Elsevier

Body Image

Volume 14, June 2015, Pages 20-28
Body Image

Social anxiety and associations with eating psychopathology: Mediating effects of fears of evaluation

https://doi.org/10.1016/j.bodyim.2015.02.003Get rights and content

Highlights

  • Social anxiety relates to drive for thinness, body dissatisfaction, and bulimic symptoms.

  • Fear of negative evaluation mediates between social anxiety and eating pathology.

  • Body mass index may moderate the above relationship.

  • Fear of positive evaluation does not mediate between social anxiety and eating domains.

Abstract

Recent work suggests unique relations among features of social anxiety disorder and eating disorder pathology. Thus, it may be important to determine specific facets of social anxiety that account for the relation between it and eating disorders. Given the similarities in social-evaluative concerns in both sets of symptoms, we hypothesized that fears of both positive and negative evaluation would each independently account for the relationship between social anxiety symptoms and eating pathology among college females (N=167). Results were partially supportive of hypotheses. Fear of negative evaluation independently accounted for a significant portion of the relationship between social anxiety and each domain of eating pathology that was tested, which included Drive for Thinness, Body Dissatisfaction, and Bulimic Symptoms. Body mass index appeared to play a moderating role on the relationship between fear of negative evaluation and body dissatisfaction, but not drive for thinness or bulimia symptoms. Clinical implications including diagnostic and treatment considerations will be discussed.

Introduction

The high correlation between social anxiety disorder (SAD) and various forms of eating pathology has been well documented (e.g., Hudson et al., 2007, Silgado et al., 2010, Swinbourne and Touyz, 2007). Many studies examining the comorbidity of SAD and eating disorders (EDs) have found that SAD tends to precede EDs in age of onset, leading researchers to suspect that SAD may serve as a risk factor for the later development of EDs (Brewerton et al., 1995, Bulik et al., 1997). Consistent with this notion, SAD tends to be highly overrepresented in ED populations (Kaye et al., 2004). There is a need for an increased understanding of the overlap between social anxiety and eating pathology, particularly in clinical settings, as false negative detection rates for EDs may be as high as 80% in clinics specializing in treatment of anxiety disorders (Becker, DeViva, & Zayfert, 2004).

A growing body of research has attempted to identify cognitive and behavioral explanatory mechanisms that account for this relationship (Hinrichsen, Wright, Waller, & Meyer, 2003). SAD may be a particularly relevant risk factor for EDs due to overlap in reported concerns, such as a desire to ‘fit in’ or to avoid the loss of social resources (i.e., mates, allies, etc.; P. Gilbert, 2001). Individuals with social anxiety, especially females, may be more likely to ascribe to social constructions such as the ‘thin ideal’ if they believe it will protect them from negative social outcomes (Utschig, Presnell, Madeley, & Smits, 2010). A number of studies have reported that fear of negative evaluation (FNE) may increase vulnerability to psychological symptoms beyond social anxiety, and FNE specific to body-image or eating concerns may, in part, account for the link between social anxiety and eating pathology (Bulik et al., 1991, N. Gilbert and Meyer, 2003, N. Gilbert and Meyer, 2005). Utschig et al. (2010) reported that FNE, a core component of SAD (Heimberg, Brozovich, & Rapee, 2014), was uniquely associated with bulimic pathology over and above risk factors included in the widely-cited dual-pathway etiological model proposed by Stice and Agras (1998). The dual-pathway model posits that internalization of the thin ideal and perceived societal pressures to be thin lead to body dissatisfaction which, in turn, leads to both negative affect and problematic dieting behaviors, which together result in problematic eating behaviors. Utschig et al. (2010) reported that FNE was positively related to many elements in the dual-pathway model, including pressure to be thin, internalization of the thin ideal, and negative affect. This pattern suggests that there may be an ED-specific component to FNE. Consistent with these findings, a recent longitudinal study conducted by DeBoer et al. (2013) found that FNE predicted future body dissatisfaction and ED symptoms, as well as internalization of the thin ideal among women with a relatively high BMI, defined as one standard deviation (2.17) above their sample mean of 21.72. These results suggest that BMI may play an important role in how FNE relates to the expression of eating pathology. Although there is accumulating evidence for a role of FNE in ED symptomatology, it remains to be seen whether other core facets of social anxiety may also help to explain its link with eating pathology. Other studies examining mechanisms linking social anxiety to eating pathology (e.g., Menatti et al., 2013, Silgado et al., 2010) are restricted to bulimic pathology and focus on perfectionism as the mediating variable. The authors know of no published work that has simultaneously examined multiple facets of social anxiety as potential explanatory mechanisms of the social anxiety–eating pathology link.

