Writing Development in Children withHearing Loss, Dyslexia, or OralLanguage Problems

1Writing Development in Children with Hearing Loss, Dyslexia, or Oral Language ProblemsImplications for Assessment and Instruction

Edited by

Barbara Arf

Julie Dockrell

Virginia Berninger

1Oxford University Press is a department of the University ofOxford. It furthers the Universitys objective of excellence in research,scholarship, and education by publishing worldwide.

Oxford NewYorkAuckland Cape Town Dar es Salaam Hong Kong KarachiKuala Lumpur Madrid Melbourne Mexico City NairobiNew Delhi Shanghai Taipei Toronto

With offices inArgentina Austria Brazil Chile Czech Republic France GreeceGuatemala Hungary Italy Japan Poland Portugal SingaporeSouth Korea Switzerland Thailand Turkey Ukraine Vietnam

Oxford is a registered trademark of Oxford University Pressin the UK and certain other countries.

Published in the United States of America byOxford University Press198 Madison Avenue, NewYork, NY 10016

Oxford University Press 2014

All rights reserved. No part of this publication may be reproduced, stored ina retrieval system, or transmitted, in any form or by any means, without the priorpermission in writing of Oxford University Press, or as expressly permitted by law,by license, or under terms agreed with the appropriate reproduction rights organization.Inquiries concerning reproduction outside the scope of the above should be sent to theRights Department, Oxford University Press, at the address above.

You must not circulate this work in any other formand you must impose this same condition on any acquirer.

Library of Congress Cataloging-in-Publication DataWriting development in children with hearing loss, dyslexia, or oral language problems: implications for assessment and instruction / edited by Barbara Arf, Julie Dockrell, and Virginia Berninger. pages cmIncludes bibliographical references and index.ISBN 97801998272821. ChildrenWriting. 2. Children with disabilities. 3. Child development. I. Arf, Barbara, editor of compilation.LB1139.W7W75 2014371.90446dc23 2014004456

9 8 7 6 5 4 3 2 1Printed in the United States of Americaon acid-free paper

This book is dedicated to Karen, Michael, Stephen, Ray and Vanessa and all the young writers who collaborated over the years in our studies as well as those who participated in the writing studies of all the contributors to this volume. Our greatest thanks goes to them, for their personal contribution to the work of writingresearchers.

CONTENTS

Preface xiContributors xiiiIntroduction xvii

PART ONE:Models and Perspectives on Writing Development1. Cognitive Processes in Writing:AFramework 3

John R.Hayes and Virginia W. Berninger2. Linguistic Perspectives on Writing Development 16

Ruth A. Berman3. Two Metaphors for Writing Research and Their Implications for

WritingInstruction 33Pietro Boscolo

PART TWO:The Impact of Oral Language Problems on Written TextProduction

Children with Hearing Loss4. Spelling in Deaf Children with Cochlear Implants:Implications

forInstruction 45Heather Hayes, Rebecca Treiman, and Ann E.Geers

5. Spelling Acquisition in French Children with Cochlear Implants: ACase-StudyInvestigation 55Sophie Bouton and Pascale Col

6. Spelling Abilities in Hebrew-Speaking Children with Hearing Loss 70Ronit Levie, Dorit Ravid, Tal Freud, and Tova Most

7. The Influence of Verbal Working Memory on Writing Skills in Children withHearing Loss 85Barbara Arf, Federica Nicolini, and Elena Pozzebon

8. Composing Academic Essays:Using Dictation and Technology to ImproveFluency 100John Albertini, Michael Stinson, and Argiroula Zangana

[ v i i i ] Contents

Children with Oral Language Difficulties 9. Examining Early Spelling and Writing Skills:AComparative Analysis of

Kindergarteners with Speech and Oral Language Impairments and Their Typically Developing Peers 112Cynthia S.Puranik, Stephanie Al Otaiba, and Feifei Ye

10. Morphological Awareness and Spelling Difficulties in French-SpeakingChildren 130Monique Snchal

11. Writing Abilities of Pre-adolescents with and without Language/Learning Impairment in Restructuring an Informative Text 143Orna Davidi and Ruth A.Berman

12. Writing Development of Spanish-English Bilingual Students with Language Learning Disabilities:New Directions in Constructing IndividualProfiles 158Robin L.Danzak and Elaine R.Silliman

13. Written Narratives from French and English Speaking Children with LanguageImpairment 176Judy S.Reilly, Josie Bernicot, Thierry Olive, Jol Uz, Beverly Wulfeck, Monik Favart, and Mark Appelbaum

Children with Dyslexia14. A Review of Dyslexia and Expressive Writing in English 188

Emma Sumner, Vincent Connelly, and Anna L.Barnett15. Written Spelling in French Children with Dyslexia 201

Sverine Casalis16. Written Spelling in Spanish-Speaking Children with Dyslexia 214

Francisca Serrano and Sylvia Defior17. The Writing Development of Brazilian Children with Dyslexia:

AnEvidence-Based Clinical Approach 228Jane Correa

18. Expressive Writing in Swedish 15-Year-Olds with Reading and WritingDifficulties 244sa Wengelin, Roger Johansson, and Victoria Johansson

19. Improving Expressive Writing in Children with Learning Disabilities: TheEffects of a Training Focused on Revision 257Martina Pedron, Anna Maria Re, Chiara Mirandola, and Cesare Cornoldi

PART THREE:Linking Research to Practice in Oral and Written Language Assessment and Intervention20. Integrating Language Assessment, Instruction, and Intervention in an

Inclusive Writing Lab Approach 273Nickola Wolf Nelson

C o n t e n ts [ i x ]

21. Integrating Oral and Written Language into a New Practice Model: Perspectives of an Oral Language Researcher 301Elaine R.Silliman

22. Integrating Writing and Oral Language Disorders: Perspectives of a WritingResearcher 313Vincent Connelly

23. The Role of Oral Language in Developing Written Language Skills: Questionsfor European Pedagogy? 325Julie E. Dockrell and Barbara Arf

Bridging Research and Practice: Conclusions 336

Index 343

PREFACE

Today, in addition to the 10% of children who do not have access to schooling of any kind, we know that millions more are in school but are not achieving minimal levels of learning

(Bernard, 1999, p.v).

Language problems are considered to be a significant barrier to learning, partici-pating in academic and professional activities, and developing as a person. Spoken language and written language are representational systems, which allow us to understand the world and express our views and ideas. Both reading and writing are tools of personal empowerment and a means of social and cultural development (UNESCO, 2013). Educational systems should equip children with writing skills which allow them to engage in wider communication and progress in the work-force. However, children with oral language problems frequently fail to develop developmentally appropriate writing skills, and, as such, are at disadvantage.

This handbook focuses on the way oral language acts as a barrier to the produc-tion of written texts and the strategies that can be used to ameliorate these prob-lems. To locate the childrens problems with language alone is insufficient. Such an explanation fails to clarify all the mechanisms which limit childrens text produc-tion and how oral language problems may interact with the structure of the oral and written language that the child is learning.

Improving the quality of education children and young adults with oral lan-guage and speech problems receive is the ultimate goal. To achieve this goal, we need to understand how the language learning mechanisms interact with (a)the linguistic, social, and cultural factors that characterize the oral language learning environment and (b)the task and motivational factors that characterize the writ-ten language learning environment (Boscolo, chapter 3, this volume; Danzak & Silliman, chapter12, this volume). Models and views of writing development that we present in Part I of this volume (Hayes & Berninger; Berman and Boscolo) underpin this analysis, which is presented in Part II. Chapters in Part II describe the

[ x i i ] Preface

main characteristics of writing difficulties in children with hearing loss, speech and language impairment, and dyslexia. Authors from three continents and nine coun-tries contributed their research work to extend our understanding of the problems that the children face. The product is a rich overview of writing problems across different types of oral language difficulties in different languages.

Another important theme of this volume is oral language and writing assess-ment. Part III develops this issue with links to intervention. Berman and Connelly show how the measures we examine (such as syntactic complexity, word diversity and writing fluency) must consider both the nature of the language learners profile and the features of the language. Some measures may be more sensitive to improve-ments and developmental changes in one language and less in another language, requiring language-specific scaffolding (Reilly etal., chapter13, this volume).

Finally, decisions about the assessment of the childrens writing must consider the most sensitive measures for the question at hand, be that static or dynamic assessment or standardized or natural tasks. Nelson suggests that dynamic assess-ment, at least in situations where assessment is linked to intervention, is the more powerful tool. Some authors emphasize the importance of considering authentic and meaningful writing practices in writing assessment (Boscolo, chapter 3, and Silliman, chapter21) and to examine childrens ability to use language across dif-ferent genres (Danzak & Silliman, chapter12; Silliman, chapter21). Other authors present a more cognitive approach to language and writing assessment where skills are assessed by standardized tests.

The authors have worked with us to present their arguments succinctly and draw out implications for researchers and practitioners alike. We would like to thank all the authors for their patient and hard work on the chapters. We have learned much from all authors and hope our readers will learn from and value their work.

Barbara Arf, Julie Dockrell, and Virginia W.Berninger

REFERENCES

UNESCO (2013). Literacies for the 21st century. Downloaded from:http://unesdoc.unesco.org/images/0022/002230/223029E.pdf

Bernard, A. (1999). The child-friendly school:a summary. Paper prepared for UNICEF.

