljungberg et al. (2013) study reviewed in the memory section, this study appears to demonstrate that learning a second language (rather than being bilingual) could contribute to cognitive reserve in aging (Bak et al., 2014). These three studies demonstrate the difficulties of conducting large-scale cohort studies to examine the effects of bilingualism on cognition in aging. Specifically, because it is not clear what particular aspect of bilingualism may contribute to the hypothesized benefit, it is important to attempt to assess a number of relevant variables (e.g., language proficiency, age of acquisition) and to take into account variables known to affect cognition (e.g., education). Furthermore, as the field advances, certain issues are arising that research indicates may be relevant (e.g., immigration status). Given that no individual study is capable of measuring every potentially relevant variable, we must be cautious when interpreting the findings of single studies. This issue is particularly relevant for the studies discussed in the following section, those that examine the effects of bilingualism on dementia.

The Benefits of Bilingualism in Dementia Dementia is a general term for a decline in mental ability that is severe enough to interfere with daily life. Dementia is not a specific disease but a characteristic of a number of diseases and disorders. The most common form of dementia is Alzheimer’s disease, but other forms include frontotemporal dementia, Lewy body dementia, and vascular dementia. Alzheimer’s disease is a progressive, late-life neurodegenerative disorder, characterized by problems with long-term memory, executive functions, attention, and behavior. Given the aging population, Alzheimer’s disease is a significant health concern. An estimated 500,000 Canadians (Alzheimer Society of Canada, 2010), and 36 million people worldwide (Alzheimer’s Disease International, 2010) currently have Alzheimer’s disease, and it is estimated that within a generation this number will more than triple, reaching 115 million people (Alzheimer’s Disease International, 2010). There is currently no cure for Alzheimer’s disease; thus, most efforts focus on preventing or delaying the symptoms of the disease. Research has shown that those with greater cognitive reserve are likely to go more years without experiencing cognitive symptoms compared with patients with lower cognitive reserve (e.g., Liu, Cai, Xue, Zhou, & Wu, 2013; Querbes et al., 2009). Recent research suggests that bilingualism may protect against the onset of dementia in Alzheimer’s disease and related disorders. The general outcome of this line

314 

 Duncan and Phillips

of research is mixed, with some studies providing evidence for a positive effect of bilingualism on later life outcomes (e.g., see studies by Alladi et al., 2013; Bialystok, Craik, & Freedman, 2007), some finding mixed results that depend specifically on immigrant status or the possible nature of the first language–second language relationship (e.g., Chertkow et al., 2010) or education level (Gollan, Salmon, Montoya, & Galasko, 2011), and some studies finding no effect of language group (e.g., Brewster et al., 2014; Crane et al., 2009; Yeung, St. John, Menec, & Tyas, 2014). A number of the studies will be reviewed in more detail, to illustrate for the reader some of the methods used and the challenges faced by researchers in this area. Some of the issues that will be discussed include the limited assessment of other variables that may be related to bilingualism (e.g., immigration status), the assessment of language group (e.g., self-report compared with objective testing of language ability), and the limitations imposed by retrospective versus prospective studies. These studies are broadly divided into three sections: (a) retrospective studies that look at the estimated age at onset of dementia symptoms and age at diagnosis in dementia patients, (b) studies that follow large cohorts of healthy older adults and document which language group has the highest rate of conversion to dementia, and (c) experimental studies comparing cognitive performance or neurophysiological measures of monolingual and bilingual adults with dementia. The pioneering study on the effects of bilingualism on dementia was conducted by Bialystok et al. (2007), who examined the association between bilingualism and age at diagnosis of dementia. The monolingual and bilingual groups did not differ in terms of cognitive abilities (as assessed by a commonly used short cognitive screening exam called the Mini-Mental State Examination [MMSE]) and were matched on occupational status (as assessed by a five-point scale by Human Resources and Skills Development Canada). Bilingual patients had significantly fewer years of education (10.8 years vs. the monolingual group mean of 12.4 years). Finally, a panel of judges (with experience in bilingualism research) decided the language group to which each participant belonged. Importantly, results indicated that the bilingual dementia patients had an onset of symptoms 4.1 years later than the monolingual dementia patients, and visited the clinic for the first time 3.2 years later than the monolingual group. Although the patient group was mixed (i.e., composed of those with frontotemporal dementia, dementia with Lewy bodies, etc.), this major finding was also significant for the subset of patients with probable Alzheimer’s disease: Bilingual Alzheimer’s disease patients had an age of onset of symptoms that was 4.3 years later than monolingual patients (although information on MMSE, education, and occupational status for the Alzheimer’s disease subgroup is not given). This study provides an example of how difficult it is to parse out variables related to bilingualism: Bilingual status was confounded with immigrant status, as 81 of the 93 bilingual patients, (compared with 13 of the 91 monolingual patients) were immigrants. As mentioned earlier in the chapter, it is possible that being an immigrant could contribute to cognitive reserve through a number of different manners (e.g., diet, health behaviors).



Bilingualism in Healthy Aging and Dementia 

 315

However, the study was the first of its kind (to explore a potential beneficial impact of bilingualism on dementia), and the results indicated that the issue certainly warranted further exploration.

Immigration As discussed in an editorial by Fuller-Thomson, and Kuh (2014), there is an emerging body of evidence suggesting that immigrants have better health and cognitive outcomes than nonimmigrants. To assess the possible influence of immigration status on dementia symptoms, or its interaction with bilingualism, Chertkow et al. (2010) examined the age at diagnosis of Alzheimer’s disease and age at symptom onset for a cohort of 632 monolingual, bilingual, and multilingual (three or more languages) participants. This research took place in Montreal, Canada, where bilingualism and multilingualism are common. Residents can be bilingual for different reasons, with some being nonimmigrant native Canadians speaking the two official languages of Canada, others being immigrants who have a native language and have learned English and/or French. According to this study, for the group as a whole, those who spoke three or more languages (but not bilingual speakers per se) experienced a protective effect in relation to age at diagnosis or age at symptom onset, but there was no significant effect of speaking only two languages (i.e., bilingualism). However, when the analyses were limited to the immigrant patients, being either bilingual or multilingual delayed the diagnosis of Alzheimer’s disease by almost 5 years. Interestingly, there was a trend toward the same effect in nonimmigrant bilingual adults whose first language was French, but not for those whose first language was English. Thus, this study showed that the relationship between the number of languages spoken and its impact on the onset of symptoms is not straightforward and may interact with important factors such as immigration, or, in the case of the native French versus native English analyses, perhaps culture-related or language-use variables. One way to examine the effects of bilingualism on dementia without conflating it with immigration is to look at bilingualism in a group of nonimmigrants. Crane and colleagues (2009) followed a large group of healthy Japanese American men who were born in Hawaii and recorded who became diagnosed with dementia. All of the patients were second generation (i.e., their parents had immigrated), which eliminates any possible direct effect of immigration status. Additionally, this large cohort study overcomes the downfalls of the previously reviewed retrospective chart studies, which relied on assessing age of symptom onset and age at diagnosis retrospectively. The sample of 2,299 men were asked to rate their written and spoken Japanese abilities; all of them were fluent in English, but their abilities in Japanese varied from none at all to completely fluent. The study showed that there was no significant difference in the prevalence of dementia (Alzheimer’s disease or vascular dementia) between monolingual (English) speakers and bilingual speakers. Similarly, Sanders, Hall,

316 

 Duncan and Phillips

Katz, and Lipton (2012) found no benefit of bilingualism in a large cohort of fluent English speakers for those with low (0–11 years) or intermediate (12–15 years) levels of education.

Assessment of Language Group Variables Studies of bilingualism in healthy aging tend to use experimental designs, and the assessment of the level or degree of bilingualism is often more thorough than in the large cohort studies reviewed here (i.e., Chertkow et al., 2010; Crane et al., 2009). Many of the papers discussed used self- or caregiver report to assess bilingualism. Some of the studies did not collect any data on fluency, proficiency, or age of acquisition (e.g., Alladi et al., 2013; Chertkow et al., 2010), others collected data on language fluency but did not report its effects on conversion to dementia (i.e., Bialystok et al., 2007; Ossher, Bialystok, Craik, Murphy, & Troyer, 2013; Yeung et al., 2014). Objective measures of bilingualism are a more accurate way to assess proficiency and to categorize participants into language groups. Gollan, Salmon, Montoya, and Galasko (2011) assessed the impact of degree of bilingualism (by comparing naming in a bilingual speaker’s first and second language) on age at onset of symptoms (assessed subjectively by family members) and age at diagnosis (taken from medical records). This study is the only one to date that has directly assessed language dominance as a factor in cognitive reserve. The researchers found that the degree of bilingualism was positively correlated with age at symptom onset and age at diagnosis only for Spanish-dominant bilingual speakers, and not English-dominant bilingual speakers. That is, for Spanish-dominant bilingual speakers, the more equally bilingual a person was, the later their age of onset and age of diagnosis. Importantly, the outcome variables did not correlate with subjective measures of bilingualism, suggesting that use of self-report may not accurately reflect actual language proficiency or its relationship with cognitive outcomes.

Education In the previously mentioned study by Gollan and colleagues (2011), their finding of delayed age at symptom onset and age at diagnosis was found only in Spanishdominant bilingual speakers (rather than English-dominant bilingual speakers). The Spanish-dominant group had significantly lower levels of education, and statistical analyses showed that the benefit associated with bilingualism was robust only in bilingual speakers with low education level. The authors hypothesized that this is because those with higher levels of education hit a ceiling in cognitive reserve and that bilingualism could not add any further benefit. Other studies factoring in education have found mixed results. One study found no benefit for those with low to intermediate levels of education, and a reverse effect (bilingual speakers were more likely to convert



Bilingualism in Healthy Aging and Dementia 

 317

to dementia) for those with higher levels of education (Sanders et al., 2012). However, others have found that a subset of noneducated and illiterate bilingual speakers showed a similar delay in onset of dementia (compared with monolingual speakers) as a larger, educated cohort of bilingual speakers (Alladi et al., 2013). Still others have found a delay in the onset of dementia symptoms in a group of bilingual speakers who were less educated than their monolingual counterparts (Bialystok et al., 2007). In that case, the authors suggested that the fewer years of education may be more reflective of a lack of access to education than a lack of ability, given that many of the bilingual speakers immigrated from Europe and would have had their adolescence and/or early adulthood disrupted by World War II (Bialystok et al., 2007). Regardless, it appears that the impact of bilingualism on the expression of dementia may be moderated in a complex way by education and other factors that relate to cognitive reserve.

Neuroanatomical To examine whether bilingualism might contribute to cognitive reserve by protecting against neuropathological changes, Schweizer, Ware, Fischer, Craik, and Bialystok (2012) assessed 20 monolingual and 20 bilingual patients with probable Alzheimer’s disease and found that although the two groups were matched on cognitive ability, the bilingual patients showed greater atrophy in Alzheimer’s disease–relevant brain areas than did monolingual patients. They concluded that bilingualism contributed to cognitive reserve, which delayed the onset of Alzheimer’s disease by requiring greater amounts of brain pathology before the disease clinically manifests, supporting the neural compensation hypothesis of cognitive reserve.

Conclusion The studies in this newly developing research area can mostly be divided into several types: experimental studies examining cognitive function with detailed tasks in older monolingual and bilingual adults (e.g., Bialystok et al., 2008), clinical studies that have retrospectively examined the language status and onset of dementia from medical files in already diagnosed dementia patients (e.g., Chertkow et al., 2010), and large cohort studies that have followed healthy older adults and have observed dementia prevalence or cognitive decline (e.g., Yeung et al., 2014). With the exception of the experimental studies, these latter studies were not initially designed to examine the impact of language status on later life outcomes. In both classes of studies, given their retrospective nature, the assessment of potentially important language variables, variables related to cognitive reserve, and/or culturally relevant variables like immigration status or the degree of assimilation or acculturation into a person’s new environment can be cursory or missing altogether.

318 

 Duncan and Phillips

With regard to language-related variables, the studies we have reviewed have shown that the degree of bilingualism may have an impact, as well as age of acquisition, and amount of time spent speaking both languages. In fact, two of the studies provide evidence that learning a second language later in life (rather than being bilingual per se) could contribute to cognitive reserve (Bak et al., 2014; Ljungberg et al., 2013). We would suggest that learning a second language later in life could reflect a general openness to stimulating cognitive challenges. Variables like education and occupational status were included in many of the studies to account for their potential impact on cognitive reserve. Unfortunately, the relationship between education, bilingualism, and cognitive reserve is unclear, as some studies showed a bilingual benefit in those with little or no education (e.g., Alladi et al., 2013; Kavé et al., 2008), whereas others did not (e.g., Sanders et al., 2012). Finally, participation in social activities has been shown to contribute to cognitive reserve (Engelhardt et al., 2010). Depending on the country in question, and whether the person in question is an immigrant or native to the country, being multilingual could promote a person’s ability to participate in either a majority culture (e.g., Chertkow et al., 2010) or reflect access to formal education and/or media (e.g., Alladi et al., 2013; Perquin et al., 2013). As illustrated in this chapter, bilingualism is a multifaceted phenomenon that varies in important ways across individuals, groups, and cultures. Although promising, the current studies indicate that much more research is needed to examine what aspect of “bilingualism” may lead to increased cognitive reserve—whether the hypothesized benefit comes from frequent switching between languages, from learning a second language within or outside of the language “critical period,” or from variables related to the degree to which a person is integrated into his or her culture and thus has opportunities to engage in enriching activities. Future research will have to use more in-depth measurement of relevant language behaviors to fully understand the impact language use has on brain plasticity. Regardless, if a clear relationship can be established between bilingualism and increased cognitive reserve, then effort put into clarifying how to take advantage of this benefit is warranted.

