Updating evidence on facial metrics: A Bayesian perspective on sexual dimorphism in facial width-to-height ratio and bizygomatic width

Facial width-to-height ratio (fWHR) is an extensively studied morphological measure, whichwas presumably shaped by sexual selection and has been linked to a wide range of perceptualand physiological traits. Underpinning these associations is the premise that fWHR is largerin men, which empirically exhibits a mixed and equivocal pattern in the literature due tovariation in measurement, large sample sizes revealing small but significant differences, anda lack of control of body size. In Study 1, in a sample of 1,949 faces, we used a Bayesianhierarchical model that incorporates prior information to simultaneously estimate sexualdimorphism in fWHR, adjusted for body size, across five measurement types. While wefound larger fWHR in women, comparing this effect to variability in fWHR due to imagecapture settings revealed no robust evidence of sex differences in fWHR. In Study 2, weinvestigated sex differences in facial width specifically (also adjusted for body size), againincorporating prior information, and confirmed men have greater face width than women.Advances in this area can be made by shifting focus away from arbitrary ratios like fWHR todirect measures like facial width – as well as carefully considering prior evidence of existingassociations.