Thought–shape fusion in anorexia nervosa: an experimental investigation

https://doi.org/10.1016/S0005-7967(01)00085-7Get rights and content

Abstract

Cognitive biases and cognitive distortions have been implicated as important factors in the development and maintenance of many disorders. The concept of thought–shape fusion (TSF) in eating disorders was developed by Shafran, Teachman, Kerry, and Rachman (British Journal of Clinical Psychology 38 (1999) 167) as a variant of thought–action fusion, described by Shafran, Thordarson and Rachman (Journal of Anxiety Disorders 10 (1996) 379). TSF occurs when thinking about eating certain types of food increases a person's estimate of their shape and/or weight, elicits a perception of moral wrongdoing, and/or makes the person feel fat. Shafran et al. (1999) examined both the psychometric and experimental properties of TSF in an undergraduate sample. This paper reports an extension of this work to a clinical group (N=20) of patients with anorexia nervosa. After completing a set of relevant questionnaires, participants were asked to think about a food which they considered extremely fattening. They were then asked to write out the sentence, “I am eating ———.”, inserting the name of the fattening food in the blank. After being asked to rate their anxiety, guilt, feelings about their weight, morality, etc., participants were given the opportunity to neutralize their statement in any way they chose. The majority of the participants neutralized in ways consistent with the findings of Shafran et al. (1999). The results are discussed in terms of cognitive–behavioural formulations of eating disorders, and of the influence of cognitive biases and cognitive distortions on the processing of information relevant to food, weight and shape in anorexia nervosa.

Introduction

There is evidence that people with anorexia nervosa have various cognitive distortions. The role of cognitions in this disorder has been acknowledged for a long time. Hilde Bruch (1962), on the basis of her clinical observations, highlighted the importance of maladaptive thinking in the development of anorexia nervosa. The presence in these patients of dysfunctional attitudes towards weight, eating, body size, shape and weight are clinically well known, and indeed are a defining feature of this disorder (e.g. Russell, 1983). Garner and Bemis (1982) proposed a formal cognitive–behavioral model akin to Beck's model of depression (Beck, 1976), and identified a set of dysfunctional cognitions. These included: relative abstraction (“I am very special if I am thin”), dichotomous reasoning (“If I put on one pound, I'll go on and put on enormous weight”); overgeneralization (“I was unhappy when I was at normal weight. So I know putting on weight is going to make me unhappy”); magnification (“I have gained a pound. So I won't be able to wear shorts again”); superstitious thinking (“If I enjoy anything, it will be taken away”); and personalization (“They were laughing. I am sure they were laughing at me”). Mizes (1992) has provided evidence of a variety of cognitive distortions in people with eating disorders.

Shafran, Teachman, Kerrry, and Rachman (1999) have postulated a specific cognitive distortion in patients with eating disorders: thought–shape fusion (TSF). This was developed on the basis of a distortion that had previously been identified for obsessive–compulsive patients—namely, thought–action fusion (TAF; Rachman, Shafran, Mitchell, Trant, & Teachman, 1996; Shafran, Thordarson, & Rachman, 1996). Essentially, this distortion consists of the belief that having a thought about a negative event increases the probability of the occurrence of the feared event, and the related belief that having a negative thought is morally the same as carrying out the corresponding negative action. Evidence has been reported that shows the existence of TAF in those with obsessive–compulsive disorder (Amir, Freshman, Ramsey, Neary, & Bartholomew, 2001, Shafran et al., 1996). TSF was hypothesized as a comparable cognitive distortion relevant to the psychopathology of eating disorders. This distortion was assumed to occur when merely thinking about eating a “forbidden” food increases the person's estimate of shape/weight, elicits a sense of moral wrongdoing, and makes the person feel fat (Shafran et al., 1999). Shafran et al. (1999) carried out two studies with undergraduate samples. They developed a questionnaire to measure TSF, consisting of 33 items. The first study showed that the questionnaire had high internal consistency, and also showed that TSF was associated with the psychopathology of eating disorders. The second study showed that the distortion can be elicited under experimental conditions. Subjects were asked to write a sentence about eating a forbidden food and then to imagine eating it. This led to the predicted TSF effect, and also to negative emotional reactions, and an urge to engage in correcting/neutralizing behaviour. The latter was found to lead to reductions in the negative emotional reactions. Shafran et al. (1999) stated: “(W)e believe that we have identified a cognitive distortion that may play some part in the maintenance of eating disorder psychopathology. The precise nature of that role and the clinical implications of the distortion warrant further investigation” (pp. 177–178).

The present study was undertaken as an extension of the work of Shafran et al. (1999). The aim was to explore the occurrence and extent of TSF in a clinical sample of patients with anorexia nervosa, using broadly the same experimental paradigm used in the second study of Shafran et al. (1999).