CONTRIBUTORS

John AlbertiniNational Technical Institute for the DeafRochester Institute of TechnologyRochester, NewYork

Stephanie Al OtaibaDepartment of Teaching and LearningSouthern Methodist UniversityDallas, Texas

Mark AppelbaumDepartment of PsychologyUniversity of California, San DiegoSan Diego, California

Barbara ArfDepartment of Developmental Psychology and SocialisationUniversity of PadovaPadova, Italy

Anna L.BarnettDepartment of PsychologyOxford Brookes UniversityOxford, United Kingdom

Josie BernicotDepartment of PsychologyUniversit de Poitiers-CNRSPoitiers, France

Virginia W.BerningerDepartment of Educational PsychologyUniversity of WashingtonSeattle, Washington

Ruth A. BermanDepartment of LinguisticsTel Aviv UniversityTel Aviv, Israel

Pietro BoscoloDepartment of Developmental Psychology and SocializationUniversity of PadovaPadova, Italy

Sophie BoutonLaboratoire de Neurosciences Cognitivescole Normale SuprieureParis, France

Sverine CasalisURECAUniversit de LilleVilleneuve-dAscq, France

Pascale ColLaboratoire de Psychologie CognitiveAix-Marseille UniversitMarseille, France

[ x iv ] Contributors

Vincent ConnellyDepartment of PsychologyOxford Brookes UniversityOxford, United Kingdom

Cesare CornoldiDepartment of General PsychologyUniversity of PadovaPadova, Italy

Jane CorreaInstituto de PsicologiaUniversidade Federal do Rio de JaneiroRio de Janeiro, Brazil

Robin L.DanzakDepartment of Speech-Language PathologySacred Heart UniversityFairfield, Connecticut

Orna DavidiSchool of Cultural StudiesTel Aviv UniversityTel Aviv, Israel

Sylvia DefiorFacultad de PsicologiaUniversity of GranadaGranada, Spain

Julie E. DockrellDepartment of Psychology and Human DevelopmentInstitute of EducationLondon, United Kingdom

Monik FavartDepartment of PsychologyUniversit de Poitiers-CNRSPoitiers, France

Tal FreudDepartment of Communications Disorders and School of EducationTel Aviv UniversityTel Aviv, Israel

Ann E.GeersDallas Cochlear Implant Program, Callier Advances Research CenterUniversity of Texas at DallasDallas, Texas

Heather HayesProgram in Audiology and Communication SciencesWashington University School of MedicineSt. Louis, Missouri

John R.HayesDepartment of PsychologyCarnegie Mellon UniversityPittsburgh, Pennsylvania

Roger JohanssonCentre for Languages and LiteratureLund UniversityLund, Sweden

Victoria JohanssonCentre for Languages and LiteratureLund UniversityLund, Sweden

Ronit LevieDepartment of Communications Disorders and School of EducationTel Aviv UniversityTel Aviv, Israel

C o n t r i b u to r s [ xv ]

Chiara MirandolaDepartment of General PsychologyUniversity of PadovaPadova, Italy

Tova MostDepartment of Communications Disorders and School of EducationTel Aviv UniversityTel Aviv, Israel

Nickola Wolf NelsonDepartment of Speech Pathology and AudiologyWestern Michigan UniversityKalamazoo, Michigan

Federica NicoliniServizio di Audiologia e FoniatriaUniversit di PadovaTrevisoTreviso, Italy

Thierry OliveDepartment of PsychologyUniversit de Poitiers-CNRSPoitiers, France

Martina PedronDepartment of General PsychologyUniversity of PadovaPadova, Italy

Elena PozzebonServizio di Audiologia e FoniatriaUniversit di PadovaTrevisoTreviso, Italy

Cynthia S.PuranikDepartment of Communication Science and DisordersUniversity of PittsburghPittsburgh, Pennsylvania

Dorit RavidDepartment of Communications Disorders and School of EducationTel Aviv UniversityTel Aviv, Israel

Anna Maria ReDepartment of General PsychologyUniversity of PadovaPadova, Italy

Judy S. ReillyDepartment of PsychologySan Diego State UniversitySan Diego, California

Monique SnchalDepartment of PsychologyCarleton UniversityOttawa, Ontario

Francisca SerranoFacultad de PsicologiaUniversity of GranadaGranada, Spain

Elaine R.SillimanDepartment of Communication Sciences and DisordersUniversity of South FloridaTampa, Florida

Michael StinsonNational Technical Institute for the DeafRochester Institute of TechnologyRochester, NewYork

Emma SumnerDepartment of PsychologyGoldsmiths, University of LondonLondon, United Kingdom

[ xv i ] Contributors

Rebecca TreimanDepartment of PsychologyWashington University in St. LouisSt. Louis, Missouri

Jol UzCRTL-Centre Hospitalier H.LaboritPoitiers, France

sa WengelinDepartment of SwedishUniversity of GothenburgGteborg, Sweden

Beverly WulfeckSchool of Speech, Language, and Hearing SciencesSan Diego State UniversitySan Diego, California

Feifei YeDepartment of Communication Science and DisordersUniversity of PittsburghPittsburgh, Pennsylvania

Argiroula ZanganaInstitute for Language and Speech ProcessingAthens, Greece

INTRODUCTION

In his Nobel Prize lecture, V.S. Naipaul, described the difficulties he experienced growing up in a world that he initially did not understand and explains how he progressively discovered and understood this world:When Ibecame a writer those areas of darkness around me as a child became my subjects. The land; the aborigi-nes; the New World; the colony; the history; India; [..] knowledge and ideas came to me [..], principally from my writing (2013). Writing is a wonderful tool, which has a unique role in our development. Through writing people communicate with each other and themselves, learn, discover themselves and build their identities and establish roles in society. Yet writing is also an extremely complex activity, an activ-ity that is a struggle for many students and professionals alike (Dockrell, 2014). Difficulties with written expression are currently one of the most common learn-ing problems, involving between 6.9% and 14.9% of the school aged population, depending on the formula used to identify written expression disorders (Katusic, Colligan, Weaver, & Barbaresi, 2009). For children and young people with difficul-ties related to oral language the prevalence of writing difficulties increases signifi-cantly (Mayes & Calhoun, 2006), yet there is much to learn about the relationship between oral and written language.

For students with oral language difficulties text production can be particularly challenging, yet there have been few attempts to draw together the impact of dif-ferent oral language problems on the production of written text. This book aims to illuminate the nature of the writing process through its relationship with oral lan-guage and oral language difficulties, and to understand how language systems and educational contexts may contribute to shape this relationship in different ways. Our aim is to provide a bridge between research and practice by presenting current research evidence to help guide and support practitioners and researchers alike. Models of writing represent the framework through which the writing process and writing difficulties are examined and understood. Over the last 30years psycholo-gists have studied the ways in which the cognitive system supports the writing pro-cess and how writing changes and develops over time (Hayes & Flower, 1980). This research has led to the construction of models of expert writing (Hayes & Flower, 1986)and models of writing development (Berninger, 2012).

[ xv i i i ] Introduction

Cognitive models capture part of the writing process by focusing on the infor-mation processing demands that are placed on the writer. But the writer produces text within a social context that can support text production or can present specific demands on the writer. The contexts in which writing occurs and the values and meanings writing has for the writer are an important component of our under-standing of writing and writing problems (see Danzak & Silliman, chapter12, this volume).

In the first part of this book we present three different perspectives from which writing difficulties can be understood. In chapter1, Hayes and Berninger present a new cognitive framework for understanding the writing process. Berman, in chapter2, discusses how a linguistic perspective should guide our research work, analyses of written expression and understanding of developmental difficulties. In addition, Berman uses cross-linguistic data to broaden our conceptualiza-tion of the writing process. Boscolo, in chapter3, discusses the theoretical and instructional implications of a cognitive perspective and a socio-cultural perspec-tive to our understanding of the writing process. Research from a cognitive per-spective has increased our understanding of the processes which underpin text production, resulting in teaching and training packages, which can be used to improve performance. However, to impact on learning and on an effective and authentic use of writing, we also need to understand writing as a social act. Some of the chapters in the second and third part of this volume specifically address this issue (see Danzak & Silliman, chapter12, Nelson, chapter20, and Silliman, chapter21). This entails both an understanding of the context and situations in which the written message is produced and of the written text as a product of an authentic communication act. As Boscolo emphasizes, this component is often forgotten in writing instruction.

The chapters in this book consider situations where the language system has been compromised, and present current research on writing difficulties in the area of deafness, language impairment and dyslexia. Clinicians and educators often work with children who have these difficulties, but struggle in identifying the nature of their writing problems. In many cases problems with language and writing result from a complex mixture of cognitive and linguistic difficulties, which are addressed by this handbook.

The book focuses on the ways in which aspects of the language system can impact on oral/written language difficulties:(a)difficulties with oral and written language learning which are caused by a difficult access to speech-sounds (deaf-ness) ( Johnson & Goswami, 2010); (b)difficulties with oral and written language learning, which involves the language learning mechanisms necessary for devel-oping grammatical and semantic and pragmatic linguistic representations of spo-ken language (as in Specific Language Impairment) (Bishop & Snowling, 2004; Dockrell & Connelly, 2013; Dockrell, Lindsay, & Connelly, 2009); (c)difficulties with oral and written language learning that selectively involve the mechanisms required to process phonological information and the phonological structure of

i n t r o d u Ct i o n [ x ix ]

words (such us in some cases of dyslexia, see Bishop & Snowling, 2004; Peterson & Pennington, 2012).

