NORMAL READING ABILITIES AND SPECIFIC READING DISABILITIES (DEVELOPMENTAL DYSLEXIA) : A CROSS-LINGUISTIC STUDY

RiPLA

RIVISTA DI PSICOLINGUISTICA APPLICATA

Direttore Scientifico/Editor-in-ChiefMARIA ANTONIETTA PINTO

Redazione/Editorial officeDipartimento di Psicologia dei Processi di Sviluppo e

Socializzazione · Università di Roma “Sapienza”Via dei Marsi 78 · i 00185 Roma

Segretario di Redazione/Editorial SecretaryDott. Sergio Melogno · Via dei Marsi 78 · i 00185 Roma

[email protected]

Direttore Onorario / Honorary EditorRENZO TITONE

Prof. Emerito/Emeritus Prof. Università di Roma“Sapienza” e Toronto (Canada)

Comitato Scientifico / Scientific BoardThomas D. Baldwin (Un. di Milano · Italia), Beatrice Benelli (Un. di Padova · Italia), Cristina Caselli (Ist. di Psicologia c.n.r. Roma · Italia), Marcel Danesi (Un. of Toronto · Canada), Antonella Devescovi (Un. di Roma “Sapienza” · Italia), Laura D’Odorico (Un. di Milano “Bicocca” · Italia), Alessandra Fasulo (Un. di Roma “Sapienza” · Italia), Emilia Ferreiro (Ist. Politecnico Nac. · Mexico), Lola Gonzales Gil (Un. de Sevilla · España), David Lasagabaster (Un. del País Vasco · España), Giuseppe Mininni (Un. di Bari · Italia), Margherita Orsolini (Un. di Roma “Sapienza” · Italia), Maria Da Graça Pinto (Un. de Porto · Portugal), Clotilde Pontecorvo (Un. di Roma “Sapienza” · Italia), Tatiana Slama-Cazacu (Un. of Bucharest · Romania), Stefania Stame (Un. di Bologna · Italia), Traute Tæschner (Un. di Roma “Sapienza” · Italia), Arturo Tosi (Royal Holloway, Un. of London · uk), FrancescaTrusso (Un. di Roma “Sapienza” · Italia), Jean Vivier

(Un. de Caen · France).

*« Rivista di psicolinguistica applicata » is an International Peer-Reviewed Journal

indexed by APA PsycinFO.The eContents is Archived with Clockss and Portico.

R I V ISTADI PSICOLI NGU ISTICA

A PPLICATA

xI

1-2 · 2011

PISA · ROMA

FABRIZIO SERRA EDITORE

MMxI

GESTO E PAROLA in PROSPETTiVA SEMiOTiCA,EVOLUTiVA ED inTERCULTURALE

GESTURE AnD SPEECH in A SEMiOTiC,DEVELOPMEnTAL AnD inTERCULTURAL PERSPECTiVE

Numero monotematico a cura di / Special Issue edited by

Maria Antonietta Pinto · Olga Capirci

Amministrazione e abbonamenti Fabrizio Serra editore

Casella postale n. 1, succursale n. , i 1 Pisa,tel. + 542332, fax + 574888, [email protected]

Periodico quadrimestraleI prezzi ufficiali di abbonamento cartaceo e/o Online sono consultabili

presso il sito Internet della casa editrice www.libraweb.netPrint and/or Online official subscription rates are available

at Publisher’s web-site www.libraweb.net.

I pagamenti possono essere effettuati tramite versamento su c.c.p. n. 11 o tramite carta di credito (American Express, Visa, Eurocard, Mastercard)

Uffici di Pisa : Via Santa Bibbiana 28, i 56127 Pisa,tel. + 542332, telefax + 574888, [email protected]

Uffici di Roma : Via Carlo Emanuele I 48, i 15 Roma,tel. + 670493456, telefax + 0670476605, fse.roma@ libraweb.net

*

Autorizzazione del Tribunale di Pisa n. del ..1Direttore responsabile : Lucia Corsi

Sono rigorosamente vietati la riproduzione, la traduzione, l’adattamento,anche parziale o per estratti, per qualsiasi uso e con qualsiasi mezzo effettuati,

compresi la copia fotostatica, il microfilm, la memorizzazione elettronica,ecc., senza la preventiva autorizzazione scritta della Fabrizio Serra editore®, Pisa · Roma.

Ogni abuso sarà perseguito a norma di legge.

Proprietà riservata · All rights reserved© Copyright 2011 by Fabrizio Serra editore®, Pisa · Roma

www.libraweb.net

issn 1592-1328issn elettronico 1724-0646

SOMMARIO / CONTENTS

articoli / research articles

Sandra Petroni, Affective learning and computational systems. New perspectives on second language learning 11Elena nuzzo, Camilla Bettoni, Insegnamento mirato in varietà avanzate d’italiano l2 : una sperimentazione didattica nel quadro della teoria della proces- sabilità / Developmentally-moderated instruction with advanced l2 Italian learn- ers: a pilot study within the framework of Processability Theory Paola Pettenati, Daniela Vacchini, Silvia Stefanini, Maria Cristi- na Caselli, Parole e frasi nel primo vocabolario di bambini bilingui Italiano- Spagnolo / Words and sentences in Italian-Spanish bilingual children’s first vocabulary Stéphanie Bellocchi, Mireille Bastien-Toniazzo, Normal reading abilities and specific reading disabilities (developmental dyslexia) : a cross-linguistic study Anna Claudia Ticca, Acciones despreferidas en la interacción mediada médico- paciente en Yucatán / Dispreferred activities in mediated doctor-patient interact- ions in Yucatan Antonio Aiello, Marcello Secchi, la prospettiva discorsiva applicata allo studio delle rappresentazioni “mediali” delle professioni : il caso dell’assistente sociale / Discursive perspective applied to the study of mass-media representat- ions of professional: the case of the Social Worker 1Giuseppe Mininni, Stefania Stame, In memory of Tatiana Slama-Cazacu 1

« rivista di psicolinguistica applicata », xi, 1-2/2011

NORMAL READING ABILITIES AND SPECIFIC READING DISABILITIES

(DEVELOPMENTAL DYSLExIA) : A CROSS-LINGUISTIC STUDY

Stéphanie Bellocchi* · Mireille Bastien-Toniazzo**University of Bologna · University of Provence, Aix-Marseille

Abstract : Although the majority of research on developmental dyslexia has been carried out in English, a number of studies have been performed in different languages. Some of the findings suggest that learning to read in deep orthographies, such as English, is harder than in shallow orthographies such as Italian (Seymour, Aro, Erskine 2003). However, very few direct comparative studies have been carried out in order to highlight differences or similarities between different orthographies (e.g. Ziegler, Perry, Ma-Wyatt, ladner, Schulte-Körne 2003). The present study aims at directly comparing reading-aloud abilities between French and Italian dyslexics by using a similar methodology for de-fining groups and measure performances. Results show a specific reading disorder in dyslexics and a lexicality effect which is not influenced by the type of orthography. By using a direct cross-linguistic comparison, this study provides evidence that lexicality effect is a general, non language-specific feature of the reading process, at least among French and Italian subjects.Keywords : Reading acquisition - Developmental dyslexia - Cross-linguistic studies – Shallow and deep orthographies – French – Italian.

introduction

Reading is a highly complex task involving the rapid coordination of visual, phonological, semantic and linguistic processes in order to access meaning.