References Abutalebi, J., Canini, M., Della Rosa, P. A., Sheung, L. P., Green, D. W., & Weekes, B. S. (2014). Bilingualism protects anterior temporal lobe integrity in aging. Neurobiology of Aging, 35, 2126–2133. http://dx.doi.org/10.1016/j.neurobiolaging.2014.03.010 Alladi, S., Bak, T. H., Duggirala, V., Surampudi, B., Shailaja, M., Shukla, A. K., . . . Kaul, S. (2013). Bilingualism delays age at onset of dementia, independent of education and immigration status. Neurology, 81, 1938–1944. http://dx.doi.org/10.1212/01.wnl.0000436620.33155.a4 Alzheimer’s Disease International. (2010). World Alzheimer Report 2010. Retrieved from http:// www.alz.co.uk/research/worldreport Alzheimer Society of Canada. (2010). Rising tide: The impact of dementia on Canadian society. Retrieved from http://www.alzheimer.ca/~/media/Files/national/Advocacy/ASC_Rising_Tide_Full_ Report_e.pdf



Bilingualism in Healthy Aging and Dementia 

 319

Bak, T. H., Nissan, J. J., Allerhand, M. M., & Deary, I. J. (2014). Does bilingualism influence cognitive aging? Annals of Neurology, 75, 959–963. http://dx.doi.org/10.1002/ana.24158 Barulli, D., & Stern, Y. (2013). Efficiency, capacity, compensation, maintenance, plasticity: Emerging concepts in cognitive reserve. Trends in Cognitive Sciences, 17, 502–509. http://dx.doi.org/ 10.1016/j.tics.2013.08.012 Belleville, S., Rouleau, N., & Van der Linden, M. (2006). Use of the Hayling task to measure inhibition of prepotent responses in normal aging and Alzheimer’s disease. Brain and Cognition, 62, 113–119. http://dx.doi.org/10.1016/j.bandc.2006.04.006 Bennett, D. A., Schneider, J. A., Arvanitakis, Z., Kelly, J. F., Aggarwal, N. T., Shah, R. C., & Wilson, R. S. (2006). Neuropathology of older persons without cognitive impairment from two community-based studies. Neurology, 66, 1837–1844. http://dx.doi.org/10.1212/01.wnl.0000219668.47116.e6 Bennett, D. A., Schneider, J. A., Wilson, R. S., Bienias, J. L., & Arnold, S. E. (2005). Education modifies the association of amyloid but not tangles with cognitive function. Neurology, 65, 953–955. http://dx.doi.org/10.1212/01.wnl.0000176286.17192.69 Bialystok, E. (1999). Cognitive complexity and attentional control in the bilingual mind. Child Development, 70, 636–644. http://dx.doi.org/10.1111/1467-8624.00046 Bialystok, E., Craik, F., & Luk, G. (2008). Cognitive control and lexical access in younger and older bilinguals. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34, 859–873. http://dx.doi.org/10.1037/0278-7393.34.4.859 Bialystok, E., Craik, F. I. M., & Freedman, M. (2007). Bilingualism as a protection against the onset of symptoms of dementia. Neuropsychologia, 45, 459–464. http://dx.doi.org/10.1016/ j.neuropsychologia.2006.10.009 Bialystok, E., Craik, F. I. M., Klein, R., & Viswanathan, M. (2004). Bilingualism, aging, and cognitive control: Evidence from the Simon task. Psychology and Aging, 19, 290–303. http://dx.doi.org/ 10.1037/0882-7974.19.2.290 Bialystok, E., Craik, F. I. M., & Ruocco, A. C. (2006). Dual-modality monitoring in a classification task: The effects of bilingualism and ageing. Quarterly Journal of Experimental Psychology, 59, 1968–1983. http://dx.doi.org/10.1080/17470210500482955 Bialystok, E., Craik, F. I. M., & Ryan, J. (2006). Executive control in a modified antisaccade task: Effects of aging and bilingualism. Journal of Experimental Psychology: Learning, Memory, and Cognition, 32, 1341–1354. http://dx.doi.org/10.1037/0278-7393.32.6.1341 Bialystok, E., & Luk, G. (2012). Receptive vocabulary differences in monolingual and bilingual adults. Bilingualism: Language and Cognition, 15, 397–401. http://dx.doi.org/10.1017/ S136672891100040X Bialystok, E., Luk, G., Peets, K. F., & Yang, S. (2010). Receptive vocabulary differences in monolingual and bilingual children. Bilingualism: Language and Cognition, 13, 525–531. http://dx.doi.org/ 10.1017/S1366728909990423 Bialystok, E., & Martin, M. M. (2004). Attention and inhibition in bilingual children: Evidence from the dimensional change card sort task. Developmental Science, 7, 325–339. http://dx.doi.org/ 10.1111/j.1467-7687.2004.00351.x Brewster, P. W. H., Melrose, R. J., Marquine, M. J., Johnson, J. K., Napoles, A., MacKay-Brandt, A., . . . Mungas, D. (2014). Life experience and demographic influences on cognitive function in older adults. Neuropsychology. Advance online publication. Retrieved from http://psycnet.apa.org/ journals/neu/28/6/846.pdf Cahn-Weiner, D. A., Malloy, P. F., Boyle, P. A., Marran, M., & Salloway, S. (2000). Prediction of functional status from neuropsychological tests in community-dwelling elderly

320 

 Duncan and Phillips

individuals. The Clinical Neuropsychologist, 14, 187–195. http://dx.doi.org/10.1076/ 1385-4046(200005)14:2;1-Z;FT187 Chertkow, H., Whitehead, V., Phillips, N., Wolfson, C., Atherton, J., & Bergman, H. (2010). Multilingualism (but not always bilingualism) delays the onset of Alzheimer disease: Evidence from a bilingual community. Alzheimer Disease and Associated Disorders, 24, 118–125. http://dx.doi.org/ 10.1097/WAD.0b013e3181ca1221 Costa, A., Hernández, M., & Sebastián-Gallés, N. (2008). Bilingualism aids conflict resolution: Evidence from the ANT task. Cognition, 106, 59–86. http://dx.doi.org/10.1016/j.cognition. 2006.12.013 Crane, P. K., Gibbons, L. E., Arani, K., Nguyen, V., Rhoads, K., McCurry, S. M., . . . White, L. (2009). Midlife use of written Japanese and protection from late life dementia. Epidemiology, 20, 766–774. http://dx.doi.org/10.1097/EDE.0b013e3181b09332 Crawford, J., Bryan, J., Luszcz, M., Obonsawin, M., & Stewart, L. (2000). The executive decline hypothesis of cognitive aging: Do executive deficits qualify as differential deficit and do they mediate age-related memory decline? Aging, Neuropsychology and Cognition, 7, 9–31. http:// dx.doi.org/10.1076/anec.7.1.9.806 Engelhardt, H., Buber, I., Skirbekk, V., & Prskawetz, A. (2010). Social involvement, behavioural risks and cognitive functioning among older people. Ageing & Society, 30, 779–809. http://dx.doi.org/ 10.1017/S0144686X09990626 Fotenos, A. F., Mintun, M. A., Snyder, A. Z., Morris, J. C., & Buckner, R. L. (2008). Brain volume decline in aging: Evidence for a relation between socioeconomic status, preclinical Alzheimer disease, and reserve. Archives of Neurology, 65, 113–120. http://dx.doi.org/10.1001/archneurol.2007.27 Fuller-Thomson, E., & Kuh, D. (2014). The healthy migrant effect may confound the link between bilingualism and delayed onset of Alzheimer’s disease. Cortex, 52, 128–130. http://dx.doi.org/ 10.1016/j.cortex.2013.08.009 Goffaux, P., Phillips, N. A., Sinai, M., & Pushkar, D. (2008). Neurophysiological measures of task-set switching: Effects of working memory and aging. The Journals of Gerontology: Series B. Psychological Sciences and Social Sciences, 63, 57–66. http://dx.doi.org/10.1093/geronb/63.2.P57 Gold, B. T., Johnson, N. F., & Powell, D. K. (2013). Lifelong bilingualism contributes to cognitive reserve against white matter integrity declines in aging. Neuropsychologia, 51, 2841–2846. http://dx.doi.org/10.1016/j.neuropsychologia.2013.09.037 Gollan, T. H., Montoya, R. I., & Werner, G. A. (2002). Semantic and letter fluency in Spanish–English bilinguals. Neuropsychology, 16, 562–576. http://dx.doi.org/10.1037/0894-4105.16.4.562 Gollan, T. H., Salmon, D. P., Montoya, R. I., & Galasko, D. R. (2011). Degree of bilingualism predicts age of diagnosis of Alzheimer’s disease in low-education but not in highly educated Hispanics. Neuropsychologia, 49, 3826–3830. http://dx.doi.org/10.1016/j.neuropsychologia.2011.09.041 Good, C. D., Johnsrude, I. S., Ashburner, J., Henson, R. N. A., Friston, K. J., & Frackowiak, R. S. J. (2001). A voxel-based morphometric study of ageing in 465 normal adult human brains. NeuroImage, 14, 21–36. http://dx.doi.org/10.1006/nimg.2001.0786 Green, D. (1998). Mental control of the bilingual lexico-semantic system. Bilingualism: Language and Cognition, 1, 67–81. http://dx.doi.org/10.1017/S1366728998000133 Hilchey, M. D., & Klein, R. M. (2011). Are there bilingual advantages on nonlinguistic interference tasks? Implications for the plasticity of executive control processes. Psychonomic Bulletin & Review, 18, 625–658. http://dx.doi.org/10.3758/s13423-011-0116-7 Kamat, R., Ghate, M., Gollan, T. H., Meyer, R., Vaida, F., Heaton, R. K., . . . Marcotte, T. D., & the HIV Neurobehavioral Research Program (HNRP) Group. (2012). Effects of Marathi-Hindi bilingualism



Bilingualism in Healthy Aging and Dementia 

 321

on neuropsychological performance. Journal of the International Neuropsychological Society, 18, 305–313. http://dx.doi.org/10.1017/S1355617711001731 Kavé, G., Eyal, N., Shorek, A., & Cohen-Mansfield, J. (2008). Multilingualism and cognitive state in the oldest old. Psychology and Aging, 23, 70–78. http://dx.doi.org/10.1037/0882-7974.23.1.70 Kousaie, S., & Phillips, N. A. (2012). Ageing and bilingualism: Absence of a “bilingual advantage” in Stroop interference in a nonimmigrant sample. Quarterly Journal of Experimental Psychology, 65, 356–369. http://dx.doi.org/10.1080/17470218.2011.604788 Kousaie, S., Sheppard, C., Lemieux, M., Monetta, L., & Taler, V. (2014). Executive function and bilingualism in young and older adults. Frontiers in Behavioral Neuroscience, 8, 250. http:// dx.doi.org/10.3389/fnbeh.2014.00250 Liu, Y., Cai, Z.-L., Xue, S., Zhou, X., & Wu, F. (2013). Proxies of cognitive reserve and their effects on neuropsychological performance in patients with mild cognitive impairment. Journal of Clinical Neuroscience: Official Journal of the Neurosurgical Society of Australasia, 20, 548–553. http:// dx.doi.org/10.1016/j.jocn.2012.04.020 Ljungberg, J. K., Hansson, P., Andrés, P., Josefsson, M., & Nilsson, L.-G. (2013). A longitudinal study of memory advantages in bilinguals. PLoS ONE, 8, e73029. http://dx.doi.org/10.1371/journal. pone.0073029 Luk, G., & Bialystok, E. (2013). Bilingualism is not a categorical variable: Interaction between language proficiency and usage. Journal of Cognitive Psychology, 25, 605–621. http://dx.doi.org/ 10.1080/20445911.2013.795574 Luk, G., Bialystok, E., Craik, F. I. M., & Grady, C. L. (2011). Lifelong bilingualism maintains white matter integrity in older adults. The Journal of Neuroscience, 31, 16808–16813. http://dx.doi.org/ 10.1523/JNEUROSCI.4563-11.2011 Luk, G., Green, D. W., Abutalebi, J., & Grady, C. (2011). Cognitive control for language switching in bilinguals: A quantitative meta-analysis of functional neuroimaging studies. Language and Cognitive Processes, 27, 1479–1488. http://dx.doi.org/10.1080/01690965.2011.613209 Ossher, L., Bialystok, E., Craik, F. I. M., Murphy, K. J., & Troyer, A. K. (2013). The effect of bilingualism on amnestic mild cognitive impairment. The Journals of Gerontology: Series B. Psychological Sciences and Social Sciences, 68, 8–12. http://dx.doi.org/10.1093/geronb/gbs038 Perani, D., & Abutalebi, J. (2005). The neural basis of first and second language processing. Current Opinion in Neurobiology, 15, 202–206. http://dx.doi.org/10.1016/j.conb.2005.03.007 Pereira, F. S., Yassuda, M. S., Oliveira, A. M., & Forlenza, O. V. (2008). Executive dysfunction correlates with impaired functional status in older adults with varying degrees of cognitive impairment. International Psychogeriatrics, 20, 1104–1115. http://dx.doi.org/10.1017/S1041610208007631 Perquin, M., Vaillant, M., Schuller, A.-M., Pastore, J., Dartigues, J.-F., Lair, M.-L., & Diederich, N., & the MemoVie Group. (2013). Lifelong exposure to multilingualism: New evidence to support cognitive reserve hypothesis. PLoS ONE, 8(4), e62030. http://dx.doi.org/10.1371/ journal.pone.0062030 Querbes, O., Aubry, F., Pariente, J., Lotterie, J. A., Démonet, J. F., Duret, V., . . . & the Alzheimer’s Disease Neuroimaging Initiative. (2009). Early diagnosis of Alzheimer’s disease using cortical thickness: Impact of cognitive reserve. Brain: A Journal of Neurology, 132, 2036–2047. http://dx.doi.org/10.1093/brain/awp105 Raz, N., Gunning, F. M., Head, D., Dupuis, J. H., McQuain, J., Briggs, S. D., . . . Acker, J. D. (1997). Selective aging of the human cerebral cortex observed in vivo: Differential vulnerability of the prefrontal gray matter. Cerebral Cortex, 7, 268–282. http://dx.doi.org/10.1093/ cercor/7.3.268

