Looking good. BMI, attractiveness bias and visual attention

Abstract

The aim of this study was to study attentional bias when viewing one's own and a control body, and to relate this bias to body-weight and attractiveness ratings. Participants were 51 normal-weight female students with an unrestrained eating style. They were successively shown pictures of their own and a control body for 30 s each, while their eye movements (overt attention) were being measured. Afterwards, participants were asked to identify the most attractive and most unattractive body part of both their own and a control body. The results show that with increasing BMI and where an individual has given a relatively low rating of attractiveness to their own body, participants attended relatively more to their self-identified most unattractive body part and the control body's most attractive body part. This increasingly negative bias in visual attention for bodies may maintain and/or exacerbate body dissatisfaction.

Introduction

Body dissatisfaction has been shown to play an important role in eating disorders (Cash & Deagle, 1997; Stice & Shaw, 2002). It can be defined as a “negative subjective evaluation of one's physical body, such as figure, weight, stomach, and hips” (Stice & Shaw, 2002, p. 985). It has been proposed (Cash & Deagle, 1997; Williamson, Muller, Reas, & Thaw, 1999) that a negative body image is best “conceptualized as a complex form of a cognitive bias” (Williamson et al., 1999, p. 568). One domain of cognitive bias is attention, and a frequently used paradigm to study this bias is the modified Stroop task. On a body Stroop task, eating disorder patients consistently show an increased interference effect (Dobson & Dozois, 2004). However, it is important to note that the Stroop effect does not necessarily reflect an attentional bias (see Williams, Mathews, & MacLeod, 1996). The current study aims to study attentional bias for the body in a more direct way and relates this bias to body-weight (body mass index, BMI) and attractiveness ratings.

Jansen, Nederkoorn, and Mulkens (2005) studied what female eating disorder symptomatic and normal participants look at when they are presented with their own or someone else's body. Each participant was shown a picture of her own body and of someone else's body for 30 s each, while eye movements were registered. Eye movements are generally considered to closely follow visual attention and to be guided by attention (Kowler, 1995; Mogg, Millar, & Bradley, 2000). When looking at their own body, eating symptomatic women attended more to their self-identified most unattractive body part than to their self-identified most attractive body part, whereas lean controls showed the opposite pattern: they attended more to their most attractive body part than to their most unattractive body part. When looking at the other body, eating symptomatic women attended more to the perceived most attractive body part than to the perceived most unattractive body part, whereas lean controls showed the opposite pattern: they attended more to the most unattractive body part than to the most attractive body part. Thus, eating symptomatic women appeared to have a dysfunctional way of looking at their own and another woman's body (see also Freeman et al., 1991; Janelle, Hausenblas, Fallon, & Gardner, 2003), which may be a causal and/or a maintenance factor in their body dissatisfaction. Note that causal status can only be determined in an experiment in which the visual attentional bias is manipulated.

In the current experiment, we build on the above findings by examining whether women with a relatively high BMI show a dysfunctional way of looking at their own and another's body, even when their BMI is within the normal range. It was found in previous studies that BMI is related to body dissatisfaction (Annis, Cash, & Hrabosky, 2004; Faith, Leone, & Allison, 1997; Schwartz & Brownell, 2004). Therefore, we hypothesize that with increasing BMI, participants attend relatively more to the perceived most unattractive part of their own body, and the perceived most attractive part of a control body. Further, we examine whether this dysfunctional way of looking is also associated with relatively low ratings of their own body.