In alphabetical languages, learning to read, especially at the very early stages, is fundamentally a process of matching distinctive visual symbols to units of sound and is thus a phonological recoding. However, despite the similar developmental trajectory of phonological representation across many European languages, rea-ding acquisition itself varies markedly across languages : different spoken langua-ges result in different systems of phonological representation and these differences affect the acquisition of literacy. In particular, the linguistic factor most likely to influence reading acquisition is the nature of the writing system (Seymour ).

Variation between European orthographies, which are all alphabetic, invol-ves two dimensions : consistency of relationship between letters (graphemes) and sounds (phonemes) and syllabic structure. Regarding the former, European ortho-graphies are called “shallow” or “transparent” when they provide a single distinct

- Latest version received January 11.* Corresponding author : Dr. Stéphanie Bellocchi, Department of Psychology, V.le C. B. Pichat ,

1 Bologna (Italy) ; [email protected].** Mireille Bastien-Toniazzo, Laboratoire Parole et Langage, UMR , CNRS & University of

Provence. , Avenue Pasteur. BP , 1 Aix-en Provence, Cedex 1. France ; [email protected].

70 stéphanie bellocchi · mireille bastien-toniazzo

written symbol for each phoneme. On the contrary, they are classified as “deep” or “opaque” when the same letter represents different phonemes in different con-texts, or when different letters represent the same phoneme (Frost, Katz, Bentin 1). European orthographies are now classified along a complex and continuous dimension of orthographic depth, with Finnish and Italian on the shallow end, En-glish on the deep end and French between these two extremes. A large number of monolingual studies carried out on relatively consistent writing systems showed a high accuracy performance on word and pseudoword reading by the end of Grade 1 (Ziegler, Goswami ). For example, Italian children’s word reading accuracy is about % compared with % for pseudowords (Cossu, Gugliotta, Marshall 1) ; French children’s accuracy is about % and %, respectively (Sprenger-Charol-les, Siegel, Bonnet 1). Regarding the latter dimension, European languages ha-ve different syllabic structures (Seymour ). For example, Romance languages (Italian, Spanish, Portuguese and French) are characterized by a simple structure with a majority of open CV syllables and few initial or final consonant clusters. On the contrary, Germanic languages (German, English, Scandinavian languages) have a more complex syllabic structure, with more closed CVC syllables and more consonant cluster at the onset or in coda positions. This linguistic characteristic influences the decoding process, as shown by Seymour, Aro and Erskine (). In particular, the influence of the syllabic structure primarily emerges in pseudoword reading, which is supposed to imply a decoding process. In their study, Seymour et alii (), showed that the lexicality effect (the advantage, in terms of error rates and reading speed, for words over pseudowords), for both accuracy and speed, was small in languages with a simple syllabic structure such as Italian and French and relatively large in orthographies characterized by a complex syllabic structure such as English and German. The grapheme-phoneme decoding is more difficult to acquire in the context of a language with a complex phonology than in a language with a simple phonology.

As suggested by Ziegler and Goswami () beginning readers are faced with another problem, the so-called granularity of spelling-to-sound mappings. In this perspective, development of reading in transparent orthographies is based on small grain-size units because of their relatively consistent grapheme-to-phoneme mappings of their script which is different to deep orthographies for which readers are forced to develop orthographic units of varying sizes (Ziegler, Goswami ). This implies that the development of grapheme-phoneme recoding skills takes longer in deep orthographies. However, this view is still a matter of debate. For example, Paulesu () suggested that, even in transparent writing systems, units of intermediate size influence reading performances, especially in adult readers. In fact, this consideration can also be made for child readers. Actually, recent fin-dings showed word frequency effect in Italian child readers (Barca, Ellis, Burani ; Colombo, Pasini, Balota ; Marcolini, Burani, Colombo ). Furthermore, Marcolini et alii (ibid.) found lexicality effect in children aged : their results sug-gest that lexical information is exploited even in the development of reading in a transparent writing system, at least by the age of . Moreover, in this latest study, authors found also that children were slower and less accurate in rejecting pseu-dowords derived from high-frequency words than pseudowords derived from low-frequency words, confirming data on adult readers (Arduino, Burani ). Recent

normal reading abitlies and specific reading disabilities 71

findings about an orthography of intermediate depth, Portuguese, showed that in this case, grapheme-phoneme conversion is not as predominant for phonological recoding as it is in shallow ones and that skilled readers in those orthographies may switch from smaller segment-by-segment decoding to larger unit or lexicon-related processing (Lima, Castro 1). Those results were also previously found by Content and Peereman (1) who suggested that for French readers reliance on grapheme-phoneme conversion is modulated by task condition. The authors observed that when reading lists were composed of words and pseudowords the effects of length and regularity were enhanced whereas the effect of frequency was attenuated, suggesting that this specific condition might trigger smaller units of processing.

Finally, there is another important and complex factor affecting the development of reading strategies across alphabetic orthographies : educational policy. Formal instruction in reading usually begins with primary school education. However, the age at which children start school varies among countries (Seymour ). For example, in UK, children enter school at about years of age, in Italy and France at about . In some countries, children may receive some reading instruction at ho-me or at kindergarten. Moreover, educational policy also influences teaching me-thods, ranging from holistic meaningful approaches to analytic ones. Regarding methods, there may be a centralized approach (all schools follow the same method and materials), as in Greece, or the possibility for individual teachers to choose their own preferred methods, as in Italy or in France. For this reason, the role of teaching may have an effect on comparisons between languages (Seymour ).

Several researchers have argued that models of reading development should be tested across languages varying in orthographical consistency (e.g. Aro, Wimmer ). Not only are cross-linguistic studies useful for understanding the mechani-sms involved in typical reading, but they also contribute to increasing our know-ledge on specific disabilities such as developmental dyslexia. Developmental dysle-xia is defined as a specific disability in learning to read and spell in spite of normal intelligence, adequate instruction, socio-cultural opportunity and in the absence of sensory defects in vision and hearing (W.H.O. 1).