A challenge for both practitioners and researchers is the plethora of terms used to describe a child who has a specific set of problems. Different labels are used to refer to the same group of children both within and across countries (Dockrell, Lindsay, Letchford, & Mackie, 2006). Labels also often identify heterogeneous groups of children. For example, specific language impairment often includes children who have both problems with the structural aspects of the language system and prob-lems with social communication (Conti-Ramsden & Botting, 1999). Sometimes a group identified for research purposes does not reflect the reality in practice or illus-trates significant comorbidity with other problems (between dyslexia and dyscal-culia, or dyslexia and other reading difficulties, Wengelin etal., chapter18; Pedron etal., chapter19; language impairment and speech sound disorders, Puranik etal., chapter9; dyslexia and language impairment, Peterson & Pennington, 2012). These differences in nomenclature can be problematic for the field, but detailed descrip-tions of the population under study allow informed comparisons. In this book the terminology used by the authors reflects both their country of origin and research focus. In each case details of the population will help the reader establish how the research can be embedded within their own practice or research framework. The variety of labels and definitions used in this book also represent the complexity of identifying and understanding oral and written language difficulties. As Davidi and Berman ( chapter 11) and Danzak and Silliman ( chapter 12) highlight, the term Language Learning Disability emphasizes the linkages between spoken language and literacy learning and suggests that we are examining a learning problem, not just a linguistic problem, a conclusion that many of the chapters illustrate.

Part II provides studies that consider writing at different levels, word, sentence, text and discourse. To date, most research work in the field of written language prob-lems has focused on the single word level, spelling. Spelling is important because it is the code writers need to discover and use to write (Ehri, 2005), but also because it represents one of the greatest constraints to writing. Children who do not mas-ter spelling processes continue to meet difficulties in text production (Berninger, Nielsen, Abbott, Wijsman, & Raskind, 2008)and problems in mastering spelling taxes the writing process such that children not only produce short and inaccurate texts, but avoid writing, leading to further delays in writing development (Graham & Harris, 2000).

Various chapters in this book focus on spelling and spelling difficulties, in the section on deafness (Hayes etal., chapter4, Bouton & Col, chapter5, and Levie etal., chapter6), oral language difficulties (Puranik etal., chapter9, and Snchal, chapter10) and dyslexia (Casalis, chapter15, and Serrano & Defior, chapter16). The chapters on spelling show how children with writing problems make use of regularities in the language to spell words, in a similar fashion to typically develop-ing children. Different forms of language knowledge (phonological, orthographic, and morphological) are exploited by these children to discover and use these

[ xx ] Introduction

regularities. For example, Hayes et al. in chapter 4, Levie et al. ( chapter 6), Arf etal. ( chapter7), Snchal ( chapter10) and Casalis ( chapter15), emphasize the role of morphology in spelling in combination with phonology and orthography. These findings are important since they are derived from studies on different lan-guage systems (Hebrew, French, English and Italian) and on different populations (children with deafness, oral language problems and dyslexia).

As spelling is also a component of text production, spelling difficulties are also treated in chapters which focus is on text production (see for example Arf etal., chapter7, Reilly etal., chapter13, Sumner etal., chapter14, and Wengelin etal., chapter18). Difficulties with text production are related to difficulties with spelling in children with dyslexia (Sumner etal., chapter14 and Wengelin etal., chapter18), children who are deaf (Arf etal., chapter7) and in children with oral language problems (Reilly etal., chapter13). However, writing and written communication goes beyond spelling, involving the processing of words, sentences, ideas, and dis-course structures (Arf & Boscolo, 2006; Berninger, Nielsen, Abbott, Wijsman, & Raskind, 2008; Dockrell, Lindsay, Connelly, & Mackie, 2007). Analyses of spelling alone fails to capture all the components in writing that are linked to oral language, and the ways in which writing develops over time.

Many chapters in this book demonstrate how difficulties in writing are often at a grammatical and discourse level (Albertini etal., chapter8, Davidi & Berman, chapter 11, Danzak and Silliman, chapter 12, Reilly et al., chapter 13, Wengelin et al., chapter 18). These writing difficulties are discussed as both an expression of problems with the development of oral language knowledge (e.g., grammatical morphology, see Reilly etal., chapters13) and in terms of basic language learning mechanisms affecting concurrently both oral and written language production (see for example, Arf etal., chapter7).

Some of the chapters in part II (in particular Arf et al., chapter 7, Davidi & Berman, chapter11, Danzak & Silliman, chapter12, and Reilly etal., chapter13) point to the value of multi-level text analyses. Although this is an extremely com-plex and time consuming activity, examining written expression, and its difficul-ties, at word, sentence and text level is a particularly useful way to capture variation in writing and to identify strengths and challenges in the writing performance of children who show language problems. The multilevel analysis of writing can also provide a more comprehensive evaluation of the ways in which language problems at word (spelling and vocabulary), sentence (grammar) and text (discourse) level interact in the production of the written text (Puranik, Lombardino, & Altmann, 2008; Wagner etal., 2011).

Implications of this research for assessment and intervention are discussed in Part III. Difficulties in learning to write pose challenges for both the writer and the teacher, in particular when teachers and their students are building on weak oral language skills. The final chapters in the book discuss how the integration of oral and written language in assessment and intervention can change our ways of analyzing writing problems, providing appropriate instruction and empowering

i n t r o d u Ct i o n [ xx i ]

learning environments. The integration of oral and written language is examined in the context of instructional and educational intervention (Nelson, chapter20), in terms of future instructional approaches (Silliman, chapter21) and at a more conceptual level, as the result of progress in the field of writing research (Connelly, chapter22). Finally, Dockrell and Arf ( chapter23) discuss the pedagogical ques-tions that emerge from this volume.

REFERENCES

Arf, B., & Boscolo, P. (2006) Causal coherence in deaf and hearing students written narratives. Discourse Processes, 42, 271300.

Berninger, V. (Ed.) (2012). Past, present, and future contributions of cognitive writing research to cognitive psychology. NewYork, NY:Psychology Press/Taylor Francis Group.

Berninger, V.W., Nielsen, K.H., Abbott, R.D., Wijsman, E., & Raskind, W. (2008). Writing problems in developmental dyslexia: Under-recognized and under-treated. Journal of School Psychology, 46(1). doi:10.1016/j.jsp.2006.11.008

Bishop, D. V. M., & Snowling, M. J. (2004). Developmental dyslexia and specific language impairment:Same or different? Psychological Bulletin, 130(6), 858886.

Conti-Ramsden, G., & Botting, N. (1999). Classification of children with specific language impairment: Longitudinal considerations. Journal of Speech Language and Hearing Research, 42(5), 11951204.

Dockrell, J. E. (in press). Developmental Variations in the Production of Written Text: Challenges for Students who Struggle with Writing. In Stone, Silliman, Ehren, & Wallach, (Eds.), Handbook of Language and literacy, (2nd ed.). Guildford Publications.

Dockrell, J.E., & Connelly, V. (2013). The role of oral language in underpinning the text gen-eration difficulties in children with specific language impairment. Journal of Research in Reading. doi:10.1111/j.1467-9817.2012.01550.x

Dockrell, J.E., Lindsay, G., & Connelly, V. (2009). The impact of specific language impairment on adolescents written text. Exceptional Children, 75(4), 427446.

Dockrell, J.E., Lindsay, G., Connelly, V., & Mackie, C. (2007). Constraints in the production of written text in children with specific language impairments. Exceptional Children, 73, 147164.

Dockrell, J.E., Lindsay, G., Letchford, B., & Mackie, C. (2006). Educational provision for chil-dren with specific speech and language difficulties:perspectives of speech and language therapy service managers. International Journal of Language & Communication Disorders, 41(4), 423440. doi:10.1080/13682820500442073|issn 1368-2822

Ehri, L. C. (2005). Learning to read words: Theory, findings, and issues. Scientific Studies of Reading, 9(2), 167188. doi:10.1207/s1532799xssr0902_4

Graham, S., & Harris, K.R. (2000). The role of self-regulation and transcription skills in writ-ing and writing development. Educational Psychologist, 35(1), 312. doi:10.1207/s15326985ep3501_2

Hayes, J., & Flower, L. (1980). Identifying the organisation of writing processes. In L. Gregg & R. Sternberg (Eds.), Cognitive processes in writing (pp. 330). Hillsdale, NJ:Erlbaum.

Hayes, J, & Flower, L. (1986). Writing research and the writer. American Psychologist, 41, 11061113.

