Responsibility and perfectionism in OCD: an experimental study

Abstract

Cognitive models of obsessive–compulsive disorder (OCD) suggest a number of different variables that may play a role in the development and maintenance of obsessive compulsive symptoms [Freeston, M. H., Rhéaume, J., & Ladouceur, R. (1996) Correcting faulty appraisals of obsessional thoughts. Behaviour Research and Therapy, 34, 433–446]. This study's aim was to verify the effect of perfectionism and excessive responsibility on checking behaviors and related variables. Twenty-four moderately perfectionistic subjects (MP) and 27 highly perfectionistic subjects (HP) were submitted to a manipulation of responsibility (low and high). After each manipulation, they had to perform a classification task during which checking behaviors were observed. Results indicate that more checking behaviors (hesitations, checking) occurred in the high responsibility condition than in the low responsibility condition for subjects of both groups. After executing the task in the high responsibility condition, HP subjects reported more influence over and responsibility for negative consequences than MP subjects. These results suggest that high perfectionistic tendencies could predispose individuals to overestimate their perceived responsibility for negative events. Furthermore, perfectionism could be conceived as playing a catalytic role in the perception of responsibility. Results are discussed according to cognitive models of OCD.

Introduction

Behavior therapy is faced with some limits for the treatment of obsessive–compulsive disorder (OCD). Almost 25% of patients refuse this type of treatment and approximately 25% do not benefit from it (Foa, Steketee, Grayson, & Doppelt, 1983). Cognitive therapy is considered as an alternative or as a complement to traditional behavioral treatments (e.g. van Oppen & Arntz, 1994; van Oppen et al., 1995; Freeston et al., 1996). In these circumstances, it is important to know and understand the cognitive variables that are involved in OCD in order to find effective ways of correcting cognitive distortions. Dysfunctional perfectionism and excessive responsibility have been identified as part of the five main cognitive variables associated with OCD (McFall & Wollersheim, 1979; Freeston et al., 1996); the other principal variables being overestimation of the importance of thoughts, overestimation of danger and the belief that anxiety caused by thoughts is unacceptable.

Recent definitions of responsibility and perfectionism contribute to our understanding of the role these variables play in OCD. Excessive or inflated responsibility has been defined as the belief which is pivotal to bring about or prevent subjectively crucial negative outcomes. They may be actual, that is, having consequences in the real world and/or at a moral level (Salkovskis et al., 1996). This definition has been empirically supported using a semi-idiographic questionnaire (e.g. Rhéaume, Ladouceur, Freeston, & Letarte, 1995c) as well as with experimental manipulations of responsibility (e.g. Ladouceur et al., 1995; Ladouceur, Rhéaume, & Aublet, 1997).

The unidimensional definition of perfectionism used in this study is: `the belief that a perfect state exists that one should try to attain' (Pacht, 1984). According to this perspective, which is particularly pertinent in the study of OCD (Rhéaume, Freeston, Dugas, Letarte, & Ladouceur, 1995a), perfection does not exist and the attempt to attain this perfect state would be associated with psychopathology. Hamachek (1978) points out that perfectionism can be a positive personality trait and distinguishes between sane and pathological perfectionism. The Perfectionism Questionnaire was devised to measure this construct with respect to the distinction between functional and dysfunctional perfectionism. The criterion and convergent validity of this instrument were established with questionnaire (Rhéaume, Freeston, & Ladouceur, 1995) and behavior manipulation studies (Rhéaume et al., 1995b).

