par Genard, Nathalie 
;Alegria Iscoa, Jesus ;Mousty, Philippe
Référence Neuro-psy, 14, 5, page (183-190)
Publication Publié, 1999-11
;Alegria Iscoa, Jesus ;Mousty, Philippe Référence Neuro-psy, 14, 5, page (183-190)
Publication Publié, 1999-11
Article révisé par les pairs
| Résumé : | This study highlights the importance that the writing system of a language may have on the classification and the etiology of the surface and phonological subtypes of developmental dyslexia. Phonological dyslexics were expected to be poor on pseudoword reading relative to irregular word reading, while surface dyslexics were expected to display the opposite pattern of performance. The regression method introduced by Castles and Coltheart (1993) was used to isolate these two extreme subgroups from a homogeneous sample of 75 French-speaking dyslexic children. The dyslexic sample was compared to a chronological age comparison group (CA group) and to a reading level comparison group (RL group). The CA comparison gives the opportunity of determining whether an individual's reading is abnormal and where difficulties lie. The RL comparison identifies whether the dyslexic profile of a subject corresponds to a deviance or a developmental delay, without specific deficit. Until now, all the studies using the regression method concerned English-speaking populations. Therefore we used this method in order to allow close comparisons with the English studies. Comparative studies between French and English are of particular interest as they allows us to examine (1) the degree of universality of the data obtained with the English writing system and (2) the effect of the consistency of the grapheme-phoneme relations (higher in French than in English) on the relative proportion of the surface and the phonological profiles. The results of the CA comparison showed that the relative proportion of the two subtypes differs from the data obtained in English. While the incidence of surface dyslexia is much higher in French, the incidence of phonological dyslexia is much higher in English, suggesting that the prevalence of one subtype on the other depends on the writing system of the language. The RL comparison enabled us to replicate Manis, Seidenberg, Doi, McBride-Chang and Petersen (1996) finding that surface dyslexics present a developmental delay while phonological dyslexics suffer from a phonological impairment, suggesting a universal picture of etiology of the different subtypes of developmental dyslexia. |