[ xx i i ] Introduction

Johnson, C., & Goswami, U. (2010). Phonological awareness, vocabulary, and reading in deaf children with cochlear implants. Journal of Speech Language and Hearing Research, 53(2), 237261. doi:10.1044/1092-4388(2009/08-0139)

Katusic, S.K., Colligan, R.C., Weaver, A.L., & Barbaresi, W.J. (2009). The forgotten learn-ing disability: Epidemiology of written-language disorder in a population-based birth cohort (1976-1982), Rochester, Minnesota. Pediatrics, 123(5). doi:10.1542/peds.2008-2098

Mayes, S. D., & Calhoun, S. L. (2006). Frequency of reading, math, and writing disabili-ties in children with clinical disorders. Learning and Individual Differences, 16(2). doi:10.1016/j.lindif.2005.07.004

Naipaul, V.S. - Nobel Lecture: Two Worlds. Nobelprize.org. Nobel Media AB 2013. Web. 1 Jul 2013. http://www.nobelprize.org/nobel_prizes/literature/laureates/2001/naipaul-lecture-e.html

Peterson, R. L., & Pennington, B. F. (2012). Developmental dyslexia. Lancet, 379(9830), 19972007. doi:10.1016/s0140-6736(12)60198-6

Puranik, C. S., Lombardino, L. J., & Altmann, L. J. P. (2008). Assessing the microstructure of written language using a retelling paradigm. American Journal of Speech-Language Pathology, 17(2), 107120.

Wagner, R.K., Puranik, C.S., Foorman, B., Foster, E., Wilson, L.G., Tschinkel, E., & Kantor, P. T. (2011). Modeling the development of written language. Reading and Writing, 24(2), 203220. doi:10.1007/s11145-010-9266-7

PART ONE

Models and Perspectives on WritingDevelopment

CHAPTER1

Cognitive Processes in Writing

A Framework

JOHN R . H AYES A ND VIRGINI A W. BER NINGER

In this chapter, we present a framework relating the cognitive processes that writ-ers in general use when they create written texts, the mental resources that these cognitive processes can draw on, and the task environment in which these cogni-tive processes operate. We begin by clarifying how this approach to cognition con-trasts with that most familiar to professionals who work with individuals who have sensory, motor, or language disabilities that affect their oral or written expression. We then explain the benefits of a theoretical framework of cognition specific to the writing process and describe the details of this framework. Finally, we encour-age professionals and researchers who work with individuals who have disabilities related to hearing, speech, and language to adapt this cognitive framework to those individuals disabilities and capabilities, evaluate the adaptations, and share the results. Such results should extend knowledge of cognition during writing for writ-ers in general to writers with specific sensory, motor, or language limitations.

Typically, IQ tests are used to assess cognition in individuals with a variety of disabilities. IQ stands for intelligence quotient, which is not what these tests really measure. To begin with, they do not assess all aspects of human intelligence, but rather specific kinds of cognitive abilities, for example, verbal reasoning or non-verbal reasoning, which are related to school achievement or specific kinds of vocational aptitude. More importantly, they have not yielded quotients (mental age divided by chronological age) ever since standard scores were developed in the middle of the last century to compare an individuals current score to others of the same age and also the scores of the same individual across time. As such, these tests may be useful indices of levels to which specific kinds of cognitive abilities are

[4 ] Part I:Models and Perspectives on Writing Development

currently developed and their stability across development. They do not assess the specific cognitive operations during actual listening, speaking, reading, or writing, four language tasks that differ in which sensory or motor systems they engage and how cognitive systems are accessed and utilized. Language by ear (listening com-prehension), language by eye (reading comprehension), language by mouth (oral expression of ideas), and language by hand (written expression of ideas) are sepa-rable language systems that may function together in integrated ways and draw on both common and unique processes (Berninger & Abbott, 2010).

Thus, a framework that specifies writing-specific cognitive processes and oper-ations provides useful information that goes beyond what IQ tests provide and is more relevant to planning instructional treatment to improve the written expres-sion of ideas, which is not fully identical with understanding ideas in others spo-ken or written language or expressing ones own ideas orally. We hope that this framework will be useful to those who work with or study special populations of writers who may be deaf, have difficulty processing the speech they hear or pro-ducing speech others can understand, or have specific disabilities in learning to understand or construct spoken or written texts with words, syntax, and discourse structures.

The framework, based on over three decades of research on typical writing, is presented graphically in Figure1.1. The framework has three levels. The bottom, or resource, level represents general cognitive resources that writers may draw on as they compose. The middle, or process, level represents the cognitive processes that writers may use to create texts together with the task environment in which these processes operate. The top, or control, level represents the factors that control operations at the process level.

THE FRAMEWORK

Resource Level

We have included four resources at the resource level:attention, long-term mem-ory, working memory, and reading. These are resources that are used by many activities including writing. By attention we mean the ability to maintain focus on a task in the face of distraction. This ability is also often referred to as executive func-tion or executive control and is a resource that the top-level control processes can draw upon. Focused attention (inhibit what is not relevant) is often measured by the Stroop task (Stroop, 1935); the participant is asked to name the print color of a sequence of color words printed in a different color than the color named by the word. For example, the participant may see the word RED printed in green, the word BLUE printed in yellow, and so on. Many find it difficult to say the print colors (e.g., green, yellow) and resist reading the distracting color word (e.g., red, blue). Performance on the Stroop task and related tasks of attention and executive func-tion improves from early childhood into the 20s (Diamond, 2006). Development

C o g n i t i v e P r o C e s s e s i n W r i t i n g : A F r A m e W o r k [ 5 ]

of the writers attention is a resource that enables executive function control and may have an important impact on the writers choice of writing strategy.

Long-term memory is a complex resource that stores the individuals knowledge of facts, events, motor planning, control, and execution skills, letter form access and production skills, and language including vocabulary, spelling, grammar/syntax, and discourse schema, all of which are sources of knowledge that are important for competent writing. Chenoweth and Hayes (2001) showed that the fluency with which a person writes in a language depends critically on how many years of experi-ence the person has with the language. Underdeveloped spelling knowledge and illegible or nonautomatic handwriting may also interfere with writing development in the first six grades (e.g., Berninger, Yates, Cartwright, Rutberg, Remy, & Abbott, 1992; Berninger, Cartwright, Yates, Swanson, & Abbott, 1994)and even in older writers (Hayes & Chenoweth, 2006). Writers with ample knowledge in long-term memory about the topic they are writing about produce essays of higher quality and more quickly and with less effort than the less informed writers (Caccamise, 1987; Dansac & Alamargot, 1999; Voss, Vesonder, & Spilich, 1980).

Working memory is a memory system designed to store the required informa-tion while the cognitive operations are performed to carry out a task. For example,

WritingProcesses

TaskEnvironment Collaborators

& Critks

Proposer

Task Materials

Translator

TaskInitiator

Transcriber

ProcessLevel

ControlLevel

ResourceLevel

Text-Written-So-Far

TranscribingTechnology

Writing Schemas

Planner

Attention

WorkingMemory Reading

Long-TermMemory

Evaluator

Figure1.1.A framework representing the organization of cognitive processes involved in writing. Note: The model does not include a revision process. We view revision not as a writing process but rather as a specialized writing task that makes use of the processes in the writing modelproposing, translat-ing, planning, reading, and so forthto replace an earlier text. We have included arrows to indicate some relations between processes but, to avoid visual clutter, we have not indicated all potential relations. For example, although there are important relations among evaluating, reading and the text-written-so-far, these relations have not been marked with arrows. Similarly, relations between the TWSF and translation, long-term memory and proposing, writing schemas and writing pro-cesses, and many other relations are not marked.

[6 ] Part I:Models and Perspectives on Writing Development

if we were to multiply 2 three-digit numbers mentally, we would need to remember several lists of numbersthe partial productsand, in addition, we would need memory for carrying out the arithmetic processesmultiplication and addition. Since Daneman and Carpenter (1980) developed the first measure of individual differences in working memory, measures of different kinds of working memory have been created (Ransdell & Levy, 1999). Research indicates that interfering with working memory resources can reduce the fluency and quality of writing (Daiute, 1984; Fayol, Largy, & Lamaire, 1994; Chenoweth & Hayes, 2003; Hayes, 2009).

Reading, which is another written language skill, is also an important resource for writers. Kaufer, Hayes, and Flower (1986) found that adult writers spent a substan-tial portion of their composing time reading the text they had just written. Much of this rereading focused on the sentence currently under construction. For example, a writer who had written down the first part of a sentence often reread the sentence parts already written before completing the sentence. Such rereading (reviewing to monitor what has been written so far to decide whether to revise as well as what to write next) may serve to promote construction of cohesive text. When writers are composing from sources, they typically read and reread the source texts written by others (Alamargot, Dansac, Chesnet, & Fayol, 2007). Similarly, editing and revis-ing typically involves the repeated reading of the target text the writer has written (Hayes, Flower, Schriver, Stratman, & Carey, 1987).

These resources may also interact with each other. For example, Hyn, Lorch, and Kaakinen (2002) found that writers with high working-memory capacity con-struct more accurate summaries of source texts in a writing-from-sources task than do writers with low working memory capacity.

Process Level

Protocol studies (Kaufer, et al., 1986; Chenoweth & Hayes, 2001) showed that adults typically compose texts in language bursts averaging 612 words in length depending on the skill of the writer. These bursts consist of language that the writer proposes for inclusion in the text. In the protocols, the bursts were separated by pauses that often included statements suggesting planning of the next fragment or evaluation of the text just written. Chenoweth and Hayes (2001) proposed that in adult writers language bursts are produced through the interaction of four cogni-tive processes:a proposer, a translator, an evaluator, and transcriber. See Figure1.1.

The function of the proposer is to suggest a package of ideas for inclusion in the text and to pass that package on to the translator. The proposer can take input from the planner, from the task environment, from long-term memory, and from the text written so far. Ideas suggested by the proposer are in nonverbal form.

