One central disturbance across eating disorders is that eating behavior is abnormal in some way. To understand the causes and characteristics of eating disorders, researchers, at Columbia and elsewhere, have conducted experiments to probe aspects of eating, and often these experiments rely upon pictures (images) of food. For example, by asking participants to rate how much they like, or want to eat, a set of photographed foods, we can learn about rewarding properties of different foods for groups of people with eating disorders, as compared with individuals without eating disorders (healthy volunteers). This type of study is informative because how rewarding food is could affect what, and how much, is eaten. Other studies have measured attention to food images, or brain activity while food images are viewed. These studies aim to understand whether people with eating disorders engage certain cognitive processes when faced with food, which could impact eating.
Unfortunately, to date, eating disorders research involving food images has yielded some inconsistent findings. For instance, some have found increased attention to food stimuli among individuals with anorexia nervosa relative to healthy volunteers, while others report less attention in anorexia nervosa, and others find no differences at all between groups. Brain imaging studies tend to report differences in brain activity between those with eating disorders and healthy volunteers, when viewing food images. However, the location and nature (whether activity is increased or decreased) of these differences varies substantially across studies.
To have confidence that a given study finding is true (e.g. greater attention to food in anorexia nervosa), it needs to be observed across multiple other studies (an important concept in the scientific process referred to as replication). The current lack of replication means that it remains unclear whether and how groups differ in their responses to food, and, by extension, which processes could explain eating disorder symptoms.
One potential cause of the discrepant findings in food image research is the use of different food images across studies. Images vary widely in a number of ways, such as:
- how healthy a depicted food is considered to be
- how filling the food is estimated to be
- how colorful the image is
These factors can influence responses such as liking (ratings), and attention, and unless these factors are the same across different studies, it is likely studies will continue to generate varying results.
In our research over the years, we have created and relied upon a standardized food choice task. We have used this task in a range of studies and groups, including those with anorexia nervosa and bulimia nervosa. The set of food images we developed for our research, called Food Folio, is now publicly available and can be found here (with two sample photos pictured on the left here). Food Folio images have been rated by healthy controls on a range of dimensions (e.g., healthiness, tastiness, how filling the depicted foods are), and analyzed in terms of nutrient content and physical image properties (e.g. how light the food appears against the background). This information accompanies the image set, promoting improved matching of image characteristics across studies. Our hope is that by sharing the Food Folio resource with others, we can enhance efforts towards potentially replicable findings, to in turn get closer to understanding mechanisms of illness.
The sharing of Food Folio forms part of a broader effort to engage in open science practices. Open science is a movement encouraging the sharing of research materials for free, to promote collaboration amongst different research groups, and enhance replication. We are proud to be involved, and are always reviewing the ways in which we can share our resources and standardize our methods, to improve finding replication, and help further understanding of eating disorders.
© The Feed, 2013-present. Unauthorized use and/or duplication of this material without express and written permission from this blog’s authors is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to the article’s author and The Feed with appropriate and specific direction to the original content.
