Facial attractiveness: Is the whole more than the sum of its parts?

Abstract

Background: In order for aesthetic surgery to participate in the movement toward evidence-based medicine, a system that quantifies aesthetic outcomes must be developed based on reliable and reproducible data. Previous studies have investigated the attractiveness of averaged (or composite) faces. Data has shown that these faces are deemed more attractive than their constituent faces. Objectives: The authors seek to elucidate the mechanism behind the phenomenon in which more attractive faces included in a composite increase the overall attractiveness rating. Methods: Ten composites, five male and five female, were generated with resources provided by the www.faceresearch.org research group. The five composites contained the top two, three, and five most attractive faces, a composite of all faces, and the five least attractive faces. An online survey was conducted in which participants were asked to rank each of the five faces in order of attractiveness. Results: Analysis of variance (ANOVA) on ranks was performed on the data provided by the 245 participants. Attractiveness increased from the top two to the top three to the top five composite faces, and then decreased from the top five to all faces to the bottom five composite faces. This trend was present and statistically significant (with a P value