Multiple patterns of dyslexia in a shallow orthography: Principles for a diagnostic screening of developmental dyslexias.

Dyslexia is a term that refers to a variety of reading impairments, each resulting from damage to a different component of the reading model. This view is common in the study of acquired dyslexias, but is less accepted in the realm of developmental dyslexia. In this study we examine whether we can identify various types of developmental dyslexia (Friedmann & Coltheart, in press) in Italian readers, by using a reading test that includes stimuli that are sensitive to the various types of dyslexia. The new test is aimed to detect not only surface, deep, and phonological developmental dyslexias, but also dyslexias that result from a deficit of the orthographic-visual analyzer, which involves three main processing units: letter identification, letter position detection, and letter-to-word binding (Ellis & Young, 1988). METHOD Participants. 262 primary school children (from 2nd to 5th grade) were assessed (50% Male). Materials. We developed and administered a new reading test that includes lists of 165 words, 40 nonwords, and 25 word pairs, created to detect the various types of dyslexia. A specific coding scheme for the classification of reading errors was developed (adapted from the TILTAN battery: (Friedmann & Gvion, 2003). In addition, we ran standard measures of reading skills (Cornoldi & Colpo, 2006; Zoccolotti et al. , 2005). Analysis. Poor readers (N = 52; 20%) and skilled readers (N= 210) were identified on the basis of the standard measures. An analysis of error types was carried out on the skilled readers’ performance, to obtain normative profiles for each grade. RESULTS Accuracy for skilled readers was 90%, and 75% for poor readers. The agreement between standard tests and the new assessment procedure in identifying poor readers was 94%: The new diagnostic assessment detected 11 poor readers who were not identified with the usual diagnostic procedure. In Italian skilled readers, all expected types of errors were found. Percentage distributions of errors for word lists showed that 41% of errors can be referred to the functions of the orthographic-visual analyzer (Letter Position, Attentional, and Visual Dyslexias), 18% can be identified as typical of Vowel Dyslexia and 27% can be considered as Surface errors. Other errors (about 14%) can be linked to dysfunctions of sublexical reading route. As for poor readers, two different groups were identified: Group A (about 50%), that showed an error distribution similar to skilled readers, but a lower level of accuracy; Group B, in which children showed a pattern of errors corresponding to a specific impairment in one component of the process (Figure 1) CONCLUSIONS The new screening method allowed us to detect also in Italian the several types of dyslexia described by Friedmann and Coltheart (in press). The new assessment seems to help in discriminating, among poor readers, children with a general lower level of accuracy (Group A), but with no specific impairment, and children with major impairment in one component of the process (Group B). Further research, will allow us to verify the hypothesis that this two profiles can be associated to different developmental trajectories.