Reasoning biases and delusions

Abstract

We know little about the formation and maintenance of delusional beliefs. Two main approaches have dominated the scant literature. These seek to account for delusions as primarily disturbances of perception (Maher, 1988) or as differences in reasoning (Garety, 1991). The concern here is with reasoning biases. Garety and Hemsley (1994) have proposed a model in which delusions’ are caused by a "failure to utilise previously acquired information". This leads to people with delusions exhibiting characteristic information processing biases in reasoning (i.e. hastiness and overconfidence). The aim of the present research was to compare the performance on reasoning tasks of people with delusions with that of psychiatric and normal control subjects in order to examine whether these subjects exhibited die proposed characteristics of delusional thought. The reasoning tasks were manipulated in both the form of reasoning (deductive, probabilistic etc.) and in content to examine the effect of reasoning with different types of material (neutral or emotional).The results of the six studies demonstrated both abnormal and normal reasoning by people with delusions. These people were no more confident than control subjects in the certainty of the correctness of their answers (Experiment 2). Nor were people with delusions excessively swayed by information currently present in the environment (Experiments 1 and 5) which is a supposed consequence of the inability to use past experience. However, people with delusions were shown to be hasty in their decisions relative to comparison subjects (Experiment 5). This hastiness was further exaggerated when the material reasoned with was self referent in content (Experiment 6). In addition, people with delusions were significantly poorer at reasoning on one of the most researched paradigms the Wason Selection Task (Experiments 3 and 4). The relevance of these findings for theories of delusions was examined.