The anatomic site responsible for infantile autism is unknown. Based upon converging neuropsychological/neuropathological evidence that the temporal lobe may be the primary site of dysfunction in autism, we have adapted discrimination and memory tasks that distinguish amygdalar from hippocampal damage in monkeys (Murray/Mishk in l982). This neuropsychological paradigm was applied to 18 autistic subjects matched very closely to 18 Down's syndrome (DS) controls on the basis of age, sex and scores on the Raven Progressive Matrices. A subject was considered autistic when he/she achieved a score within 1 SD of the mean for the autistic subgroup of a 2,000 subject standardization sample (Autism Behavior Checklist, ABC). On a delayed nonmatching-to-sample task (DNMTS) cross-modal scores correlated with ABC body and object use (r= -0.42, p<0.05) and cross-modal with ABC social and self-help skills (r= -.45, p < 0.05). These correlation scales contain several items in common with the Kluver-Bucy syndrome, a neurobehavioral abnormality in animals and humans associated with damage to the amygdala. On the visual DNMTS, a significant group x task interaction was found (F= 6.68, DF= 2, 34, p < .01) indicating that DS performed better than autistics. Impairment of the visual DNMTS in monkeys results from lesions in both the amygdala and the hippocampus. There were no significant differences observed between the two groups on the spatial location tasks reflective of the hippocampal lesions alone. Taken together, these results suggest a disportionate functional impairment of the amygdala in autism.
Read full abstract