Trait impressions as overgeneralized responses to adaptively significant facial qualities: evidence from connectionist modeling

Connectionist modeling experiments tested anomalous-face and baby-face overgeneralization hypotheses proposed to explain consensual trait impressions of faces. Activation ofa neural network unit trained to respond to anomalousfaces pre-dicted impressions ofnormal adultfaces varying in attractiveness as well as several elderly stereotypes. Activation ofa neural network unit trained to respond to babies' faces predicted impressions ofadults varying in babyfaceness as well as 1 elderly ste-reotype. Thus, similarities ofnormal adultfaces to anomalous faces or babies 'faces contribute to impressions of them quite apartfrom knowledge of overlapping social stereotypes. The evolutionary importance ofappropriate responses to unfit individu-als or babies is presumed to produce a strong response preparedness that is overgeneralized to faces resembling the unfit or babies. We are enjoined not to judge a book by its cover, and we are cautioned that beauty is only skin deep. These warnings suggest that our natural proclivity is in fact to judge people by their appearance and to prefe