The translator takes ideas from the proposer and represents them as grammatical strings of language; that is, it translates nonverbal ideas into a verbal form of expres-sion (Chenoweth & Hayes, 2003). The translator may also take language strings

C o g n i t i v e P r o C e s s e s i n W r i t i n g : A F r A m e W o r k [ 7 ]

presented in visual or auditory form that were coded in verbal long-term memory as language and transform them into new language strings (Hayes & Chenoweth, 2007). Research by Chenoweth and Hayes (2001, 2003) suggests that the fluency with which the translator operates depends on the writers linguistic experience and on the amount of verbal working memory that is available to the writer. A series of studies by Hayes and his colleagues (summarized in Hayes, 2009) attributes the fact that texts are composed in such a choppy fashion, that is, by putting together a sequence of fragmentary language bursts, to the high demands that the transla-tor makes on available working memory resources. For adults, at least, translation appears to be the bottleneck limiting fluency.

The transcriber takes the grammatical strings produced by the translator and turns them into written text. For beginning and developing writers in cross-sectional studies (Berninger etal., 1992; Berninger etal., 1994)transcription (handwriting and spelling), if not age appropriate, may pose special challenges for the translation process. Of these, spelling had the most consistent longitudinal influences across adjacent grade levels from first to seventh grade on written composing (Abbott, Berninger, & Fayol, 2010). For those who initially struggle with transcription, stud-ies have shown that, with appropriate instruction and practice, these tasks become more automated and demand fewer cognitive resources (e.g., Berninger & Amtmann, 2003). However, Hayes and Chenoweth (2006) found that even in adults the pro-cess of transcription still places demands on working-memory resources.

The evaluator can examine the outputs of any of the other processes and pass judgment on their adequacy. For example, the evaluator may reject an idea that has been proposed before it is translated into language; it may reject a translated language string before it is transcribed, and it may demand the revision of already transcribed language. Hayes (2011) suggests that the evaluation process may be minimal or absent in some of the writing strategies that very young writers may adopt. However, these same children may respond to specific requests to evaluate texts even if they may not do so in their self-regulated, independent writing.

The Task Environment

The task environment includes the immediate social and physical factors that influ-ence the writing processes. The social task environment includes concurrent inputs from collaborators and critics (lets do this, why did you do that?), a teachers admonition to finish up quickly, or simply the background of conversation in a classroom or workplace. Because we represent the social factors as the immediate social environment (what people are doing right now in the writers presence) one might argue that we have left out the very important influences that society and culture have on the writer. We dont think this is true. We believe that these influ-ences are represented in the writers long-term memory and in the task environ-ment. Socially determined factors such as the social and physical structure of the

[8 ] Part I:Models and Perspectives on Writing Development

classroom and the nature of the writing technology are all represented in the task environment. Long-term memory contains the writers knowledge of genre and of writing strategies. It contains the writers understanding of how audiences are likely to respond to particular language. It contains the writers episodic knowledge:the memory of the writers interaction with the social and physical world. If the influ-ence of society and culture were not represented in long-term memory and the task environment, it is not clear how they could affect the writer.

The physical task environment includes the task materials, the transcribing technology, and the text written so far (TWSF). The task materials might include a written assignment sheet or, in the case of a writing-from-sources task, graphics and/or texts that the writer must refer to while completing the assignment. In a second-language writing task, the task materials might include a dictionary.

The nature of the transcription technology can influence the conduct of the writing task in substantial ways. Researchers have investigated whether children produce lon-ger texts of better quality if they dictate text orally rather than writing them on paper. For example, De La Paz and Graham (1995) found that, if primary school children dictated rather than wrote their texts, text quality improved significantly. Connelly, Gee, and Walsh (2007) compared fifth and sixth graders essays written by hand and by keyboard. Students wrote significantly faster by hand than keyboard. Handwritten essays were significantly superior to typed essays on six analytic scoring categories: (1) ideas and development, (2) organization, (3) unity and coherence, (4) sentence struc-ture, (5) grammar, and (6) punctuation. Similarly, Hayes and Berninger (2010) found that children in second, fourth, and sixth grades proposed more ideas for inclusion in an essay when they transcribed the ideas by hand rather than by keyboard. One would expect that writing in the interactive social media, for example, Twitter and Facebook, might share more features with conversation than with formal school writing. Other research found that practice within a particular transcription mode (handwriting or typing) improved writing in that mode. Handwriting practice improved childrens writing of high quality texts by hand (Jones & Christensen, 1999). For eighth and ninth graders with low typing skills, typing practice improved the quality of typed texts, but not the length or quality of handwritten texts (Christensen, 2004).

Kaufer et al. (1986) found that as college and graduate students composed, they fre-quently re-read to TWSF. Most of this re-reading (more than 80%) involved re-reading the early parts of the sentence currently under construction. For example, while com-posing an essay about a trip, a writer may write down that their experiences on the trip to Asia made . . . and pause. Then, while attempting to complete this sentence, the writer may re-read the written fragment one or more times. Usually this was followed by an addition extending the initial fragment. For example, the writer might add me appreciate . . . and pause again. In about one fourth of cases, re-reading was followed by a revision. For example, the writer might replace made with helped me to. . . These observations suggest that the re-reading of the TWSF serves a coordinating function. Since, as we noted earlier, sentences are composed in parts, it is possible that writers may lose track of text features such as tone, number, and tense as they move from

C o g n i t i v e P r o C e s s e s i n W r i t i n g : A F r A m e W o r k [9 ]

composing one part to the next. Re-reading the TWSF may well help the adult writer to keep such features consistent across sentence parts and thus to maintain text coher-ence. Hayes (2011) has suggested that attention to the TWSF may develop over the school years and contribute to the increasing coherence of writers texts as they mature. He also suggests that very young writers may ignore the content of the TWSF and attend only to its quantity using the quantity to determine if they have written enough.

Control Level

The task initiator may be a teacher who assigns an essay in class, a boss who assigns a writing task at work, or it may be the writer herself who decides to write a story or a journal entry. Usually, the task initiator will influence the planner by specifying the topic, the audience, or other features of the text to be written.

The planner is responsible for setting goals for the writing activity. These goals may be quite simple in young writers. For example, the primary school students may start with the single goal of writing about a particular topic. More advanced writers may plan a sequence of topics and subtopics together with the sequence in which these topics should be addressed. Still more advanced writers may set goals for tone and the intended impact on the audience.

Writing schemas represent the writers beliefs about the properties that the text to be produced should have (genre knowledge) and also beliefs about how to go about producing that text (strategic knowledge). Writing schemas vary from writer to writer and change within writers as the writers develop. The strategies specified by the writing schemas determine the selection of writing processes, how the writ-ing processes operate and how the writing processes interact with each other and with the task environment.

A recent study illustrates the relation between writing schema, the writing pro-cesses, and text structure. Hayes (2011) analyzed the structure of a sample of first to ninth grade childrens expository texts (from Fuller, 1995)and concluded that most could be produced by one of three strategies. The simplest strategy, that Hayes called flexible-focus, might be thought of as stream-of-consciousness writing. This strategy does not require the proposer to maintain focus on a general topic. Figure1.2 shows an essay that this strategy would typically produce. With this strategy, there is no evaluation of the quality of the output of the proposer, translator, or transcriber. The only evaluation involves examining the TWSF to see if enough has been written.

A second strategy, the fixed-topic strategy, is the most common strategy in grades one to five. With this strategy, every statement proposed must reference a single topic. Figure1.3 shows an essay that this strategy would typically produce. Further, unlike the flexible-focus strategy, the evaluation process does evaluate the quality of the output of the other three processes.

A third strategy, the topic-elaboration strategy, is the most common strategy in grades six through nine. With this strategy, the proposer maintains focus on a

[ 10 ] Part I:Models and Perspectives on Writing Development

general topic but may introduce subtopics related to the main topic. Figure 1.4 shows an essay that this strategy would typically produce. These three strategies produce texts with distinctive structures. The writing schema, then, selects and organizes writing processes used to produce text and thus impacts the properties of the text that is written.

Wallace and Hayes (1989) used the writing schema concept to analyze revision in freshman college students. They noted that when freshmen revise, they tend to revise locally. They read sentence 1, perhaps several times, and then revised it. Then they read sentence 2 and revised it, and so on, sentence by sentence, through the text. In contrast, more experienced writers typically revised globally. To prepare themselves to revise a text, they evaluated the whole text and commented on global features of the text such as its organization or the adequacy of the introduction or the conclusion. Wallace and Hayes (1989) speculated that the reason for the differ-ence between freshmen and more experienced writers might be a difference in their schema for revision and that perhaps the freshman schema could be modified by instruction. To test this hypothesis, they designed eight minutes of instruction that

I Like coloring because its not boring

I like coloring cats

I have a black cat at home

His name is Inky

Coloring

Coloring

My Cat

My Cats Name

Figure1.2.An example of a flexible-focus essay (from Fuller, 1985).

I like Ashley cus she is nice (1)I like Ashley cus she plays with me (2)Ashley is my friend (3)I like people and Ashley is one (4)She is nice (5)

Ashley

321 4 5

Figure1.3.An example of a fixedtopic essay (from Fuller, 1985).

