Material specific serial memory deficit in adolescent dyslexics.

The present experiment was concerned with an assessment of possible serial, as opposed to general, memory deficit in dyslexia. Previous studies had consistently confounded general and serial memory assessments. Fifteen specifically dyslexic adolescents and 15 normal-reading controls were administered four separate memory tasks. In the general memory for verbal material task they were shown 20 words, each for three seconds, and were then given a deck of 40 word cards and required to pick out the 20 words just seen. In the serial memory version of the verbal task they were again shown 20 words and subsequently required to reproduce the order in which they had been shown. Two comparable (general and serial) procedures were also administered using faces as the to-be-remembered materials. Results showed comparable performances of the dyslexics and controls on both versions of the face memory task and on the general memory version of the words task. On the serial reproduction version of the verbal task, however, dyslexics were found to be significantly impaired relative to controls. Thus, adolescent dyslexics appear to have a memory impairment which is specific for both type of material (verbal) and type of memory (serial). The results are compatible with Orton's (1937) speculation regarding "sequecing" and "memory" in dyslexia and with the view that the defect resides within the neuropsychological processes of the language-dominant hemisphere.