Cognitive-behavioral models of OCD suggest that dysfunctional beliefs such as over-inflated personal responsibility, the overestimation of threat, the need to control thoughts, and perfectionism, underlie ineffective strategies in the management of intrusive thoughts, images and impulses ( OCCWG, 1997). Research has indicated that the vast majority of the population experience intrusions at times, and that the difference between common intrusive thoughts and "obsessions" is in terms of the frequency, intensity, duration, discomfort and misinterpretations elicited by the thoughts, rather than in the content of the intrusions (Rachman & de Silva, 1978; Salkovskis & Harrison, 1984). This raises the question of why particular intrusions lead to dysfunctional reactions such as anxiety, preoccupation and the activation of maladaptive beliefs and appraisals in certain individuals.
Current cognitive models of OCD implicate self-perceptions as leading to dysfunctional responses to particular intrusions (for a review, see Doron & Kyrios, 2005). For example, Rachman (1997, 1998) argued that catastrophic interpretations of the personal significance of intrusive thoughts are the main cause of the development and maintenance of obsessions. The content of the obsession is of critical concern in this process by serving as the initial trigger for dysfunctional appraisals in particular individuals (Rachman, 1997, p. 795, italics in original). Similarly, Clark and Purdon (1993; Purdon & Clark, 1999) proposed that the appraisal of a thought as inconsistent with an individual's sense of self, or their beliefs and values (i.e., as ego-dystonic) contributes to the exacerbation of obsessions. Indeed, following on from the work of Guidano and Liotti (1983) and Bhar (2004; Bhar & Kyrios, 2000), Clark (2004) argued that if "obsession-prone individuals have a preexisting ambivalent or fragile self-view, then unwanted intrusive thoughts that are completely contrary to this self view are more likely to be interpreted as highly significant or threatening" (p. 141). Finally, Salkovskis (1985, 1999) implicated the importance of OCD of an over inflated responsibility, defined as an individual's tendency to believe that they may be pivotally responsible for causing or failing to prevent harm from occurring to themselves or others. This implies such individuals may excessively value social and moral aspects of self.
Empirical investigations also support the link between self-structures and sensitivity to intrusive thoughts. For example, Rowa, Purdon, Summerfeldt and Antony (2005) found that for individuals with OCD, more upsetting obsessions were evaluated as more meaningful and more contradictory of valued aspects of the self than less upsetting obsessions. Bhar (2004; Bhar & Kyrios, 2000) found that individuals with OCD showed higher levels of self-ambivalence (i.e., an uncertain, preoccupied and dichotomous self-concept) than non-clinical controls, although they did not differ from individuals with other anxiety disorders. Thus, there is support for the contention that OC symptoms are related to intrusive thoughts that contradict valued aspects of self, particularly those that are not held securely.
Recently, Doron and Kyrios (2005; see chapter 2 of this thesis) proposed that enduring cognitive-affective structures, consisting of particular structures of self and coupled with specific beliefs about the world, may underlie vulnerability for OCD. Initial evidence has supported the association of particular views of the world with OC phenomena (see chapter 4 of thesis). Negative perceptions of one's own character, moderated by beliefs that the world is a just place where "people get what they deserve" added to the prediction of OC symptoms severity scores in a non-clinical cohort. Also, perceiving oneself as active in the prevention of harm moderated by beliefs that negative events can be avoided in the world, added significantly to the prediction of OC symptom severity scores. These results remained significant after controlling for OC related cognitions and affective variables. This study further examines evidence for the model, investigating the relationship of specific self-structures to OC phenomena.