Unlike non-human primates, humans are highly tolerant of out-group strangers, as evidenced by the ethnographic and archaeological records; however, very little research has addressed when and why people build relationships with out-group individuals. What selection pressures might have favored inter-group relationship building in humans? As reducing temporal variation in resource access has been crucial in the human foraging ecology, I suggest that relationships with out-group members may provide access to non-local resources and buffer resource shortfalls striking entire communities. The relevance of out-group relationships is not limited to small-scale or prehistoric contexts either, as out-group members can also provide access to difficult-to-access resources like market items. My research program is focused on the above question; in my dissertation, I address three components of it: Which social and ecological factors favor out-group relationships? What do people look for when picking partners from out-groups? When people value out-group members, will they avoid behavior that would inflict costs on these individuals?

Among three populations of horticulturalists in Bolivia, I measured valuation for out-group members using a non-anonymous economic game in which participants could be generous to in-group and out-group strangers. I found that participants were highly generous to both in-group and out-group strangers. Participants invest more in out-group relationships when they have less non-local (i.e., market) resource access, as proxied by their subjective socioeconomic status relative to others in their community. As has been demonstrated in the literature on in-group partner choice, participants prefer strangers perceived to be "good people" (a term associated with cooperative qualities in the Bolivian context). I also expected that out-group partner choice would track opportunities for resource access; indeed, participants also give more to strangers from out-groups perceived to have more market or political resource access. That said, the effect of resource access on out-group valuation and out-group partner choice varied by proxy.

In a 55-country sample, I found that participants were less likely to condone corrupt acts, which generate costs for others, when they identified with larger groups of individuals, a proxy for valuation of these individuals. However, there was a caveat: participants with local identities were as opposed to corrupt acts as those with country, continent, or world identities, while those with regional identities were the most willing to condone corrupt acts. Consistent with my suggestion that out-group members buffer actors against variation in resource access, participants experiencing the greatest degree of resource shortfalls were more likely to not find corruption permissible; however, this was also true of those experiencing the least amount of shortfall.

In sum, valuation for out-group partnerships does track resource access in three populations of horticulturalists and a sample from 55 counties, but the effect varies by measure in instructive ways. For example, the negative relationship between market items owned and out-group valuation in one population may reflect experiences of discrimination in market contexts, rather than a lack of utility for out-group relationships per se. As discussed in the conclusion, my ongoing and future work capitalizes on these variable effects as it delves further into the nature of out-group relationships in the Bolivian context.