Many cognitive deficits are associated to dyslexia. The core symptoms, in addi-tion to poor reading skills, include weak phonological processing, spelling difficul-ties and untidy writing (Vellutino, Fletcher ; for a comprehensive description of this disorder, see also Hulme, Snowling ). Other symptoms, such as unsta-ble visual perception, clumsiness and inattention have also been reported (Stein, Walsh 1). Dyslexia is also frequently associated with other learning disabilities, mainly language disabilities (Snowling ; ) and attention deficits (Penning-ton ).

Regarding the debate concerning the main causes of dyslexia (for a review, see Vellutino, Fletcher, Snowling, Scanlon ), the most accepted explanation refers to a core deficit at the phonological level of processing. In particular, the so-called phonological theory asserts that dyslexics have a specific impairment in representa-tion, storage and/or retrieval of speech sounds which prevents the correct acqui-sition of the grapheme-phoneme correspondence necessary for learning to read an alphabetic system (Snowling ; ). Different hypotheses have been put forward about the nature of phonological deficits ; however, the causal role of pho-


nology in dyslexia is largely prevailing (Ramus, Rosen, Dakin, Brian, Castellote, White, Frith ).

Although the majority of the studies on dyslexia have been carried out in En-glish, a number of researches have investigated it in different languages. For exam-ple, it has been reported that French dyslexics read % of pseudowords correctly compared to % for the chronological age-matched (CA) controls (Sprenger-Charolles, Colé, Lacert, Serniclaes ). The same study also showed that the phonological deficit manifested by phonological dyslexics emerged only as slow pseudoword reading. An orthographic deficit in surface dyslexics was found for irregular word processing time and irregular-word spelling accuracy and it also emerged, unexpectedly, in pseudoword spelling accuracy. These results were re-cently replicated (Sprenger-Charolles, Colé, Kipffer-Piquard, Pinton, Billard ).

Similarly, it has also been shown that Italian dyslexic children may be characteri-zed by a reading speed deficit, with relatively preserved accuracy (Tressoldi, Stella, Flagella 1 ; Zoccolotti et al. 1). Exploring their eye movement patterns while reading a meaningful text, De Luca, Di Pace, Judica, Spinelli and Zoccolotti (1) showed a highly fractionated text scanning with prevailing small amplitude saccades. This pattern was also present when dyslexics read lists of words and pseudowords (De Luca, Borrelli, Judica, Spinelli, Zoccolotti, ). Moreover, when reading a list of words and pseudowords, Italian dyslexics were found to be similarly impaired on both types of stimuli ( Judica, De Luca, Spinelli, Zoccolotti ). This result excludes a selective deficit in using the nonlexical route, often reported in English children (Castles, Coltheart, 1). The presence of a similar deficit in word and pseudoword reading and of a sequential analysis of written text could suggest that in languages with a shallow orthography dyslexics behave like “surface dyslexics” relying more on the non-lexical procedure of reading. However, recent findings sho-wed that Italian dyslexics seem to also use lexical reading and rejected the idea of an over-reliance on non-lexical processing (Barca, Burani, Di Filippo, Zoccolotti ).

More generally, dyslexic children of primary school age learning transparent orthographies (such as Italian) seem to have less difficulties with word and pseu-doword reading accuracy than dyslexics learning deeper languages, especially En-glish (for a review, see Caravolas, ). Regarding reading fluency, dyslexics learning more consistent orthographies read more quickly than their English counterparts (Landerl, Wimmer, Frith, 1).

However, it is necessary to take into account some methodological considera-tions and controversial issues to better understand general vs language-specific processes involved in typical and atypical reading acquisition.

First of all, a number of comparison studies which investigate the development of reading skills across different languages require assessments which depend upon standardized psychometric measures determined by historical, clinical and educa-tional traditions that differ between cultures (Seymour, ). This means that the way in which literacy and dyslexia are operationally defined is not equivalent for all different countries. In Italian, for example, developmental dyslexia is usually diagnozed mainly on the basis of the extreme reading slowness, although accuracy problems are present. On the contrary, in French the diagnosis is mostly done on the basis of the inaccurate reading alone although reading speed and spelling are compromized. These differences are reflected by the use of different parameters


(e.g. fluency vs accuracy) and standardized tests among the different countries.Another problem with these cross-language comparisons is that most of them

used a monolingual research design, as observed by Ziegler, Perry, Ma-Wyatt, Lad-ner and Schulte-Körne (). This creates two limits : 1) these studies did not di-rectly compare performance of dyslexics from different languages and ) the com-parison was not done on identical tasks and stimulus material. Moreover, in the analysis of their own English-German comparison study, Landerl, Wimmer and Frith (1) suggested that the conclusion that dyslexia seems to be more severe in English-speaking countries might have been amplified by the presence of a very large number of words that were irregular in English but not in German. Given that irregular words produce errors and slow down reading performance even for skilled adult readers (e.g. Seidenberg, Waters, Barnes, Tanenhaus 1), including too many of these stimuli in a cross-language comparison study could be a pro-blem. Ziegler and colleagues () carried out a cross-language comparison study where they compared German- and English-speaking dyslexic children on a num-ber of important marker effects of the reading process (like lexicality effect, length effect, body neighborhood, effect etc.). In this study, authors used lists of cognates (words with identical meaning and similar orthography and phonology between two languages) and lists of pseudo-words, mainly identical across languages (e.g. fot/Fot, lank/lank etc.). It emerged that the similarities between the two languages are far bigger than their differences. Both German and English dyslexics showed a reading speed deficit, a pseudoword reading deficit which was greater than the word reading deficit and a slow and serial phonological decoding procedure. Mo-reover, a direct comparison study between English and Czech (a transparent or-thography) dyslexics in which reading speed was measured in number of syllables per second (instead of mean time of word reading), showed no differences betwe-en the two languages (Caravolas ; Caravolas, Volin, Hulme ).

Finally, another methodological issue to consider is the importance of com-paring the performance of dyslexic children to that of two control groups, one matched on chronological age and one matched on reading age. The reading age matched group is required to ensure that a given deficit is not simply a consequen-ce of a developmental delay (poorer reading experience). The difference between dyslexics and normal readers with the same reading level can show a causal link between developmental dyslexia and a given deficit (Casalis ; Goswami ). However, some studies did not consider this latest control group, thus limiting the conclusions about the specificity of the developmental reading disorders.

Taking into account all these important considerations, the main goal of the present study was to investigate reading characteristics of developmental dyslexia in languages with transparent (Italian) or deeper orthography (French), through a direct comparison method. In particular, we explored how the structures of dif-ferent orthographies could influence the performance on word and pseudoword reading (fluency and accuracy) by using similar stimuli and the same method to measure the reading skills (number of syllables per second for speed, number of words/pseudowords incorrectly read for accuracy). To this end, we constructed some new experimental lists of words and pseudowords, trying not to include irregular words in the French material and to create pseudowords not too similar to familiar words, as suggested by Ziegler et alii (). We also attempted to ex-


plore the link between reading skills and developmental dyslexia by comparing the performance of dyslexic children to that of two control groups, one matched by chronological age and one matched by reading age. Finally, we aimed at testing the existence of an association between the experimental reading task and the standar-dized tests used to measure reading skills in each language.

