Elsevier

Appetite

Volume 68, 1 September 2013, Pages 98-104
Appetite

Research report
Get your own mirror. Investigating how strict eating disordered women are in judging the bodies of other eating disordered women

https://doi.org/10.1016/j.appet.2013.04.015Get rights and content

Highlights

  • We investigate how women rate their own and other women’s attractiveness.

  • Women high and low in eating disorder symptoms (“high/low symptomatics”) are compared.

  • Photographs of the own body and of other bodies are used for attractiveness ratings.

  • High symptomatics rate other high symptomatics as equally (un)attractive but thinner.

  • Low symptomatics rate other low symptomatics as less attractive but equally thin.

Abstract

Previous research has shown that eating disordered women lack a self-serving body image bias and largely make self-defeating social comparisons. These factors influence how eating disordered women feel about their bodies, and may also influence treatment for disordered eating. In group mirror exposure therapy, women inevitably compare their own bodies to other women’s bodies. Yet, how eating disordered women view their own bodies in relation to other women’s bodies has never been investigated. This study investigated how subjects high in eating disorder symptoms (“high symptomatics”) view the bodies of other women and of other high symptomatics specifically. Twelve high symptomatics and 13 low symptomatics viewed photos of, and rated the attractiveness of, their own and other participants’ bodies. The results show that low symptomatics rated both other women’s bodies and other low symptomatic bodies specifically as less attractive, but not as fatter, than their own bodies. In contrast, high symptomatics rated both other women’s bodies and other high symptomatic bodies specifically as equally (un)attractive, but as thinner than their own bodies. These results suggest that high symptomatics lack a self-serving body image bias when it comes to aspects of weight specifically. Considering weight is a self-relevant dimension to eating disordered women, the impact of these self-defeating comparisons may be especially negative and may impede treatment progress.

Introduction

For women with an eating disorder, looking into the mirror can be damaging, as these women are likely to scrutinise their body and focus on their disliked body parts (e.g., Fairburn, 2008, Jansen et al., 2005). Yet a mirror can also be a useful tool for the treatment of eating disorder patients. During ‘mirror (or body) exposure,’ the individual stands in front of a mirror, for an extended period of time, possibly over several sessions. Different variants of mirror exposure exist, including: asking the individual to describe his or her body in a neutral way (e.g., Delinsky and Wilson, 2006, Hilbert et al., 2002, Jansen et al., 2008, Moreno-Dominguez et al., 2012) which may de-emphasise negative evaluations; asking the individual to describe his or her body in a positive way (e.g., Luethcke et al., 2011, Stice and Presnell, 2007) which may instill new, positive evaluations; or exposing the individual without guidance at all (‘pure exposure’) which may work via the traditional extinction route (e.g., Moreno-Dominguez et al., 2012).

To date, it appears that mirror exposure can be beneficial in treating eating disorder patients. However, it is important to recognise that there may be risks associated with mirror exposure. This risk is due in part to the peculiar way that eating disorder patients see their own and others’ bodies. In a study of Jansen, Smeets, Martijn, and Nederkoorn (2006) a panel of independent judges rated the attractiveness of eating disordered and control participants. Despite the fact that control participants and eating disordered participants did not differ in body mass index (BMI), the eating disordered participants received lower attractiveness ratings from the panel. Moreover, these ratings also matched their own ratings. On the contrary, control participants received somewhat higher ratings from the panel, but these ratings were lower than the control participants’ own ratings. Further, a second panel rated the most attractive and most unattractive body parts of the participants (Jansen et al., 2006). The results showed that there was larger agreement between the panel-identified and self-identified most unattractive body parts in the eating disordered participants compared to the agreement between the panel and the control participants. The researchers concluded that eating disordered individuals are more realistic and lack a “self-serving body image bias” that may buffer against body dissatisfaction and disordered eating. Similar to depressed patients (e.g., Brewin, 1993) they may also be “sadder but wiser,” specifically in terms of their body image. In essence, the control subjects were the ones with the perceptual distortion, albeit a healthy, protective one.

Eating disordered individuals may also see others women’s bodies differently than healthy women do. Jansen et al. (2005) exposed eating disordered and control participants to pictures of their own bodies and to pictures of other women’s bodies. Visual attention was measured using eye movement registration. The results showed that eating disordered participants attended more to their self-identified unattractive body parts whereas control participants attended more to their self-identified attractive body parts. In contrast, when looking at other bodies, eating disordered participants attended more to the attractive body parts whereas control participants attended more to the unattractive body parts of other bodies. These findings also indicate the presence of the self-serving body image bias in control participants, and the lack thereof in eating disordered participants. Furthermore, inducing a negative bias (focusing on the self-defined unattractive body parts) in healthy women leads to a decrease in body satisfaction (Smeets, Jansen, & Roefs, 2011). Thus, the tendency of eating disordered participants to focus on their unattractive body parts may be causal to body dissatisfaction (Smeets et al., 2011).

Finally, it is important to note that, in clinical practice, mirror exposure is often conducted in group format. In these sessions, one patient stands in front of the mirror and describes his or her body and the other patients are invited to respond to his or her descriptions. Although mirror exposure, and group therapy in general, is aimed at improving individual well-being, group mirror exposure inevitably introduces social influences. The social comparison theory may shed light on how these social factors may affect eating disordered individuals’ experience of therapy. The social comparison theory posits that individuals compare themselves to others in order to gauge their standing in a particular aspect of the self (Festinger, 1954). Individuals may make social comparisons on various dimensions (e.g., appearance) and with various ‘targets’ (e.g., one’s friends). Moreover, comparisons may be in the upward direction (i.e., with a target who perceivably fares better on the particular dimension) or in the downward direction (i.e., with a target who perceivably fares worse on the particular dimension). Upward comparisons (e.g., “I’m less beautiful than her”) may lead to feelings of distress or lower self-worth, whereas downward comparisons (e.g., “I’m more beautiful than her”) may lead to feelings of increased self-worth and satisfaction (Festinger, 1954). Also, social comparisons made on self-relevant dimensions and with similar others have a greater impact on one’s self-worth than those made on irrelevant dimensions and with dissimilar others (Wood, 1989).

