LAUSR

We aimed at investigating whether typical and atypical young readers extract vertices (viewpoint-invariant line junctions) in reading, as has been shown for fluent adult readers. In an identification task, we presented partly deleted printed letters, words, and pseudowords, preserving either the vertices or the midsegments of the letters. This allowed assessing the occurrence of a vertex effect, that is, more errors when vertices are partly removed, keeping the midsegments intact, than in the reverse situation. In Experiment 1, the vertex effect was observed on words and pseudowords in three groups of typical readers: 48 adults, 56 beginning readers (Grades 2 and 3), and 42 more advanced readers (Grades 4 and 5). Yet, the effect was smaller in the beginning readers, in relation to their irregular word reading skills. In Experiment 2, we compared 40 children with dyslexia with children selected from Experiment 1 to match them on either chronological age (30 CA controls) or reading level (42 RL controls). Although all groups displayed a vertex effect on words and pseudowords, dyslexic children presented a smaller effect than CA controls without differing from RL controls. The whole result pattern suggests that vertices play an important role in the recognition of written strings not only in skilled adult readers but also in young readers, in relation to their actual reading skills rather than to a specific reading deficit.