The specificity of weak central coherence: The neuropsychological profile of autistic patients in contrast to depressive patients, schizophrenic patients and control subjects

Following the hypothesis of Frith (1), the neurocognitive deficits of autistic patients are due to weak central coherence. Accordingly, autistic patients are supposed to outperform control subjects on tasks which require piecemeal processing and underperform on tasks that require global or configurational processing of information. We investigated 15 autistic patients, 16 control subjects, 17 depressive patients and 16 schizophrenic patients. All participants underwent neuropsychological investigation including specific tests sensitive to central coherence. Autistic patients demonstrated better results in tests depending on local or piecemeal processing (mosaic test (WAIS), embedded figures test) relative to their performance in other tests but failed to outperform the control groups. They were impaired on a test in which success depends on global interpretation (fragmented picture test) relative to control subjects and depressive patients (p=0.05). We found evidence for weak central coherence in autistic patients in visual tasks but only limited evidence for its specificity.