Elsevier

Behaviour Research and Therapy

Volume 87, December 2016, Pages 117-127
Behaviour Research and Therapy

Modifying interpretation biases in body dysmorphic disorder: Evaluation of a brief computerized treatment

https://doi.org/10.1016/j.brat.2016.09.005Get rights and content

Highlights

  • Tested a 4-session computerized interpretation bias modification program for BDD.

  • IBM increased benign and reduced threat biases, relative to placebo control (PC).

  • IBM led to reductions in symptoms among those high in pre-treatment symptoms.

  • Compared to PC, IBM reduced reactivity to in vivo appearance-related stressor task.

  • The current study provides preliminary support for the efficacy of IBM for BDD.

Abstract

Individuals meeting diagnostic criteria for body dysmorphic disorder (BDD; N = 40) were enrolled in a randomized, four-session trial comparing interpretation bias modification (IBM) training designed to target social evaluation- and appearance-related interpretation biases with a placebo control training condition (PC). Sessions took place over the course of two weeks (two sessions per week). Analyses indicated that, relative to the PC condition, IBM led to a significant increase in benign biases and reduction in threat biases at post-treatment. IBM also led to greater reductions in BDD symptoms compared to PC, though this effect was present at high but not low levels of pre-treatment BDD symptoms. Additionally, compared to PC, IBM led to lower urge to check and lower fear in response to an in vivo appearance-related stressor (having their picture taken from different angles), though the latter effect was present only among those reporting elevated fear at pre-treatment. The effects of treatment on interpretation biases and BDD symptoms were largely maintained at a one-month follow-up assessment. Moderated-mediation analyses showed that change in threat bias mediated the effect of condition on post-treatment symptoms for individuals high in pre-treatment BDD symptoms. The current study provides preliminary support for the efficacy of IBM for BDD.

Introduction

Body Dysmorphic Disorder (BDD) is a disturbance of body image characterized by a distressing and/or impairing preoccupation with a perceived anomaly in one's physical appearance (e.g., facial features, skin, hair); these individuals engage in time consuming compulsive behaviors or mental acts in response to intrusive appearance-related thoughts (e.g., excessive grooming, mirror checking, camouflaging, skin picking, comparing with others; APA, 2013). Individuals with BDD have markedly impaired psychosocial functioning and report lower quality of life than healthy controls (Marques et al., 2011a, Phillips, 2000, Phillips et al., 2008). If left untreated, BDD is a chronic and unrelenting disorder associated with high rates of being housebound, psychiatric hospitalization, and suicidality (Phillips, 2007, Phillips et al., 2013). Epidemiological studies investigating the point prevalence of BDD within the general population have found prevalence rates between 1.7% and 2.4% (Buhlmann et al., 2010, Koran et al., 2008, Rief et al., 2006). Existing cognitive-behavioral and pharmaceutical treatment approaches have demonstrated efficacy in treating BDD (meta-analyses indicate large average effect sizes for CBT d = 1.78, and SSRIs d = 0.92; e.g., Williams, Hadjistavropoulos, & Sharpe, 2006), though there is room for improvement. Modular CBT yielded an 81% response rate in a recent study, with participants experiencing an average symptom decrease of 55% at post-treatment (Wilhelm et al., 2014). This treatment was somewhat time-consuming, involving 22 sessions delivered over 24 weeks. It is important that researchers continue to explore novel treatment approaches for this population, particularly treatments that are easily disseminated, efficient, and cost-effective.

