What’s in a face? A lot, apparently. Ever get the feeling when you meet someone for the first time, that there is just something about this person you really don’t like? Although most of the time we are not consciously aware of it when it happens, humans pay great attention to facial features when evaluating who to trust, who not to trust, the desirability of a mate, and even the potential efficacy of political leaders. Recent scholarship suggests that this is a consequence of psychological adaptations designed to detect reliable cues in facial structure that have ancestrally correlated with personality traits and behavior. For example, one study shows that genetically conditioned facial attributes such as facial width-to-height ratio reliably predict a tendency to deceive and cheat others, as well as propensity for aggression.(1) Other studies have shown that subjects are surprisingly good at distinguishing cheaters from cooperators based merely on having viewed pictures of people’s faces.(2) So there seems to be evidence both that facial cues reliably correlate with behavior and personality traits, and that people are good at using these cues to infer such traits. In other words, stable environmental cue structures (e.g. facial width-to-height ratio) has reliably correlated with fitness-relevant behavior in others (e.g. deception), such that natural selection has favored adaptations for detecting such cues and for regulating behavior contingent upon their detection.

Why should it be the case that facial features have been evolutionarily reliable indicators of personality traits? The short answer is that sexually dimorphic facial features develop around puberty and are triggered by underlying changes in hormones such as testosterone. As you probably know, testosterone itself is strongly related to aggression, sensitivity to threat, relatively selfish behavior in economic games, and it has also been found that increases in testosterone tend to correlate negatively with empathy toward others.(3) Given the fact that particular facial cues have been evolutionarily reliable indicators of certain personality traits, it is no surprise that psychological adaptations should be designed to attend to these cues when regulating motivation during inter-personal interactions. The fitness benefits of attending to such adaptively relevant cues are obvious, especially in the realm of cooperation and mating.

This may be easy enough to accept in the realm of mating and even cooperation, but what about political behavior? A recent article by Brian R. Spisak and colleagues presents evidence that people rely on visual cues in the face when choosing a political leader, and that people tend to prefer masculine faces in wartime and feminine faces in peacetime.(4) This is a provocative claim, so let’s consider the theory and evidence. Spisak and colleagues begin by establishing that the challenge of finding, establishing, and maintaining leadership for collective action was a reproductively significant adaptive problem for our ancestors. Furthermore, not all collective action is the same, which suggests that “different leadership prototypes” may be more effective in different situations. The authors therefore argue that, “A key adaptive challenge then is for group members to identify an individual to follow in any particular situation.”

Given the prevalence of this adaptive problem in ancestral environments, what cues might psychological adaptations seek when evaluating support for a potential leader? Spisak et al. argue that this is essentially a process in which leaders are chosen who most closely match the leadership prototype that is best suited for the specific context of collective action. The two leadership prototypes that the authors consider are represented by facial masculinity and facial femininity, while the contexts of collective action that they consider are warfare and peacetime. The choice of collective action context seems intuitive (war vs. peace), but why have Spisak and colleagues chosen facial masculinity and femininity as leadership prototypes? The answer again has to do with the underlying personality traits with which these facial cues reliably correlate. As the authors put it, the “differentiated constellations of skills required for competition or cooperation seem to parallel phenotypic associations with hormones such as testosterone and estrogen.” In other words, testosterone reliably correlates with facial masculinity and dominance, while estrogen reliably correlates with facial femininity and cooperativeness, and of course there may be certain contexts in which a dominant leader may be more effective, and certain contexts in which a cooperative leader may be more effective.

Thus, the importance of context. Spisak et al. hypothesize that in the context of warfare, masculine faces will be preferred over feminine faces as a consequence of the fact that facial masculinity reliably correlates with the personality traits that are likely to have been most useful during ancestral inter-group conflict. In contrast, in the context of peacetime, feminine faces should be preferred over masculine ones for opposite reasons. The authors show that it is not the sex of the candidate that matters; instead, “masculine-females are preferred as leaders over feminine-males for war and the converse for peace.” Spisak and colleagues also show that the effect holds cross-culturally, by comparing effects in Western and East Asian population samples.

Taken together, these studies suggest that humans possess psychological adaptations that are designed to actively track cues in the environment such as facial masculinity/femininity, which ancestrally correlated with fitness-relevant behavior and personality traits. Importantly, these systems operate outside of conscious awareness. In our interactions with others, we do not consciously and deliberately scan for variables such as the degree of genetic relatedness and facial width-to-height ratio and then use such variables to compute an optimal level of cooperation; instead, natural selection designs systems that operate below consciousness and produce output that we experience as attraction, trust, or avoidance. As recent findings suggest, these systems operate at all levels of social interaction, from the interpersonal to the political.

Anthony C. Lopez is a Ph.D. candidate at Brown University studying international relations and political psychology. As an affiliate with the Center for Evolutionary Psychology, his research seeks to incorporate evolutionary psychology into dynamic models of political behavior. Currently, he is investigating war as the product of an evolved coalitional psychology, as well as examining the relationship between inter-group conflict and intra-group cooperation from an adaptationist perspective. Anthony maintains a blog, Evolutionary Politics, aimed at collecting and reviewing current research in this area.


1. Haselhuhn, M.P. & Wong, E.M. 2012. Bad to the bone: Facial structure predicts unethical behavior. Proceedings of the Royal Society, Biological Sciences-B; Short, et al. 2012. Detection of propensity for aggression based on facial structure irrespective of face race. Evolution and Human Behavior, 33: 121-129.

2. Yamagishi et al. 2003. You can judge a book by its cover: Evidence that cheaters may look different from cooperators. Evolution and Human Behavior, 24: 290-301; Verplaetse et al. 2007. You can judge a book by its cover: The sequel. Evolution and Human Behavior, 28: 260-271.

3. Archer, J. 2006. Testosterone and human aggression: An evaluation of the challenge hypothesis. Neuroscience and Biobehavioral Reviews, 30: 319-345; Zak, P.J. 2009. Testosterone administration decreases generosity in the ultimatum game. PLoS ONE, 4(12); Van Honk, J. et al. 2011. Testosterone administration impairs cognitive empathy in women depending on second-to-fourth digit ratio. Proceedings of the National Academy of Sciences, 108(8): 3445-3452.

4. Spisak, B.R. et al. 2012. Warriors and peacekeepers: Testing a biosocial implicit leadership hypothesis of intergroup relations using masculine and feminine faces. PLoS ONE, 7(1).

Leave a Reply