Method

Participants

Two language groups of children, French and Italian, participated in the present study.

French sampleThe French sample was composed of groups : one experimental group of dy-

slexics ( females and 1 males ; mean age in months = 11, ; SD = , ; mean age of reading in months = , ; SD = ,) and two control groups without reading pro-blems, the first being a chronological-age matched group (CA) ( children, females and 1 males ; mean age in months = 11,1 ; SD = , ; mean age of reading in mon-ths = 1,1 ; SD = 1,) and the second a reading-level matched group (RL) ( children, females and 1 males ; mean age in months = , ; SD = , ; mean age of rea-ding in months = , ; SD = 1,). The total French sample was composed of participants, all native French-speakers with normal or correct-to-normal vision.

Dyslexics were recruited from two classes from schools in the South of France specialized in teaching to dyslexic children. Children were all diagnozed as dyslexi-cs by clinical psychologists. The criteria for inclusion in the sample were manifest reading deficits during a standard reading test, an IQ level within normal limits and no sensory deficit regarding vision or hearing.

Participants in control groups were recruited from the same primary schools, from Grade 1 to Grade , according to reading levels (Grade 1, or ) and chrono-logical age (Grade or ) of the dyslexic children. The criteria for inclusion were normal reading levels, IQ levels within normal limits and no sensory deficit regar-ding vision or hearing.

French groups’ reading and cognitive performancesFor all participants, reading level was assessed by means of a standardized

French subtest (lecture et déchiffrement- Reading and decoding) drawn from the K.ABC-Batterie pour l’Examen Psychologique de l’Enfant (Test for child psychological assessment), (French version by ECPA 1) which provides both standard scores and reading age. This last characteristic allowed us to match subjects’ reading level to dyslexics. The material consisted of a list of letters, regular and mainly irregular words of increasing difficulty. Participants were required to read them aloud as well as they could.

Table 1 provides means and standard deviations of the reading aloud perfor-mance for French participants.

Cognitive assessments of control groups were performed using the Raven’s Co-loured Progressive Matrices (French version by ECPA ). Cognitive assessments of the dyslexics were carried out by the clinical psychologists at the moment of the diagnosis using WISC-III (French version by ECPA 1). All control participants


were at least at the th percentile for their respective age level. Dyslexics’ mean standard score for Total IQ was , (SD = ,), for Verbal IQ : , (SD = ,) and for Performance IQ : , (SD = 1,).

Italian sampleAs the French sample, the Italian sample was composed of groups : one experi-

mental group of dyslexics ( females and 1 males ; mean age in months = 11,1 ; SD = 1,) and two control groups without reading problems, the first of these being a chronological-age matched group (CA) ( children, females and 1 males ; mean age in months = 11, ; SD = 1, ) and the second a reading-level matched group (RL) ( children, females and 1 males ; mean age in months = , ; SD = ,). The total sample was composed of Italian participants, all native Italian-speakers with normal or correct-to-normal vision.

Italian dyslexics were recruited from the Centro Regionale per le Disabilità Linguistiche e Cognitive in Età Evolutiva – ASL of Bologna (Regional center for linguistic and cognitive impairment in developmental age) and from the Servizio di Neuropsichiatria Infantile of Carpi – AUSL di Modena (Center of Developmental Neuropsychiatry in Carpi, Modena) where they all received a diagnosis of develop-mental dyslexia by clinical psychologists. The criteria for inclusion in the sample were manifest reading deficits during a standard reading test, an IQ level within normal limits and no sensory deficits regarding vision or hearing.

Participants belonging to the control groups were all recruited from mainstre-am primary schools where L.A.D.A. (laboratory for the Assessment of learning Di-sabilities) - Department of Psychology - University of Bologna, was conducting a screening activity for assessing learning disabilities. Also in this case normal rea-ders were recruited from Grade 1 to Grade , according to the reading level (Grade 1 or ) and chronological age of the dyslexic children (Grade , or ). The criteria for inclusion were normal reading levels, an IQ level within normal limits and no sensory deficits regarding vision or hearing.

Italian groups’ reading and cognitive performancesEach participant’s reading abilities were evaluated using three standard reading

achievement tests : Task 4 and 5 (speed and accuracy in word and pseudoword rea-ding) from the Batteria per la valutazione della Dislessia e Disortografia Evolutiva (Test for the assessment of developmental dyslexia and dysorthographia) (Sartori, Job, Tressoldi 1). The third task required to read aloud a meaningful passage (Prove di lettura MT per la scuola elementare ; Cornoldi, Colpo 1) where text difficulty varied depending on school level. Speed and accuracy were rated.

As it is known that in transparent orthographies like Italian the parameter mo-stly indicative of reading abilities is speed (Wimmer 1 ; Wimmer, Mayringer,

Dyslexics RL CA

Mean SD Mean SD Mean SD

K.ABC subtest-std scores 62,4 10,6 102,9 7,2 102,9 8,9

Table 1. Means and standard deviations of K.ABC subtest(Reading and decoding) standard scores


Landerl ), it was also considered as the most important in our study. Since a specific Italian test which provides age of reading does not exist, we also used a speed parameter to match reading level subjects to dyslexics.

Table provides means and standard deviations of the reading aloud perfor-mance for Italian participants.

Dyslexics RL CA


Words speed (syll/sec) 1,2 0,4 1,2 0,4 3,4 1,0

Pseudowords speed (syll/sec) 1,0 0,3 0,9 0,2 1,9 0,6

Text speed (syll/sec) 1,8 1,2 1,7 0,5 3,7 0,9

Table . Means and standard deviations of words, pseudowordsand text reading performance for each Italian group.

Cognitive assessments of controls were performed using an Italian version of the Kaufman Brief Intelligence Test (K-BIT, Kaufman and Kaufman 1) (Bonifacci, San-tinelli, Contento ). This test comprises Vocabulary and Matrices subtests and gives standardized measures of Verbal (VIQ), Performance (PIQ) and Composite Full Scale IQ. The dyslexics’ cognitive assessment was carried out by the clinical psychologists at the moment of the diagnosis using WISC-R (Wechsler 1 ; Italian version by Orsini 1). All the groups manifested standard scores within normal limits in relation to Italian norms (see Table ).

Dyslexics RL CA


Total IQ 102,2 9,5 109,2 9,3 102,9 8,0

Verbal IQ 101,6 8,8 103,5 9,6 102,6 10,4

Performance IQ 104,1 13,2 112,7 8,6 102,5 12,1

Table . Means and standard deviations of Total, Verbal and Performance IQfor each Italian group.