The cognitive-behavioral model of BDD proposes that biases in the perception, interpretation, and memory of appearance-related stimuli serve to maintain the disorder (Buhlmann & Wilhelm, 2004). Research suggests that individuals with BDD tend to interpret normal visual input (e.g., their image in the mirror) and social situations in a negative or threatening manner (Buhlmann et al., 2002, Fang and Wilhelm, 2015). For instance, these individuals tend to erroneously process others' emotional expressions, often misinterpreting neutral facial expressions as angry, contemptuous, or disgusted (Buhlmann and Wilhelm, 2004, Buhlmann et al., 2006, Buhlmann et al., 2011). Of note, this processing deficit is specifically observed when participants are asked to imagine the facial expression directed toward themselves and not when they imagine the facial expression directed towards others (e.g., a friend; Buhlmann et al., 2006). Falsely perceiving ambiguous emotional expressions as threatening or rejecting may maintain the disorder by reinforcing the individual's distorted self-image and ideas of social undesirability (Buhlmann et al., 2011, Buhlmann et al., 2013, Buhlmann et al., 2006). Further, these interpretive biases might paradoxically serve to elicit the feared consequences. For instance, an individual who perceives others' facial expressions as contemptuous or disgusted might be curt or unfriendly, leading others to adopt unfavorable perceptions of them.

Individuals with BDD also tend to endorse negative interpretation biases for ambiguous social and appearance-related information (e.g., interpreting laughter or looks from others as mocking or judgment about their appearance); these biases are stronger in patients with BDD than those with OCD or healthy controls (Buhlmann et al., 2002). In another study, individuals high in BDD symptoms exhibited negative biases for self-referent, but not others-referent, ambiguous situations, rating negative outcomes as more likely to occur to them than individuals low in symptoms (Clerkin & Teachman, 2008). Additionally, the high-BDD symptom group appeared to lack a normal self-enhancement bias, which is thought to foster and maintain self-esteem.

Preliminary research utilizing a single-session training program suggests that negative emotion recognition biases can be normalized through computerized training. Specifically, Buhlmann, Teachman, and Kathmann (2011) presented BDD participants with pictures of emotional faces (angry, disgusted, happy, neutral, sad, scared, and surprised) and trained them to correctly identify each expression via feedback following each trial. Findings indicated that training improved BDD participants' ability to identify neutral and scared facial expressions. However, the durability of training effects and the potential effects of training on BDD symptoms were not evaluated.

The tendency to process ambiguous information within one's environment as threatening is theorized to contribute to the etiology and maintenance of anxiety and emotional disorders (McNally, 1996). Research has shown that these problematic cognitive biases can be targeted and modified via brief computerized training programs; though these interventions have generally demonstrated small effects, IBM shows more promise than other CBM approaches such as attention bias modification (Hallion & Ruscio, 2011), and seems particularly effective for certain conditions, such as social anxiety (Cristea et al., 2015, Menne-Lothmann et al., 2014). One paradigm that has been used to induce healthier interpretations involves participants reading and imagining themselves in emotionally ambiguous scenarios that are resolved in non-threatening ways. Mathews and Mackintosh (2000) pioneered this approach, showing that computerized interpretation training can be used to induce benign (i.e., harmless/benevolent) or anxious interpretations after a single training session. Beard and Amir (2008), and later Amir and Taylor (2012), expanded on this methodology by developing an 8-session program in which participants with social anxiety were taught to associate ambiguous, socially-relevant sentences (e.g., “People laugh after something you said.”) with words indicating a benign interpretation (e.g., “funny”) and reject threat-related words (e.g., “embarrassing”) paired with these sentences. Participants indicated whether each word was related or unrelated to the sentence and were provided positive or negative feedback based on their responses. The benign training condition reported increases in benign and reductions in threatening interpretations and social anxiety symptoms, relative to a control condition.

Interpretation-training programs have shown efficacy in reducing trait anxiety (Mathews, Ridgeway, Cook, & Yiend, 2007), anxiety sensitivity (MacDonald et al., 2013, Steinman and Teachman, 2010), height fear (Steinman & Teachman, 2014), maladaptive perfectionism (Yiend, Savulich, Coughtrey, & Shafran, 2011), and anger reactivity (Hawkins & Cougle, 2013). Though several studies have shown promising findings, recent meta-analyses examining the general effectiveness of CBM protocols have been mixed; protocols with the strongest and most reliable effects typically instruct participants to imagine themselves in the training scenarios, provide participants feedback about their response accuracy during training, and involve multiple training sessions (Menne-Lothmann et al., 2014).1

