Women with bulimia nervosa (BN), dieters and non-dieting control participants were questioned about spontaneous imagery linked to concern with food and eating, weight and shape. The downward arrow technique was used to access any associated negative or core beliefs, which were examined for belief, distress and content. A semi-structured interview with open and closed questions was used. Negative self (core) beliefs were successfully accessed, and responses to the interview items had good test–retest and good inter-rater reliability. Patients with BN reported significantly more negative self (core) beliefs than those in the other two groups. Only a very small number of core beliefs about other people or the world in general were reported. Emotional belief ratings appeared to be higher overall than rational belief ratings. Patient's negative self-beliefs contained themes of “self-value”, followed by “failure”, “self-control” and “physical attractiveness”, in descending order of frequency. The findings are discussed in relation to existing research, and implications for cognitive theories of bulimia nervosa and clinical practice are briefly discussed.
The first cognitive theory of bulimia nervosa (BN), outlined by Fairburn, Cooper, and Cooper (1986), is primarily a model of maintenance, and does not specify in great detail how the disorder develops. Subsequent theoretical developments have emphasised negative core beliefs as a potential causal factor (for a review see Cooper, 2005). The term ‘core-beliefs’ refers to fixed, absolute beliefs that are strongly held by an individual. They can be positive as well as negative. It has been suggested that three types of core belief are important; those related to the self, others and the world (Padesky & Greenberger, 1996). All three are negative when linked to psychopathology, and are associated with great distress when activated. They may be held rationally (i.e. when you think about how you feel logically) as well as emotionally (i.e. when you think about how you feel deep inside, regardless of what you know to be true logically). Beliefs that are strongly held emotionally are often difficult to modify, and may be held with a greater degree of conviction than rational beliefs in those with eating disorders (Turner & Cooper, 2002).
A semi-structured interview study of women with BN showed that core beliefs about the self were negative and unconditional, and that their content was concerned with themes of failure and uselessness (Cooper, Todd & Wells, 1998). Self-report sentence completion suggested that such beliefs, termed “negative self-beliefs” by the authors, may be comparatively more important than other types of core belief (i.e. world and other beliefs) in BN patients (Cooper, Todd & Cohen-Tovée, 1996). Using the Eating Disorders Belief Questionnaire, Cooper, Cohen-Tovee, Todd, Wells and Tovee (1997) found that women with BN scored higher than controls on a negative self-beliefs subscale, a measure composed of ten core beliefs (e.g. I'm no good, I'm unlovable) thought to be particularly relevant to this group.
Much remains to be understood about core beliefs in BN. There has been relatively little description in the literature of the content of core beliefs, of the degree of distress associated with them, or of both the rational and emotional belief that characterises them. The majority of studies have investigated only self related, and not other or world related core beliefs. To date, core beliefs have typically been accessed using verbal strategies — either self-report questionnaires or semi-structured interviews. It has been suggested that imagery may provide a quick and effective route to deeper levels of meaning (e.g. Barnard & Teasdale, 1991), and use of spontaneous imagery to access core beliefs is now well established in anxiety disorders (Holmes & Hackmann, 2004). To date, its usefulness in BN has not been evaluated. The current study aimed to (1) determine if the core beliefs (self, other and world) of patients with BN can be accessed via their spontaneously occurring imagery, (2) provide a detailed description of the content and other characteristics (distress, rational and emotional belief) of any core beliefs identified.
