Browsing by Subject "Cooperation"

This dissertation looks at how to identify and promote cooperation in a multiagent system, first theoretically through the lens of computational game theory and later empirically through a human subject experiment. Chapter 2 studies the network dynamics leading to a potential unraveling of cooperation and identify the subset of agents that can form an enforceable cooperative agreement with. This is an important problem, because cooperation is harder to sustain when information of defection, and thus the consequent punishment, transfers slowly through the network structures from a larger community. Chapter 3 examines a model that studies cooperation in a broader strategic context where agents may interact in multiple different domains, or games, simultaneously. Even if a game independently does not give an agent sufficient incentive to play the cooperative action, there may be hope for cooperation when multiple games with compensating asymmetries are put together. Exploiting compensating asymmetries, we can find an institutional arrangement that would either ensure maximum incentives for cooperation or require minimum subsidy to establish sufficient incentives for cooperation. Lastly, Chapter 4 studies a two-layered public good game to empirically examine whether community enforcement through existing bilateral relationships can encourage cooperation in a social dilemma situation. Here, it is found that how the situation is presented matters greatly to real life agents, as their understanding of whether they are in a cooperative or a competitive, strategic setting changes the level of overall cooperation.

Social trust has long been of interest as a micro-level factor that is predictive of prosocial attitudes and behaviors. In this dissertation I argue that the relationship between social trust and prosocial behaviors not one-to-one. My argument is informed by social psychological theory and work from the cognitive sciences on the importance of cultural factors in individual decision-making. Specifically, my theoretical argument is that widespread beliefs about the trustfulness of people in general, cultural trust, moderates the relationship between social trust and prosociality. I test this theory in three studies: 1) analysis of interview data on social trust and self-reported prosocial behavior, 2) an experimental study involving a public goods game, and 3) a cross-national, comparative study of observational data. The results of these studies have important implications for the studies of social trust and prosocial behaviors. The first study suggests that whether an individual believes people in general are trustworthy or not has more to do with perceptions shaped by cultural factors rather than experiences. The second and third studies clarify the role of cultural trust in the social trust-prosociality relationship. Namely, that high trustors are inclined to behave cooperatively, but this depends on their inferring from cultural trust what the likely behaviors others will be. The findings of these dissertation studies demonstrate that social trust is an important, but insufficient predictor of prosociality. A cultural environment characterized by low social trust can induce an equilibrium where even high trustors who would prefer to act prosocially behave selfishly, inferring others will not reciprocate their cooperation. They imply that for the maintenance of important public goods the information environment may be as important a place for interventions to promote cooperation as are education and institutional arrangements.

This dissertation presents three studies of the social and reproductive behavior and social structure of male chimpanzees (Pan troglodytes schweinfurthii) in Gombe National Park, Tanzania.

I. In many species of non-human primates, males cooperate and form friendly social bonds while simultaneously competing with each other for dominance rank and mating opportunities. While several studies reveal clear links between female bonds and correlates of fitness in female philopatric primate species, few have investigated whether bonds among males have measurable fitness benefits. Further, no studies in chimpanzees have investigated the fitness effects of cooperative exchange, or contrasted the fitness effects of bond formation and cooperative exchange. Here, I investigate whether 1) male social bonds, 2) position in affiliative and coalition networks, or 3) strategic exchange with other individuals of grooming for support or tolerance facilitate fitness benefits in one population of wild, free-ranging chimpanzees. I generated measures of social connectedness, coalition formation, and grooming effort for each male in two-year periods from 1990 to 2011, and employed mixed models to determine whether, controlling for current rank and age, these measures predicted 1) rank change (a correlate of future reproductive output) and 2) reproductive success within each period. I found that rank change was associated with betweenness in the network of coalition formation, but only weakly with social bonds and not at all with position in the network of social relationships. I further found that rank change was predicted by grooming effort, although this relationship depended on male dominance rank. Surprisingly, reproductive success was not associated with social connectedness or with betweenness in coalitionary or social relationship networks. Instead, grooming effort strongly predicted reproductive success. Thus it appears that males that occupy central positions in the coalition network, and those that groom others at a high rate, are more likely to rise in rank (if they are low-ranking to begin with). However, males that successfully sire offspring groom others at a high rate. These results suggest that, unlike in cercopithecine primates, social bonds do not enhance fitness in male chimpanzees, and instead males rely on grooming and coalition formation to improve their rank and reproductive success.

