A vast body of research suggests that tailoring health messages to a particular audience or individual can have significant persuasive effects. Tailoring enhances the recipient's sensitivity to message relevancy and leads to more thoughtful consideration of the message. Another line of research suggests that the more similar a referent, the more likely one is to assume that that their risk is similar to the referent. Digital media technology now enables a heightened level of visual tailoring in health communication campaigns. Software can digitally “morph” a subject's facial features with a spokesperson's features to create a more similar-looking spokesperson without the subject recognizing the alteration. Those who view the morphed image should report increased perceptions of similarity, identification and cancer risk compared to those who view an un-morphed image. This study tests these hypotheses in the arena of skin cancer with a sample of young adult females, a population with a rapidly growing incidence of disease and low awareness of risk.

We designed an informational health brochure about the risks of sun exposure and skin cancer. The brochure featured an image of a college-age female alongside a written narrative of her experience with skin cancer and the behavior changes that others should adopt to lower their risk of skin cancer. Half of the subjects were shown the brochure with the spokesperson's face morphed with their own face and half were shown the brochure with the un-morphed face. Independent samples t-tests and ANOVA were used to assess whether the experimental condition significantly influenced outcomes including behavioral intentions, risk perception, emotion, character empathy and identification.

Ninety-seven female undergraduates, predominantly Caucasian, participated in the study. Participants in the morphed condition reported higher levels of empathy with the spokesperson than those in the unmorphed condition. Three empathy items were significant: subjects in the morphed condition were more concerned for the spokesperson, the misfortunes of the spokesperson upset them more, and they felt more pity for the spokesperson. Subjects in the morphed condition also felt more anxious while reviewing the brochure, possibly because they were feeling more personally vulnerable to skin cancer. Although not statistically significant, those in the morphed condition reported higher levels of skin cancer risk and had higher intentions of avoiding the sun and covering their body than subjects in the control group. None of the subjects in the morphed condition recognized their own facial image in the spokesperson's image, as confirmed by a debriefing exercise.

This preliminary study is the first to suggest promising results from facial similarity manipulation for the domain of health communications and message tailoring. This study also improves upon previous research by testing the target photograph in a natural context - an informational brochure about cancer health. Although the facial similarity tailoring produced limited effects on the main health outcome of cancer risk perception, it is encouraging that the majority of measures worked in the desired direction, particularly for a manipulation that was designed as a subconscious priming effect.

Second AACR International Conference on the Science of Cancer Health Disparities— Feb 3–6, 2009; Carefree, AZ