Devine, S., Germain, N., Ehrlich, S. & Eppinger, B. (in press in Psychological Science). Changes in the prevalence of thin bodies biases young women's judgements about body size
- The task code
- The raw data and the analysis file(s)
- Materials for our presentation at the Canadian Society for Brain, Behaviour and Cognitive Science (CSBBCS) annual meeting, June 2021.
Looking for the preregistration?
Want to run the task just for fun?
Body image researchers agree that an overly slender female body is internalized by young women as the most suitable shape (the thin ideal), via repeated exposure to media imagery but the cognitive process (or processes!) by which this internalization and normalization of thinness occurs is still uncertain. We asked around 400 young women to repeatedly judge whether or not a randomly computer-generated female body was overweight. By presenting thin bodies more often over time, we found that women’s conceptualization of thinness shifted in accordance with the prevalence of thin bodies. Put simply, a body of average weight was more likely to be judged as overweight when it was surrounded by thin bodies. Conversely, when the ratio of thin-to-overweight bodies were balanced, no such shift in concept occurred. These results suggest that women’s judgements of other women’s bodies are biased by an overrepresentation of thinness.
Among the many forces that contribute to body dissatisfaction, the overrepresentation of thin bodies in visual media has received much of the blame, but the cognitive mechanism(s) underlying this internalization of the thin ideal are still not well understood. In this study, we propose that prevalence-induced concept change may be one of the cognitive mechanisms that explains how beauty standards shift in Western society. Prevalence-induced concept change is defined as a process where we expand our concepts as the examples of that concept appear less often.
Our first hypothesis attempted to directly test this within the context of body image, so we expected that when the prevalence of thin bodies in the environment increased, the concept of overweight would expand to include bodies that would otherwise be judged as normal in size.
We also hypothesized that women’s judgements about their own bodies would be affected by this change in prevalence.
We conducted a pre-registered online experiment with young women (N = 419), where participants were assigned to one of two conditions. In the increasing prevalence condition (marked throughout with the color red), the presentation of thin bodies increased over the task, and in the stable prevalence condition (denoted with the color blue), the proportion of large to thin bodies was always 50/50. For each trial, participants saw a body for half a second, and were asked to judge that body as overweight or not. Bodies were taken from a validated set of images, and the BMIs they represent ranged from emaciated to morbidly obese. Participants made a total of 800 judgments overall, and they also were asked to indicate the body that best matched their own weight and whether or not they would judge that body as overweight. If you’d like to learn more about how the task works or what the stimuli look like, I invite you to check out the pre-registration or the provided code repository.
To analyze our main hypothesis, we fit a logistic multilevel model to the data, where the main outcome variable was whether or not a participant judged a body as overweight or not on a given trial. The variables within the model included the trial number (how far along participants were in the task), the body’s size, and the condition (either increasing or stable). Most importantly, we found a three-way interaction between Condition, Trial, and Size, which means that when the prevalence of thin bodies in the environment increased, participants judged more ambiguous bodies (average bodies) as overweight than when the prevalence remained fixed.
As for our second hypothesis, we did not find a significant effect of condition on participants’ categorical judgements of the self. Lastly, for our third hypothesis, we found a small, but statistically significant, effect of condition on participants’ self-image but this effect was in the opposite direction than predicted: when the prevalence of thin bodies increased, participants judged themselves as thinner than they did at the beginning of the task and compared to when the prevalence remained stable. So while our two exploratory hypotheses aren’t supported, we noted a presence of individual sensitivities to concept change relating to judgements about women’s bodies and their own. This leaves us a path for future research, where we could try to account for individual sensitivity and resilience to prevalence-induced concept change. We’d also like to try this out with adolescents—who are particularly sensitive to both body image and concept change—with men, and with clinical samples.
Overall, our results provide novel empirical evidence for the degree to which visual media can manipulate women’s perception of what constitutes a normal body in a relatively short amount of time! On a more holistic level, these results also speak to why we should strive to portray a more honest distribution of women’s bodies in our media.