322 

 Duncan and Phillips

Rosselli, M., Ardila, A., Araujo, K., Weekes, V. A., Caracciolo, V., Padilla, M., & Ostrosky-Solí, F. (2000). Verbal fluency and repetition skills in healthy older Spanish-English bilinguals. Applied Neuropsychology, 7(1), 17–24. http://dx.doi.org/10.1207/S15324826AN0701_3 Sanders, A. E., Hall, C. B., Katz, M. J., & Lipton, R. B. (2012). Non-native language use and risk of incident dementia in the elderly. Journal of Alzheimer’s Disease, 29, 99–108. Schroeder, S. R., & Marian, V. (2012). A bilingual advantage for episodic memory in older adults. Journal of Cognitive Psychology, 24, 591–601. http://dx.doi.org/10.1080/20445911.2012.669367 Schweizer, T. A., Ware, J., Fischer, C. E., Craik, F. I. M., & Bialystok, E. (2012). Bilingualism as a

contributor to cognitive reserve: Evidence from brain atrophy in Alzheimer’s disease. Cortex, 48, 991–996. http://dx.doi.org/10.1016/j.cortex.2011.04.009 Sorel, O., & Pennequin, V. (2008). Aging of the planning process: The role of executive functioning. Brain and Cognition, 66, 196–201. http://dx.doi.org/10.1016/j.bandc.2007.07.006 Staff, R. T., Murray, A. D., Deary, I. J., & Whalley, L. J. (2004). What provides cerebral reserve? Brain: A Journal of Neurology, 127, 1191–1199. http://dx.doi.org/10.1093/brain/awh144 Stern, Y. (2009). Cognitive reserve. Neuropsychologia, 47, 2015–2028. http://dx.doi.org/10.1016/ j.neuropsychologia.2009.03.004 Sylvain-Roy, S., & Belleville, S. (2015). Interindividual differences in attentional control profiles among younger and older adults. Aging, Neuropsychology, and Cognition: A Journal on Normal and Dysfunctional Development, 22, 259–279. http://dx.doi.org/10.1080/13825585.2014.926305 Troyer, A. K., Graves, R. E., & Cullum, C. M. (1994). Executive functioning as a mediator of the relationship between age and episodic memory in healthy aging. Aging, Neuropsychology and Cognition, 1, 45–53. http://dx.doi.org/10.1080/09289919408251449 Wilson, R. S., Boyle, P. A., Yu, L., Barnes, L. L., Schneider, J. A., & Bennett, D. A. (2013). Life-span cognitive activity, neuropathologic burden, and cognitive aging. Neurology, 81, 314–321. http:// dx.doi.org/10.1212/WNL.0b013e31829c5e8a Wodniecka, Z., Craik, F. I. M., Luo, L., & Bialystok, E. (2010). Does bilingualism help memory? Competing effects of verbal ability and executive control. International Journal of Bilingual Education and Bilingualism, 13, 575–595. http://dx.doi.org/10.1080/13670050.2010.488287 Yeung, C. M., St. John, P. D., Menec, V., & Tyas, S. L. (2014). Is bilingualism associated with a lower risk of dementia in community-living older adults? Cross-sectional and prospective analyses. Alzheimer Disease & Associated Disorders, 28, 326–332. http://dx.doi.org/10.1097/ WAD.0000000000000019

V Conclusion

Simona Montanari and Elena Nicoladis

Concluding Remarks and Future Directions Research on bilingualism has come a long way from the early, oft-cited work by Ronjat (1913) and Leopold (1939–1949) in the first half of the 20th century. This volume illustrates how research on bilingualism has evolved throughout the past decades. For example, studies have been extended to subjects across the lifespan, from infancy (and even the prenatal period), to childhood, to adulthood, and to old age. Moreover, as societies have become increasingly more mobile and multilingual, issues related to educating young bilingual children as well as the consequences of bilingualism on literacy have become key topics. New areas of investigations have also emerged to explore how the bilingual experience shapes cognitive and emotional processes in different stages of life. Taken together, the chapters in this book underscore the breadth of the research on bilingualism today, as well as the complexity of studying this topic without considering the host of factors that characterize bilingualism. In Chapter 1 of this volume, Genesee pointed out that bilingualism is characterized by variability not only across ages and stages of life but also across populations, communities, and individuals. All contributions in this volume demonstrate indeed that bilinguals can vary tremendously, both in terms of the language experiences they have and the language outcomes they achieve. One important source of variability is the oft-unacknowledged relevance of sociopolitical and sociocultural factors. This issue was first raised by Lambert (1974) when contrasting why French–English bilingualism in Canada was often associated with positive outcomes whereas other bilingual groups in other countries did not always fare well. He argued that bilingualism in Canada was a case of additive bilingualism, in which bilingual speakers were motivated to seek out opportunities to maintain both of their languages. Negative outcomes were more likely associated with subtractive bilingualism, situations in which people are motivated to learn one language, usually the majority language of the community, at the expense of the other. Montrul (Chapter 8), for example, showed that bilingualism is rarely attainable among immigrants in the United States because learning the majority (and prestigious) language often means losing the first minority (and stigmatized) language. Even in the U.S. educational context, bilingualism may be seen an obstacle to academic achievement for immigrant students. For majority language students, it may be seen as an optional benefit but not as a critical measure of achievement (Chapter 11). Furthermore, bilingual speakers who code-switch might still be perceived as having incomplete knowledge of both languages, and, in the extreme cases, as having a language disability (see Chapter 10). Either directly or indirectly, this sociopolitical reality favors English monolingualism, with clear negative consequences on bilingualism. http://dx.doi.org/10.1037/14939-018 Bilingualism Across the Lifespan: Factors Moderating Language Proficiency, E. Nicoladis and S. Montanari (Editors) Copyright © 2016 by the American Psychological Association and Walter de Gruyter, GmbH. All rights reserved.

326 

 Montanari and Nicoladis

In contrast, studies that are set in sociopolitical realities that are more positive toward bilingualism (e.g., Singapore, Canada, Wales, and other multilingual areas in Europe) demonstrate that competence in two languages is more easily attainable when social, political and educational forces are in favor. In these contexts, many individuals have the opportunity to become bilingual. At the same time, social factors—in particular, socioeconomic status, education, and immigration status among others— play a crucial role for the attainment of bilingualism. Those individuals with higher means and more education are indeed more likely to develop and maintain high levels of proficiency in two languages from childhood to adulthood (e.g., Chapter 7) and to experience the cognitive benefits linked to bilingualism throughout the lifespan (e.g., Chapters 13–16). Furthermore, in realities in which multilingualism and multiculturalism are the norm (i.e., Singapore), the benefits of bilingualism are even more obvious. Yow, Patrycia, and Flynn (Chapter 5) reported indeed that positive attitudes toward multilingualism and the active use of code-switching result in increased levels of bilingualism in young children. This sociopolitical reality produces individuals who are competent in and use two or more languages on a daily basis, with no negative consequences on overall development or academic achievement. Taken together, these contributions make clear that findings and interpretations of findings related to bilingualism are dependent on the sociopolitical realities in which the studies are set. Characteristics of the sociocultural context in which bilingual speakers are raised and operate also influence how languages develop, are used, and are maintained. Quay and Montanari (Chapter 2), for example, pointed out that evidence of language differentiation—which has been traditionally taken as sign of the separate development of two languages according to a monolingual perspective—cannot be found in contexts that do not require language separation (e.g., bilingual children cannot be expected to have translation equivalents for words they only hear in one language). In these contexts, the bilingual child will typically show distributed vocabulary learning and code-mixing in line with the speakers in his/her own sociolinguistic milieu. Similarly, Fennell, Tsui, and Hudon (Chapter 3) showed that bilingual infants demonstrate certain speech perception skills only when bilingual, but not monolingual, speakers produce those words—the former reproducing more faithfully the children’s typical language environment. Conboy and Montanari (Chapter 4) and Unsworth (Chapter 6) further showed that differences from monolingual patterns in bilingual acquisition reflect appropriate responses to sociocultural properties of the input, for example, the extent of exposure to infant-directed speech, to mixed language input, and to accented input. In the context of adult second-language acquisition, Birdsong and Vanhove (Chapter 9) pointed out that, besides age, sociopsychological factors such as learners’ socialization patterns might crucially condition bilingual outcomes. Overall, the studies presented in this volume underscore the need to examine bilingualism taking into account the host of sociocultural factors that crucially affect the learning, use, and maintenance of two languages. This means that future studies should move beyond perspectives limited to linguistics or psycholinguistics and



Concluding Remarks and Future Directions 

 327

become more interdisciplinary in nature, bringing the sociopolitical and sociolinguistic realities in which the studies are set more to light and under scrutiny. The research field of family language policies described by Quay and Montanari (Chapter 2) is a good example of how child language acquisition studies can be examined through a language policy perspective in the study of parents’ language decisions, practices, and beliefs and their consequences on child outcomes. Work in this area of investigation closes the gap between macrosocial issues (e.g., language ideology and language attitudes) and microsocial issues (e.g., interactional patterns in bilingual homes), bringing together ideology, management, and practice. More studies of this nature are needed to fully understand topics that are strongly constrained by social forces such as language learning and use—whether this involves multiple languages or only one language. Indeed, this approach should be taken in monolingualism research as well. Studies have shown, for instance, that socioeconomic conditions and other sociocultural factors crucially affect language outcomes even among monolingual speakers (Hart & Risley, 1995; Hoff, 2006; Vasilyeva, Waterfall, & Huttenlocher, 2008) and that monolingual development is not as universal and monolithic as traditionally depicted in language acquisition theories. Thus, it becomes fundamental to examine competence in one or more languages in different sociopolitical realities and under differing input conditions. The studies reviewed in this volume also make clear that research on bilingualism is geographically limited and includes primarily middle-class bilinguals from Europe and North America (with the exception of Chapter 5). This means that what is known about bilingualism is mostly on the basis of research involving individuals from middle or upper socioeconomic backgrounds in Western cultures, and speaking only a tiny fraction of bilingual speakers’ possible language combinations. Given that social and cultural variables strongly condition bilingual outcomes, it is imperative that more studies are conducted in sociopolitical and sociolinguistic realities that are very different from those investigated. For instance, research is needed in areas in which majority and minority languages are reversed from the case of current studies (e.g., English and French as minority rather than majority languages) and among bilingual speakers of diverse socioeconomic backgrounds. Most important, bilingualism should be studied in regions in which bilingualism and multilingualism are the norm, as in India, Africa, and certain areas of Asia (Bhatia & Ritchie, 2004). Studies in these areas—where bilingual and multilingual individuals outnumber monolingual individuals—may reveal intriguing patterns of language development and use and may uncover cognitive, emotional, and academic advantages (or disadvantages) different from those documented so far. This geographically broader research would also allow the study of bilingualism in the context of language pairs that are typologically different from those studied in extant work, revealing new types and degrees of cross-linguistic interactions than those documented in the current research. By expanding the geography of bilingualism research, future studies and theories of bilingualism can truly move away from a monolingual perspective, and examine

328 

 Montanari and Nicoladis

what is typical for bilinguals in a determined sociopolitical reality independent of what is typical for monolinguals. As put by Genesee in Chapter 1, monolingualism is not the only norm; in fact, monolingualism might gradually become the exception in our increasingly mobile, multicultural, and multilingual world. Kroll, Gullifer, and Zirnstein (Chapter 12)—as well as other contributors—already foreshadow that new models of language processing and literacy that accommodate the presence of two or more languages will soon become the norm because of the prevalence of bilingualism. Future work should also move in this direction. At the same time, more research is needed from a true lifespan perspective, because, as shown in this volume, some effects of bilingualism are only relevant at particular ages and stages of life. Although research on early bilingualism, in particular simultaneous bilingualism, has been particularly fertile, studies on early sequential bilingual speakers are scarcer. More studies should be devoted to bilingual development in this population, as child second-language learners probably outnumber simultaneous learners in many geographical areas (Mackey, 2004; Chapter 8, this volume). Similarly, studies on bilingual adolescents are overall lacking. As Quay and Montanari (Chapter 2) pointed out in their contribution, this population is important to study because, often, early bilingualism may be lost in later years when teenagers refuse to speak the minority language because of societal attitudes (Caldas & Caron-Caldas, 2002). Similarly, research has only recently began to examine bilingualism in older adults, and, as argued by Duncan and Phillips (Chapter 16), much more needs to be learned to assess the impact of dual language competence on social, cognitive, and linguistic processing during aging. Clearly, bilingualism is a dynamic phenomenon that can only be understood if conceptualized as ever evolving and transforming the human experience. Only this conceptualization can produce theoretical accounts of bilingualism that move beyond static descriptions and show the fluctuating nature of linguistic knowledge—in one, two, or more languages—across the lifespan. It is also important that new research findings on bilingualism—such as those reviewed in this volume—are more easily transmitted to the public at large and translated into practice. Bilingualism is indeed still surrounded by false beliefs and misunderstandings, not only among the public at large but even among educators, administrators, and clinicians (Sorace, 2007). For instance, it is often believed that dealing with two languages at the same time is too much of a burden for the infant’s brain and it may delay overall language development because of increased demands on language learning capacities. It is also often assumed that languages compete for resources in the brain at the expense of general cognition. Thus, speech pathologists might advise parents to stop using one language with their bilingual child when a language or cognitive disability exists, believing that dual language input exacerbates the language problems and takes away from language gains in one language alone (Kohnert & Derr, 2012). Similarly, it is not uncommon that educators blame bilingualism for academic weaknesses, as it is often the case with low-performing Spanish-speaking students in the United States (Lopez, 2009). Even current educational policies passed in U.S.



