Date Published: February 9, 2018
Publisher: Public Library of Science
Author(s): Anne Thaler, Michael N. Geuss, Simone C. Mölbert, Katrin E. Giel, Stephan Streuber, Javier Romero, Michael J. Black, Betty J. Mohler, Cosimo Urgesi.
Previous literature suggests that a disturbed ability to accurately identify own body size may contribute to overweight. Here, we investigated the influence of personal body size, indexed by body mass index (BMI), on body size estimation in a non-clinical population of females varying in BMI. We attempted to disentangle general biases in body size estimates and attitudinal influences by manipulating whether participants believed the body stimuli (personalized avatars with realistic weight variations) represented their own body or that of another person. Our results show that the accuracy of own body size estimation is predicted by personal BMI, such that participants with lower BMI underestimated their body size and participants with higher BMI overestimated their body size. Further, participants with higher BMI were less likely to notice the same percentage of weight gain than participants with lower BMI. Importantly, these results were only apparent when participants were judging a virtual body that was their own identity (Experiment 1), but not when they estimated the size of a body with another identity and the same underlying body shape (Experiment 2a). The different influences of BMI on accuracy of body size estimation and sensitivity to weight change for self and other identity suggests that effects of BMI on visual body size estimation are self-specific and not generalizable to other bodies.
In the last decades, the mean body mass index (BMI) has significantly increased worldwide. Between 1975 and 2014, obesity prevalence has more than doubled, resulting in 10.8% of men and 14.9% of women being obese . According to the international classification of body weight , obesity is defined as a BMI of 30 and above, overweight as a BMI between 25 and 29.9, normal weight as a BMI between 18.5 and 24.9, and underweight as a BMI below 18.5. Excess body weight is associated with a variety of diseases, such as cardiovascular diseases and diabetes, and can reduce life expectancy by up to eight years . In line with the Health Belief Model , it has been suggested that a disturbed ability to accurately identify own weight and/or changes in own weight may contribute to overweight [5–7]. The goal of the current research was to test whether one’s personal body size predicts the accuracy of estimating own body size, the sensitivity to weight changes, and the desired body weight. By comparing body size estimation (BSE) of self and another identity, we additionally addressed whether distortions in BSE were due to a general bias in estimating body size, or were self-specific, and further whether they could be explained by cognitive-affective factors.
The current set of experiments investigated the relationship between personal body size, indexed by BMI, and the accuracy of BSE, the sensitivity to weight changes, and the ideal body weight of self and other in a non-clinical population of female individuals varying in BMI. Experiment 1 tested whether one’s personal body size predicts BSE of a body with own identity. A depictive BSE task was used to assess an explicit representation of one’s body. Following Moelbert et al. , depictive BSE tasks mainly recruit explicit representations, namely of how participants think their body looks like in terms of body weight. In two well-established psychophysical experiments, participants were presented with their personalized avatars varying in weight in a virtual environment and responded whether the body presented corresponded to their actual body size (one-interval forced choice paradigm with two response possibilities; 1AFC) and adjusted the avatar until it matched the size they perceived their actual body to be, as well as the size of their ideal (desired) body (Method of Adjustment; MoA). Additionally, to test whether distortions in BSE could be explained by cognitive-affective factors, measurements of self-esteem, eating behaviour, and attitudes towards own body shape and weight were obtained using several validated questionnaires.
Experiment 1 investigated whether personal BMI predicts the accuracy of BSE as predicted by a contraction bias [26,31], and/or influences the sensitivity to positive and negative changes in BMI relative to participants’ estimated own body size. For each participant, personalized 3D bodies (avatars) with varying body weight were created. In the 1AFC task, participants judged whether the presented avatar (-20% to +20% of their actual BMI, at 5% intervals) corresponded to their actual body size or not by answering to the question ‘Is that your body? Yes/No’ (for example data, see Fig 1). The avatar with the highest proportion of yes-answers (peak) reflects each participant’s estimated body size. The percentage change of yes-answers indicates how sensitive participants were to positive and negative changes in BMI relative to their estimated body size. In the MoA task, participants adjusted the size of their personalized avatar until it matched the size of their actual body and their ideal body. The MoA task was employed in addition to the 1AFC task for two reasons. First, a smaller step size in the MoA task allowed a more precise measure of BSE than the peak in the 1AFC task. Second, the MoA task additionally provided a measure of each participant’s ideal body size and body dissatisfaction as represented by the discrepancy between estimated and ideal body size. The discrepancy between estimated and actual physical body size in both the 1AFC and MoA task was used as a measure of distortion of BSE. Further, several questionnaires were administered to assess cognitive-affective factors related to own body size to additionally examine whether those factors could explain distortions in BSE.
