Effects of Variety of Dynamic Facial Expressions on Attractiveness Judgments with Successive Viewing on Digital Media

It is widely known in laboratory experiments and everyday life that smiling increases attractiveness. However, negative facial expressions, such as angry faces, may also act as an indicator of honesty in certain situations and may increase facial attractiveness. While facial emotions are now relatively easy to simulate in various digital visual media, how facial expressions lead to the attractiveness of a person (or an avatar) remains largely unclear. In the present study, we examined whether judgments of facial attractiveness would depend solely on the positivity/negativity and frequency of expressions. Participants rated the attractiveness of faces with neutral facial expressions in static images before and after observing the dynamic facial expressions. One-third of the faces appeared twice as a smiling face (Two Smile condition), another third appeared twice as an angry face (Two Angry condition), and the remaining third appeared once as a smiling face and once as an angry (One Each condition). The results showed that the attractiveness ratings of the static face images were higher in the Two Smile and One Each conditions than in the Two Angry condition, but there was no significant difference between the Two Smile and One Each conditions. An additional analysis showed that changes in attractiveness were significantly greater when the participant and the face being rated were different genders. These results are partially consistent with the hypothesis that facial expressiveness, or the variety of facial expressions, may act as an honest signal and increase facial attractiveness. Our findings may also have implications for producing more attractive characters in avatars and robots.