Concluding Remarks and Future Directions 

 329

states such as California, Arizona, and Massachusetts that block the use of primary languages in school (i.e., they outlaw bilingual education) are based on the premise that educating English language learners only in English is the most effective way to promote their linguistic competence and, ultimately, their academic success (Genesee & Lindholm-Leary, 2012). And yet, we have learned from research with bilingual speakers that any assessment of bilingual proficiency should eschew traditional, monolingually normed tests to evaluate the bilingual speaker’s total linguistic knowledge in all domains, whether this involves the use of a single language in a context or the mixing of both languages in another (Baker, 2011). These monolingually oriented social, clinical, and educational realities deeply influence bilingual speakers’ language and educational outcomes, often at the expense of bilingualism. Therefore, it is imperative that future work bridge the gap between the scientific approach to the study of bilingualism and what many people, including professionals and policy makers, believe about life with two languages. The latest research in psychology and neuroscience indicates indeed that there are no foundations to the belief that monolingualism is somehow the biological norm. In fact, as Genesee puts it in Chapter 1 of this volume, “the uniquely human neurocognitive capacity for learning language is prepared to learn not just one language but two (or more) languages with relative facility–provided the learning conditions are adequate” (p. 12). Whether the learning conditions are “adequate” will depend not only on the attitudes of parents, teachers, and individuals in the child’s immediate environment but also— and perhaps most importantly—on the approaches adopted within the broader sociopolitical and educational context in which bilinguals operate. It should be a universal goal to pursue and foster an understanding of bilingualism that is based on the most current scientific evidence, forcing public policymakers to reassess monolingual perspectives and embrace the multilingual nature of the humans of the 21st century.

References Baker, C. (2011). Foundations of bilingual education and bilingualism (5th ed.). Clevedon, England: Multilingual Matters. Bhatia, T. K., & Ritchie, W. C. (Eds.). (2004). The handbook of bilingualism. Malden, MA: Blackwell. Caldas, S., & Caron-Caldas, S. (2002). A sociolinguistic analysis of the language preferences of adolescent bilinguals: Shifting allegiances and developing identities. Applied Linguistics, 23, 490–514. http://dx.doi.org/10.1093/applin/23.4.490 Genesee, F., & Lindholm-Leary, K. (2012). The education of English language learners. In K. Harris, S. Graham, & T. Urdan (Eds.), APA educational psychology handbook (Vol. 3, pp. 499–526). Washington, DC: American Psychological Association. Hart, B., & Risley, T. (1995). Meaningful differences in the everyday experience of young American children. Baltimore, MD: Brookes. Hoff, E. (2006). How social contexts support and shape language development. Developmental Review, 26, 55–88. http://dx.doi.org/10.1016/j.dr.2005.11.002

330 

 Montanari and Nicoladis

Kohnert, K., & Derr, K. (2012). Language intervention with bilingual children. In B. Goldstein (Ed.), Bilingual language development and disorders in Spanish-English speakers (2nd ed., pp. 337–356). Baltimore, MD: Brookes. Lambert, W. E. (1974). Culture and language as factors in learning and education. In F. E. Aboud & R. D. Meade (Eds.), Cultural factors in learning (pp. 91–122). Bellingham, WA: Western Washington State College. Leopold, W. (1939–1949). Speech development of a bilingual child: A linguist’s record. Vol. I–IV. Evanston, IL: Northwestern University.

Lopez, M. H. (2009). Latinos and education: Explaining the attainment gap. Washington, DC: Pew Hispanic Research Center. Mackey, W. F. (2004). Bilingualism in North America. In T. K. Bhatia & W. C. Ritchie (Eds.), The handbook of bilingualism (pp. 607–641). Malden, MA: Blackwell. Ronjat, J. (1913). Le developpement du langage observé chez un enfant bilingue [Language development observed in the bilingual child]. Paris, France: Edouard Champion. http://dx.doi.org/ 10.3726/978-3-653-03449-3 Sorace, A. (2007). The more, the merrier: Facts and beliefs about the bilingual mind. In S. Della Sala (Ed.), Tall tales about the mind and brain: Separating fact from fiction (pp. 193–203). Oxford, England: Oxford University Press. http://dx.doi.org/10.1093/acprof:oso/ 9780198568773.003.0014 Vasilyeva, M., Waterfall, H., & Huttenlocher, J. (2008). Emergence of syntax: Commonalities and differences across children. Developmental Science, 11, 84–97. http://dx.doi.org/10.1111/ j.1467-7687.2007.00656.x

Index Abrahamsson, N., 146, 165, 175 Abutalebi, J., 311–312 Academic achievement. See also Dual language (DL) programs ––and bilingualism, 13, 183 ––and bilingualism outcomes, 206–215 ––and language proficiency, 216–218 ACC (anterior cingulate cortex), 297 Accents ––infants’ accommodation of, 53–54 ––and phonological perceptions of infants, 55–56 ––social impressions of, 176 Accentuation, word recognition and, 66 Accountability, language proficiency and, 219 Achievement. See Academic achievement; Reading achievement Acquired distinctiveness, 58 Acquisition, language. See Language acquisition Activation ––cross-language, 7, 225–232 ––language. See Language activation ––lexical, 226–232, 290, 291 Adaptation, bilingualism as cognitive, 264–265 Additive bilingualism, 325 Adi-Japha, E., 273–274 Adjective placement, code-switching and, 189 Adoptees, international, 154–157 Adults ––bilingualism studies with, 14–16 ––bilingual reading by. See Bilingual reading ––code-switching by. See Code-switching by adults ––cognitive effects of bilingualism for. See Cognitive effects of bilingualism for adults ––cognitive reserve of. See Cognitive reserve ––language attrition for, 142–145 ––proficiency factors for, 125, 131–132, 134 AEM (anticipatory eye movement) tasks, 49 Age of onset of acquisition (AoA) ––and nature-of-bilingualism view, 165–166 ––in non-nativelikeness prediction, 164–166 ––and ultimate language attainment, 166–175

Age of second-language acquisition, 163–177 ––and bilingual attainment, 6 ––and bilingual reading ability, 235 ––critical period hypothesis of, 164–175 ––discontinuity prediction in, 166–175 ––and first language attrition, 145–157 ––and memory of adults, 15 ––non-nativelikeness prediction in, 164–166 ––by school-age children, 12–14 ––social context for, 175–176 Aging ––effects of bilingualism on, 15, 307–313 ––effects of cognitive control on, 287–288 Albareda-Castellot, B., 45–46, 49 Alfi-Shabtay, I., 171–172 Allen, S. E., 12 Allerhand, M. M., 312–313 Altarriba, J., 233, 293–294 Alzheimer’s disease ––and bilingualism, 314–315, 317 ––and cognitive reserve, 288, 306 ––described, 313 ––diagnosis of, 314–315 ––episodic memory in, 309 ––onset of symptoms of, 16, 288, 306, 315 ––prevalence of, 313 Ameel, E., 131 Amygdala, 297–298 Anderson, R. T., 147–148 Andrés, P., 309 Anterior cingulate cortex (ACC), 297 Anticipatory eye movement (AEM) tasks, 49 AoA. See Age of onset of acquisition Arju, T., 36 Aslin, R. N., 52 Attention ––joint, 34 ––nonlinguistic attentional tasks, 276 ––selective, 13, 269, 276–277 Attentional control. See also Cognitive control ––of adults, 289, 307, 308 ––of children, 26, 72 ––and executive functions, 26, 253, 254 ––of infants, 50, 253, 254, 261 ––on nonlinguistic tasks, 72, 289 ––and phonetic perception, 50

332 

 Index

Attentional Network Task, 253 Attention switching, 264–265 Attrition of first language. See First language attrition Atypical constructions, cross-language influence on, 28 Au, T. K., 154, 155 Auer, P., 187 Automatic rhythm clustering, 251 Baayen, R. H., 173n4 Bae, J., 214 Bailey, B., 187 Bajo, M. T., 230, 238 Bak, T. H., 312–313 Baker, C., 187 Barrett, K. C., 83 Basnight-Brown, D. M., 293–294 Bearse, C., 219 Bedore, L. M., 107, 131 Behavioral symptoms, Alzheimer’s disease, 288 Benson, E. J., 183 Berberich-Artzi, J., 273–274 Berman, R., 142 Between-language competition, 289, 291–292 BFLA children. See Bilingual first-language acquisition children Bialystok, E., 13, 16, 26, 109, 168, 253, 275, 286, 311, 314, 317 Biculturalism, 295–296 Bilingual attainment, 6, 163. See also Ultimate language attainment (UA) Bilingual benefit theory, 307 Bilingual development, 34, 37 Bilingual first-language acquisition (BFLA) children. See also Simultaneous bilingual speakers ––defined, 23 ––ELSA children vs., 33 ––family language policies for, 37 ––infant-directed speech for, 34 ––language choice for, 25 ––views on, 36, 37 Bilingualism ––and academic achievement in dual language programs. See Dual language (DL) programs ––additive vs. subtractive, 325 ––and aging, 307–313 ––as cognitive adaptation, 264–265

––consequences of, 3 ––and dementia, 313–317 ––dementia and degree of, 316 ––early. See Early bilingualism ––influences of languages on one another in, 165 ––as multifaceted phenomenon, 305, 318 ––and multilingualism, 9 ––self-reports of, 215 ––stigmatization with, 36–37 ––views of, 328–329 Bilingualism studies, 9–17 ––with adults, 14–16 ––communicating results of, 328–329 ––evolution of, 325 ––future directions for, 326–328 ––geographic limitations on, 327–328 ––with infants and young children, 10–12 ––with school-age learners, 12–14 ––on variation in bilingual speakers, 16–17 Bilingual knowledge, distributed characteristic of, 109, 113–114, 128–129 Bilingual language acquisition, milestones in, 249–250 Bilingual language acquisition paradox, 249 Bilingual outcomes ––and academic achievement in DL programs, 206–215 ––age and, 4 ––language input and, 32–33 Bilingual proficiency. See also Language proficiency ––assessments of, 206–207 ––difficulties in attaining, 219–220 ––and long-term student outcomes, 216–218 ––of native other language speakers, 205 ––self-ratings of, 214–215 Bilingual readers, monolingual vs., 238–239 Bilingual reading, 225–239 ––cross-language lexical activation in, 226–232 ––and cross-language processes within sentence context, 232–236 ––interaction between languages in, 236–238 ––language selection in, 229–232 ––and reading code-switched text, 232–236 Bilingual speakers. See also Early sequential bilingual speakers; Simultaneous bilingual speakers ––defined, 3

Index 

––heterogeneity/variability of, 3, 4, 9, 16–17, 30, 325 ––identities of, 196 ––monolingual vs. See Monolingual speakers, bilingual vs. ––as native other language speakers, 205 ––receptive, 153 ––successive, 14–15 ––variability of attrition in, 144–145 Bilingual tagging, 189 Birdsong, D., 169–171, 174 Block, N., 208, 218 Blom, J. P., 185–186 Blumenfeld, H. K., 290, 291, 297 Boas, Franz, 185 Bogulski, C. A., 237 Bohman, T. M., 107, 112 Bornstein, M. H., 125 Bosch, L., 45–48, 52, 56, 125, 131, 279 Bowers, E. P., 106 Brain ––cognitive control in, 297–298 ––emotion processing in, 294, 297–298 ––functional ability and damage to, 306 ––language switching in, 289 ––links between cognitive control and emotion in, 285–286 ––of monolingual vs. bilingual speakers, 288 ––structural impact of bilingualism on, 311–312 ––structural reorganization of, 252 Bregman, N. J., 275 Bridges, K., 105, 126 Brosseau-Lapré, F., 24 Burling, R., 10 Burns, T. C., 43–44, 47 Burridge, A., 104 Byers-Heinlein, K., 43–44, 46, 54, 55, 58, 70, 73, 126, 263, 265 Bylund, E., 144, 150, 151 CA (conversational analysis) approach, 187 Canada, views of bilingualism in, 325 Caregiver report questionnaires ––on word comprehension, 67–68 ––on word production, 72–73 Carlson, S. M., 277 CB (constraint-based) approaches, 194–195 Ceiling effects, 170 Census studies of discontinuity, 167–168 CF (constraint-free) approaches, 194

 333

Chametzky, R. A., 190 Chan, P.-S., 126 Chen, J., 126 Chertkow, H., 315 Chiappe, P., 214 Chiat, S., 134 Children. See also specific groups, e.g.: Young children ––code-switching by. See Code-switching by children ––cognitive effects of bilingualism for. See Cognitive effects of bilingualism for children ––in dual language programs. See Dual language (DL) programs ––language attrition for, 145–154 ––language input and development for. See Language input ––proficiency factors for, 125, 126, 128–129, 131, 132 CHL component, 190 Chomsky, N., 187 Chomskyan view of language acquisition, 142 Chondrogianni, V., 113–114 Choras, 36 CLAN (Computerized Language Analysis), 86 Clinical studies of bilingualism and aging, 317 Clitics, 188, 192 Cobo-Lewis, A. B., 128 Code-meshing, 187 Code-mixing ––code-switching vs., 81–82 ––defined, 23–24, 81–82 ––in early bilingualism, 24–25 ––by grandparents, 36 ––within utterances, 11 ––by young children, 11 Code-switched text, reading, 232–236 Code-switching (in general) ––and bilingual development, 34–35 ––code-mixing vs., 81–82 ––defined, 5, 81–82, 183, 232 ––intersentential, 183 ––intrasentential, 73, 88, 183, 233 ––worldwide views of, 3 Code-switching by adults, 183–196 ––and children, 14, 16, 183–185 ––early research on, 183 ––function of, 82, 183 ––grammatical studies of, 188–195 ––and language attrition, 142, 143