Dinosaurs

I like dinosaurs because they are big. (1) Andthey are scary. (2) I like Rex. (3) He was verybig. (4) He ate meat. (5) Triceratops is a verynice dinosaur. (6) He ate plants. (7) He hadthree horns on his face. (8) He had a shield onhis neck. (9) Stegosaurus was a plant eatertoo. (10) He had (unfinished) (11)

1 2 3

4 5 7 8 9 11

106

Dinosaurs

Figure1.4.An example of a topic-elaboration essay (from Fuller, 1985).

C o g n i t i v e P r o C e s s e s i n W r i t i n g : A F r A m e W o r k [ 1 1 ]

contrasted a local revision schema with a global revision schema. They found that the instruction resulted in a significant increase in global revision.

The fact that Wallace and Hayes (1989) found that revising a schema could be modified by instruction doesnt imply that all writing schema can be modified in the same way. The flexible-focus schema may be the only schema that a young writer can manage given her limited attentional resources. Hayes (in press) sug-gests that changes in writing schema in the primary grades may await developmen-tal enhancement of executive function.

INTEGRATING LEVELS OF THE FRAMEWORK

Now that we have discussed the parts of the framework separately, we will try to tie them together with an example of a fourth grade student writing an essay in class. Suppose that a teacher, acting as task initiator, asks a student, Susan, to write about something she likes. This request leads Susan to set a goal. She decides to write on a topic: her classmate Alice. Now she must adopt a writing schema. Since she is in 4th grade, lets assume that she chooses the fixed-topic strategy described earlier. With this strategy, Alice will be the topic of all of her sentences. To start composing, Susan retrieves knowledge about Alice from long-term memory. She proposes the idea Alice is my friend, translates it into language, and evaluates it as appropriate for the essay. Finally, she transcribes the idea, but because Susan in the fourth grade, spelling and handwriting are still difficult for her. These activi-ties place heavy demands on her working memory resources so that, at this point, she has few working-memory resources left to devote to other writing processes. Having written one sentence, she starts the cycle again, proposing and writing She plays with me, She is fun to play with, and so on. While she is writing, the task environment may help or hurt. Looking around the classroom may remind her of things to write about Alice. On the other hand, the voices of her classmates may reduce her available working memory (Salame & Baddeley, 1982). After sev-eral cycles, Susan examines the text she has written so far and decides that she has written enough for an essay and decides to stop. In this imagined writing incident, Susan didnt draw on reading as a resource as may be typical of fourth grade writ-ers in this kind of writing task. However, as we have noted, adult writers would typically read and re-read the TWSF as they translate ideas into text and as they evaluate what they have written.

APPLYING THE COGNITIVE FRAMEWORK TO CLINICAL POPULATIONS

In exploring the applications of this framework to writing practice and research, it will be important to study samples for whom the sensory, motor, and language

[ 12 ] Part I:Models and Perspectives on Writing Development

bases of their diagnosed disorders are well defined, with careful documenta-tion of whether an individual has speech only, language only, or speech and lan-guage impairments and whether some degree of hearing acuity problems may be present.

Deaf

Factors to consider in evaluating the resources (lower level Figure 1.1) a writer who is deaf has to bring to the writing process is age of identification, which is related to when intervention may have begun (e.g., as an infant versus toddler) and the nature of the intervention (e.g., sign language, oral method, cochlear implantation or total communication). These factors are related to whether lan-guage develops in age-appropriate ways despite an auditory sensory impairment and thus provides resources in the form of vocabulary and syntax knowledge. For example, cued speech (see Bouton & Col, this volume) may prepare writers who are deaf for the phonology involved in written spelling. Onset of deafness and age of identification can also affect the development of the childs working-memory system, which plays an important role in shaping the writing process of children with hearing impairment, at the word and sentence level (see Arf, Nicolini, & Pozzebon, this volume).

Speech Problems

Factors to consider, which may affect both resources and cognitive processes (lower and middle level of Figure1.1), are whether the individual has speech-sound dis-order and thus difficulty in processing the sounds of heard speech or speech articu-lation disorder (produced speech is not intelligible to others). Either could affect development of spelling skills (transcription) that supports translation of ideas into language and also learning word meanings through interacting with others in the language-learning environment (see Puranik, Al Otaiba, & Ye, this volume).

Selective Language Impairment (Language-Learning Disability)

The nature of the language impairment will probably affect the nature of writing problems encountered. Word finding problems may result in impaired composing fluency (shorter language bursts and written texts). Syntax problems may result in shorter sentences and sentences with grammar errors and thus the quality of writing may suffer. At the resource level, the language learned and used in writing will affect the complexity of syntax and the nature and rate of grammar errors (see Reilly etal., this volume).

C o g n i t i v e P r o C e s s e s i n W r i t i n g : A F r A m e W o r k [ 13 ]

Inferential thinking problems (understanding what is implied but not stated in language) may interfere with the proposing of ideas that may result in shorter and less-well-developed texts.

Future Developments

It is a step forward, however, to acknowledge, as this volume does, that individuals with aural sensory and oral motor and aural and oral language problems will also have special assessment and instructional needs in learning to write. This chapter makes the point that cognitive processes (and related resources and controls) are as relevant to treatment planning as the sensory, motor, and language bases for the disabilities these individuals have.

REFERENCES

Abbott, R., Berninger, V., & Fayol, M. (2010). Longitudinal relationships of levels of language in writing and between writing and reading in grades 1 to 7. Journal of Educational Psychology, 102, 281298.

Alamargot, D., Dansac, C., Chesnet, D., & Fayol, M. (2007). Parallel processing before and after pauses: A combined analysis of graphomotor and eye movements dur-ing procedural text production. In M. Torrance, L. v. Waes, & D. Galbraith (Eds.), Writing and cognition: Research and applications (Vol. 20, pp. 1329). Amsterdam, Netherlands:Elsevier.

Berninger, V., & Abbott, D. (2010). Listening comprehension, oral expression, reading compre-hension and written expression:Related yet unique language systems in grades 1, 3, 5, and 7. Journal of Educational Psychology, 102, 635651.

Berninger, V. W., Yates, C., Cartwright, A., Rutberg, J., Remy. E., & Abbott, R. (1992). Lower-level developmental skills in beginning writing. Reading and Writing: An Interdisciplinary Journal, 4, 257280.

Berninger, V., Cartwright, A., Yates, C., Swanson, H.L., & Abbott, R. (1994). Developmental skills related to writing and reading acquisition in the intermediate grades:Shared and unique variance. Reading and Writing:An Interdisciplinary Journal, 6, 161196.

Berninger, V., & Amtmann, D. (2003). Preventing written expression disabilities through early and continuing assessment and intervention for handwriting and/or spelling problems:Research into practice. In H. L.Swanson, K. Harris, and S. Graham (Eds.), Handbook of research on learning disabilities (pp. 345363). NewYork, NY:Guilford.

Caccamise, D. J. (1987). Idea generation in writing. In A. Matsuhashi (Ed.) Writing in Real Time? (pp. 224253). Norwood, NJ:JAI Press Limited.

Chenoweth, N., & Hayes, J.R. (2001). Fluency in writing. Generating text in L1 and L2. Written Communication, 18, 8098.

Chenoweth, N.A., & Hayes, J.R. (2003). The inner voice in writing. Written Communication, 20(1), 99118.

Christensen, C. A. (2004). Relationship between orthographic-motor integration and com-puter use for the production of creative and well-structured text. British Journal of Educational Psychology, 74, 551564.

[ 14 ] Part I:Models and Perspectives on Writing Development

Connelly, V., Gee, D., & Walsh, E. (2007). A comparison of keyboarded and handwritten com-positions and the relationship with transcription speed. British Journal of Educational Psychology, 77, 479492.

Daneman, M., & Carpenter, P.A. (1980) Individual differences in working memory and read-ing. Journal of Verbal Learning and Verbal Behavior, 19, 450466.

Dansac, C., & Alamargot, D. (1999) Accessing referential information during text com-position: When and why? In M. Torrance & D. Galbraith (Eds.), Knowing what to write: Conceptual processes in text production (pp. 7997). Amsterdam, Netherlands:Amsterdam University Press.

De La Paz, S., & Graham, S. (1995). Dictation:Applications to writing for students with learn-ing disabilities. In T. Scruggs & M. Mastropieri (Eds.), Advances in learning and behav-ioral disorders (Vol. 9, pp. 227247). Greenwich, CT:JAI Press.

Diamond, A. (2006). The early development of executive functions. In E. Bialystok & F. I.M. Craik (Eds.), Lifespan cognition: Mechanisms of change (chap. 6, pp. 7095). London, England:Oxford University Press.

Daiute, C. (1984). Performance limits on writers. In R. Beach & L. S. Bridwell (Eds.), New Directions in Composition research (pp. 205224). NewYork:Guilford Press.

Fayol, M., Largy, P., & Lamaire, P. (1994) Cognitive overload and orthographic errors:When cognitive overload enhances subject-verb agreement errors, a study in French written language. Quarterly Journal of Experimental Psychology, 47, 437464.

Fuller, D.F. (1995). Development of topic-comment algorithms and test structures in written compo-sitions of students in grades one through nine. Unpublished doctoral dissertation. Seattle, WA:University of Washington.

Hayes, J.R. (2009). From idea to text. In R. Beard, D. A.Myhill, & J. Riley (Eds.), International handbook of writing development (pp. 6579). London, England:Sage.

