The effect of perceived competence on congressional elections reported by [ Todorov et al. ][1] likely reflects differences in their babyfacedness. In their Perspective, [Zebrowitz and Montepare][2] review research to support this explanation, with suggestions for future research on appearance biases.
[1]: http://www.sciencemag.org/cgi/content/short/308/5728/1623
[2]: http://www.sciencemag.org/cgi/content/full/308/5728/1565