The goal is the determine whether the phonological and surface profiles of dyslexia stem from a common deficit (of phonological
nature) or from multiple deficits. The longitudinal approach and the resort to techniques of selective training as well as the
analysis of possible impairments of speech perception in dyslexic children are used to distinguish between these two hypotheses, which
lead to different predictions as regards the evolution of the deficit patterns. Comparative studies with, on the one hand,
patients displaying acquired deficits of reading and, on the other hand, illiterate adults allow one to examine possible analogies and
determine the specific nature of developmental disturbances.