What do people want in a friend? Ancestral cooperation shaped contemporary friend preferences
– by Adar Eisenbruch
How do people choose their friends? Given how ubiquitous friendships are, you might think that the answer to this question would be obvious and well-established. In reality, we’re still learning a lot about friendship formation. There are two broad theories of why people choose the friends that they choose. The standard in social psychology is called the “instrumentality principle,” which states that people are attracted to potential friends with traits that are useful for their current goals. Want to play basketball? You’re going to want tall friends. Hungry? Time to hang out with that guy who went to culinary school. Need help with your math homework? The nerd down the hall suddenly seems pretty cool.
Evolutionary psychologists have suggested another theory, which my colleagues and I have called “social taste buds.” On this theory, people have evolved friend preferences, just like we have evolved food preferences and mate preferences. Since friendship likely evolved as a form of long-term, ongoing cooperative relationship, we hypothesize that people want friends with traits that would have contributed to cooperative partner value in ancestral environments. For example, cues of generosity, trustworthiness, health, and the ability to produce resources as a forager should all be attractive in a potential friend.
We recently tested this theory in two studies reported in Evolution and Human Behavior. In Study 1, we had pairs of unacquainted, same-sex college students come to the lab. After a half-hour conversation, the participants privately rated each other on friend desirability and on traits that might predict friend desirability (e.g. trustworthiness, attractiveness, ability as a hunter-gatherer). At the end of the study, the participants had the opportunity to leave their email address for the other person so that they could meet up outside the lab. We actually exchanged their email addresses if both participants wanted to.
Overall, ratings of ancestral productivity (the ability to produce resources as a hunter-gatherer), attractiveness, prosociality, dominance, social status, nurturance (the ability to take care of kin), and knowledge of what the other sex likes were all positive predictors of friend desirability. Two preferences in particular support the social taste buds theory. First is the preference for physically attractive friends, which was the single strongest preference. This seems like a strange thing to want in a same-sex friend – what good is a friend’s attractiveness to you? But physical attractiveness is likely a summary judgment of partner value. Attractiveness is based on a combination of cues of health, maturity, strength, etc. Since so many of the traits that made someone a good mate likely also made them a good cooperative partner, we suggest that humans evolved a preference for physical attractiveness in same-sex relationships too.
The second is the preference for ancestral productivity, that is, the ability to produce resources as a hunter-gatherer, which was stronger among men than women. This is another bizarre trait for undergraduates to want in a friend – the campus dining halls don’t make students forage for their dinner. It makes sense, however, if friend preferences evolved as cooperative partner preferences, and men faced stronger selection pressures for choosing good hunting partners.
Another sex difference was also striking. Recall that the participants had the opportunity to leave their contact info for the other participant so that they could continue their relationship outside the lab. Men were significantly more likely than women to leave their contact information for the other person (50% vs. 30%). This wasn’t because the men found their partners more desirable as friends. Instead, at any level of friend desirability, men were more likely than women to initiate a real-world friendship. This is consistent with the hypothesis that men have evolved a looser friendship style than women in order to facilitate building a flexible network of cooperative partners and allies. Previous studies have found that men prefer to have more, but less intimate, friendships than women, but this is the first observation of men’s lower friendship standards affecting real relationship formation.
Notice that many of the preferences we observed could be explained by either theory. A preference for friends high in prosociality or social status, for example, could be due to ancestral selection pressures or due to the relevance of those traits for current goals. We therefore conducted a second study that compared participants’ friend preferences to the traits that they said would be useful for their current goals. Participants first spent a few minutes writing about their most important current goal, then they rated 11 traits on how much a friend having each trait would help them achieve their goal. The participants then rated same-sex faces for how much they wanted to be friends with each person. The faces had also been rated on the same 11 traits, so we could calculate the effect of each trait on participants’ friend desirability judgments and compare it to the trait’s goal-relevance.
Attractiveness, extraversion, and ancestral productivity were more attractive in a potential friend than could be explained by their goal-relevance, suggesting that the preferences for those traits were not motivated by current goals. Overall, there was little evidence that preferences were calibrated to current utility, but stronger evidence of preferences like those observed in Study 1 irrespective of their current goal-relevance.
Of course, the instrumentality principle and the social taste buds theory are not mutually exclusive. These studies provided clear evidence of evolved friend preferences, but it’s very likely that people sometimes choose their friends based on evolved preferences and sometimes to serve their current goals. This would match how people choose foods; sometimes your dietary goals lead you to order a salad, and sometimes your evolved preferences win and you order an ice cream sundae. Researchers can now proceed to the interesting question of how these two types of preferences are integrated with each other.