Computerized treatment protocols are cost-effective, allow for easy administration, and do not require patient-therapist contact. The latter is especially relevant for individuals with BDD, as shame and apprehension about potential stigma are common barriers to treatment for this population (Marques, Weingarden, LeBlanc, & Wilhelm, 2011c). Given that BDD is characterized by interpretation biases (Buhlmann et al., 2002) and shares clinical features with other disorders that have been successfully targeted via CBM (e.g., social anxiety; Coles et al., 2006), it is possible that individuals with BDD may also benefit from similar training techniques. It is also conceivable that individuals with more severe symptoms will have biases that are more resistant to change, as beliefs about the importance of appearance and related perceptions of social consequences may be more deeply ingrained. IBM may work best for individuals with less severe symptoms or as an adjunctive treatment strategy in tandem with CBT and/or medications.

Premo, Sarfan, and Clerkin (2016) recently recruited individuals with elevated BDD symptoms and randomized them to two sessions of positive/benign interpretation bias modification or a control condition. The format of training was modeled after Mathews and Mackintosh (2000); participants were asked to read and imagine themselves in scenarios tapping themes related to BDD concerns (e.g., perfectionism, over-importance of appearance in daily interactions, belief that others are judging them based on their looks). Results indicated that, relative to the control condition, positive/benign training altered biased appearance-relevant interpretations in the expected direction, though it did not impact emotional reactivity to two in vivo appearance-related stressors (mirror gazing and picture tasks). These null findings with regard to emotional response may be attributable to low dosage of training (two 10–15 min sessions), training materials used (e.g., scenarios or format of the treatment and control conditions), low symptom severity of the sample, or other features of the study design.

The current study extends previous research on IBM training by examining its utility for BDD. All participants completed a 4-session treatment program (two 25-min sessions per week for two consecutive weeks). They were randomly assigned to one of two conditions: (1) positive IBM training designed to target social evaluation- and appearance-related interpretation biases characteristic of the disorder, or (2) a placebo control training (PC). In order to evaluate change due to training, participants were administered pre- and 1-week post-treatment measures of BDD-related interpretation biases and past-week BDD symptom severity. To evaluate the specificity of training effects, measures of past-week depression and anxiety were also administered. Further, participants' response to a novel in vivo task designed to evoke BDD-related concerns (having their picture taken from different angles) was examined to assess generalizability of effects. Finally, to examine the durability of treatment effects, participants completed follow-up questionnaires 1-month after the post-treatment assessment.

The primary hypotheses were: (1) Following training, participants in the IBM condition would report reduced threat interpretations and increased benign interpretations compared to the PC condition; (2) Relative to the PC group, participants in the IBM group would show a reduction in past-week BDD symptom severity; (3) Participants in the IBM group would show reductions in reactivity (urges to check appearance and peak fear) to the in vivo stressor task relative to PC; and (4) The effect of condition on BDD symptoms would be mediated by changes in biases. Exploratory analyses were also conducted to examine whether pre-treatment symptom severity moderated the effect of condition on outcome variables.

Section snippets

Participants

Participants (N = 40) were recruited from undergraduate psychology courses at a large southeastern university (N = 32) and the surrounding community via flyers and advertisements placed in local businesses (N = 8)2. The study was advertised as a computerized treatment for appearance concerns. One student participant per condition withdrew from the study because they were no

Baseline comparisons

Table 1 presents baseline demographic and clinical characteristics by group and Table 3 shows means and standard deviations for dependent variables at each time point. Group comparison tests revealed no significant differences between groups at baseline (ps > 0.05).

Interpretation biases

Linear regression analyses predicting post-treatment threat interpretations revealed main effects of pre-treatment threat interpretations (β = 0.30, t = 2.59, p = 0.014, sr2 = 0.088) and condition (β = 0.64, t = 5.52, p < 0.001, sr2

Discussion

The current study sought to determine the efficacy of a brief, four-session interpretation bias modification (IBM) program designed to reduce social evaluation- and appearance-related interpretation biases in individuals with BDD. IBM successfully reduced participants' tendency to make negative/threat interpretations in favor of positive/benign interpretations when presented with ambiguous scenarios, and these effects were especially pronounced at high levels of pre-treatment BDD symptoms.