334 

 Index

––as language use, 185–188 ––and lexical development by infants, 73 Code-switching by children, 81–96 ––and adults, 14, 16, 183–185 ––functions of, 4, 16, 82–83 ––in language and social environment, 84–85 ––and language input, 107 ––naturalistic observation study of, 85–94 Cognates ––infants’ processing of, 48 ––and interactive influences of languages, 131–132 ––and language proficiency, 125 ––in parallel activation studies, 227, 231–232 ––and phonetic discrimination, 58 ––in phonological perception tests, 56 ––and verbal fluency, 310–311 ––and vocabulary development, 110 Cognition ––effects of bilingual reading on, 238 ––interactive influences of languages on, 132–133 ––social, 277–278 Cognitive adaptation, bilingualism as, 264–265 Cognitive control ––of bilingual adults, 286–292, 296–298 ––and bilingualism in infancy, 252, 254–257 ––components of, 286–287 ––and emotion, 285–286, 296–298 ––influence of language on, 285 ––language as filter for, 299 ––linguistic and nonlinguistic, 289–291 ––and prediction of word features, 235–236 ––and word processing, 71–72 ––and word segmentation, 66 Cognitive costs ––of bilingual reading, 234–236 ––of language switching. See Switching costs Cognitive decline, 305, 311 Cognitive effects of bilingualism for adults, 13, 285–299 ––cognitive control, 286–292, 296–298 ––and effect of biculturalism on emotion, 295–296 ––emotional memory, 295 ––emotion processing, 292–295 ––inhibitory control, 287–288 ––and interaction of linguistic/nonlinguistic cognitive control, 290–291

––interactions of cognitive control and emotion, 296–298 ––language activation and interaction, 288–290 ––nonlinguistic, 16 Cognitive effects of bilingualism for children, 13, 269–279 ––cognitive flexibility, 273–274 ––executive functions, 276 ––future research directions, 279 ––and linking of language selection with cognitive function, 270–273 ––nonlinguistic, 16 ––social cognition, 277–278 ––visuospatial tasks, 275–276 Cognitive effects of bilingualism for infants, 249–265 ––executive functions, 252–257 ––habituation patterns, 263–264 ––language discrimination, 251–252 ––learning of linguistic structural regularities, 257–262 ––memory flexibility, 263–264 ––and milestones in bilingual language acquisition, 249–250 ––and view of bilingualism as cognitive adaptation, 264–265 ––visual language discrimination, 262 ––word learning strategies, 262–263 Cognitive flexibility, 273–274 Cognitive function, language selection and, 270–273 Cognitively stimulating activities, 306, 307 Cognitive reserve, 305–318 ––and aging, 307–313 ––and bilingual benefit theory, 307 ––and cognitive control, 288 ––defined, 288, 306–307 ––and dementia, 15, 313–317 Cognitive screening test scores, 312 Cognitive system, influence of language on, 285 Cohen-Mansfield, J., 312 Cohort studies ––of aging, 312–313, 317 ––of dementia, 315–317 Comeau, L., 25, 83 Commins, N. L., 184 Communication ––of bilingualism study results, 328–329

Index 

––code-mixing as strategy for, 25 ––cognitive flexibility for improved, 274 Community attitudes, about bilingualism, 279 Competency, language, 83, 95, 96 Competition resolution, 286, 288 Complexity, linguistic, 129–130 Comprehension ––defined, 67 ––language, 26, 229–230 ––reading, 234–236 ––word, 66–71 Computerized Language Analysis (CLAN), 86 Conboy, B. T., 27, 69 Conceptual vocabularies, 26 Conflict monitoring, 286, 288 Confusion, language, 14, 43, 82, 94, 249 Conrad, M., 294 Constraint-based (CB) approaches, 194–195 Constraint-free (CF) approaches, 194 Conversational analysis (CA) approach, 187 Conversational code-switching, 186 Core, C., 113 Correlation coefficients, 171 Costa, A., 235, 263 Covert components (minimalist model), 190 Covert movements, 190 CPH/L2A (critical period hypothesis for L2 acquisition), 164–175 Crago, M., 28, 104 Craik, F. I. M., 13, 311, 317 Crane, P. K., 315 Crib bilingualism. See Infants Critical period hypothesis for L2 acquisition (CPH/L2A), 164–175 ––census studies related to, 167–168 ––deficits with, 176 ––described, 163, 164 ––discontinuity prediction in, 166–175 ––in foreign language classroom, 175 ––nature-of-bilingualism view in, 165–166 ––non-nativelikeness prediction in, 164–166 ––task-based studies related to, 168–174 Critical periods, 6, 164, 166 Cross-domain comparisons of language input, 113–114 Cross-language activation, 7, 225–232 Cross-language processes ––in bilingual reading, 225 ––within sentence context, 232–236 Cross-language synonyms. See Doublets

 335

Cross-linguistic influence ––for bilingual children, 273 ––on grammatical development, 27–28 ––on vocabulary scores, 113–114 Cultural identity, of bilingual speakers, 286 Culture ––and interpretation of emotional events, 295–296 ––and social cognition of children, 277 Cummins, J., 184 Curdt-Christiansen, X. L., 36 Dale, P. S., 27 Danielson, D., 56, 58 Datta, H., 50 Davison, M. D., 104 Day care, language input from, 106–107 Deary, I. J., 312–313 Deficit perspective on bilingualism, 10, 13 Degner, J., 294 De Houwer, A., 26, 27, 32–33, 37, 64–65, 104–105, 125 De Jong, E. J., 219 DeKeyser, R., 166, 167, 170–172 Dementia ––and cognitive reserve, 288 ––defined, 313 ––effects of bilingualism on, 15, 313–317 ––executive functions of patients with, 307 Demographics ––of schools with DL programs, 207–208 ––of students in DL programs, 204–206 Determiners, 27, 112, 114 Deuchar, M., 24, 29 Development. See also Language development; Lexical development ––bilingual, 34, 37 ––grammatical, 12, 27–28 ––neurocognitive, 14 ––phonological, 12, 29–30, 33 ––reading, 213, 214 ––theory-of-mind, 277–278 Developmental bilingual education, 204 Dialects, 74 Differentiated language system hypothesis, 23, 250, 261 Differentiation ––language, 23, 25, 326 ––phonetic, 70–71 ––phonological, 28–29 ––visual, 45–46, 56, 58, 65, 262, 265

336 

 Index

Dijkstra, T., 227 Direct object scrambling, 112–114 Disambiguation heuristics, 262–263 Discontinuity prediction, 166–175 ––census studies of, 167–168 ––controversy around, 174–175 ––described, 166–167 ––task-based studies of, 168–174 Discourse separation, 25 Discrimination. See also Rhythm discrimination ––language, 44–45, 251–252, 262 ––phonetic, 46–51, 57, 58 ––phonological, 57, 154–155 ––tone, 51 Distributed characteristic of bilingual knowledge, 109, 113–114, 128–129 Distributional learning, 47 Diversity ––lexical, 95, 107 ––sampling, 59 DLPFC (dorsolateral prefrontal cortex), 289 DL programs. See Dual language programs Domain-general cognitive factors, 16, 17, 237 Domain-general executive functions, 254, 257, 260 Dorsolateral prefrontal cortex (DLPFC), 289 Doublets, 68–70, 250. See also Translation equivalents Dovidio, J. F., 176 Dual language (DL) programs, 203–220 ––academic achievement and bilingualism outcomes in, 206–215 ––demographics of students in, 204–206 ––English mainstream programs vs., 203–204 ––future research directions on, 218 ––goals of, 203 ––and language proficiency/academic achievement, 216–218 ––school and program variables in, 207–209 ––student backgrounds and outcomes in, 210–215 Dual lexical resources, 24 Durgunoglu, A. Y., 131 Dutch language, gender in, 112 Early bilingualism, 23–37. See also Cognitive effects of bilingualism for infants ––current perspectives on, 36–37 ––dual lexical resources in, 24

––and family language policies, 31–32 ––and grammatical development, 27–28 ––and language choice, 24–25 ––and language input environment, 32–36 ––and literacy, 225 ––and phonological differentiation/ development, 28–30 ––and vocabulary size, 25–27 Early lexical development in infants and toddlers, 63–74 ––input factors in, 73–74 ––word comprehension, 66–71 ––word processing, 71–72 ––word production, 72–73 ––word recognition, 66–67 ––word segmentation, 63–66 Early posterior negativity (EPN) components, 294 Early second-language acquisition (ESLA) children, 33 Early sequential bilingual speakers. See also Proficiency moderating factors ––future research on, 328 ––heritage speakers as, 146 ––language attrition for, 150, 151 Early vocalizations, 29 Early words, 29 Ecke, P., 144 Educational factors, in language retention, 156 Educational system. See also Teachers ––critical period hypothesis and, 175 ––language input research, 116 ––policies on bilingual education, 328–329 Education level ––and cognitive reserve, 306, 307, 318 ––and dementia, 314, 316–317 ––of parents, 134, 210–211, 216–217 Eilers, R. E., 133 Eisenberg, A., 270 EL (embedded language), 194, 195 Electrophysiological studies, of emotion processing, 294 Elementary schools, DL programs in, 204 Ellis, A. E., 254 ELs. See English learners Elston-Güttler, K. E., 231 Embedded language (EL), 194, 195 Emotion ––and cognitive control, 285–286, 296–298 ––effect of biculturalism on, 295–296

Index 

––influence of language on, 285 ––language as filter for, 299 Emotional aspects, of family language policies, 32 Emotional events, interpretation of, 295–296 Emotional memory, 294–295 Emotional processing, 292–295, 298–299 Emotional Stroop task, 293–294 Emotional valence, 294, 297 Encoding, sound, 70–71 Engagement, social, 34, 306 English as L2 classrooms, 107, 114 English learners (ELs) ––backgrounds of, 216–217 ––dual language education for, 204, 205 ––educational level of parents of, 211 ––language proficiency of, 212–213 English mainstream programs ––academic achievement of students in, 207–208, 218 ––dual language programs vs., 203–204, 207–209 ––language proficiency for students in, 207–209 Environment ––language, 32–36, 84–85 ––learning, 12 ––social, 84–85, 95 Environmental cues, for language selection, 231–232 Episodic memory, 309–310 EPN (early posterior negativity) components, 294 Equality, linguistic, 185 Equivalence constraint, 188 ERP research. See Event-related potential research ESLA (early second-language acquisition) children, 33 Ethnic background, student outcome and, 210 Event-related potential (ERP) research ––on cross-language processes, 233, 235–236 ––on parallel activation, 228–229, 232 ––on phonetic perception, 49–50 ––on word form recognition, 53 ––on word processing, 72 ––on word recognition, 66–67, 69 Executive functions (executive functioning) ––and aging, 307 ––and bilingualism in childhood, 276

 337

––and bilingualism in infancy, 252–257 ––and bilingual reading, 237 ––effects of bilingualism on, 26, 308 ––and visuospatial tasks, 275–276 Experimental studies, on aging, 317 Experimental tasks, for measuring word comprehension, 68 Exposure, language. See Language exposure Expressive language, 107 Expressive vocabulary scores, 72–73 Eyal, N., 312 Eye-gaze tracking, 69 False belief tasks, 277 Familiar words ––infants’ recognition of, 52–53 ––in phonological perception tests, 55–56 ––in word segmentation tests, 64 Family language. See also Home language; Minority language ––attrition of, 142, 153 ––incomplete acquisition of, 141 ––parental and child use of, 104–105 Family language policies (FLP), 327 ––for BFLA children, 37 ––influence of older children on, 35 ––influence on early bilingualism of, 31–32 Federmeier, K. D., 233 Fennell, C. T., 46, 54, 55, 58, 70, 126, 265 FEP students. See Fluent English proficient students Ferko, D., 214 Fernald, A., 69, 110 Fernández, S. C., 24, 68 Ferré, P., 293 First language (L1), influence of second language on, 165, 228 First language attrition, 141–157 ––and age effects/availability of input, 145–156 ––defined, 141–142 ––for heritage speakers, 145–153 ––in immigration context, 142–145 ––for international adoptees, 154–156 Fischer, C. E., 317 Fishman, J. A., 32 Flanker Task, 287 Flege, J. E., 176n5 Flexibility ––cognitive, 273–274 ––memory, 263–264

338 

 Index

Flores, C., 152–153 FLP. See Family language policies Fluency ––and language attrition, 142–143 ––reading, 15–16, 312–313 ––semantic, 112–113, 310 ––verbal, 236, 310 Fluent English proficient (FEP) students ––backgrounds of, 216–218 ––educational level of parents of, 211 ––program variables and rate of, 209 Fluidity, of home language input, 126–127 fMRI (functional magnetic resonance imaging) studies, 289 fNIRS (functional near-infrared spectroscopy), 50 Foong, J., 53 Foreign language classrooms, 175 Foucart, A., 235 Fraga, I., 293 Francis, N., 275 Free morpheme constraint, 188, 191, 194 Frequency argument, 47–48 Frequency-dependent hypothesis, 228 Fuller-Thomson, E., 315 Functional magnetic resonance imaging (fMRI) studies, 289 Functional near-infrared spectroscopy (fNIRS), 50 Fundamental frequency, 67 Galasko, D. R., 316 Gámez, P. B., 107 García, G. E., 131 García, O., 187 García, T., 293 García-Sierra, A., 34 Garratt, L. C., 275 Gathercole, V. C. M., 104, 127, 212 Gatt, D., 125–126, 134 Gender, 112. See also Grammatical gender Gender morphology, language attrition and, 147–148 Generalizability, of language acquisition theories, 3–4 Generalization, memory, 263, 265 Genesee, F., 11, 12, 23, 25, 28, 83, 104 Gerken, L., 47 Gianico, J. L., 293–294 Gildersleeve-Neumann, C., 29

