Cooperation In Human Groups Research Articles

Between-group cooperation in bonobos.

Bonobos provide insight into the origins of partner-specific cooperation in human groups.

Scaffolding cooperation in human groups with deep reinforcement learning.

Effective approaches to encouraging group cooperation are still an open challenge. Here we apply recent advances in deep learning to structure networks of human participants playing a group cooperation game. We leverage deep reinforcement learning and simulation methods to train a 'social planner' capable of making recommendations to create or break connections between group members. The strategy that it develops succeeds at encouraging pro-sociality in networks of human participants (N = 208 participants in 13 groups) playing for real monetary stakes. Under the social planner, groups finished the game with an average cooperation rate of 77.7%, compared with 42.8% in static networks (N = 176 in 11 groups). In contrast to prior strategies that separate defectors from cooperators (tested here with N = 384 in 24 groups), the social planner learns to take a conciliatory approach to defectors, encouraging them to act pro-socially by moving them to small highly cooperative neighbourhoods.

Do descriptive social norms drive peer punishment? Conditional punishment strategies and their impact on cooperation

Peer punishment is widely considered a key mechanism supporting cooperation in human groups. Although much research shows that human behavior is shaped by the prevailing social norms, little is known about how punishment decisions are impacted by the social context. We present a set of large-scale incentivized experiments in which participants (999 American participants recruited via Amazon Mechanical Turk) could punish their partner conditional on either the level of cooperation or the level of punishment displayed by others who previously interacted in the same setting. While many participants punish independently of levels of cooperation or punishment, a substantial portion punishes free riding more severely when cooperation is more common (‘norm enforcement’), or when free riding is more severely punished by others (‘conformist punishment’). With a dynamic model we demonstrate that conditional punishment strategies can substantially promote cooperation. In particular, conformist punishment helps cooperation to gain a foothold in a population, and norm enforcement helps to maintain cooperation at high levels. Our results provide solid empirical evidence of conditional punishment strategies and illustrate their possible implications for the dynamics of human cooperation.

Self-organized division of cognitive labor.

Often members of a group benefit from dividing the group's task into separate components, where each member specializes their role so as to accomplish only one of the components. While this division of labor phenomenon has been observed with respect to both manual and cognitive labor, there is no clear understanding of the cognitive mechanisms allowing for its emergence, especially when there are multiple divisions possible and communication is limited. Indeed, maximization of expected utility often does not differentiate between alternative ways in which individuals could divide labor. We developed an iterative two-person game in which there are multiple ways of dividing labor, but in which it is not possible to explicitly negotiate a division. We implemented the game both as a human experimental task and as a computational model. Our results show that the majority of human dyads can finish the game with an efficient division of labor. Moreover, we fitted our computational model to the behavioral data, which allowed us to explain how the perceived similarity between a player's actions and the task's focal points guided the players' choices from one round to the other, thus bridging the group dynamics and its underlying cognitive process. Potential applications of this model outside cognitive science include the improvement of cooperation in human groups, multi-agent systems, as well as human-robot collaboration.

Conditional Punishment: Descriptive Social Norms Drive Negative Reciprocity

Peer punishment is widely considered a key mechanism supporting cooperation in human groups. Although much research shows that human behavior is shaped by the prevailing social norms, little is known about how punishment decisions are impacted by the social context. We present a set of large-scale incentivized experiments in which participants (N=999) could punish their partner conditional on either the level of cooperation or the level of punishment displayed by others who previously interacted in the same setting. While many participants punish independently of levels of cooperation or punishment, a substantial portion punishes free riding more severely when cooperation is more common (‘norm enforcement’), or when free riding is more severely punished by others (‘conformist punishment’). With a dynamic model we demonstrate that conditional punishment strategies can substantially promote cooperation. In particular, conformist punishment helps cooperation to gain a foothold in a population, and norm enforcement helps to maintain cooperation at high levels. Our results provide solid empirical evidence of conditional punishment strategies and illustrate their possible implications for the dynamics of human cooperation.

Network Engineering Using Autonomous Agents Increases Cooperation in Human Groups.

SummaryCooperation in human groups is challenging, and various mechanisms are required to sustain it, although it nevertheless usually decays over time. Here, we perform theoretically informed experiments involving networks of humans (1,024 subjects in 64 networks) playing a public-goods game to which we sometimes added autonomous agents (bots) programmed to use only local knowledge. We show that cooperation can not only be stabilized, but even promoted, when the bots intervene in the partner selections made by the humans, re-shaping social connections locally within a larger group. Cooperation rates increased from 60.4% at baseline to 79.4% at the end. This network-intervention strategy outperformed other strategies, such as adding bots playing tit-for-tat. We also confirm that even a single bot can foster cooperation in human groups by using a mixed strategy designed to support the development of cooperative clusters. Simple artificial intelligence can increase the cooperation of groups.

The evolution of human cooperation

Darwin viewed cooperation as a perplexing challenge to his theory of natural selection. Natural selection generally favors the evolution of behaviors that enhance the fitness of individuals. Cooperative behavior, which increases the fitness of a recipient at the expense of the donor, contradicts this logic. William D. Hamilton helped to solve the puzzle when he showed that cooperation can evolve if cooperators direct benefits selectively to other cooperators (i.e. assortment). Kinship, group selection and the previous behavior of social partners all provide mechanisms for assortment (Figure 1), and kin selection and reciprocal altruism are the foundation of the kinds of cooperative behavior observed in many animals. Humans also bias cooperation in favor of kin and reciprocating partners, but the scope, scale, and variability of human cooperation greatly exceed that of other animals. Here, we introduce derived features of human cooperation in the context in which they originally evolved, and discuss the processes that may have shaped the evolution of our remarkable capacity for cooperation. We argue that culturally-evolved norms that specify how people should behave provide an evolutionarily novel mechanism for assortment, and play an important role in sustaining derived properties of cooperation in human groups.