Hayes, J.R. (2011). Kinds of knowledge-telling:Modeling early writing development. Journal of Writing Research, 3(2), 366383.

Hayes, J.R., & Berninger, V.W. (2010). Relationships between idea generation and transcrip-tion. How the act of writing shapes what children write. In C. Bazerman, R. Krut, K. Lunsford, S. McLeod, S. Null, P. Rogers, & A. Stasnsell (Eds.), Traditions of writing research (pp. 166181). NewYork, NY:Routledge.

Hayes, J.R., & Chenoweth, N.A. (2006). Is working memory involved in the transcribing and editing of texts? Written Communication, 23(2), 135149.

Hayes, J. R., & Chenoweth, N. A. (2007). Working memory in an editing task. Written Communication, 24(4), 283294.

Hayes, J.R., Flower, L., Schriver, K.A., Stratman, J., & Carey, L. (1987). Cognitive processes in revision. In S. Rosenberg (Ed.), Advances in Applied Psycholinguistics, Volume II:Reading Writing and Language Processing (pp. 176240). Cambridge Press Cambridge England.

Hyn J., Lorch, R.F., and Kaakinen J.H. (2002). Individual differences in reading to sum-marize expository text:Evidence from eye fixation patterns. Educational Journal, 94(1), 4455.

Jones, D.A., & Christensen, C.A. (1999). Relationship between automaticity in handwriting and students ability to generate written text. Journal of Educational Psychology, 91, 16.

Kaufer, D.S., Hayes, J.R., & Flower, L.S. (1986). Composing written sentences. Research in the Teaching of English, 20(2), 121140.

Ransdell, S., & Levy, C. M. (1999). Writing, reading, and speaking memory spans and the importance of resource flexibility. In M. Torrence & G. C. Gaynor (Eds.), The cogni-tive demands of writing (Vol. 3, pp. 99113). Amsterdam, Netherlands: Amsterdam University Press.

C o g n i t i v e P r o C e s s e s i n W r i t i n g : A F r A m e W o r k [ 1 5 ]

Salame, P., & Baddeley, A.D. (1982). Disruption of memory by unattended speech:Implications for the structure of working memory. Journal of Verbal Learning and Verbal Behavior, 21, 150164.

Stroop, J.R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643662.

Voss, J.F., Vesonder, G.T., & Spilich, G.T. (1980). Text generation and recall by high-knowledge and low-knowledge individuals. Journal of Verbal Learning and Verbal Behavior, 17, 651667.

Wallace, D., & Hayes, J.R. (1989). Redefining revision for freshmen. Research in the Teaching of English, 25(1), 5466.

CHAPTER2

Linguistic Perspectives on WritingDevelopmentRU TH A . BER M A N

BACKGROUND AND PURPOSE

Written language is examined here as a style of discourse necessary for achieve-ment of linguistic literacy (Ravid & Tolchinsky, 2002). By focusing on linguistic expression, discussion disregards other important aspects of writing development, including the following:

Notationalfeaturessuchasspellingorpunctuation,whichhavecriticallinguis-tic correlates (Ravid, 2011).

Thematiccontent,althoughformandmeaningcanneverbefullyseparated,asshown by the close interrelation between linguistic expression and ideational content in comparing narrative with non-narrative discourse (Berman & Katzenberger, 2004; Longacre, 1996; Ravid & Cahana-Amitay, 2005).

Global discourse structure and organization, which develops in writingnarratives well before expository essays (Berman & Nir, 2007, 2009a), and which also involves paragraphing (Longacre, 1979; Myhill, 2009; Nir, 2008).

Analysis focuses on linguistic units involved in text analysis, with a cross-linguistic and developmental perspective, spanning the period from elementary to high school as reflecting later language development (Berman, 2004, 2007; Tolchinsky, 2004).

The chapter outlines carefully specified, quantifiable methods for evaluating written language expression, deriving from research on texts constructed by typically develop-ing students from middle childhood to adulthood (Berman & Ravid, 2009). Analyses

L i n g u i st i C P e r s P eCt i v e s o n W r i t i n g d e v e L o P m e n t [ 17 ]

referred to later are based on a large-scale cross-linguistic project investigating the text construction abilities of typically developing participants at four age-schooling levels (grade schoolers aged 910years, middle-school students aged 1213, high-schoolers aged 1617, and graduate school adults in their 20s and 30s). All participants were native speakers of seven different languages including Californian English and Israeli Hebrew (Berman, 2005; 2008).1 Participants were shown a short wordless video clip depicting unresolved situations of conflict and then asked to write and tell a story about an incident where they had been involved in a situation of problems between people (a personal-experience narrative) and to write a composition and give a talk on the topic problems between people (an expository discussion)with the four text types balanced for order.

Each of the 80 participants (20 per age group) at seven different sites thus pro-duced four different texts on the shared theme of interpersonal conflicta written and spoken narrative and a written and spoken expository text. Carefully corre-sponding elicitation procedures (detailed in Berman & Verhoeven, 2002)allowed for direct comparability across the variables of age-schooling level, genre (narra-tive/expository), and modality (speech/writing). The measures outlined below rely largely on computerized tools for language analysis in the CHILDES pro-gramswith conventions of transcription, text-line specifications, and counting of units provided by detailed, constantly updated manuals (MacWhinney, 2000; http://childes.psy.cmu.edu/manuals/clan.pdfOctober 2010). Devised origi-nally for early child language and interactive discourse, the system can accom-modate speaker-writers from infancy to adulthood in varied communicative contexts. CHILDES encompasses oral and written data from numerous languages, including non-Roman orthographies, and deals with variables from pronuncia-tion via lexicon and grammar to paralinguistic and pragmatic facets of language communication.

The chapter delineates four hierarchically ordered units of analysisword, phrase, clause, clause package (Section 1)analyzes their diagnostic value in lexi-con and syntax across the variables of age, text-type, and/or modality (Section 2); and notes the role of genre, topic, and language typology for assessment of writing (Section 3).

UNITS OF ANALYSIS (SECTION 1)

Linguistic elements relevant to analyzing written language need to be defined explicitly for specific research purposes and for each language. Four such units are delineated next, each comprising a constituent level of the next:word, phrase, clause, and clause package.

[ 18 ] Part I:Models and Perspectives on Writing Development

Words

Words are recognized as the fundamental building blocks of human language, hence, too, in analysis of written texts. Aword in written language is generally defined operationally as any string of characters separated from the next by a space. Yet defin-ing a written word is not so simple. First, even fourth graders may have difficulty in segmenting lexical items into words as stipulated by dictionary conventions in their language. Second, languages differ in what counts as a word in writing:Compare contractions in English (Im ~I am, theres ~ there is, there has), elisions in French (jai ~ je suis, lpaule ~ la tte), or the fact that in Hebrew, seven high-frequency morphemes are attached as prefixes to the next word in writingthe conjunctions and, that, the definite article the, and the prepositions meaning in, to, from, like. Conventions also differ for writing compounds:In Englishsingle, hyphenated, or two separate words (appleblossom, apple-pie, apple tree); in Hebrewgenerally two separate words; and in Dutch, German, or Swedishtypically single orthographic strings, no matter how long or freely analyzable (Berman, 2009a). Third, semanti-cally corresponding lexical items may have different forms:For example, English phrasal verbs (e.g., go up, go in, go away) have monolexemic Latinate counter-parts (ascend, enter, depart, etc.), and using the phrasal verbs increases the number of words in English texts compared with, say, French or Italian. Frozen multiword expressions (Wulff, 2008)are another problem in counting words, for example, in English off and on, on the one hand, in French il y a, parce que, and in Hebrew be-sofo el davar (in-end-its of thing=eventually), lo kol e-ken (not all that-yes=letalone).

To counter these problems, the cross-linguistic project noted in the introduc-tion adopted a baseline of shared principles plus language-particular procedures for specifying words. Inflectional and derivational morphemes counted as part of a single word across languages. CHILDES conventions marked certain strings as either one or more than one lexical element (e.g., compound nouns in English and Hebrew, idiomatic two-word verbs in English, and the seven prefixed Hebrew mor-phemes); treated them distinctly for language-internal analyses and cross-linguistic comparisons; and listed multilexemic expressions separately for each language (Berman, 2002).

Phrases

Sentences are not simply linear successions of words, but are made up of internal constituents. Words cluster syntactically in phrases and clauses, in turn combining into larger packages. In phrases, words group together as syntactic, clause-internal units,2 with a lexical element as headpronoun or noun in noun phrases (NP), verb in verb phrases (VP), adjective in adjective phrases (AP), and preposition in prepo-sitional phrases (PP). The head may be modified in NPs, by determiners, adjectives, prepositional phrases, or relative clauses (compare creatures with the many strange

L i n g u i st i C P e r s P eCt i v e s o n W r i t i n g d e v e L o P m e n t [ 19 ]

green-eyed creatures with striped suits that appear in the movie); in VPs, by auxiliaries, particles, and adverbials (e.g., take with may be taken away suddenly); the adjective in APs by adverbs (e.g., beautiful with the really most amazingly beautiful). Languages differ markedly in the ordering of phrasal heads and modifiers. For example, in English, modifiers precede the head noun if analyzed as a single word: compare green-eyed creatures with creatures with green eyes; in Hebrew, all modifiers except for numerals and other quantifiers follow the head noun; in French, adjectives some-times precede and sometimes follow the head noun.