Gillam, R. B., 107 GJTs (grammaticality judgment tasks), 168–172 Glaeser, B., 214 Gluszek, A., 176 Goetz, P. J., 277 Gold, B. T., 311 Goldstein, B., 29 Gollan, T. H., 316 Goodz, N. S., 84 Gorrell, J. J., 275 Grabois, H., 154 Grady, C. L., 311 Grammar ––knowledge of, 125, 135 ––and language acquisition, 146–147 ––and language attrition, 144, 145, 148–153 ––language attrition and errors in, 83, 143 ––and language input, 110–112, 127 ––story, 114 Grammatical constraints ––for code-switching, 14, 83, 183–184 ––language loss and errors with, 83 ––young children’s use of, 11 Grammatical development, 12, 27–28 Grammatical gender ––children’s conceptualization of, 272–273 ––complexity of and proficiency with, 129–130 ––and language input, 111–112, 114 Grammaticality judgment tasks (GJTs), 168–172 Grammatical studies of code-switching, 188–195 Grandparents, 36 Granena, G., 146, 172–174 Gray matter, 288, 311–312 Green, D. W., 290 Gregory, E., 36 Grüter, T., 110 Guiberson, M., 83 Gullifer, J. W., 231 Gumperz, J. J., 81, 185–187 Gürel, A., 144 Habituation patterns, 263–264 Hakuta, K., 168 Hall, C. B., 315–316 Hall, C. J., 144 Hammer, C. S., 104 Hancin-Bhatt, B., 131

Index 

Hansson, P., 309 Harmonious bilingual development, 37 Harper, S. N., 276 Haugen, E., 183 Head (minimalist model), 190, 192–194 Head-turn preference procedure (HPP) ––in phonemic change tolerance tests, 53–54 ––in word form recognition tests, 52–53 ––in word recognition tests, 66 Heritability conditions, for code-switching, 189 Heritage language reversal, 141–142, 152–153 Heritage speakers ––DL program outcomes for, 215 ––first language attrition for, 145–153, 156–157 Hernández, A. E., 216, 298 Heteroglossia, 187 High amplitude sucking procedures, 44, 65 High schools, DL programs in, 206 Hill, K., 47 Hoff, E., 26, 33, 104–107, 113–115, 126, 127 Hohenstein, J., 270 Holowka, S., 24 Home(s) ––influence of, on early bilingualism, 32–36 ––language input in, 104–106, 124–126 ––low-income, 211 Home language. See also Family language; Minority language ––acquisition of, 124–125 ––code-switching in children and exposure to, 93 ––in developmental bilingual education, 204 ––fluidity in, 126–127 ––quantity of input in, 104–105, 116, 205 ––semantic fluency in, 112 ––sociocultural context for, 110, 126–127 ––and socioeconomic status, 134–135 ––sources of input in, 32, 34–36 ––and vocabulary size, 26 Homographs, 227, 231 Howard, E. R., 213 HPP. See Head-turn preference procedure Huang, B. H., 175 Hudon, T. E., 53–54 Hurtado, N., 69, 110 Hybrid language practices, 187 Hyltenstam, K., 146, 154, 155, 165, 175 Ibáñez, A. J., 230 Idea Proficiency Test I-Oral Spanish, 184

 339

Identity ––cultural, 286 ––social, 187, 196 IDS (infant-directed speech), 33–34 Immersion programs, 204, 209, 214 Immigrant populations ––biculturalism in, 295–296 ––bilingualism in U.S., 325 ––cognitive reserve of, 314–315 ––in discontinuity studies, 167–174 ––in emotional memory studies, 295 ––first language attrition in, 141–153 ––language choice by, 35 ––language input in home for, 126 ––learning of majority language in, 12 Immigration status, dementia and, 314–316 Infant-directed speech (IDS), 33–34 Infants. See also Cognitive effects of bilingualism for infants; Speech perception by infants ––bilingualism studies with, 10–12 ––language attrition for, 154 ––lexical development by, 63, 73–74 ––linking of word forms to meanings by, 69–71 ––phonological differentiation by, 29 ––proficiency factors for, 131, 134 ––word comprehension by, 67–69 ––word processing by, 69 ––word production by, 72–73 ––word recognition by, 66–67 ––word segmentation by, 63–66 Inhibition ––and aging, 307 ––effects of bilingualism on, 308 ––and language production, 16 ––of parallel activation, 227, 229–231 ––of previously rewarded responses, 255, 257 ––response, 287, 288 Inhibitory control. See also Cognitive control ––of bilingual adults, 287–288 ––as cognitive effect of bilingualism, 16 ––interaction of emotion and, 285 ––and language processing, 16 ––and language switching, 289–290 ––and lexical activation, 290, 291 ––and parallel activation of languages, 289 Inhibitory control model, 289–290 Input, language. See Language input Input–proficiency–use cycle, 108 Intelligence test scores, 13, 312–313

340 

 Index

Interference suppression, 286–288 Intergenerational language transfer, 32, 36 Interlingual homographs, 227 Interlocutor sensitivity, language choice and, 25 International adoptees, language attrition for, 154–157 Intersentential code-switching, 183 Interutterance switches, 88 Intrasentential code-switching, 73, 88, 183, 233 Izzard, Eddie, 43 Jacobs, A. M., 294 Jancosek, E. G., 83 Jessel, J., 36 Jeung, C., 126 Jia, G., 106, 126 Johnson, J. S., 146, 168–171, 174 Johnson, N. F., 311 Joint attention episodes, 34 Jones, B., 187 Josefsson, M., 309 Jun, S. A., 154, 155 Jusczyk, P. W., 52, 66 Kager, R., 50 Kan, P.-F., 275 Kandhadai, P., 56, 58 Katz, M. J., 315–316 Kaushanskaya, M., 295, 296, 299 Kavé, G., 312 Kelley, A., 131 Kelly, T. P., 275 Kennedy, K. D., 35 Kenner, C., 36 Kim, H. Y., 126 King, K. A., 31 Kirova, A., 113, 114, 128–129 Kohnert, K., 131, 275 Kopeliovich, S., 32, 35 Kousaie, S., 308 Kovács, Á. M., 254–261 Krehm, M., 55, 70 Kroll, J. F., 235–236, 286 Kuh, D., 315 Kuhl, P. K., 34, 50 Kuipers, J. R., 72 Kutas, M., 233 L1 (first language), 165, 228 L2. See Second language

Labov, William, 185, 186 Lambert, M., 13 Lambert, W. E., 325 Language(s) ––code-switching by adults as use of, 185–188 ––and cognitive system, 285 ––as filter for emotion and cognitive control, 299 ––interactions between, 130–133, 225, 236–238, 288–290 Language ability, 123 Language acquisition. See also Age of secondlanguage acquisition; Critical period hypothesis for L2 acquisition (CPH/L2A) ––by BFLA children, 23 ––Chomskyan view of, 142 ––and exposure in home, 124–125 ––first language attrition vs. incomplete, 146–147 ––generalizability of theories on, 3–4 ––and language input, 104–106, 115 ––and lexical development, 63 ––milestones in bilingual, 249–250 ––monolingual vs. bilingual, 30–31, 249–250 ––rate of, 114 ––for simultaneous bilingual speakers, 10–12 ––socioeconomic factors in, 133–135 ––and vocabulary development, 109 Language activation. See also Parallel activation ––cross-, 7, 225–232 ––effects of bilingualism on, 288–290 ––and language attrition in adults, 143 ––measuring time course of, 69 Language Assessment Scales-Oral Español, 184 Language attrition, 141. See also First language attrition Language choice. See also Language selection ––and early bilingualism, 24–25 ––and emotional memory, 295 ––influence of child’s, 35 ––and lexical proficiency/exposure, 126 ––social context for, 186 Language competency, 83, 95, 96 Language comprehension, 26, 229–230 Language confusion ––and bilingualism in infancy, 43, 249 ––code-switching as sign of, 14, 82, 94 Language cues, in text, 230 Language-dependent memory hypothesis, 296

Index 

Language development ––in BFLA babies, 23 ––and code-switching by children, 95 ––and infant-directed speech, 33–34 ––influence of language input on. See Language input ––for monolingual vs. bilingual infants, 58 ––sociopolitical context for, 327 ––variability in, 103 Language differentiation, 23, 25, 326 Language discrimination, 262 Language discrimination, by infants, 44–45, 251–252 Language education programs, 203–204 Language environment ––and code-switching by children, 84–85 ––and early bilingualism, 32–36 Language exposure ––assessing monolingual and bilingual children based on, 112 ––and bilingual outcomes, 33 ––and cross-language processes, 232 ––and lexical development, 26–27 ––and morphosyntax, 111 ––and processing efficiency, 110 ––and proficiency, 124–127, 208–209 ––and quality of input, 108–109 ––and reading fluency/perceptual reading span, 15–16 ––and vocabulary acquisition rate, 110 ––and vocabulary size, 68, 73 Language group, dementia for bilingual speakers and, 316, 318 Language input, 103–117 ––and degree of code-switching by children, 83 ––and early lexical development, 73–74 ––and first language attrition, 150–156 ––future research on, 116–117 ––for international adoptees, 154 ––in language acquisition theories, 115 ––language choice based on, 25 ––and language outcomes, 109–116 ––parenting and education implications of research, 116 ––and proficiency, 124–127 ––sources of, 103–109 ––variety and richness of, 106 ––and vocabulary, 25–26 ––word segmentation from, 63–66 Language knowledge, 3

 341

Language learning, 12–13, 34 Language loss, 9, 83. See also First language attrition Language neutral skills, 129 Language output ––inhibitory capacity and, 16 ––as source of input, 108 ––switch costs associated with, 229, 230 Language performance, socioeconomic status and, 133 Language processing, 16, 110, 238 Language proficiency. See also Proficiency moderating factors ––and academic achievement, 216–218 ––and accountability, 219 ––and age of acquisition, 163 ––and code-switching, 81, 185 ––and cognitive control, 291–292 ––and cognitive effects of bilingualism, 278–279 ––defined, 123, 142 ––and emotional processing, 293, 294 ––and exposure to English in DL programs, 208–209 ––of input sources, 106 ––and memory of adults, 15 ––oral, 207–208, 211–214 ––and parallel activation, 228 ––self-reporting of, 316 ––of students in DL programs, 207–209 ––and switching costs, 289–292 ––time-on-task principle for, 208 ––written, 211, 213, 214 Language retention, by international adoptees, 154–155 Language selection. See also Language choice ––in bilingual reading, 229–232 ––by children, 269–270 ––and cognitive function, 270–273 ––cues to, 230–232 ––and executive functions, 253–254 ––for motion events, 270–272 ––for possessive constructions, 271 Language separation, 56, 58, 65 Language switching. See also Switching costs ––by adults, 14 ––brain during, 286, 297–298 ––impact on executive functions of, 253–254 ––and inhibitory control, 289–290 ––and processing, 72 ––by young children, 11

342 

 Index

Language tagging, 189 Language usage, 10–11, 107 Lanza, E., 32, 87 Lapaquette, L., 25, 83 Larson-Hall, J., 170 Late bilingual learners, 225, 226, 234–236 Late exit bilingual programs, 204 Late life recreational activities, 306 Late positive complex (LPC) components, 294 Lawrence, F. R., 104 Learning ––distributional, 47 ––language, 12–13, 34 ––of linguistic structural regularities, 255, 257–262 ––sequential bilingual, 83 ––statistical, 64 ––word, 54, 237, 262–263 Learning environment, 12 Lee, K., 277 Left interior frontal cortex, 50 Lenneberg, E. H., 163 Leopold, W., 10, 325 Levine, S. C., 107 Lewis, G., 187 Lexical activation ––cross-language, 226–232 ––and inhibitory control, 290, 291 Lexical arrays, 190 Lexical development. See also Early lexical development in infants and toddlers ––and code-switching, 82 ––and language exposure, 26–27 ––and language input, 109–110 Lexical diversity, 95, 107 Lexical entry driven approach to syntax, 189–190 Lexical errors, language attrition and, 143 Lexical insertion rules, 189 Lexical overlap, 48–49 Lexical proficiency, 126 Lexical resources, 24, 25 Lexical retrieval, 310–311 Lexicon(s). See also Vocabulary(-ies) ––for code-switching, 189–190 ––defined, 63 ––interconnections between, 71–72 ––in minimalist model, 190 Lexis collocation, 172–174 Libben, M. R., 230–231

Libnawi, A., 273–274 Lindholm-Leary, K. J., 207–208, 212, 213, 215–217, 218 Linguistic cognitive control, 289–291 Linguistic complexity, proficiency and, 129–130 Linguistic confusion. See Language confusion Linguistic equality, 185 Linguistic experience, 298–299 Linguistic signal, consistency of, 261–262 Linguistic structure ––for code-switching, 195–196 ––learning regularities of, 255, 257–262 Lipscomb, T. J., 275 Lipski, J., 187, 188 Lipton, R. B., 315–316 Literacy. See also Bilingual reading ––grandparents’ demonstrations of, 36 ––and language attrition, 152 ––models of successful, 7 ––of students in DL programs, 208, 213 Liu, H. M., 34 Liu, L., 50 Ljungberg, J. K., 309, 313 Long, M. H., 146, 167, 170, 172–174 Longitudinal studies of language attrition, 144–145, 147–148, 154 Loss, language, 9, 83. See also First language attrition Low-income homes, 211 Low-income schools, 207–208 Loxtercamp, A. L., 275 LPC (late positive complex) components, 294 L2. See Second language Luk, G., 13, 109, 311 MacArthur-Bates Communicative Development Inventory (MB-CDI), 67, 68, 73 Macizo, P., 230, 238 MacSwan, J., 184, 212 Maintenance bilingual programs, 204 Majority language. See also Societal language ––first language attrition and schooling in, 141, 145, 146 ––immigrant children’s learning of, 12, 13 ––parental and child use of, 33–34, 104–105 ––socioeconomic status and acquisition of, 134 Mandel, D. R., 66 Mani, N., 71 Marchman, V. A., 27, 69, 110

