Excessive responsibility in obsessional concerns: A fine-grained experimental analysis

Abstract

Excessive responsibility has been suggested as a central cognitive variable associated with Obsessive-Compulsive Disorder (OCD) (Rachman, 1993, Behaviour Research and Therapy, 31, 149–154; Salkovskis, 1985, Behaviour Research and Therapy, 23, 571–583; Salkovskis, 1989, Behaviour Research and Therapy, 27, 677–682; Salkovskis, 1995, Current controversies in the anxiety disorders). Several studies using questionnaires (e.g. Rhéaume, Freeston, Dugas, Letarte & Ladouceur, 1995, Behaviour Research and Therapy, 33, 785–794) and experimental manipulations (Ladouceur et al., 1995, Behaviour Research and Therapy, 33, 937–946; Lopatka & Rachman, 1995, Behaviour Research and Therapy, 33, 673–684) have shown evidence for such a link between responsibility and obsessive-compulsive symptoms. Responsibility has been defined as the belief of possessing a pivotal power to provoke or preventing crucial negative consequences (Salkovskis et al., 1992 cited in Salkovskis, 1995, Current controversies in the anxiety disorders). This definition emphasizes two related cognitive distortions: personal influence and potential negative consequences. The respective roles of each component and their potential interaction need to be clarified. The present study tests the effects of an experimental manipulation of both influence and negative consequences on perceived responsibility and checking behavior during a classification task. Seventy-seven subjects were divided into four experimental conditions: the Combined condition, the Influence condition, the Negative Consequences condition and the Control condition. After the experimental manipulation, subjects from each condition had to classify capsules in semi-transparent bottles. Results showed that personal influence is the best predictor of perceived responsibility. Although increased potential negative consequences were sufficient to trigger hesitations, combined influence and negative consequences were necessary to produce modifications. These results are consistant with the results obtained by the questionnaires (e.g. Rhéaume, Ladouceur, Freeston & Letarte, 1995a, Behaviour Research and Therapy, 33, 159–169) and previous manipulations (e.g. Ladouceur et al., 1995).