Cognitive Therapy for Cognitive Theories of Depression: Restructuring Basic Assumptions

Abstract

Cognitive models of depression have generated an impressive research literature, making important contributions to the understanding and treatment of depression. Segal and Dobson's report from a recent consensus conference provides a thoughtful summary of a meeting of the leaders in the field and will serve as a useful guide for future research. They are to be commended for organizing the conference and for disseminating its contents. As their report suggests, however, all is not entirely well in the area of cognitive models of depression. The initial euphoria over the promise of these models and the psychotherapy derived from has yielded to dysphoria over weak, inconsistent, or negative findings in tests of some of the major hypotheses. In my comments, I suggest that one way to understand and treat this despair is by engaging cognitive theories in their own treatment: cognitive therapy. Specifically, I argue that cognitive distortions and underlying assumptions, particularly those related to stress, contribute to the current problems in cognitive research on depression. Identifying these underlying assumptions, examining their impact on the cognitive perspective, and submitting to the process of collaborative empiricism and revision, may help cognitive models correct misguided beliefs and become more productive. Epictetus's assertion, Men are disturbed not by but by the which they take of them (Epictetus, The Enchiridion), emerged as one of the early and enduring rallying cries of cognitive theorists. Although this and related statements certainly were, and continue to be, effective shorthand communication for focusing attention on cognitive variables in depression, they also represent a type of thinking that has contributed to the current distressed status of the field. Indeed, it is somewhat ironic that cognitive therapists have failed to recognize and reject such statements as examples of dichotomous thinking-a cognitive error seen frequently in depression. A closer look at Epictetus's statement betrays this eitheror, dichotomous quality, pitting things against views of and declaring views to be the victor in the explanation of distress. With such statements, cognitive theorists have argued themselves into the untenable and increasingly empirically unsupported corner referred to as cognitive primacy. This position essentially looks to cognitive content, processes, and structure to explain the etiology, maintenance, remission, and recurrence of depression. Importantly, this position has been reached somewhat inadvertently and quite mistakenly, for the prominent cognitive theories of the development of depression are explicitly identified as diathesis-stress models-models holding that neither the thingsnor the views are sufficient. Beck's model clearly requires the presence of both a diathesis (which he conceptualized as a cluster of attitudes reflecting two major personality dimensions, autonomy or sociotropy; Beck, 1987) and a congruent stress (an external negative life experience which impinges on the area of importance to the individual). Similarly, the hopelessness model of depression has recently made a clear statement of its diathesis-stress position (Abramson, Metalsky, & Alloy, 1989). In this model, the vulnerability is also cognitive in nature, but viewed in terms of attributional style, specifically the tendency to attribute causes of negative life events to stable, global factors. The stress is viewed as a negative life event that matches the domain (interpersonal or achievement) in which the attributional style is biased. The cognitive primacy view constitutes an inaccurate portrayal of cognitive models as it allows only a test of a single main effect. To be fair, it should be noted that most would not subscribe to such a radical cognitive primacy perspective. But, it appears that cognitive primacy has been rejected more in theory than in practice. The attention paid to the conceptualization and measurement of stress within these cognitive models does not come close to matching the attention paid to the conceptualization and measurement of the cognitive variables. Perceptions do not arise in a vacuum; they are grounded in social situations and in basic cognitive schemas. Elaborating characteristics of schemas is one aspect of the theory but it should not be taken as the only aspect. But just as cognitive therapists encourage their clients to act on changed beliefs, so should cognitive theorists. So, for example, although the four areas covered by the consensus conference (cognitive models, cognitive assessment, cognitive vulnerability, and integration) are all important, they reflect this relative neglect, which may be in part a function of the dichotomous thinking mentioned earlier. Although there is no question that the issues concerning the diathesis component of the model are critical, a model, like a chain, is only as strong as its weakest link. At the moment, the stress link appears rather weak.