Acknowledgments

The authors have no acknowledgements for this paper.

References (66)

  • A. Mathews et al.

    Inducing a benign interpretational bias reduces trait anxiety

    Journal of Behavior Therapy and Experimental Psychiatry

    (2007)
  • K.A. Phillips et al.

    Psychometric evaluation of the yale–brown obsessive-compulsive scale modified for body dysmorphic disorder (BDD-YBOCS)

    Journal of Obsessive-Compulsive and Related Disorders

    (2014)
  • K.A. Phillips et al.

    Functional impairment in body dysmorphic disorder: A prospective, follow-up study

    Journal of Psychiatric Research

    (2008)
  • J.E. Premo et al.

    Training interpretation biases among individuals with body dysmorphic disorder symptoms

    Body Image

    (2016)
  • J.C. Rosen et al.

    Development of the body dysmorphic disorder examination

    Behaviour Research and Therapy

    (1996)
  • A.M. Shaw et al.

    Hoarding and emotional reactivity: The link between negative emotional reactions and hoarding symptomatology

    Journal of Psychiatric Research

    (2015)
  • S.A. Steinman et al.

    Modifying interpretations among individuals high in anxiety sensitivity

    Journal of Anxiety Disorders

    (2010)
  • B.J. Summers et al.

    Intolerance of uncertainty in body dysmorphic disorder

    Body Image

    (2016)
  • S. Wilhelm et al.

    Cognitive behavior group therapy for body dysmorphic disorder: A case series

    Behaviour Research and Therapy

    (1999)
  • S. Wilhelm et al.

    Modular cognitive-behavioral therapy for body dysmorphic disorder: A randomized controlled trial

    Behavior Therapy

    (2014)
  • J. Williams et al.

    A meta-analysis of psychological and pharmacological treatments for body dysmorphic disorder

    Behaviour Research and Therapy

    (2006)
  • J. Yiend et al.

    Biased interpretation in perfectionism and its modification

    Behaviour Research and Therapy

    (2011)
  • L.S. Aiken et al.

    Multiple regression: Testing and interpreting interactions

    (1991)
  • American Psychological Association (APA)

    Diagnostic and statistical manual of mental disorders

    (2013)
  • N. Amir et al.

    Lack of a benign interpretation bias in social anxiety disorder

    Cognitive Behaviour Therapy

    (2012)
  • N. Amir et al.

    Interpretation training in individuals with generalized social anxiety disorder: A randomized controlled trial

    Journal of Consulting and Clinical Psychology

    (2012)
  • M.M. Antony et al.

    Psychometric properties of the 42-item and 21-item versions of the Depression Anxiety Stress Scales in clinical groups and a community sample

    Psychological Assessment

    (1998)
  • C. Beard et al.

    Combined cognitive bias modification treatment for social anxiety disorder: A pilot trial

    Depression and Anxiety

    (2011)
  • U. Buhlmann et al.

    Modifying emotion recognition deficits in body dysmorphic disorder: An experimental investigation

    Depression and Anxiety

    (2011)
  • U. Buhlmann et al.

    Cognitive factors in body dysmorphic disorder

    Psychiatric Annals

    (2004)
  • U. Buhlmann et al.

    Interpretive biases for ambiguous information in body dysmorphic disorder

    CNS Spectrums

    (2002)
  • I.P. Clara et al.

    Confirmatory factor analysis of the Depression–Anxiety–Stress Scales in depressed and anxious patients

    Journal of Psychopathology and Behavioral Assessment

    (2001)
  • E.M. Clerkin et al.

    Perceptual and cognitive biases in individuals with body dysmorphic disorder symptoms

    Cognition and Emotion

    (2008)
  • Cited by (0)

    View full text