Using secondary data analyses to establish whether face-shape characteristics predict social judgments of faces consistently across world regions

I will combine analyses of a large open-access face-rating dataset and a face-image database (i.e., secondary data analyses) to establish whether face-shape characteristics predict social judgments of faces consistently across world regions.

A large body of work has demonstrated that social judgments of faces (i.e., the stereotypic perceptions we form about other people based on their facial appearance) influence important social outcomes. For example, people prefer to date, mate with, hire, and vote for individuals perceived as being particularly attractive. Moreover, untrustworthy-looking court defendants are more likely to receive death sentences. Consequently, a large interdisciplinary literature has developed investigating the factors that shape social judgments of faces.

An unresolved issue in this literature is the extent to which the effects of face-shape characteristics on social judgments of faces are consistent across different world regions. Findings from studies of this issue have been mixed (e.g., Perrett et al., 1998 Nature; Scott et al., 2014 PNAS) and the methods they used to assess social judgments of faces have recently been strongly criticised.

I recently led a Registered Report that was published in Nature Human Behaviour (Jones et al., 2021, 62 citations in Google Scholar) and assessed over eleven thousand participants' social judgments of faces in eleven world regions (Africa, Asia, Australia and New Zealand, Central America and Mexico, Eastern Europe, The Middle East, United States and Canada, Scandinavia, South America, United Kingdom, and Western Europe) using methods that specifically addressed limitations of methods used in previous work. Using Principal Component Analyses (PCA), this study found that social judgments of faces were underpinned by trustworthiness and dominance dimensions and that this pattern of results was highly consistent across world regions. Crucially, however, Jones et al. (2021) did not investigate relationships between these dimensions and face-shape characteristics.

The analyses reported in Jones et al. (2021) are analyses of a dataset produced by a project I led (with Lisa DeBruine and Jessica Flake) that was the first project to be conducted by the Psychological Science Accelerator (PSA; https://psysciacc.org). The PSA is a distributed network of researchers from across the world who conduct large-scale democratically selected studies. The dataset from Jones et al. (2021) is referred to within the PSA network as the PSA001 dataset (reflecting the fact it was the first dataset to be produced by the PSA) and both the PSA001 dataset and the face stimuli used to collect these data are publicly available. Consequently, I propose to carry out preregistered secondary analyses of these data to assess whether face-shape characteristics predict social judgments of faces consistently across world regions.