Shadowing by Context in Schizophrenia

Abstract

Schizophrenics, depressives, and normal controls were tested on a binaural listening task in which pairs of simultaneous continuous prose passages were presented. Subjects were required to shadow one passage of each pair and ignore the other. The passages in each pair, although differing markedly in content, were inseparable on the basis of their physical features. As predicted, the performance of schizophrenics on such a task was markedly worse than that of the other groups. The results were interpreted within the framework of Broadbent's model of selective attention (Broadbent, D. E. Decision and Stress. Academic Press, London, 1971). This makes a distinction between “pigeonholing” (response set) and “filtering” (stimulus set). The present findings are compatible with a schizophrenic defect at the pigeonholing stage. In contrast, the evidence for a defect in filtering specific to schizophrenia remains weak, due to the use of inappropriate control groups in a number of studies reviewed.