86. What Is Beauty? A Crowdsourced Evaluation of Facial Attractiveness

Abstract

Purpose: Opinions of aesthetic outcomes can vary greatly amongst plastic surgeons. Our specialty creates changes for patients. Often, change can be mistaken for improvement. We are deemed experts of attractiveness, and thus it is not acceptable to have this degree of uncertainty on what is ideal. The authors hope to quantify cosmetic results to determine if postoperative changes are indeed better or worse than preoperatively. Evolutionary psychologists have shown that humans are koinophiles that seek out mates who possess average anatomy for the population. We particularly adore cognitive averages, meaning we subconsciously generate composites of the most recent attractive faces seen. Our study attempts to identify an “ideal” through crowdsourced ratings. We hypothesized a composite of the cohort would be most attractive and investigated which features correlated with attractiveness. Methods: Our male and female cohorts each consisted of 41 standardized frontal-view monochrome photos, with one composite image derived from the other forty. We measured 46 facial features per image, calibrating measurements with facial width-to-height ratio and interpupillary distance. We utilized Amazon Mechanical Turk, a widely used crowdsourcing platform, to receive ratings of images ranging from 1-7, with 1 and 7 being least and most attractive, respectively.Statistical AnalysisAnalysis of variance tests were performed for ratings and facial feature measurements. A Spearman correlation test calculated correlation coefficients between facial measurements and ratings. Lastly, we constructed heatmaps from facial measurements and scores. Results: 2064 evaluations revealed composite male and female scores (means were 5.31 and 5.60, respectively) were the highest of their cohorts and significantly greater (both p<0.0001) than scores of the other 40 photos (means were 4.06 and 4.00, respectively). Average ear width (r=1, p=0.0004) for males and average ear height, medial eye distance, and iris-to-aperture distance (all r=1, p=0.0004) for females had the highest correlations with attractiveness. Conclusion: Our study reports that average facial anatomy and ear features (heavily linked with chin-jaw and eye angles) are highly correlated with attractiveness. This is an initial step into the field of objective cosmetic quantifications and integrating evidence-based medicine into aesthetic surgery.