Index 

Marian, V., 290, 291, 295, 296, 299, 309 Marinis, T., 113–114 Markedness model of code-switching, 186–188 Markman, E. M., 278 Martin, C. D., 234–235 Martin, M., 16 Martín, M. C., 230, 236 Martínez-Sussmann, C., 27 Math scores, of students in DL programs, 206 Matrix language (ML), 194, 195 Matrix language frame (MLF) model, 194–195 Matthews, A., 254 Mattock, K., 55, 70 Maturational effects, 166, 175 Maximal projections, 190 Maye, J., 47 Mayr, R., 33 MB-CDI (MacArthur-Bates Communicative Development Inventory), 67, 68 McAllister, H. A., 275 McClure, E., 188 ME (mutual exclusivity), 69–70, 263 Meaning(s) ––cognitive flexibility and acceptance of multiple, 274 ––linking word forms to, 69–71 ––semantic understanding of, 132–133 Mean length of utterances (MLU) ––and code-switching by children, 87, 93–95 ––and morphosyntax, 111 ––of teachers, 107 Mehler, J., 254–261 Meisel, J. M., 81, 83 Memory ––development of, 269 ––effect of bilingualism on, 308–310 ––emotional, 294–295 ––of emotional words, 293 ––episodic, 309–310 ––language-dependent hypothesis, 296 ––phonological, 74, 113 ––working, 14–15 Memory flexibility, 263–264 Memory generalization, 263, 265 Mendez-Perez, A., 107 Merge operation, in minimalist model, 190 Mesman, J., 35 Metaphorical code-switching, 186 Metrolingualism, 187

 343

Miccio, A. W., 104 Middle schools, DL programs in, 206 Milk, R. D., 185 Mills, D. L., 69 Minimal grasp, 34 Minimalist model of code-switching, 190–194 Minimal pairs, sensitivity to, 265 Minority language. See also Family language; Home language ––attrition of, 141, 145, 146 ––and bilingual development, 34–35 ––development of, 104–105 ––in home, 104–105, 124–125 ––incomplete acquisition of, 148 ––proficiency moderating factors for, 124–127 Miramontes, O. B., 184 Mismatch response (MMR), 49–50 ML (matrix language), 194, 195 MLF (matrix language frame) model, 194–195 MLU. See Mean length of utterances MMR (mismatch response), 49–50 Molero, M., 293 Molis, M., 169–171, 174 Molnar, M., 45, 47 Monolingual readers, bilingual vs., 238–239 Monolingual speakers, bilingual vs. ––in childhood, 269–270, 278 ––cognitive control of, 287–288 ––cognitive development of, 278 ––cognitive flexibility of, 273–274 ––cognitive systems of adult, 285, 288 ––cohort studies of, 312–313 ––dementia in, 313–317 ––episodic memory of, 309–310 ––executive functions of, 237, 276–277, 308 ––language acquisition milestones for, 249–250 ––language comprehension of, 69 ––language development of, 58 ––language discrimination by, 44–45 ––language proficiency of, 124 ––learning by, 12–14 ––lexical development of, 63 ––linguistic and nonlinguistic control for, 290–292 ––linking of word forms and meanings by, 70–71 ––morphosyntax for, 111, 112 ––nativelikeness criteria for, 164–166, 176 ––neuroanatomy of, 311–312, 317

344 

 Index

––phonetic discrimination by, 46–50 ––phonological perceptions of, 54–56 ––phonotactic sensitivity of, 51–52 ––prenatal speech perception by, 43–44 ––semantic understanding of meaning for, 133 ––social cognition of, 277–278 ––studies including, 9–10 ––tolerance of phonemic change for, 53–54 ––visuospatial tasks of, 275–276 ––vocabulary development of, 109–110, 125–126 ––vocabulary of, 310–311 ––vocabulary size of, 26, 68 ––word form recognition by, 52–54 ––word learning by, 69–70, 237 ––word processing by, 72 ––word segmentation by, 64 Montanari, S., 25, 33, 213, 214 Montoya, R. I., 316 Montreal, Canada, 84 Montrul, S., 144, 150, 151, 155–156 Mora-timed languages, 43 Moreno, E. M., 233, 235 Morgan-Short, K., 236 Morphology ––gender, 147–148 ––nominal, 111 ––plural, 111 ––verbal, 27, 111 Morphosyntactic systems, development of, 12, 27 Morphosyntax ––in language acquisition, 147 ––and language attrition, 145, 147–148 ––and language input, 110–112, 114 ––and language retention by international adoptees, 155–156 ––testing of discontinuity prediction in, 172–174 Morton, J. B., 276 Moses, L. J., 277 Motion events, 270–272 Multiculturalism, 326 Multi-languaging, 187 Multilingualism, 9 Multilingual speakers ––cognitive screening test scores for, 312 ––dementia in, 315 ––language choice by, 25 ––phonological development for, 33

Mutual exclusivity (ME), 69–70, 263 Myers-Scotton, C., 185, 186 Nagy, W. E., 131 Naigles, L., 270 Narrative abilities, 114, 128–129 Native English speakers (NESs) ––backgrounds of, 216 ––in dual language programs, 205–206 ––immersion programs for, 204 ––outcomes of DL programs for, 211, 214 Native language ––attrition of. See First language attrition ––discrimination of, 44–46, 252, 262 ––emotion-laden words in, 292 ––English learners proficiency in, 213 ––immigrant parents’ use of, 219–220 ––infant’s attending to, 43 ––phoneme detection in, 54, 64 ––phonetic perception of, 46–51 ––in two-way immersion education, 204 Nativelikeness criteria, 164–166 Nativelike reading, 234–236 Native other language speakers (NOLSs), 205, 211–213 Native speaker effect, 211, 219 Native speakers ––defined, 142 ––language attrition for, 156–157 ––language input from, 105–106, 127 Nativist theories of language acquisition, 115 Naturalistic observation study of codeswitching, 85–94 Nature-of-bilingualism view, 165–166 Nava, N., 185 NDWR (number of different word roots), 87, 93–95 Near-cognates, 131–132 Negation, 27, 192 Negative transfer, 30 Negotiation principle, 186–187 Nelson, C. A., 254 NESs. See Native English speakers Neural compensation, 306, 307 Neural networks, 297–298 Neural reserve, 306 Neuroanatomical effects ––of aging, 311–312 ––of dementia, 317 Neurocognitive development, 14

Index 

Newmeyer, F. J., 185 Newport, E. L., 146, 168–171, 174 Nicoladis, E., 11, 25, 83, 84, 104, 110, 154, 271, 272 Nilsson, L.-G., 309 Nissan, J. J., 312–313 NOLSs (native other language speakers), 205, 211–213 Nominal morphology, 111 Nonlinguistic attentional tasks, 276 Nonlinguistic cognitive control, 289–291 Non-nativelikeness prediction, 164–166, 176 Nonnative phoneme discrimination, 50–51 Nonnative speakers, language input from, 74, 106 Nonnative tone discrimination, 51 Nonverbal cues, 277–278 Nonword repetition (NWR) tasks, 113 Novel word–object pairings, 54–55, 70 Number of different word roots (NDWR), 87, 93–95 Numerations (lexical arrays), 190 NWR (nonword repetition) tasks, 113 Oakes, J., 184 Object expression, 152–153 O’Brien, G., 212 Oh, J. S., 154, 155 Oller, D. K., 24, 68, 114, 128, 129, 133 One object–one label heuristic, 263 One-person-one-language approach (OPOL), 25, 34 One-way immersion education (immersion programs), 204, 209, 214 Opitz, B., 294 OPOL (one-person-one-language approach), 25, 34 Oral language proficiency ––of English learners, 212–213 ––of native English speakers, 214 ––and school demographics, 207–208 ––for students in DL programs, 211–214 O’Toole, C., 125–126, 134 Output, language. See Language output Overt components (minimalist model), 190 Overt movements, 190 Palermo, F., 105 Pallier, C., 154 Palmer, C., 16

 345

Paradis, J., 11, 28, 83, 104, 110, 113–115, 128–129 Parallel activation ––in bilingual reading, 225–229 ––and cognitive effects of bilingualism for adults, 288–289 ––and emotional valence of words, 297 ––linguistic and nonlinguistic cognitive control in, 290–292 ––and theory of bilingual benefit, 307 Paraphrasing, 238 Parental discourse strategies, 34 Parents ––code-switching rates for children and, 84 ––educational levels of, 134 ––family language policies of, 31–32 ––influence of, on early bilingualism, 32–34 ––language input from, 104–105, 116, 126, 127 Parra, M., 113 Partner languages (in DL programs) ––defined, 204 ––English learners’ proficiency in, 213 ––level of exposure to, 208–209 ––native English speakers’ proficiency in, 214 ––proficiency with, 208–209, 211 ––students’ self-ratings of proficiency in, 214–215 Passive participle, 147 Paulston, C. B., 184 Peabody Picture Vocabulary Test, 86, 107, 109 Peal, E., 13 Pearson, B. Z., 24, 35, 68, 128 Peers ––code-switching by children with, 95 ––influence on language policy of, 32 ––language input from, 105, 126 Peets, K. F., 109 Peña, E. D., 107, 131 Penfield, W., 163 Perceptual attentiveness, 262 Perceptual narrowing, 46, 51, 262 Perceptual reading span, 15–16 Pérez, A. M., 131 Pérez-Leroux, A. T., 28 Pérez-Tattam, R., 127 Performance-based phenomenon, code-switching as, 83 Perpiñán, S., 144 Petitto, L. A., 24, 50 Pfaff, C., 188

346 

 Index

PFC (prefrontal cortex), 297 Phillips, N. A., 308 Phonemes, contrasting, 29, 127 Phonemic changes ––detection of, 54–56 ––processing of, 48–50 ––tolerance for, 53–54 Phonetic differentiation, 70–71 Phonetic discrimination, 46–51, 57, 58 Phonetic perception, 46–51 Phonetic regularities, 64 Phonetic representations, 49–50, 55 Phonics, profile effects for, 128 Phonological components (minimalist model), 190 Phonological detail, in novel words, 58 Phonological development, 12, 29–30, 33 Phonological differentiation, 28–29 Phonological discrimination, 57, 154–155 Phonological form, interactive influences in, 131–132 Phonological memory, 74, 113 Phonological perception, 54–56 Phonological priming effects, 71 Phonology testing, 54–56 Phonotactics, 51–52, 57, 64 Picture naming, 128 Pieper, S., 35 Pierce, L. J., 155 Pirvulescu, M., 28 Pisoni, D. B., 66 Pivneva, I., 16 Place, S., 33, 106, 115, 127 Planning, aging and, 307 Plasticity, in bilingual reading, 226 Plural morphology, 111 Polka, L., 47, 55, 70 Polylanguaging, 187 Polylingual languaging, 187 Pons, F., 49 Poplack, S., 81, 188, 189, 191, 194 Positive transfer, 30 Possessive constructions, 271 Potowski, K., 219 Poulin-Dubois, D., 26, 69 Powell, D. K., 311 Pray, L., 212 Predictions, of word features, 234–236 Preferential looking tasks, 55–56, 68–69 Prefrontal cortex (PFC), 297

Prenatal speech perception, 43–44 Preschoolers. See also Young children ––executive functions of, 253–254 ––language development of, 269 ––language input for, 106–107 ––language proficiency of, 212 ––social cognition by, 278 Prevoo, M. J. L., 35 Processing ––emotional, 292–295, 298–299 ––language, 16, 110, 238 ––and language attrition in adults, 143 ––of words. See Word processing Productive vocabulary, 110 Proficiency. See also Language proficiency ––bilingual, 205–207, 214–220 ––input–proficiency–use cycle, 108 ––lexical, 126 Proficiency moderating factors, 123–136 ––future research directions, 135–136 ––influences of languages on one another, 130–133 ––language input and exposure, 124–127 ––linguistic complexity, 129–130 ––profile effects, 128–129 ––socioeconomic level, 133–134 Profile effects ––defined, 5–6 ––and linguistic complexity, 130 ––for proficiency, 128–129 ––for vocabulary scores, 113–114 Pronunciation ––social impressions of, 176 ––testing discontinuity prediction with, 172–174 ––testing phonological perception with, 54–56 Property theories of language acquisition, 115 Prosody, 64, 65, 251 Prospective studies of dementia, 314 Puberty, language attrition at, 151–152 Public policy, bilingualism in, 329 Pure utterances, 89, 94 Putnick, D. L., 125 Quay, S., 24, 25, 29, 36 Ramirez, A. G., 185 Ramírez-Esparza, N., 34 Ramon-Casas, M., 56, 125, 131, 279 Ravid, D., 147, 171–172

Index 

Reading. See also Bilingual reading ––language cues in, 230 ––profile effects for, 128 ––by younger vs. older bilingual speakers, 15–16 Reading achievement ––in dual language programs, 206 ––and language proficiency, 216–218 ––and program type, 209 ––and school demographics, 207–208 ––and socioeconomic status, 211 Reading comprehension, 234–236 Reading development, 213, 214 Reading fluency, 15–16, 312–313 Reading scores, of students in DL programs, 206, 209 Receptive bilingual speakers, 153 Receptive vocabulary, 110 Recio, G., 294 Recreational activities, late life, 306 Referents ––associating new words to, 69–71 ––cognitive flexibility with, 274 ––identification of, 262–263 Regression analysis, 171 Repetition, word recognition and, 66 Repetitive structures, recognition of, 257–262 Representations, phonetic, 49–50, 55 Reserve, cognitive. See Cognitive reserve Residence, length of, 169n1 Response inhibition, 287, 288 Response-switching tasks, 254–255 Restructuring verbs, 193–194 Retention, language, 154–155 Retrieval, lexical, 310–311 Retrospective studies ––of aging, 317 ––of dementia, 314, 315 Reversing language shift, model for, 32 Revised hierarchical model, 293, 294 Rhythm discrimination ––by infants, 261 ––language discrimination based on, 251, 252 ––and language separation, 65 ––in speech perception, 43–46, 57 ––and word discrimination, 64 Ribot, K. M., 104, 107, 126 Rivera-Gaxiola, M., 50 Rizzi, L., 193–194 Roberge, Y., 28

