Comparing theory-driven and data-driven attractiveness models using images of real women’s faces.

Abstract

Facial attractiveness plays a critical role in social interaction, influencing many different social outcomes. However, the factors that influence facial attractiveness judgments remain relatively poorly understood. Here, we used a sample of 594 young adult female face images to compare the performance of existing theory-driven models of facial attractiveness and a data-driven (i.e., theory-neutral) model. Our data-driven model and a theory-driven model including various traits commonly studied in facial attractiveness research (asymmetry, averageness, sexual dimorphism, body mass index, and representational sparseness) performed similarly well. By contrast, univariate theory-driven models performed relatively poorly. These results (1) highlight the utility of data driven models of facial attractiveness and (2) suggest that theorydriven research on facial attractiveness would benefit from greater adoption of multivariate approaches, rather than the univariate approaches that they currently almost exclusively employ.

Modeling individual preferences reveals that face beauty is not universally perceived across cultures

“…33 Using a data-driven approach, we overcome these constraints to model the shape and complexion face features that drive attractiveness perceptions in two cultures and face ethnicities. Contrary to existing accounts, we found that attractive face features form a space that is distinct from both the average and from sexually dimorphic features (replicating Said and Todorov 6 and Holzleitner et al 7 ). Critically, we show that attractive shape and complexion face features sit at the outskirts of the natural distribution of face variations, reflecting preferences for elaborate cues 34 -e.g., redder skin in Westerners, reflecting more oxygenated blood; 35 super-stimulation (e.g., redder lips); and a yellower skin in EAs, reflecting a healthier, carotenoid-rich diet, 35 plus a significant level of idiosyncrasies across individuals.…”

“…Theories of universal beauty propose that attractive faces comprise features that are closer to the population average 3 while optimizing sexual dimorphism. 4 However, emerging evidence questions this model as an accurate representation of facial attractiveness, 5 , 6 , 7 including representing the diversity of beauty preferences within and across cultures. 8 , 9 , 10 , 11 , 12 Here, we demonstrate that Western Europeans (WEs) and East Asians (EAs) evaluate facial beauty using culture-specific features, contradicting theories of universality.…”

Stability of women's facial shape throughout the menstrual cycle

“…The broken stick criterion was used to select PCs to be used in subsequent analyses [53]. Facial asymmetry, averageness and sexual dimorphism were assessed using standard methods described in Holzleitner et al ( [54]; for more details and analysis code, see https://osf.io/drtg9/). Facial asymmetry was calculated as the Euclidean distance between each woman's original and mirrored set of shape coordinates.…”

A Generative Model of Cultural Face Attractiveness

“…creating average faces; faces with exaggerated sexual dimorphism) and then test them against human perception of attractiveness. However, the evidence suggests that current hypotheses have not captured attractiveness judgments particularly well [4,11]. Furthermore, they also failed to explain the idiosyncratic cultural preferences [2,5].…”