Stimuli for the experimental reading task

In order to directly compare word and pseudoword reading abilities of the two linguistic groups we selected real words and created pseudowords using special databases available for each language.French stimuli were composed of :

- French -letter-words selected from the Manulex database (Lété, Spren-ger-Charolles, Colé ), controlled and counterbalanced for frequency (SFI ≤ for low freq. and ≥ for high freq.) and dispersion (≥ .) (see Appendix 1). We mixed them with real words belonging to Manulex, controlled and counterbalanced for the same values of frequency and dispersion which were obtained by changing the first letter of the first previous items. We thus ob-tained words in total.


- 1 legal -letter-pseudowords created by a French database which provides fre-quency of bigrams and trigrams (Content, Radeau 1) [mean frequency of bigrams =11 ; ,1 SD] mixed with 1 -letter-pseudowords (see Appendix 1) created by changing the first letter of the previous 1. In total, pseu-dowords were obtained. We controlled positional frequency of bigrams and trigrams in order to obtain pronounceable pseudowords characterized by or-thographic regularity (Content, Radeau 1).

Italian stimuli were composed of :- Italian -letter-words selected from the database named Lessico Elemen-

tare (Marconi, Ott, Pesenti, Ratti, Tavella 1), controlled and counterbal-anced for frequency (SFI ≤ for low freq. and ≥ for high freq.) (see Appendix ). We mixed them with real words belonging to Lessico Elementare, controlled and counterbalanced for the same values of frequency and created by changing the first letter of the first . In this way we obtained words in total.

- 1 legal -letter-pseudowords created by replacing letters and/or syllables from real words which were controlled and counterbalanced for frequency. These pseudowords were mixed with 1 -letter-pseudowords which were created by changing the first letter of the first 1 (see Appendix ). We thus obtained pseudowords in total.

All these stimuli where created to be used subsequently in an experimental task of word identification in which half of the stimuli had to be created by changing the first letter of the target stimuli.

Procedure

Assessment of reading abilities was carried out individually in a quiet room for ap-proximately minutes. Parental permissions were obtained before administering the tests.

For the experimental reading task children were asked to read aloud the separate lists of words and pseudowords presented. They were also instructed to be as accu-rate and fast as they could. Half of the children in each group read the words first and then the pseudowords, whereas for the other half the order was the reverse. All the stimuli were presented one per line on a sheet.

The examiner recorded speed (number of syllables per second) using a digital chronometer, and accuracy (error rates in word and pseudoword reading).

Results

This section is divided into two parts : first, we present the results concerning the cross-linguistic analysis ran on the experimental reading task performances, whe-reas in the second part we present the relations between the measures obtained with the standardised and the experimental reading tasks.

1) Experimental reading task : the cross-linguistic comparisonFocusing on the main goal of the present study, for comparing reading perfor-

mance in different languages and different groups a mixed factor Analysis of Va-riance (ANOVA) was run, with lexicality (two levels : words and pseudowords) as


a within subjects factor, language (two levels : French and Italian) and Group (three levels : Dyslexics, CAs and RLs) as between subject factors. This ANOVA was used both for speed (number of syllables per second) and accuracy (error rate percentage of words/pseudowords read transformed in arcsin) (see Table ).

Given the heteroschedasticity problems emerging from Levene’s test [speed : words (p <.) and pseudowords (p <.1) ; accuracy : words and pseudowords (p<.1)], we considered a more stringent alpha level, that of ., for evaluating F ratios in subsequent analyses .

Results showed a lexicality effect, which means that participants were faster [F(1,11) = , ; p <.1] and more accurate [F(1,11) = , ; p <.1] at reading words than pseudowords. A general Group effect was obtained in reading speed [F(,11) = ,1 ; p <.1] and accuracy [F(,11) = , ; p <.1]. Post hoc analysis (Bonferroni) sho-wed evidence that dyslexics were significantly slower and less accurate than CA controls (speed and accuracy : p < .1) and RL controls (speed : p < ., accuracy : p <.1) . No significant language effect emerged. Moreover, no significant Group × language interaction emerged.

Dyslexics RL CA

French Italian French Italian French Italian

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

Words speed (syll/sec) 1,2 (0,8) 1,4 (0,5) 1,9 (1,3) 1,8 (0,8) 3,7 (1,1) 3,8 (1,1)

Words accuracy (% errors) 0,2 (0,2) 0,2 (0,1) 0,1 (0,2) 0,0 (0,1) 0,0 (0,0) 0,0 (0,0)

Pwords speed (syll/sec) 0,8 (0,4) 0,9 (0,3) 1,3 (0,7) 1,2 (0,4) 2,4 (0,9) 2,5 (0,9)

Pwords accuracy (% errors) 0,3 (0,2) 0,3 (0,2) 0,2 (0,1) 0,1 (0,1) 0,1 (0,1) 0,0 (0,1)

Table . Means and standard deviations of speed (syllables per second)and accuracy (error rate percentage) in words and pseudowords reading

for each Language and each Group

) Relations between standardized and experimental reading performance mea-sures

In order to test the validity of the new reading task, a bivariate correlation (Pear-son) was run between data collected from the administration of the standardized and the new experimental reading tests.

For the French sample, we ran a bivariate correlation between K.ABC subtest standardized scores and words and pseudowords decoding (speed and accuracy). Results (see Table ) showed a significant and positive correlation between K.ABC subtest standard scores and speed parameters of word and pseudoword reading. Moreover, a significant and negative correlation emerged between K.ABC subtest standard scores and accuracy parameters of word and pseudoword reading. These results suggest that the two types of measures were strongly associated and that the more accurate the children were in the K.ABC Reading and decoding subtest, the more rapid and less inaccurate they were in the experimental reading task.


Regarding the Italian sample, a bivariate correlation between standardized reading tasks [word (task ) and pseudoword (task ) reading from B.D.D.E. (Sartori, Job, Tressoldi 1)] and word and pseudoword decoding (speed and accuracy) was cal-culated. As shown in Table , significant correlations also emerged for the Italian sample : positive between speed parameters and negative between speed parame-ters and accuracy ones. In this case, as well, these results suggest the presence of a strong relation between what the standardized and the experimental test mea-sures.

French Italian

K.ABCsubtest- standard

scoresb

Words speed (Task 4) b

Pwords speed (Task 5) b

Text speed b

Words speed a -0,6** -0,9** ---0,9** ----0,8**

Words accuracy a -0,5** -0,3** -0,3* -0,1

Pwords speed a -0,6** -0,9** ----0,9** ----0,8**

Pwords accuracy a -0,5** -0,5** -0,3* ---0,3*

Notes :** = correlation significant at ,1 level (-tailed) ;* = correlation significant at , level (-tailed) ;a = experimental reading task ; b = standardized reading tasks.

