Facial cues contribute to attractiveness, including shape cues such as symmetry, averageness, and sexual dimorphism. These cues may represent cues to objective aspects of physiological health, thereby conferring an evolutionary advantage to individuals who find them attractive. The link between facial cues and aspects of physiological health is therefore central to evolutionary explanations of attractiveness. Previously, studies linking facial cues to aspects of physiological health have been infrequent, have had mixed results, and have tended to focus on individual facial cues in isolation. Geometric morphometric methodology (GMM) allows a bottom-up approach to identifying shape correlates of aspects of physiological health. Here, we apply GMM to facial shape data, producing models that successfully predict aspects of physiological health in 272 Asian, African, and Caucasian faces - percentage body fat (21.0% of variance explained), body mass index (BMI; 31.9%) and blood pressure (BP; 21.3%). Models successfully predict percentage body fat and blood pressure even when controlling for BMI, suggesting that they are not simply measuring body size. Predicted values of BMI and BP, but not percentage body fat, correlate with health ratings. When asked to manipulate the shape of faces along the physiological health variable axes (as determined by the models), participants reduced predicted BMI, body fat and (marginally) BP, suggesting that facial shape provides a valid cue to aspects of physiological health.
Over the past 10 years, Oosterhof and Todorov's valence-dominance model has emerged as the most prominent account of how people evaluate faces on social dimensions. In this model, two dimensions (valence and dominance) underpin social judgements of faces. Because this model has primarily been developed and tested in Western regions, it is unclear whether these findings apply to other regions. We addressed this question by replicating Oosterhof and Todorov's methodology across 11 world regions, 41 countries and 11,570 participants. When we used Oosterhof and Todorov's original analysis strategy, the valence-dominance model generalized across regions. When we used an alternative methodology to allow for correlated dimensions, we observed much less generalization. Collectively, these results suggest that, while the valence-dominance model generalizes very well across regions when dimensions are forced to be orthogonal, regional differences are revealed when we use different extraction methods and correlate and rotate the dimension reduction solution. PROTOCOL REGISTRATION: The stage 1 protocol for this Registered Report was accepted in principle on 5 November 2018. The protocol, as accepted by the journal, can be found at https://doi.org/10.6084/m9.figshare.7611443.v1 .