To date, research has shown that social comparisons play an important role in eating disorders and in body dissatisfaction (one of the main risk factors for developing an eating disorder; Stice, 2002). For instance, Morrison, Kalin, and Morrison (2004) have demonstrated that universalistic social comparisons predicted appearance self-esteem, body dissatisfaction, number of diets to lose weight and engagement in weight control practices. Also, Corning, Krumm, and Smitham (2006) have shown that women high in eating disorder symptoms make social comparisons more often than asymptomatic women do. Moreover, the tendency to make self-defeating social comparisons was shown to predict eating disorder symptoms in this sample. Regarding body dissatisfaction, a meta-analysis by Myers and Crowther (2009) has shown that social comparison making is related to higher levels of body dissatisfaction.

As aforementioned, social comparison theory may elucidate social processes that affect eating disordered individuals during mirror exposure. Group mirror exposure creates a social setting where social comparisons are unavoidable. Eating disordered women tend to make self-defeating social comparisons with other women (Corning et al., 2006), which they will likely do in the mirror exposure setting. Relatedly, the aforesaid experimental findings suggest that eating disordered individuals focus on the beautiful body parts of other people (Jansen et al., 2005) and lack a self-serving body image bias—both factors which will also affect social comparisons made in group therapy. In sum, as a result of these influences, eating disordered women may feel fatter or less attractive than their fellow peers, which may induce increased distress and ultimately impede the therapeutic process. Importantly, the social comparisons made in mirror exposure are likely to be especially impactful, as they target a dimension that is highly relevant to eating disordered individuals (weight) and are made with individuals who are similar to them (other eating disordered individuals).

Surprisingly, despite the potential impact of the way eating disordered individuals see and make comparisons with other women, how eating disordered individuals judge the bodies of other women (and of other eating disordered women specifically) has never been tested. Therefore, the aim of the current experiment is to investigate whether or not eating disordered participants evaluate the bodies of other participants more positively than their own bodies and whether their strict judgment is limited to their own body. To do this, female participants high in eating disorder symptoms (“high symptomatics”) rated the attractiveness of other eating disordered and control bodies. Their ratings were compared with ratings made by female participants that were low in eating disorder symptoms (“low symptomatics”). It was expected that high symptomatics would see all other bodies as more attractive than their own (hypothesis 1a), reflecting the ultimate lack of a self-serving body image bias and the tendency to make self-defeating comparisons with other women. In contrast, it was expected that low symptomatics would see all other bodies as less attractive than their own (hypothesis 1b), reflecting both their tendency to attend to the unattractive aspects of other bodies and the presence of a self-serving body image bias. It was further expected that high symptomatics would also judge the bodies of other high symptomatics as more attractive than their own (again, reflecting the ultimate lack of a self-serving bias and the tendency to selectively focus on the attractive aspects of other bodies; hypothesis 2a). It is vital to investigate this comparison specifically as it is directly relevant to group mirror exposure and clinical settings in general. Contrarily, it was hypothesised that low symptomatics would rate their own bodies as significantly more attractive than the bodies of other low symptomatics (hypothesis 2b).

Finally, in order to assess how participants think other people judge their bodies, compared to their self-assessments, participants were also asked to estimate how an ‘average Dutch male’ would rate their attractiveness, as well as the attractiveness of the other bodies. In line with the hypotheses above, it was expected that high symptomatics would expect men to rate other bodies as more attractive than their bodies (demonstrating the lack of a self-serving bias; hypothesis 3a), whereas low symptomatics would expect men to rate other bodies as less attractive than their bodies (demonstrating the ultimate self-serving bias; hypothesis 3b).

Section snippets

Participants

Participants were invited to take part in a study that was said to be about the relation between character and perception. It was made clear that participants should be willing to be photographed in underclothing. Volunteers were selected based on the highest (high symptomatics, n = 13) and lowest levels (low symptomatics, n = 13) of eating disorder symptomatology, as measured by the EDE-Q (see Assessment). One participant dropped out due to pregnancy. There were thus 25 participants in total, 12

Participant characteristics

There were no group differences in age, body mass index (BMI) or waist-to-hip ratio (WHR) at test day between the high and low symptomatic participants, ts varied between 0.44 and 2.45, ns (see Table 1). To check for group differences in eating disorder pathology, a MANOVA was conducted with Group as independent variable and the subscale and total scores of the EDE-Q as dependent variables. The effect of Group was significant, F(4, 20) = 16.89, p < 0.001, ηp2=0.77, such that the high symptomatic

Discussion

The aim of the current experiment is to investigate whether or not high symptomatics evaluate other bodies more positively than their own bodies and whether their strict judgment is limited to their own body. It is important to understand the ways in which high symptomatic individuals evaluate their own and other bodies, as previous studies have shown that the way they view their own bodies in relation to other women’s bodies may be related to disordered eating and body dissatisfaction.

References (21)

There are more references available in the full text version of this article.

Cited by (0)

Acknowledgement: The contribution of Alleva was supported by NWO grant 404-10-118: Novel strategies to enhance body satisfaction.

View full text