Clauses

The clause is a semantic and syntactic unit of linguistic structure, defined by Berman and Slobin (1994, p.660) as a single predication expressing a unified situation (an activity, event, or state), with detailed conventions for dividing texts in different languages into clauses provided in an appendix (1994, pp.660662). Aclause is most typically, but not always, identifiable as containing a single verb (e.g., They walked home); but where verbs are modified by auxiliaries or by modal or aspectual verbs (e.g., The little boy might have been taken home; Her neighbor went on talk-ing nonstop) these are taken to represent unified situations, and hence are defined as single clauses. Moreover, not all clauses may contain an overt verb:In Hebrew, Russian, and Turkish for example, copular clauses in present tense need not contain an overt verb form (compare:they are students with its Hebrew equivalent hem stu-dntim literally they students). Importantly, a clause may but need not be a com-plete sentence, since complex sentences typically consist of more than a single clause. See, for example, the following sentence, with clause-endings indicated by a square bracket:When Iwas in the seventh grade,] I had a conflict with a boy] who was in a few of my classes].

The clause has proved a reliable unit of written and spoken discourse in differ-ent languages and types of extended discoursepicture-book and personal-expe-rience narratives, and argumentative and informative texts (Berman & Verhoeven, 2002). In practice, college students easily learn to demarcate texts into clauses, with high inter-coder agreement. In principle, the clause is preferable as a basic unit of written text-based linguistic analysis to the following alternatives: (a) utterancesstretches of speech output defined by intonation, best suited to spoken, interactive discourse; (b) propositionsvague semantic and/or discursive entities without clear structural boundaries; and (c) sentencesabstract, theory-dependent linguistic constructs that are notoriously difficult to define (Chafe, 1994; Halliday, 1989). Even educated adults do not share the same idea of what constitutes a sentence, and stylistic conventions differ on whether connectives like English so, yet, however may or must start a new sentence. Importantly, for present purposes, the clause is a necessary unit of analysis for evaluating syntactic complexity beyond the level of the phrase.

[20 ] Part I:Models and Perspectives on Writing Development

Clause Packages

An elusive, yet challenging facet of written text analysis lies in the domain of clause-combining (Haiman & Thompson, 1988), or syntactic packaging (Berman & Slobin, 1994, pp.538554)groups of clauses clustered together into larger units for segmenting texts and analyzing discursive connectivity (Berman, 1998; Scott, 2004). Clause packages (CPs) are text-embedded units of two or more clauses linked by syntactic, lexical, and/or thematic relations, which provide a lin-guistically motivated level of textual analysis lying between individual clauses and global discourse organization (Berman & Nir, 2009b; Nir, 2008; Nir & Berman, 2010). Interclausal relations within CPs are usually overtly marked by conjunc-tions, or else inferred from the thematic progression of a text, as illustrated in the Appendix.

The CP was preferred for assessing clause-combining to accepted notions such as a T(erminable) Unit (Hunt, 1965), on discursive and developmental grounds (Berman & Katzenberger, 2004, pp. 6468).3 CPs consider how clause linkage functions in the text as a whole; they take account of topic shifts or maintenance; and differentiate items like and, so, but used as pragmatically motivated utterance-introducers (Berman, 1996) or segment-tagging discourse markers (Ravid & Berman, 2006) as against grammatical interclausal connectives.

DIAGNOSTICS OF DEVELOPING WRITTEN TEXT CONSTRUCTION (SECTION 2)

Application of these units of analysis proved diagnostic of school-age writing abili-ties in use of words and syntax, and as reflecting communicative appropriateness.

Word-Based Measures

Overall text length assessed by number of words has been shown to differentiate between: age-schooling levels in different languages (Berman & Slobin, 1994; Berman & Verhoeven, 2002; Malvern, Richards, Chipere, & Durn, 2004), written versus spoken texts (Berman & Nir, 2009b; Berman & Ravid, 2009), and normally developing students versus students with language-impairment (Davidi & Berman, this volume).

Qualitative word-based measures reflect the importance of vocabulary for school-age literacy development in different populations (Dockrell & Messer, 2004; Perfetti, 2007; Ravid, 2004a). The cross-linguistic project revealed con-sistent patterns across the variables of age-schooling level (fourth graders in middle childhood, pre-adolescent seventh graders, adolescent eleventh grad-ers, and university-educated adults), modality (speech/writing), and genre

L i n g u i st i C P e r s P eCt i v e s o n W r i t i n g d e v e L o P m e n t [21 ]

(personal-experience narrative/expository discussion) for five text-based criteria of lexical usageword length, lexical diversity, density, abstractness, and register.

In word length, written texts contained longer words than oral texts, and exposi-tory essays contained longer words than personal-experience narratives. Older stu-dents used significantly more polysyllabic words than younger, with words of three syllables or more mainly from high school up. These findings applied to structurally distinct languages like English, Hebrew, and Swedish, both when counted in syl-lables (Berman & Ravid, 2009; Nir-Sagiv, Bar-Ilan, & Berman, 2008)and in letters (Strmqvist, Johansson, Kriz, Ragnarsdttir, & Ravid, 2002).

Lexical diversitythe proportion of different words out of total words in a text measures lack of lexical repetitiveness in writing. The VOCD (Vocabulary Diversity) procedure in CHILDES (Malvern etal, 2004)yielded higher scores for written than spoken texts, in interaction with age-schooling level (Strmqvist etal, 2002).4

Lexical densitythe proportion of content words (nouns, verbs, adjectives) in a textis indicative of textual richness, because open-class items convey the bulk of semantic content and substantive information. Written texts are lexically denser than oral texts produced by the same student (Berman & Nir, 2010; Nir et al., 2008), with content words increasing significantly as a function of age-schooling level. Proportion of Adjectives out of total content words is also diagnostic of more richly descriptive language across the variables of modality and age-schooling level (Ravid & Levie, 2010), as is use of manner adverbs like Hebrew bi-mhirut (with speed = rapidly), be-simxa (with-joy = joyfully) (Nir & Berman, 2010). However, this requires contextual analysis, because forms like ly often have a discourse-marker function of intensifying (e.g., absolutely, really, extremely) or hedging (e.g., basically, partly, probably), suited to colloquial oral style rather than to formal written language (Ravid & Berman, 2006). Lexical diversity and density, as text-based criteria of variety and referential richness, increase significantly from middle to high school, with higher scores in expository rather than narrative and in written rather than oral texts.

Similar patterns emerged for two other context-sensitive lexical measures applied to the English- and Hebrew-language samples of the cross-linguistic projectNominal abstractness and Linguistic register. Nominal abstractness is a qualitative, but quantifiable criterion that reflects the thematic content of a text. This semantic mea-sure is based on a 10-point scale devised for Hebrew (Ravid, 2006) and collapsed to 4 levels for English (Nir-Sagiv et al., 2008). At the low end are count nouns encoding concretely imageable referents (including specific entities like John, a ball, flowers and categorial roles and generic terms like every teacher, my house, people) compared to abstract, high-register, or low frequency terms (e.g., rival, cult), often also morphologically complex (like relationship, existence), or metaphorical exten-sions of concrete terms (path to success, river of time). In both languages, high-school students used far more abstract, morphologically complex nouns than younger, more in their written essays than their narratives. Similar scales could be devised

[22 ] Part I:Models and Perspectives on Writing Development

for different languages, based on contextual evaluation of how words are used in a given text. (Compare English way in: the way home, the Milky Way, the way up, right of way, came his way.)

Linguistic register characterizes level of lexical usage, from low-level slang and sub-standard usages inappropriate in formal, school-based written contexts via every-day colloquial usage of speakers of a standard variety to high-level, elevated styles (Conrad & Biber, 2001; Eggins & Martin, 1997). This largely sociologically deter-mined aspect of language use requires distinct criteria of low versus high regis-ter for each language. For example, the Germanic-Latinate contrast between largely monosyllabic and bisyllabic, everyday words of native origin like tell, childish com-pared with rarer, more elevated terms like relate, infantile, respectively, distinguishes the language used by English-speaking adolescents from different backgrounds (Corson, 1984, 1995). When applied to all open-class vocabulary items in our English-language sample by procedures detailed in Bar-Ilan and Berman (2007), this measure yielded consistently significant differences: Written expository texts relied most on formal Latinate items, as did high school students and adults com-pared with younger students. Proportion of words from Latinate compared with native Germanic origin is less relevant in a language like Swedish, whereas standard French applies other criteria for distinguishing everyday colloquial vocabulary from the academic, written-language lexicon ( Jisa, 2004a). The English-language sample also showed the five lexical measures to converge, and they were statistically cor-related (Nir-Sagiv et al., 2008): Words of Latinate origin are generally polysyllabic, they represent a more elevated register of usage, and are often semantically abstract.

Another, highly distinctive means of evaluating level of written language is word frequency as a key factor in language knowledge and use (Bybee, 2006; Bybee & Hopper, 2001), but this is only relevant in languages that have access to large, well-established lists of lexical frequencies across both written and spoken corpora (e.g., Leech, Rayson & Wilson, 2001), of a kind unavailable for Hebrew. Besides, each measure targets different facets of lexical knowledgeword-structure, seman-tic co