Which training is the most effective for children with dyslexia? Effects on the ability of reading and spelling and the functional organisation of language in the brain

The purpose of this study was to examine the effectiveness of three different trainings for 3rd and 4th grade children with dyslexia. Each training is based on a different etiological theory. The improvement of reading and spelling should be emphasized either by improvement of the phonological awareness (rythmic training with syllables, following Buschmann (1988)), or by improvement of the temporal processing (auditory perception training with a PC), or by learning cognitive strategies and rules. Before and after the training phase the children took standardized and nonstandardized tests of reading and spelling, phonological processing, categorical perception and nonverbal intelligence (SPM).The cortical language processing was examined by means of magnetoencephalography. The mismatch field (MMF) and the localisation of dipols by auditory presentation of syllables were measured and a time-frequency-band analysis was computed by short visual presented high and low frequent words and pseudo-words (duration of presentation: 350ms, ISI: 0ms; Fast Words ).22 controls without dyslexia and 64 children with dyslexia participated in the study. The children with dyslexia were randomly allocated for the different trainings, which lasted for 5 weeks.Before the training the control-group was in all behavioral measures significantly better than the dyslexia-group. This includes a phonological processing deficit for the dyslexia-group, but there was no hint for a deficit in the temporal processing, since there were no differences in the MMF-amplitude. The source-localisation of the N260 in response of the syllable /ba/ was symmetrically distributed for the dyslexia-group, while the control-group showed the typical asymmetric dipol-distribution with the dipol in the right hemisphere being significantly more anterior than in the left hemisphere. There were significant differences betwen the two groups in the maximum amplitude of evoked power as well as in the spectral frequencies. This means a missing activation of phonological processes or a phonological processing deficit respectively.After the training all training-groups improved significantly in the spelling-test (DRT), while there was no improvement in the reading and phonological processing tests. Since there were no differential training effects and no improvements of the phonological processing, it is impossible to draw any conclusion about the etiology of dyslexia. However the rhythmic-syllabic training (following Buschmann, 1988) seems to be the most effective, since the effect was the same with only the half of training time in comparison to the other trainings. Concerning the MEG-parameters only the MMF-amplitude was increased for two groups: in the left hemisphere for the temporal-processing-training group and in the right hemisphere for the rhythmic-syllabic-training group. But even if you take these improvements as a possible measure for an improved temporal processing they do not correlate with the improved spelling-ability.With this study it was possible to show phonological processing deficits in behavioral and cortical measures. However since there were no differential training effects, it is impossible to conclude a causal connection between this phonological deficit and dyslexia