Cortical reorganization in dyslexic children after phonological training: evidence from early evoked potentials.

Brain plasticity was investigated in 14 Italian children affected by developmental dyslexia after 6 months of phonological training. The means used to measure language reorganization was the recognition potential, an early wave, also called N150, elicited by automatic word recognition. This component peaks over the left temporo-occipital cortex and its amplitude depends on linguistic expertise. N150 elicited by written words was measured both in dyslexic children before and after training and in a sample of matched normal readers during phonological, semantic and orthographic tasks. After training, dyslexic children increased their reading speed. Normal readers showed a typical left posterior N150, whereas in dyslexic children it was equally distributed across hemispheres before and shifted to left posterior sites after training. In addition, dyslexics' left posterior N150 asymmetry on the phonological task after training was significantly correlated with reading speed improvement, that is, those children who showed the greatest left shift in phonological N150 also had the greatest reading speed improvement. Source localization of the N150 component was made with both the Standard Low Resolution Electromagnetic Tomography software and the classical dipole analysis method termed Brain Electrical Source Analysis. The N150 generator lies in the left occipito-temporal cortex (Brodmann areas 39, 37 and 19) in good readers, but in right homologous areas in dyslexic children before training. After the treatment, the dyslexics' main N150 generator shifted to the left occipito-inferotemporal cortex (namely Brodmann areas 37 and 19) with small differences between tasks. The two source location methods provided consistent, converging solutions. Results add to the current literature on the phonological hypothesis of dyslexia by showing hemispheric reorganization of linguistic networks at the level of early word recognition potential. Furthermore, the present work is the first to investigate brain reorganization in a regular/transparent language like Italian.