Table . Bivariate correlation (r-values) between experimental reading tasks and standar-dized reading tasks in the French and in the Italian sample.

Discussion

Most of the research on normal reading development and dyslexia is performed in English, a language with a deep orthography and a complex syllabic structure. However, some studies suggested that generalizations drawn from these studies should be taken cautiously because it has been shown that English children have specific reading difficulties compared to children speaking other languages (e.g. Seymour, Aro, Erskine ). Recently, an increasing number of studies have also investigated reading development and dyslexia in different alphabetic languages with different orthographies, such as Italian and French. Italian has a transparent orthography and a simple syllabic structure, whereas French is characterized by an intermediate depth orthography (deeper than Italian) and simple syllabic structu-re. Cross-linguistic research is very important to determine whether the findings concerning dyslexia or reading development can be generalized to different langua-ges. However, some methodological problems have to be solved before we come to solid conclusions. For example, as clearly stated in the Introduction, a number of comparative studies across different languages is based on a type of assessment depending upon standardized psychometric measures related to historical, clinical and educational traditions that differ across cultures (Seymour ). This means that the way in which literacy and dyslexia are operationally defined is not the


same in all countries. Moreover, most of the cross-linguistic studies used mono-lingual research design, which means that they did not compare performance of dyslexic readers from different languages in identical tasks and stimuli. However, very few studies directly compared reading abilities in different languages. Some direct comparative studies showed that there are more similarities than differences between dyslexics using different orthographies (Caravolas, Volin, Hulme ; Ziegler et al. ). Starting from these considerations, one of the principle aims of the present work was to investigate reading characteristics of developmental dy-slexia in a transparent (Italian) and in a deeper (French) orthography by means of a direct comparative design. We also wanted to investigate whether the reported results in literature could be generalized to different types of orthographies.

First of all, a direct comparison between French and Italian participants’ reading performances, evaluated by specific language standardized tests, would not have been correct because of the specificity of the two languages. Hence, in order to pursue the main goal of the present work, we elaborated specific stimuli trying not to include difficult irregular words in the French material, to create same length words/pseudowords ( letters) equally counterbalanced for frequency (half low and half high, in both languages) and to create pseudowords not too similar to familiar words, that were also counterbalanced for frequency. We then considered the same parameters among languages to measure the reading skills (number of syllables per second for speed, number of words/pseudowords read incorrectly for accuracy). Afterwards, we ran a direct comparison between French and Italian par-ticipants’ performance on the experimental reading task. The speed and accuracy analysis showed that all dyslexic children were slower and less accurate on word and pseudoword reading compared to chronological age-matched and, more im-portantly, to reading level-matched controls (CA and RL). The dyslexic child had a significantly lower performance compared to the RL control child, which means that, in regard to the experimental lists of words and pseudowords, the first group is characterized by a specific decoding deficit and not by a developmental delay (Casalis ; Goswami ). This suggests that the reading deficit (speed and ac-curacy) is not simply a consequence of poor exposure to text but a clear and basic disorder underlying dyslexia.

Moreover, no difference emerged between French and Italian children, nor between normal readers or dyslexics, which might mean that their respective per-formance was not heterogeneous. In both languages, therefore, dyslexics exhibited a marked word and pseudoword reading deficit in comparison to both control groups, as explained before, and a pseudoword reading disorder greater than the word reading deficit. This last result seems to be in line with Ziegler and colle-agues’ findings () who found that similarities between German and English dyslexics were greater than differences : both showed a reading speed deficit, a pseudoword reading deficit greater than the word reading deficit and a slow and serial phonological decoding procedure. As Ziegler et alii (), we used lists of words and pseudowords with a similar number of letters ( for French and Italian) and, most importantly, we created a French word list without irregular words. It is known that the irregular words produce regularization errors and slow down re-ading performance even in skilled adult readers (e.g. Seidenberg, Waters, Barnes, Tanenhaus 1) and that French is a language characterized by the presence of


many irregular words, unlike Italian. Ziegler and colleagues () suggested that including these words may be problematic for a cross-linguistic comparison study. In addition, we also measured speed and accuracy using the same parameters for Italian and both French groups (syllables per second and error rates). In this study, dyslexic children learning a transparent orthography did not seem to have fewer difficulties with word and pseudoword reading accuracy than dyslexics learning deeper written language. As for reading fluency, dyslexics learning a more consi-stent orthography did not seem to read more quickly than their French counter-parts. The results obtained in this study might suggest that using similar materials and similar methodology may reduce differences between a deep (French) and a shallow orthography (Italian), as obtained in other studies (Caravolas, Volin, Hul-me ; Ziegler et al. ).

Results showed a lexicality effect, which means that word reading was faster and more accurate than pseudoword reading. Here, again, no differences between lan-guages emerged : the dyslexics’ pseudoword reading deficit was present in both languages with different orthography. The deficit persisted even after comparison with younger reading level controls which points to a specific disability. Seymour, Aro and Erskine () showed that the influence of the syllabic structure prima-rily emerges in pseudoword reading because it is supposed to affect the decoding process. In fact, orthographies characterized by a complex syllabic structure ma-nifested a more marked lexicality effect compared to those with a simple syllabic structure. In this study, the two languages involved, French and Italian, both have simple syllabic structure which could partly explain the absence of differences. In addition, pseudowords were created to be regular, using a special database (for French items) or replacing letters and/or syllables from real regular words (for Ita-lian items). The lexicality effect in reading extends to shallow (Italian) and interme-diate (French) orthographies, a result that was obtained for English, a deep ortho-graphy. Taken together, this evidence suggests that lexicality effect seems to be a general, non-language specific property of the reading process, at least among two writing systems with simple syllabic structure but different orthographic depth.

Regarding Italian, these results suggest that lexical information is exploited even in the reading development of transparent writing systems, as previously shown by different studies such as those carried out by Barca, Ellis and Burani (), Pagliuca, Arduino, Barca, and Burani () and Marcolini, Burani and Colom-bo (). Moreover, this effect seems to appear from the age of - years. For French, these findings are in line with those recently obtained by Lima and Castro (1) in Portuguese, another intermediate depth orthography, and, formerly, by Sprenger-Charolles, Colé, Kipffer-Piquard, Pinton and Billard () who found that lexicality effect was not greater for dyslexics than for reading level controls when comparing regular words and pseudowords.

Finally, in order to test the validity of the new reading tasks, we ran a correlation between data emerged from the administration of the standardized and the new experimental reading tests. Results showed the presence of a strong link between the two types of measures, suggesting that the experimental reading tasks created for this study have good validity in assessing reading abilities.