Psychoevolutionary models may provide a possible framework for conceptualizing the link between social anxiety and eating pathology. According to P. Gilbert (2001), socially anxious individuals tend to view their social world as being hierarchically organized, with more dominant others at the top and more submissive others at the bottom of the hierarchy. P. Gilbert (2001) argued that socially anxious individuals find themselves in a predicament in which any move that decreases their standing in this hierarchy puts them at risk of losing social resources to disastrous effect (e.g., ostracism, isolation, etc.). On the other hand, any move to improve their standing in this hierarchy could be met either with reprisal from dominant others or with an expectation that a high level of performance would continue indefinitely. Accordingly, Weeks and colleagues (e.g., Weeks et al., 2008a, Weeks et al., 2008b) have invoked P. Gilbert's (2001) model in explicating the construct of fear of positive evaluation (FPE), and several studies have since provided compelling evidence that FPE is a critical and unique component of social anxiety as well as SAD (e.g., see Weeks & Howell, 2014, for a review). FPE appears to be strongly related to, but distinct from, FNE and to contribute significant incremental variance to the prediction of social anxiety, social situation-specific negative affect, negative automatic thoughts related to social interactions, disqualification of positive social outcomes, self-consciousness, and susceptibility to embarrassment (Weeks and Howell, 2012, Weeks and Howell, 2014). According to Weeks and Howell (2012), socially anxious individuals demonstrate not only FNE, but rather a fear of evaluation in general.

Fear of evaluation may also be relevant to disordered eating behaviors resulting from internalization of the thin ideal and the drive to “fit in” in sub-cultures that place undue value on physical appearance (e.g., sororities and college women in general; Bardone-Cone et al., 2013, DeBoer et al., 2013). A novel avenue for research involves examining FPE alongside FNE as specific mechanisms linking social anxiety to eating pathology. The application of P. Gilbert's (2001) psychoevolutionary model to eating pathology may help researchers and clinicians better understand the phenomenological overlap between eating pathology and social anxiety. For example, individuals with problematic eating concerns may view others as being organized into a hierarchy similar to those with social anxiety, but with “attractive” and physically fit people at the top and “ugly” or “fat” people at the bottom (Bardone-Cone et al., 2013). Thus, an individual's move to decrease his or her standing in this hierarchy may result in a loss of social resources (e.g., mates, allies, etc.), whereas any move to increase his or her standing in the hierarchy may result in heightened social competition, reprisal from others, or an expectation that their move up the hierarchy would be maintained and therefore untenable (P. Gilbert, 2001). P. Gilbert's (2001) psychoevolutionary model applied to eating pathology may help explain behaviors such as dietary restriction (i.e., avoidance of appearing unattractive) and body-concealing (e.g., wearing baggy clothes, which may serve to hide perceived attractiveness [i.e., FPE] as well as perceived flaws [i.e., FNE]; Craighead & Smith, 2008). FPE applied to eating pathology may involve a person worrying that she will attract more positive attention to herself. Applied to an individual suffering from bulimia nervosa: an innocuous positive comment on one's appearance (e.g., “that dress fits you very well”) may lead to thoughts such as, “I hope my boss doesn’t think I’m trying to look better than her.” Furthermore, this individual may seek to avoid future positive evaluation by dressing very conservatively or in baggy or loose-fitting clothes. The experience of FNE in the context of eating pathology (aversion to feeling unattractive) may be balanced precariously against the experience of FPE (aversion to feeling too attractive). As is the case in SAD (see P. Gilbert, 2001, for a review), sufferers of eating pathology may fear both decreases as well as increases in their perceived attractiveness because each brings its own set of potentially threatening consequences.

Levinson and Rodebaugh (2012) found some support for the role of FPE in participants with eating concerns. FPE was positively related to body dissatisfaction, bulimic symptoms, shape and weight concerns, as well as eating concerns among undergraduates. It should be noted that Levinson and Rodebaugh (2012) tested two models: one model specified FNE and social appearance anxiety as simultaneous mediators between social anxiety and eating pathology, and the other specified FNE, FPE, and social appearance anxiety as vulnerabilities to both social anxiety and eating pathology. Their results indicated that the (latter) vulnerability model demonstrated superior fit compared to the mediational model. However, this interpretation of their data may be problematic for a number of reasons. First, their vulnerability model does not account for prospective findings that indicate that SAD precedes EDs in the majority of cases (Brewerton et al., 1995, Godart et al., 2002). Second, their sample was comprised of both males and females. This may be problematic due to the lack of psychometric validation of ED measures in male samples (Strother, Lemberg, Stanford, & Turberville, 2012). Indeed, the reliability of the Eating Disorder Inventory (EDI) and Eating Disorder Inventory – II (EDI-II) items tends to be lower in males than in females, and scores for the eating pathology subscales of the EDI-II (Drive for Thinness, Bulimia, Body Dissatisfaction) tend to be lower among males than females (Stanford & Lemberg, 2012). Stanford and Lemberg (2012) recommend the use of eating disorder assessment instruments that are specifically designed for use among male samples in research on eating pathology among males. Utilizing measures that have been adequately normed for both males and females is essential if a mixed-gender sample is used to make inferences based on scores on these measures (Groth-Marnat, 2009). Without male norms for certain ED measures, it may be prudent to focus on female-only samples.

