There is now extensive evidence that the learning processes of dyslexic chi
ldren show some abnormalities, generally consistent with failure to complet
ely automatise skills. Two studies are reported in which a group of adolesc
ent dyslexic children and a group of normal children matched for age and IQ
undertook long-term training on a keyboard spatial task and a choice react
ion task respectively. It was concluded that, following extended training,
the dyslexic children had normal "strength" of automatisation (as assessed
by resistance to unlearning, by ease of relearning after one year, and by d
ual task performance) but that their initial and their final performance (a
s assessed by speed and accuracy) were impaired. The results are consistent
with the hypothesis that dyslexic children suffer from cerebellar deficit.