II. Animals face both costs and benefits associated with living in groups. When the costs of membership exceed the benefits, group fissions can occur. Fissions are documented in a number of animal species, but are comparatively rare in male philopatric primates. One of the few likely cases occurred in chimpanzees in 1973 in Gombe National Park, Tanzania, when the main study community split into two separate communities, Kasekela and Kahama. Over the next four years, the Kasekela community killed the adult males and one female of the Kahama community. Here we use social network analysis to explore the process of community fission in chimpanzees by examining association, grooming, and ranging patterns. We found that the two communities split from one original cohesive community, although one with incipient subgrouping patterns. Subgrouping patterns in the grooming and association networks began to increase sharply beginning in 1971, and this period closely coincided with a dominance struggle between three high-ranking males and with a peak in operational sex ratio. Finally, we found a relationship between post-split community membership and previous association, grooming and ranging patterns in most periods of analysis, a tendency that became more pronounced as the fission approached. Thus, analysis suggests that the community began to split during a time of unusual sex ratio and a protracted dominance struggle, and that individuals remained with others with whom they preferentially associated in the previous years. These results are contrasted with group fissions in other taxa, and provide clues to the costs and benefits of group membership in chimpanzees.

III. In sexually reproducing animals, male and female reproductive strategies often conflict. In some species, males use aggression to overcome female choice, but debate persists over the extent to which this strategy is successful. Previous studies of male aggression toward females among wild chimpanzees have yielded contradictory results about the relationship between aggression and mating behavior. Critically, however, copulation frequency in primates is not always predictive of reproductive success. We analyzed a 17-year sample of behavioral and genetic data from the Kasekela chimpanzee community in Gombe National Park, Tanzania, to test the hypothesis that male aggression toward females increases male reproductive success. We examined the effect of male aggression toward females during ovarian cycling, including periods when the females were sexually receptive (swollen) and periods when they were not. We found that, after controlling for confounding factors, male aggression during a female’s swollen periods was positively correlated with copulation frequency. However, aggression toward swollen females was not predictive of paternity. Instead, aggression by high-ranking males toward females during their nonswollen periods was positively associated with likelihood of paternity. This indicates that long-term patterns of intimidation allow high-ranking males to increase their reproductive success, supporting the sexual coercion hypothesis. To our knowledge, this is the first study to present genetic evidence of sexual coercion as an adaptive strategy in a social mammal.

Previous research argues that people receive positive evaluations from third parties for engaging in peer punishment, which lead to reputational rewards. Recent work challenges this, however, finding instead that punishers can potentially receive negative evaluations from others. These conflicting arguments present a set of open questions regarding third-party evaluations of punishers and the factors shaping them. This dissertation offers three studies intended to improve knowledge surrounding these questions. These studies contribute to research in this area by using data from real-world settings to test experimental findings from the laboratory, and by considering how punishers can use verbal justifications to shape third-party evaluations of their actions.

The first of these studies, contained in Chapter 2, uses data from therapeutic communities for addiction treatment (TCs) to determine if people generally receive reputational rewards for engaging in peer punishment. Using social network analysis of data from four TCs, it asks whether residents of these communities are more likely to be identified as “role models” by others when they engage in punishment. Results offer mixed support for this being the case, with punishers being more likely to receive role model nominations from peers in two communities but no more likely in the other two. Simultaneously, results indicate that punishers are more likely to receive role model nominations from supervisory staff. Findings from this study show how punishers may or may not receive reputational rewards from peers and from authority figures, and suggest the importance of local institutional and cultural contexts in how third parties evaluate punishment.

Chapter 3 identifies the importance of social and relational factors in how people construct justifications for peer sanctions. Using social network analysis and automated text analysis on data from one TC, it examines how social proximity, status differences, and friendships shape the amount of explanatory effort made in justifications of rewards and punishment. Results indicate that these factors have different effects on the amount of explanatory effort given in justifications for rewards versus punishment. This finding suggests that TC residents use rewards and punishment to build and enact different kinds of relationships with their peers, supporting recent arguments advanced by TC scholars.

Using two online vignette experiments, Chapter 4 considers how third-party evaluations change when punishers mention specific social actors when justifying punishment. These experiments measured participants’ evaluations of punishers who mentioned themselves, a friend, a stranger, or a collective when justifying their actions. Results show that observers evaluate punishers more positively when punishers offer justifications that mention collectives rather than themselves. Punishers who reference individuals other than themselves also receive more positive evaluations, though this finding requires further investigation. Additional results show that people generally prefer punishment for norm violations that affect broader groups rather than specific individuals. However, this preference appears to have no effect on how third parties evaluate punishers based on whom they mention in justifications.

Overall findings from this dissertation contribute to research on cooperation, punishment, and “metanorms” surrounding the propriety of peer punishment. Findings from select chapters make additional contributions to other areas of inquiry, including sociological theories of accounts and justifications and clinical research on therapeutic communities as a form of addiction treatment. These contributions are discussed in detail in their respective chapters, as well as in the concluding piece of this dissertation.