The eating pathology literature has yet to deeply explore the potential role of FPE in its current models. Though it has already been demonstrated that FNE may play a unique and important role in the development of bulimic symptoms (Utschig et al., 2010), FPE has been largely neglected in eating pathology literature despite its well-documented high correlations with FNE (Carter et al., 2012, Weeks et al., 2008a). On the other hand, the emerging literature examining FPE has yet to consider its potential role in eating pathology. Clinically, this may translate into diagnostic considerations, including specialized assessment of eating disorders/symptoms, in patients who present with clinically elevated levels of FNE and FPE. If FPE is indeed a common trait across both social anxiety and eating pathology, it may also mean that interventions designed to target FNE and FPE would be effective for both SAD and ED symptoms.

The current study sought to investigate the potentially unique roles of FNE and FPE in explaining the relationship between social anxiety and ED symptoms among females. An abundance of research has found that SAD tends to precede eating pathology in age of onset (Brewerton et al., 1995, Bulik et al., 1997, Kaye et al., 2004), and it has also been proposed that SAD may reflect a genetic vulnerability to eating disorders (Kaye et al., 2004). Thus, we designed our mediation model in such a way that social anxiety predicts eating pathology. However, given that our study design is cross-sectional, it bears noting that causal predictions are precluded. Specifically, we hypothesized that social anxiety would account for unique variance in three domains of eating pathology (i.e., bulimic symptoms, body dissatisfaction, and drive for thinness). Given the strong role FNE has been reported to play in eating pathology (DeBoer et al., 2013, Utschig et al., 2010) and the theoretical basis for the putative role of FPE in the relation between social anxiety and eating pathology, we also hypothesized that FNE and FPE would each account for unique variance in these relationships. In other words, we hypothesized that, after partialing the variance in social anxiety associated with fears of evaluation (FNE/FPE), the association between social anxiety and each domain of eating pathology would be reduced, suggesting that fear of evaluation in general is a critical component related to the high co-occurrence of social anxiety and eating pathology. Despite limitations of testing mediational models in cross-sectional studies, we believe that it is appropriate in this case, given that there is a reasonable case for temporal precedence of social anxiety in relation to eating pathology (Brewerton et al., 1995, Bulik et al., 1997, Kaye et al., 2004). Furthermore, we expected that FNE and FPE would each independently contribute to this indirect effect.

Section snippets

Participants

These hypotheses were tested in a sample of 167 female undergraduates who completed study questionnaires in exchange for academic credit. The study measures were embedded within a larger questionnaire battery containing measures of personality, health behaviors, and mood so that participants would not be alerted to the purpose of the study.

Measures

The Social Interaction Anxiety Scale (SIAS; Mattick & Clarke, 1998) assesses levels of anxiety experienced in interactions with others. Internal consistency

Preliminary Analyses

Sample characteristics, along with means (M), standard deviations (SD), and correlations for the independent and dependent variables, are presented in Table 1, Table 2, respectively. The a paths for both FNE and FPE were significant. Social anxiety significantly predicted both FNE (b = .40, t = 10.74, p < .001; a path1) and FPE (b = .66, t = 9.36, p < .001; a path2).

Drive for Thinness

When controlling for BMI, which was a significant predictor of drive for thinness (b = .22, t = 2.72, p = .007), regression analyses revealed that

Discussion

The aim of this study was to examine FNE and FPE simultaneously, as unique explanatory variables of the associations between social anxiety and drive for thinness, body dissatisfaction, and bulimic symptoms. Whereas prior research has shown that FNE may serve as an important link between social anxiety and various domains of eating pathology, we hypothesized that this relationship could be better accounted for by adding FPE into the model. Enhanced knowledge pertaining to the putative role of

References (56)

  • R.P. Mattick et al.

    Development and validation if measures of social phobia scrutiny fear and social interaction anxiety

    Behaviour Research and Therapy

    (1998)
  • F.R. Schneier et al.

    The social anxiety spectrum

    Psychiatric Clinics of North America

    (2002)
  • E. Stice et al.

    Predicting onset and cessation of bulimic behaviors during adolescence: A longitudinal grouping analysis

    Behavior Therapy

    (1998)
  • A.C. Utschig et al.

    An investigation of the relationship between fear of negative evaluation and bulimic psychopathology

    Eating Behaviors

    (2010)
  • J.W. Weeks et al.

