EN
Event-related potentials were used to examine if the brain response (N400 component) in dyslexic children is modulated by phonological or semantic priming, similarly to age-matched controls. ERPs were recorded while the children listened to word lists in which the semantic and phonological congruency of the terminal words were manipulated. Dyslexics exhibited a dissociation of priming effects depending on whether semantically or phonologically loaded primes were used. An enhancement of N400 amplitude to semantically incongruent words was observed, although this effect was reduced and delayed compared to that seen in controls. Direct comparison of the performance of the two groups in the semantic priming task revealed that they differed only in their response to semantically incongruent words. In the phonological priming task, rather than an enhancement of the N400 amplitude found in controls, dyslexics displayed a reduction of the N400 to the incongruent condition in comparison to the congruent condition. In this task, the studied groups differed in both the phonologically congruent and incongruent conditions. These results suggest that when faced with phonological priming, dyslexics have problems with both matching for similarities (integration into context) and incongruency detection. In the case of semantic priming, the integration of semantic context seems relatively intact in dyslexics, but they experience diffi culties in detecting the shift from one semantic category to another.