Avatars That Look Like You and Healthy Behaviors

By Jorge Ruiz, MD; and Allen D. Andrade, MD

Jorge Ruiz, MD is director of the Laboratory of E-learning and Multimedia Research and associate director for Education and Evaluation. Allen D. Andrade, MD is associate director of the Laboratory of E-learning and Multimedia Research and GRECC Investigator and both are at the Geriatric Research, Education, and Clinical Center (GRECC), Bruce W. Carter Veterans Affairs Medical Center, Miami, Florida.

Can an avatar that resembles you influence your behavior? (Click here for the link to the article by Dr. Ruiz and Dr. Andrade: http://www.fedprac.com/Article.aspx?ArticleId=hrHiCg0q/fw=.) There is growing evidence suggesting just that. Manipulating an avatar’s appearance to resemble an individual is now possible. Jeremy Bailenson at Stanford University found that college students reported increases in exercise 24 hours after viewing an avatar resembling them exercising. We investigated whether these findings translate into physical activity gains in sedentary, overweight veterans. We objectively assessed veterans’ levels of physical activity with accelerometers. We randomly assigned them into 1 of 3 groups: viewing an avatar that looked like them exercising, another viewing an avatar of another person exercising, and a third to watch a static figure exercising. Veterans who viewed an avatar resembling themselves exercising significantly improved their physical activity 1 week after the intervention. (http://www.ncbi.nlm.nih.gov/pubmed/22357026) 

What’s the explanation? Psychological theories as well as neuroscientific evidence underpin these findings. According to social cognitive theory, individuals may acquire a behavior through observation of another’s performance of the behavior (modeling). Modeling the behavior may influence the individual’s cognitive development and build self-efficacy. Importantly, responses to a model are dependent on its attractiveness, believability, and credibility. A self-model resembling the individual performing the behavior and experiencing its consequences would positively influence that behavior. This influence could work at the conscious level, but it is most likely unconscious. In the 2 avatar studies, the subjects did not recognize themselves. The dual-coding model of health behavior suggests that in addition to conscious processes, unconscious processes may influence behaviors. It seems that avatars can subliminally influence the viewer’s mind.  

Can we take advantage of these findings to promote behavior change? We certainly need more research that includes more patients and other behaviors (overeating, smoking, etc.). These findings open up exciting possibilities for the use of avatars to foster healthy behaviors.