To conclude, this study seems to provide behavioural evidence that, at least among two orthographies with simple syllabic structure but different orthogra-


phic depth, lexicality effect seems to be a general, non language-specific property of the reading process : this occurs both in normal readers and in dyslexics. More evidence is needed regarding the comparison between orthographies with diffe-rent (or more complex) syllabic structures. Since differences may also arise from standardized psychometric measures used for assessing reading performance in different cultures (Seymour ), direct comparison studies involving similar me-thodology should be used to guarantee the generalization of results obtained with different orthographies. This could improve our comprehension of the language-specific or non-language-specific characteristics regarding typical and atypical rea-ding development.

Aknowledgments

Data presented in this paper are part of a wider study reported in a previously submitted article.

Authors thank all the children, their parents and their motivated teachers for their helpfulness, especially French teachers Isabelle Gras and Fouziya Lutz, for allowing data collection in their schools of Pélissanne and Avignon. We are also very grateful to the psychologists and speech therapists of the Centro Regionale per le Disabilità Linguistiche e Cognitive in Età Evolutiva -AUSL of Bologna and the Servizio di Neuropsichiatria Infantile of Carpi - AUSL of Modena, for allowing data collection on Italian dyslexic children.

We thank Stéphanie Ducrot and Claude Bastien for their precious suggestions on stimuli creation and methodological issues. We are also grateful to Silvana Contento for her support.

Finally, the authors thank the two anonymous referees for their useful sugge-stions and comments.

References

Arduino, L. S., Burani, C. (), Neighborhood effects on nonword visual processing in a lan-guage with shallow orthography, « Journal of Psycholinguistic Research », , pp. -.

Aro, M., Wimmer, H. (), learning to read : English in comparison to six more regular orthog-raphies, « Applied Psycholinguistics », , pp. 1-.

Barca, L., Burani, C., Di Filippo, G., Zoccolotti, P. (), Italian developmental dys-lexic and proficient readers : where are the differences ?, « Brain and Language », , pp. -1.

Barca, L., Ellis, A. W., Burani, C. (), Context-sensitive rules and word naming in Italian children, « Reading and Writing : An Interdisciplinary Journal », , pp. -.

Bonifacci, P., Santinelli, C., Contento, S. (), Un test breve di Intelligenza : il K-BIT, « Bollettino di Psicologia Applicata », , pp. -1.

Caravolas, M. (), The Nature and Causes of Dyslexia in Different languages, in The Science of Reading. A Handbook, Eds. M. J. Snowling, C. Hulme, Oxford, Blackwell, pp. -.

Caravolas, M., Volin, J., Hulme, C. (), Phoneme awareness is a key component of alpha-betic literacy skills in consistent and inconsistent orthographies : evidence from Czech and English children, « Journal of Experimental Child Psychology », (), pp. 1-1.

Casalis, S. (), Prédire et dépister précocement les dyslexies de l’enfant. Quelques questions théoriques et méthodologiques, « Rééducation orthophonique », , pp. 1-.

Castles, A., Coltheart, M. (1), Varieties of developmental dyslexia, « Cognition », , pp. 1-1.

Colombo, L., Pasini, M., Balota, D. A. (), Dissociating the influence of familiarity and


meaningfulness from word frequency in naming and lexical decision performance, « Memory & Cognition », , pp. 11-1.

Content, A., Peereman, R. (1), Single and multiple process models of print to speech conver-sion, in Analytic approaches to human cognition, Eds. J. Alegria, D. Holender, J. Morais, M. Radeau, Amsterdam, North-Holland, pp. 1-.

Content, A., Radeau, M. (1), Données statistiques sur la structure orthographique du Fran-çais, « Cahiers de Psychologie Cognitive- European Bulletin of Cognitive Psychology », n. hors série, pp. 1-.

Cornoldi, C., Colpo, G. (1), Prove di lettura MT per la scuola elementare. Firenze, O.S.Cossu, G., Gugliotta, M., Marshall, J. C. (1), Acquisition of reading and written spell-

ing in a transparent orthography : two non-parallel processes ?, « Reading and Writing : An Interdisciplinary Journal », , pp. -.

De Luca, M., Borrelli, M., Judica, A., Spinelli, D., Zoccolotti, P. (), Reading words and pseudo-words : an eye movement study of developmental dyslexia, « Brain and Lan-guage », , pp. 1-.

De Luca, M., Di Pace, E., Judica, A., Spinelli, D., Zoccolotti, P. (1), Eye movement patterns in linguistic and non-linguistic tasks in developmental surface dyslexia, « Neuropsycho-logia », , pp. 1-1.

Frost, R., Katz, L., Bentin, S. (1), Strategies for visual word recognition and orthographic depth : a multilingual comparison, « Journal of Experimental Psychology : Human Percep-tion and Performance », 1, pp. 1-11.

Goswami, U. (), Why theories about developmental dyslexia require developmental designs, « Trends in Cognitive Sciences », (1), pp. -.

Hulme, C., Snowling, M. J. (), Developmental Disorders of language, learning and Cog-nition, Oxford (UK), Wiley-Blackwell.

Judica, A., De Luca, M., Spinelli, D., Zoccolotti, P. (), Training of developmental surface dyslexia improves reading performance and shortens eye fixation duration in reading, « Neuropsychological Rehabilitation », 1, pp. 1-1.

Kaufman, A. S., Kaufman, N. (1), K.ABC-Batterie pour l’Examen Psychologique de l’Enfant, Paris, Editions du Centre de Psychologie Appliquée.

Kaufman, A. S., Kaufman, N. L. (1), Kaufman brief intelligence test, Circle Pines, MN, AGS.Landerl, K., Wimmer, H., Frith, U. (1), The impact of orthographic consistency on dys-

lexia : a German-English comparison, « Cognition », , pp. 1-.Lété, B., Sprenger-Charolles, L., Colé, P. (), MANUlEX : A grade-level lexical data-

base from French elementary school readers, « Behavior Research Methods, Instruments, and Computers », (1), pp. 1-1.

Lima, C. F., Castro, S. L. (1), Reading strategies in orthographies of intermediate depth are flexible : modulation of length effects in Portuguese, « European Journal of Cognitive Psycho-logy », (), pp. 1-1.

Marcolini, S., Burani, C., Colombo, L. (), lexical effects on children‘s pseudoword reading in a transparent orthography, « Reading and Writing : An Interdisciplinary Journal », (), pp. 1-.

Marconi, L., Ott, M., Pesenti, E., Ratti, D., Tavella, M. (1), lessico Elementare – Dati statistici sull’italiano scritto e letto dai bambini delle elementari, Bologna, Zanichelli.

Pagliuca, G., Arduino, L., Barca, L., Burani, C. (), Fully transparent orthography, yet lexical reading aloud : the lexicality effect in Italian, « Language and Cognitive Processes », , pp. -.