Section snippets

Participants

Participants were 20 inpatients in the Eating Disorders Unit at the Bethlem Royal Hospital in Kent. All participants had received a diagnosis of anorexia nervosa according to the criteria given in the Diagnostic and Statistical Manual, 4th Edition (American Psychiatric Association, 1994). Participants, all females, had a mean age of 27.7 (SD=7.8) years, and a mean body mass index (BMI) of 14.8 (SD=2.0).

Procedure

Participants were seen individually for this experiment. After consenting to participate in

Results

Questionnaire results are given in Table 1. Pearson correlations were computed for these measures and revealed a significant relationship between TSF and BDI (r=0.59, p<0.01), and marginally significant relationships between TSF and TAF (r=0.43, p<0.07), and between TSF and MOCI total scores (r=0.34, p<0.13).

Of the 20 participants who completed writing out the sentence, 15 chose to neutralize. Of these, four participants wrote out a new sentence to replace the old one; three changed the

Discussion

The results of this experiment show that inpatients with anorexia nervosa have the cognitive bias of TSF, as demonstrated by both psychometric data and experimental procedures. Participants who chose to neutralize (75% of the sample) experienced significant reductions in anxiety, guilt, likelihood of weight gain, feelings of fatness and urges to neutralize. The small number of participants who did not choose to neutralize appeared to experience similar reductions—but also appeared to have

Acknowledgements

The authors wish to thank Roz Shafran for the materials from her earlier work and for her valuable comments and feedback on this study.

References (22)

  • R. Shafran et al.

    Thought–action fusion in obsessive–compulsive disorder

    Journal of Anxiety Disorders

    (1996)
  • Cited by (41)

    • Central Autonomic Network Alterations in Anorexia Nervosa Following Peripheral Adrenergic Stimulation

      2023, Biological Psychiatry: Cognitive Neuroscience and Neuroimaging
    • Cognitive distortions in anorexia nervosa and borderline personality disorder

      2018, Psychiatry Research
      Citation Excerpt :

      Unlike in current literature (Coelho et al., 2012; Shafran et al., 1999), this study only examined the psychometric properties of the cognitive distortions, yet not the experimental induction of TSF or TAbF using a respective disorder-specific sentence paradigm. Inductions of TSF, for example, in which individuals imagined the consumption of high-calorie food, has been found to increase state-TSF in individuals with AN (Radomsky et al., 2002). In order to test disorder specificity more extensively, it would have been wiser to present either group with the respective sentence paradigm in order to make a direct comparison.

    • Modifying thought-action fusion via a single-session computerized interpretation training

      2017, Journal of Obsessive-Compulsive and Related Disorders
      Citation Excerpt :

      Even when TAF was broken down into TAF moral and TAF likelihood, these associations held, indicating that each subtype is significantly associated with OC symptoms. In addition to the relevance of TAF for OC symptoms and obsessive compulsive disorder (OCD), TAF is a transdiagnostic cognitive error that has been observed across a number of emotional disorders characterized by distressing, unwanted, or uncontrolled intrusive cognitions, including generalized anxiety disorder (GAD; Muris, Meesters, Rassin, Merckelbach, & Campbell, 2001; Thompson-Hollands, Farchione, & Barlow, 2013), depression (Abramowitz et al., 2003; Rachman, Thordarson, Shafran, & Woody, 1995; Shafran & Rachman, 2004), disordered eating (Radomsky, De Silva, Todd, Treasure, & Murphy, 2002; Shafran & Robinson, 2004), schizotypy (Lee, Cougle, & Telch, 2005; Muris & Merckelbach, 2003), and schizophrenia (Kabakci, Demir, Demirel, & Sevik, 2008). Therefore, developing an effective intervention for TAF has clinical utility across a wide range of problems.

    • A model of disturbed eating behavior in men: The role of body dissatisfaction, emotion dysregulation and cognitive distortions

      2016, Psychiatry Research
      Citation Excerpt :

      TSF has also been investigated in experimental studies: It has been shown that the imagination of eating fattening/forbidden foods may directly lead to distorted cognitions which influenced self-perception, in terms of feeling fat, concerns about weight gain and the experience of moral wrong-doing. This effect has been demonstrated mostly in female samples (Coelho et al., 2008, 2010; Radomsky et al., 2002). To conclude, the current research including prospective studies suggests a strong association between BD and ED pathology in men (e.g., Dakanalis et al., 2015a; 2015c; Lavender and Anderson, 2010).

    • Subjective experience of sensation in anorexia nervosa

      2013, Behaviour Research and Therapy
    View all citing articles on Scopus
    View full text