 347

Roberts, L., 163 Rolstad, K., 184 Romo, H. D., 35 Ronjat, J., 10, 325 RO set (set of rights and obligations), 186 Ruby, M., 36 Rumiche, R., 104 Rvachew, S., 55, 70 Sabbagh, M. A., 277 Salmon, D. P., 316 Sampling diversity, 59 Sánchez-Casas, R., 293 Sánchez-Walker, N., 144 Sanders, A. E., 315–316 Sankoff, D., 188, 189, 191, 194 Saunders, W., 212 Sauve, D., 11, 12 Scheizer, T. A., 317 Schelletter, C., 131 Schmid, M., 141, 144, 145 School(s) ––influence on language policy of, 32 ––language input from, 106–107 ––language learning in, 12–13 ––low-income, 207–208 ––support for bilingual proficiency in, 219–220 ––variables in DL programs based on, 207–208 ––views of code-switching in, 183–185 School-age children ––bilingualism studies with, 12–14 ––language proficiency of, 212–213 ––vocabulary size of, 26 Schroeder, S. R., 309 Schwartz, M., 32 Scutellaro, A., 48 Sebastián-Gallés, N., 45–49, 52, 56, 262, 263 Secondary schools, DL programs in, 206, 219, 220 Second language (L2). See also Age of secondlanguage acquisition; Critical period hypothesis for L2 acquisition (CPH/L2A) ––cognitive benefits of learning, 313 ––English as L2 classrooms, 107, 114 ––influence of, on first language, 165, 228 ––and language attrition, 144, 150 ––literacy of adults in. See Bilingual reading Second-language immersion programs, 12 Second-language learners, 9, 205. See also Native other language speakers

348 

 Index

Segregated schools, 207–208 Selection, language. See Language selection Selective attention, 13, 269, 276–277 Select operation (minimalist model), 190 Semantic fluency, 112–113, 310 Semantic priming effects, 71 Semantics ––as cue to language selection, 230–231 ––interactive influences of languages on, 132–133 Semilingualism, 184 Sentence context ––cognitive costs of predicting word features in, 234–236 ––cross-language processes within, 232–236 ––and inhibition of parallel activation, 230–231 Separation, language, 56, 58, 65 Sequence approach to code-switching, 187, 188 Sequential bilingual learning, 83 Serrano, R., 213 SES. See Socioeconomic status Set of rights and obligations (RO set), 186 Set-shifting, 307 Shafer, V. L., 50 Shapero, D., 275 Shared characteristics, of bilingual knowledge, 114 Shorek, A., 312 Siblings ––influence on early bilingualism of, 35 ––language attrition studies with, 148–149 ––language input from, 105, 126, 127 Silva-Corvalán, C., 148–149 Silva-Pereyra, J., 50 Simon task, 253, 286, 287 Simultaneous bilingual speakers. See also Bilingual first-language acquisition (BFLA) children ––defined, 9 ––heritage speakers as, 146, 149 ––language acquisition theories for, 10–12 ––language attrition for, 151 ––phonological differentiation by, 29 ––proficiency moderating factors for. See Proficiency moderating factors ––speech perception by. See Speech perception by infants ––working memory of, 14–15 Singapore, code-switching in, 3, 84–95

Singh, L., 53, 71, 263–264 Situational code-switching, 186 Situational shifting, 81 Slobin, D. I., 269 Smithson, L., 110 Social activities, participation in, 318 Social cognition, 277–278 Social context ––for code-switching by adults, 185–188 ––for second-language acquisition, 175–176 Social engagement, 34, 306 Social environment, 84–85, 95 Social identity formation, 187, 196 Socialization, 30, 31 Societal language. See also Majority language ––parental and child use of, 104–105, 116 ––schooling and use of, 36, 107 ––siblings use of, 35, 105 Socioaffective factors, in language retention, 156 Sociocultural argument for language separation, 56, 58 Sociocultural context ––bilingualism in, 43, 325 ––for grammatical development, 28 ––for language knowledge, 3 ––for language proficiency, 126–127 ––for phonological development, 30 ––and vocabulary development, 110 Socioeconomic status (SES) ––and cognitive reserve, 306 ––and executive functioning, 276–277 ––and language development, 327 ––and language proficiency, 133–134, 216–217 ––of native other language speakers, 205 ––and student outcomes of DL programs, 210–211 Sociopolitical context ––for bilingualism, 325–327 ––for language knowledge, 3 Sorace, A., 144 Sounds, encoding of, 70–71 Spatial role-taking ability, 275 Speaking, thinking for, 269, 272 Speas, M., 190 Speech ––code-switching in text vs., 234 ––infant-directed, 33–34 ––spontaneous, 86–87, 91–93

Index 

Speech pathologists, 328 Speech perception by infants, 43–59 ––limitations of research on, 58–59 ––phonetic perception, 46–51 ––phonology, 54–56 ––phonotactics, 51–52 ––rhythm discrimination, 43–46 ––word form recognition, 52–54 Speech repetition structures, 65–66 Speech sound acquisition, 28–29 Speech therapy, 123 Spell-Out operation (minimalist model), 190 Spolsky, B., 32 Spontaneous speech, 86–87, 91–93 Stadthagen-González, H., 127 Statistical learning, 64 Stevens, G., 167, 168 Stigmatization, 36–37, 176, 325 Story grammar, 114 Stowell, T., 190 Stress patterns, 65. See also Prosody Stress-timed languages, 43 “Stretched L” pattern, 167, 169, 175 “Stretched 7” pattern, 167, 169, 175 “Stretched Z” pattern, 167 Stroop task ––emotional, 293–294 ––and linguistic/nonlinguistic cognitive control, 290 ––measuring cognitive control with, 286, 287 ––with older adults, 308 Subjunctive, 147 Substitutions, 29 Subtractive bilingualism, 325 Successive bilingual speakers, working memory of, 14–15 Sundara, M., 47, 48 Superscript insertion mechanism, 189 Sutton, T. M., 293–294 Swingley, D., 56 Switching costs ––and language proficiency, 289–292 ––in language selection, 229–230 ––in reading of code-switched text, 233 Switch tasks, 70 Syllable-timed languages, 43 Syntactic equivalence, 188 Syntactic priming, 232

 349

Syntax ––in code-switching, 190 ––in language acquisition, 147 ––and language attrition, 143–144 System morpheme principle, 195 Tabouret-Keller, A., 83 Taeschner, T., 10, 23, 27 Tagging, language, 189 Tannenbaum, M., 32 Task-based studies of discontinuity, 168–174 Teachers ––language input from, 106–107 ––views of bilingualism by, 328–329 ––views of code-switching by, 183–185 Tense-aspect and mood (TMA) system, 148–149 Terminal (minimalist model), 190 Thal, D. J., 27 Theory-of-mind development, 277–278 Thierry, G., 72, 297 Thinking for speaking, 269, 272 Thomas, E. M., 111 Thordardottir, E., 110, 111, 113–115 Time-on-task principle, 208 Titone, D., 15, 16, 230–231 TMA (tense-aspect and mood) system, 148–149 Toddlers ––code-mixing and language choice by, 25 ––cognitive control and word processing by, 71–72 ––executive functions of bilingual, 253–254 ––grammatical development of, 27 ––input factors in lexical development for, 74 ––language attrition for, 154 ––language input and outcomes for, 104–106 ––linking of word forms to meanings by, 70 ––vocabulary size of, 26 ––word comprehension of, 68, 69 ––word processing by, 69, 71–72 ––word production by, 72–73 Tone discrimination, 51 Tones, perceptual correlates of, 155 Total conceptual vocabulary, 68 Total vocabulary score, 68 Tracking, educational, 184–185 Transitional bilingual classes, 107 Transition theories of language acquisition, 115 Translanguaging, 187, 188

350 

 Index

Translating, in bilingual reading, 238 Translation equivalents. See also Doublets ––and cognitive flexibility, 274 ––in early bilingualism, 24 ––emotional valence and access to, 297 ––and language acquisition milestones, 249, 250 ––and processing of emotional content of words, 293 ––and proficiency moderating factors, 125, 131 ––and vocabulary size, 25, 26 Translingual practice, 187 Tsao, F. M., 34 Two-way immersion education, 204, 209, 212–213 Ultimate language attainment (UA), 164, 166–176. See also Bilingual attainment Unitary language system hypothesis, 10–12, 23, 249, 261 United States ––bilingualism in immigrant populations, 325 ––challenges with bilingualism, 219–220 ––value of bilingualism, 3 ––views on code-switching, 3 Unsworth, S., 107, 112–115 U.S. Census Bureau, 205 Usage-based theories of language acquisition, 4, 115 Utterances ––defined, 86–87 ––mean length of, 87, 93–95, 107, 111 ––pure, 89, 94 ––switches within, 88 Valdés, G., 183–184, 186 Van Hell, J. G., 235–236 Vanhove, J., 171–172, 173n4, 174 Van IJzendoorn, M. H., 35 Vasilyeva, M., 106 Ventureyra, V., 154 Verbal fluency, 236, 310 Verbal morphology, 27, 111 Verbs, restructuring, 193–194 Verschik, A., 32 Vihman, M. M., 29, 53, 270 Visual differentiation, 45–46, 56, 58, 65, 262, 265 Visual language discrimination, 262

Visual stimuli, executive control tests with, 255, 257 Visuospatial tasks, 275–276 Vocabulary(-ies). See also Lexicon(s) ––conceptual, 26 ––and language input, 114, 125–127 ––of monolingual vs. bilingual adults, 310–311 ––receptive, 110 ––translation equivalents in, 24 ––and use of minority language at home, 104 Vocabulary knowledge ––and shared phonological forms, 131–132 ––and socioeconomic status, 134–135 Vocabulary reasoning, 128 Vocabulary scores ––and language input, 105–107 ––profile effects on, 113–114 Vocabulary size ––and bilingualism as cognitive adaptation, 265 ––and early bilingualism, 25–27 ––and grammatical development, 27 ––and language exposure, 73 ––of monolingual vs. bilingual infants, 68 ––and phonological memory, 74 Vocalizations, early, 29 Vogt, H., 183 Voice onset time (VOT), 166 Voicing contrast, 29 Volterra, V., 10, 23, 27 Von Holzen, K., 71 VOT (voice onset time), 166 Ware, J., 317 Wei, L., 187 Weikum, W. M., 45–46 Weinreich, Uriel, 183 Welsh, S. N., 104 Welsh language, initial consonants in, 53, 67 Wentz, J., 188 Werker, J. F., 43–47, 49, 54, 55, 56, 58, 263, 265 Weyant, J. M., 176 White matter, 288, 311 Whitford, V., 15, 16 Williams, C., 187 Windsor, J., 275 Within-language competition, 289–291 Wittgenstein, Ludwig, 285 Word boundaries, 51 Word comprehension, 66–71

Index 

Word forms ––linking meanings to, 69–71 ––recognition of, 52–54, 57 ––words vs., 52 Word learning, 54, 237, 262–263 Word–object associations, 54–55, 70, 265 Word order patterns ––cross-language influence on, 28 ––and early bilingualism, 27 ––in language attrition studies, 152–153 Word processing ––of emotion-laden words, 292–295, 298–299 ––by infants and toddlers, 71–72 ––and language exposure, 15 ––and lexical development, 74 Word recognition ––by infants, 66–67 ––and inhibitory capacity, 16 ––and language selection cues, 230–231 ––mispronunciation effect in, 56 ––parallel activation for, 226, 227 Words ––early, 29

 351

––interpretations of, 274 ––production of, 72–73 ––segmentation of, 63–66 ––word forms vs., 52 Working memory, 14–15 Written language proficiency, 211, 213, 214 Wu, Y. J., 297 Xu, F., 277 Yamamoto, M., 33 Yang, S., 109 Yavas, F., 127 Yeung, H. H., 49 Yoshida, K. A., 47 Yoshinaga-Itano, C., 83 Young children, 10–12. See also Preschoolers; Toddlers Yow, W. Q., 278 Yu, Y. H., 50 Zagona, K., 192 Zirnstein, M., 235–236 Zwanziger, E. E., 12

About the Editors Elena Nicoladis, PhD, is a professor in the psychology department at the University of Alberta in Edmonton, Alberta, Canada. She received her PhD in developmental psychology at McGill University. Her dissertation focused on the language choice of preschool French–English bilingual children. Dr. Nicoladis’s research focuses on several areas. One area focuses on language acquisition in bilingual children, in particular, how bilinguals use their two languages. The bilinguals in her studies include not only French–English bilinguals but also bilinguals who speak other minority languages, like Hindi and Chinese, as well as English. Another area focuses on the relationship between language and thought in development, including how language use and/or structure might affect children’s thinking. A third area raises questions about how manual gestures are related to language use and development on the one hand and cognitive development on the other. Simona Montanari, PhD, is an associate professor in the Department of Child and Family Studies at California State University, Los Angeles. She received her PhD in linguistics from the University of Southern California, specializing in early trilingual development. Her research, which has focused on the acquisition of phonology, lexicon, morphosyntax and communicative competence in two or three languages, has been published in the Journal of Child Language, in Bilingualism: Language and Cognition, and in the International Journal of Bilingualism, among others. Dr. Montanari has also been involved in the creation and implementation of the first public Italian/English dual language program in the Glendale Unified School District (Los Angeles County), and has been studying its educational outcomes since its launch in 2009. At California State University, Los Angeles, Dr. Montanari teaches graduate and undergraduate courses on language development and second-language acquisition in childhood, and she is currently involved in research on phonological development in Spanish–English dual language preschoolers. She also continues to work as a consultant for the Glendale Unified School District and has designed common-core-standards-aligned curricula for its Italian/English, French/English, and German/English dual-language programs.