    The Fear of Positive Evaluation Scale: Assessing a proposed cognitive component of social anxiety

    Journal of Anxiety Disorders

    (2008)
  • J.W. Weeks et al.

    Exploring the relationship between fear of positive evaluation and social anxiety

    Journal of Anxiety Disorders

    (2008)
  • A.M. Bardone-Cone et al.

    Anxiety, appearance contingent self-worth, and appearance conversations with friends in relation to disordered eating: Examining moderator models

    Cognitive Therapy and Research

    (2013)
  • S. Bonn et al.

    How valid are Web-based self-reports of weight?

    Journal of Medical Internet Research

    (2013)
  • T.D. Brewerton et al.

    Comorbidity of axis I psychiatric disorders in bulimia nervosa

    Journal of Clinical Psychiatry

    (1995)
  • E.J. Brown et al.

    Validation of the Social Interaction Anxiety Scale and the Social Phobia Scale across the anxiety disorders

    Psychological Assessment

    (1997)
  • C.M. Bulik et al.

    An analysis of social anxiety in anorexic, bulimic, social phobia, and control women

    Journal of Psychopathology and Behavioral Assessment

    (1991)
  • C. Bulik et al.

    Eating disorders and antecedent anxiety disorders: A controlled study

    Acta Psychiatrica Scandinavica

    (1997)
  • M.M. Carter et al.

    Comparing fear of positive evaluation to fear of negative evaluation in predicting anxiety from a social challenge

    Journal of Experimental Psychopathology

    (2012)
  • L.W. Craighead et al.

    Eating disorders: Bulimia nervosa and binge eating

  • L.B. DeBoer et al.

    Associations between fear of negative evaluation and eating pathology during intervention and 12-month follow-up

    Cognitive Therapy and Research

    (2013)
  • L.B. DeBoer et al.

    Anxiety and disordered eating

    Cognitive Therapy and Research

    (2013)
  • I. Duemm et al.

    The addition of sociotropy to the dual pathway model of bulimia

    Canadian Journal of Behavioural Science

    (2003)
  • D.E. Farrar et al.

    Multicollinearity in regression analysis: The problem revisited

    The Review of Economics and Statistics

    (1967)
  • Cited by (33)

    • Bi-directional associations between depressive symptoms and eating disorder symptoms in early adolescence

      2022, Body Image
      Citation Excerpt :

      Adolescents who experience depressive symptoms may have higher levels of social anxiety and consequently be more prone to experiencing body-related ED symptoms. This is supported by research that has found social anxiety to predict shape and weight dissatisfaction (Pawijit, Likhitsuwan, Ludington, & Pisitsungkagarn, 2017) and drive for thinness (Menatti, DeBoer, Weeks, & Heimberg, 2015). Given an association between depression and social anxiety is well-established (Belmans, Bastin, Raes, & Bijttebier, 2019; de Jong, Sportel, de Hullu, & Nauta, 2012), and that social anxiety is identified as a significant predictor of ED symptoms (Kerr-Gaffney, Harrison, & Tchanturia, 2018), future research should explore how social anxiety influences the longitudinal relationship between depression and ED symptoms in early adolescence.

    • Fear of positive evaluation and social anxiety: A systematic review of trait-based findings

      2020, Journal of Affective Disorders
      Citation Excerpt :

      In general, the 18 studies (a few produced latent and bivariate relations) reporting bivariate correlations between self-reported FPE and FNE were slightly smaller than studies reporting latent correlations. In the community samples of adult participants, effect sizes for bivariate relations varied from the small, (Kortink et al., 2018; Wang et al., 2012), moderate (Levinson & Rodebaugh, 2012; Weeks et al., 2008a; Weeks et al., 2010; Weeks, 2015), and large range (Cook et al., 2019; Fetzner et al., 2016; Lipton et al., 2016; Menatti et al., 2015a; Weeks et al., 2008b; Yap et al., 2016). Overall, the typical effect size for the bivariate correlation between self-reported FPE and FNE (measured by the BFNE-S) in adult community samples appears to be between .25 and .60.

    • Appearance-based rejection sensitivity as a mediator of the relationship between symptoms of social anxiety and disordered eating cognitions and behaviors

      2017, Eating Behaviors
      Citation Excerpt :

      Due to consistent findings, maladaptive emotion regulation strategies and negative self-evaluative fears are plausible mechanisms driving this link. However, as Menatti et al. (2015) noted, more research is needed to pinpoint additional explanatory variables of the social anxiety-disordered eating relationship. They suggested this so that we can have a greater understanding of the factors implicated in this relationship, particularly since social anxiety and disordered eating are such broad constructs that encompass a variety of cognitive, behavioral, and emotional characteristics.

    View all citing articles on Scopus
    View full text