Experiment 2a aimed at investigating whether the effect of participants’ BMI on the results of the psychophysical experiments in Experiment 1 were due to estimating the size of one’s own body or generalize to a body with another identity. The same psychophysical methods and sets of avatars were used as in Experiment 1, with one important exception: participants viewed the set of avatars with their own body shape, but the texture was altered such that the avatar looked like another person. The manipulation of texture was intended to target only identification with the avatar while keeping body shape constant across the experiments. Previous studies have shown that swapping the texture of the body from self to other significantly reduces identification with the avatars [40,46]. Further, by maintaining the body shape information, which is a strong cue used for estimating body weight, we controlled for factors such as individual body shape characteristics and visual experience with the body shape, that would have otherwise been confounding variables when comparing the results of BSE of self (Experiment 1) and other (Experiment 2a). If BMI predicts the accuracy of BSE and the sensitivity to weight changes due to factors associated with own identity, then we expect that participants’ BMI does not predict the size estimation of a body with another identity. However, if the results of Experiment 1 were driven by general biases in BSE, then the same pattern should extend to size estimates of a body with another identity.
Experiment 2b investigated whether the difference in results in Experiment 1 and 2a might be due to judging the body size of an avatar with own identity, or due to the exposure phase that was introduced before the psychophysical tasks in Experiment 2a. The same design was used as in Experiment 2a, with the exception that participants saw their own personalized avatar (own shape, own photo-realistic texture) in the exposure phase and were made explicitly aware that this was their body. Participants judged whether the presented bodies corresponded to the memorized body’s size rather than to their physical body size.
Previous literature suggests that a disturbed ability to accurately identify own weight and/or changes in own weight may contribute to overweight [5–7] and further, distortions in estimating own body size might be due to biases that generalize to others’ bodies [26,27]. The current research aimed at investigating whether own body size predicts the accuracy of BSE and the sensitivity to weight changes using a depictive body size estimation task assessing an explicit representation of one’s body. Further, by manipulating the identity of the judged bodies (self vs. other), we systematically investigated whether distortions in BSE were due to general biases in estimating body sizes or were self-specific, and whether they could be explained by cognitive-affective factors as assessed by several validated questionnaires. Our results show that the accuracy of own BSE was indeed predicted by personal BMI in Experiment 1, such that low BMI individuals underestimated their body size and high BMI individuals overestimated their body size. Further, with increasing BMI, participants were less sensitive to weight changes of bodies bigger than their estimated own body size. Crucially, this was only the case when participants estimated the size of a body they identified with (Experiment 1 and 2b), not however when they estimated the size of a memorized body with another identity (Experiment 2a). This suggests, that the effect is most likely due to conceptual factors linked to own identity or emotional self-concept, such as cognitive-affective factors, rather than due to a general bias in estimating body size.
There were three main results reported in the current set of studies. First, the accuracy of BSE is influenced by personal BMI. Participants with higher BMI overestimated their body size, while participants with lower BMI underestimated their body size. Second, sensitivity to changes in body size was predicted by participants’ BMI for bodies bigger than their estimated own body size. Participants with a higher BMI were more willing to accept an even heavier body as their own than participants with a lower BMI. Third, these effects were only apparent when participants estimated the size of a body with their own identity, not when they estimated the size of a body with another identity but the same underlying body shape. The different influence of BMI on BSE for self and other identity suggests that the effect of own BMI on BSE is self-specific and not generalizable to other bodies. Given the increase in BMI worldwide, it is crucial to better understand how own body size can influence the estimation of own body size and the ability to detect weight gain. Training programs using for example avatars matched for certain bodily dimensions in virtual environments could improve the ability to recognize own weight and weight gain.