Paulesu, E. (), On the advantage of ‘‘shallow’’ orthographies : number and grain size of the orthographic units or consistency per se ?, « Developmental Science », , pp. –.

Pennington, B. F. (), From single to multiple deficit models of developmental disorders, « Co-gnition », 11 (), pp. -1.

Ramus, F., Rosen, S., Dakin, S. C., Brian, L. D., Castellote, J. M., White, S., Frith,


U. (), Theories of developmental dyslexia : insights from a multiple case study of dyslexic adults, « Brain », 1, pp. 1-.

Raven, J. (), Progressive matrices Raven, Paris, Editions du Centre de Psychologie Appli-quée.

Sartori, G., Job R., Tressoldi, P. E. (1), Batteria per la valutazione della Dislessia e della Disortografia Evolutiva, Firenze, O.S.

Seidenberg, M. S., Waters, G. S., Barnes, M. A., Tanenhaus, M. K. (1), When does irregular spelling or pronunciation influence word recognition ?, « Journal of Verbal Learning and Verbal Behavior », , pp. -.

Seymour, P. H., Aro, M., Erskine, J. M. (), Foundation literacy acquisition in European orthographies, « British Journal of Psychology », , pp. 1-1.

Seymour, P. H. K., (), Early reading development in European orthographies, in The Science of Reading. A Handbook, Eds. J. M. Snowling, C. Hulme, Oxford, Blackwell, pp. -1.

Snowling, M. J. (), Dyslexia, Owford, Blackwell.Snowling, M. J. (), language skills and learning to read : the dyslexia spectrum, in Dyslexia,

Eds. M. J. Snowling, J. Stackhouse, Chichester (West Sussex), Whurr, pp. 1-1.Sprenger-Charolles, L., Colé, P., Kipffer-Piquard, A., Pinton, F., billard, C.

(), Reliability and prevalence of an atypical development of phonological skills in French-speaking dyslexics, « Reading and Writing : An Interdisciplinary Jounal », (), pp. 11-. [IF=

Sprenger-Charolles, L., Colé, P., Lacert, P., Serniclaes, W. (), On subtypes of developmental dyslexia : evidence from processing time and accuracy scores, « Canadian Journal of Experimental Psychology », , pp. -1.

Sprenger-Charolles, L., Siegel, L. S., Bonnet, P. (1), Reading and spelling acquisition in French : the role of phonological mediation and orthographic factors, « Journal of Experimen-tal Child Psychology », , pp. 1-1.

Stein, J., Walsh, V. (1), To see but not to read : the magnocellular theory of dyslexia, « Trends Neuroscience », , pp. 1-1.

Tressoldi, P. E., Stella, G., Flagella, M. (1), The development of reading speed in Ital-ians with dyslexia : a longitudinal study, « Journal of Learning Disabilities », , pp. 1-1.

Vellutino, F. R., Fletcher, J. M. (), Developmental Dyslexia, in The Science of Reading. A Handbook, Eds. M. J. Snowling, C. Hulme, Oxford, Blackwell, pp. -.

Vellutino, F. R., Fletcher, J. M., Snowling, M. J., Scanlon, D. M. (), Specific reading disability (dyslexia) : what have we learned in the past four decades ?, « Journal of Child Psychology and Psychiatry », (1), pp. -.

Wechsler, D. (1), WISC-R. Wechsler Intelligence Scale for Children-Revised, New York, Psy-New York, Psy-chological Corporation (Trad. It. : Orsini, A., (1), WISC-R. Contributo alla taratura ita-liana, Firenze, O.S.).

Wechsler, D. (1), Echelle d’Intelligence pour Enfants et Adolescents, 3ème édition, Paris, Edi-tions du Centre de Psychologie Appliquée.

Wimmer, H., (1), Characteristics of developmental dyslexia in a regular writing system, « Ap-plied Psycholinguistics », 1, pp. 1-.

Wimmer, H., Mayringer, H., Landerl, K. (), The double deficit hypothesis and difficul-ties in learning to read a regular orthography, « Journal of Educational Psychology », (), pp. -.

World Health Organisation (1), The ICD-10 classification of mental and behavioural disorders. Clinical descriptions and diagnostic guidelines, Geneva, W.H.O.

Ziegler, J. C., Goswami, U. (), Reading Acquisition, Developmental Dyslexia, and Skilled Reading Across languages : a Psycholinguistic Grain Size Theory, « Psychological Bulletin », 11 (1), pp. -.

Ziegler, J. C., Goswami, U. (), Becoming literate in different languages : similar problems, different solutions, « Developmental Science », (), pp. -.


Ziegler, J. C., Perry, C., Ma-Wyatt, A., Ladner, D., Schulte-Körne, G. (), De-velopmental dyslexia in different languages : language specific or universal ?, « Journal of Experi-mental Child Psychology », (), pp. 1-1.

Zoccolotti, P., De Luca, M., Di Pace, E., Judica, A., Orlandi, M., Spinelli, D. (1), Markers of developmental surface dyslexia in language (Italian) with high grapheme-phoneme correspondence, « Applied Psycholinguistics », , pp. 11-1.


Appendix 1 – List of French words and pseudowords

words words pseudowords

1. vache. glace. lapin. terre. singe. homme. lampe. tarte. fille1. piste11. bombe1. pente1. riche1. belle1. fable1. fuite1. mille1. reste1. maçon. galet

1. cache. place. sapin. serre. linge. pomme. rampe. carte. bille. liste1. tombe. tente. biche. selle. table. suite. ville. veste. façon. valet

1. moter. raque. vosse. matte. déger. conge. plare. soume. prode1. parce11. poter1. faque1. posse1. fatte1. kéger1. ponge1. slare1. voume1. trode. marce

Appendix 2 – List of italian words and pseudowords

words words pseudowords

1. fondo. gatto. letto. madre. testa. forma. nonno. passo. carta1. gente11. freno1. mento1. buona1. sorta1. tacco1. pasto1. cappa1. morso1. santo. peste

1. mondo. fatto. tetto. padre. festa. treno. vento. suono. porta. sacco1. norma. tonno. masso. sarta. lente. tasto. zappa. sorso. manto. veste

1. volco. lorpa. farge. gunfo. solva. torlo. silte. vorro. taggo1. valvo11. golco1. zorpa1. narge1. munfo1. colva1. sorlo1. vilte1. sorro1. naggo. ralvo

composto in carattere dante monotype dallafabriz io serra editore, p i sa · roma.

stampato e r ilegato nellatipografia di agnano, agnano p i sano (p i sa) .

*

Giugno 2011(cz2/fg3)

NORMAL READING ABILITIES AND SPECIFIC READING DISABILITIES (DEVELOPMENTAL DYSLEXIA) : A CROSS-LINGUISTIC STUDY

Documents