In the past few years, the concept of excessive responsibility has received a lot of attention (Cottraux, 1990, Salkovskis, 1989; Rachman, 1993; Rhéaume, Ladouceur, Freeston, & Letarte, 1994; Tallis, 1994; van Oppen & Arntz, 1994; Rhéaume et al., 1995). Salkovskis (1985), Salkovskis (1989) made a great contribution to this theory by proposing a theoretical model whereby an excessive sense of responsibility is at the core of OCD. According to this model, obsessional patients would appraise intrusive thoughts as a function of possible harm to themselves or others. This excessive sense of responsibility would produce automatic negative thoughts, and discomfort would arise. The individual would then attempt to reduce the anxiety through cognitive neutralization or compulsive behavior (e.g. checking repetitively). Many studies support this model. In a number of clinical studies, the presence of an excessive sense of responsibility was observed in OC patients (Salkovskis, 1989; van Oppen et al., 1995; Ladouceur, Léger, Rhéaume, & Dubé, 1996). Furthermore, questionnaire studies comparing OC patients to control subjects support the existence of a link between responsibility and OC-type behaviors (Freeston, Ladouceur, Gagnon, & Thibodeau, 1992Freeston, Ladouceur, Gagnon, & Thibodeau, 1993; Rhéaume et al., 1995a). Finally, two recent experimental studies manipulated the level of perceived responsibility. Lopatka and Rachman (1995) succeeded in changing the perceived responsibility for negative consequences in thirty compulsive checkers. As expected, the lowering of responsibility was associated with a significant drop in discomfort and need to check. Lastly, two studies experimentally manipulated responsibility in nonclinical subjects (Ladouceur et al., 1995, Ladouceur et al., 1997). Results showed that checking behaviors were more frequent in the group receiving high responsibility instructions compared to controls.

On the other hand, although it has been suggested for many years that perfectionism also plays a key role in OCD (Hamachek, 1978; Burns, 1980; Pacht, 1984; Rasmussen & Eisen, 1989), this variable has not yet been fully studied. At the clinical level, Ladouceur et al. (1996) observed that perfectionism was a common characteristic in a group of OC patients without manifest compulsions, while responsibility was clearly less apparent in those same subjects. Correlational studies have shown a significant link between perfectionism and OC symptoms (Hewitt & Flett, 1991; Hewitt, Flett, & Turnbull, 1992). Furthermore, two correlational studies using the Maudsley OC symptom checklist with analogue subjects have shown that participants with OC tendencies were more perfectionistic than noncompulsive individuals (Frost, Steketee, Cohn, & Griess, 1994) and were also more perfectionistic than nonanxious controls (Gershuny & Sher, 1995). Finally, in a recent experimental study using a variety of tasks with nonclinical subjects, dysfunctional perfectionists obtained higher scores on the Padua Inventory and performed more poorly in precision and decision making tasks compared to functional perfectionists. These results support the link between perfectionism and OC symptoms (Rhéaume et al., 1995b).

At this point in time, the links between perfectionism and responsibility remain obscure. Studies conducted by Rhéaume et al. (e.g. Rhéaume et al., 1995a; Rhéaume, Ladouceur, & Freeston, 1998) have shown that responsibility and perfectionism are good independent predictors of OC symptoms. Nonetheless, little is known about the nature of the relationship between these two important factors. Considering the recent developments in cognitive therapy (e.g. van Oppen et al., 1995; Ladouceur et al., 1996; Freeston et al., 1997), it appears necessary to investigate how perfectionism and responsibility are linked together and to OC symptoms, in order to develop specific cognitive interventions adapted to this population.

The aim of the current study is to explore the links between perfectionism and excessive responsibility. This relationship will be studied by increasing and lowering perceived responsibility in subjects showing different degrees of perfectionism. We will verify if perfectionism, together with an excessive sense of responsibility, has an impact on the appraisal of intrusions and whether they predispose individuals to show OC tendencies. It is expected that highly perfectionistic subjects will show more OC-type behaviors and will attribute more responsibility to themselves than moderately perfectionistic subjects. In the high responsibility condition, we should observe, for both groups, more OC-type behaviors than in the low responsibility condition. Furthermore, this increase in checking and other related behaviors should be more marked in highly perfectionistic subjects than in moderately perfectionistic subjects.