The Sanctioning Dilemma: A Quasi-Experiment on Social Norm Enforcement in the Train

Numerous laboratory experiments have established peer-sanctioning as an important driver of norm compliance and cooperation in human groups. However, systematic evidence of peer-sanctioning occurring in the field is still rare. Here we present results from a quasi-experimental field study investigating the enforcement of the silence norm in the train. We let a confederate play music on his/her mobile phone in an open-plan train car and measure the time until a negative sanction occurs (if any). The silence norm is enforced in 45 of 90 cases, enforcement rates do not differ across silent- and non-silent-area cars, and the more passengers are in a car, the more likely is the silence norm enforced. Passengers’ propensities to enforce the silence norm are in line with predictions derived from the asymmetric volunteer’s dilemma (VOD). The higher a passenger’s net benefit from enforcing the silence norm is, the more likely is the passenger to negatively sanction the norm breaker. Our findings extend the validity of results from laboratory experiments which conceive the second-order free-rider problem as a VOD.

Generosity is a sign of trustworthiness—the punishment of selfishness is not

Peer-punishment is an important determinant of cooperation in human groups. It has been suggested that, at the proximate level of analysis, punitive preferences can explain why humans incur costs to punish their deviant peers. How punitive preferences could have evolved in humans is still not entirely understood. A possible explanation at the ultimate level of analysis comes from signaling theory. It has been argued that the punishment of defectors can be a type-separating signal of the punisher's cooperative intent. As a result, punishers are selected more often as interaction partners in social exchange and are partly compensated for the costs they incur when punishing defectors. A similar argument has been made with regard to acts of generosity. In a laboratory experiment, we investigate whether the punishment of a selfish division of money in a dictator game is a sign of trustworthiness and whether punishers are more trustworthy interaction partners in a trust game than non-punishers. We distinguish between second-party and third-party punishment and compare punitive acts with acts of generosity as signs of trustworthiness. We find that punishers are not more trustworthy than non-punishers and that punishers are not trusted more than non-punishers, both in the second-party and in the third-party punishment condition. To the contrary, second-party punishers are trusted less than their non-punishing counterparts. However, participants who choose a generous division of money are more trustworthy and are trusted more than participants who choose a selfish division or participants about whom no information is available. Our results suggest that, unlike for punitive acts, the signaling benefits of generosity are to be gained in social exchange.

Individual heterogeneity and costly punishment: a volunteer's dilemma

Social control and the enforcement of social norms glue a society together. It has been shown theoretically and empirically that informal punishment of wrongdoers fosters cooperation in human groups. Most of this research has focused on voluntary and uncoordinated punishment carried out by individual group members. However, as punishment is costly, it is an open question as to why humans engage in the punishment of wrongdoers even in one-time-only encounters. While evolved punitive preferences have been advocated as proximate explanations for such behaviour, the strategic nature of the punishment situation has remained underexplored. It has been suggested to conceive of the punishment situation as a volunteer's dilemma (VOD), where only one individual's action is necessary and sufficient to punish the wrongdoer. Here, we show experimentally that implementing the punishment situation as a VOD sustains cooperation in an environment where punishers and non-punishers coexist. Moreover, we show that punishment-cost heterogeneity allows individuals to tacitly agree on only the strongest group member carrying out the punishment, thereby increasing the effectiveness and efficiency of social norm enforcement. Our results corroborate that costly peer punishment can be explained without assuming punitive preferences and show that centralized sanctioning institutions can emerge from arbitrary individual differences.

Cooperation Dynamics in a Multiethnic Society

The importance of ethnic diversity in determining social outcomes and reducing generalized cooperation is increasingly well documented. Theory suggests that cooperation in human groups may depend on reciprocal altruism and frequency of contact, yet these factors have not been linked with ethnic diversity. This study explores how fine-scale components of cooperation—social exclusivity and reciprocity—relate to broad-scale social conditions—ethnic diversity and ethnic stratification—in villages in Tamil Nadu’s Palani hills. Both ethnic diversity and ethnic stratification are associated with declines in indirect reciprocity, although stratification has a larger effect. In addition, stratification is linked to increased social exclusivity. Moreover, measures of direct reciprocity in the form of agricultural labor exchanges are uncorrelated with both diversity and stratification. These results imply (1) that ethnic stratification is more detrimental to cooperation than is ethnic diversity, (2) that social excl...

The virtues of gossip: Reputational information sharing as prosocial behavior.

Reputation systems promote cooperation and deter antisocial behavior in groups. Little is known, however, about how and why people share reputational information. Here, we seek to establish the existence and dynamics of prosocial gossip, the sharing of negative evaluative information about a target in a way that protects others from antisocial or exploitative behavior. We present a model of prosocial gossip and the results of 4 studies testing the model's claims. Results of Studies 1 through 3 demonstrate that (a) individuals who observe an antisocial act experience negative affect and are compelled to share information about the antisocial actor with a potentially vulnerable person, (b) sharing such information reduces negative affect created by observing the antisocial behavior, and (c) individuals possessing more prosocial orientations are the most motivated to engage in such gossip, even at a personal cost, and exhibit the greatest reduction in negative affect as a result. Study 4 demonstrates that prosocial gossip can effectively deter selfishness and promote cooperation. Taken together these results highlight the roles of prosocial motivations and negative affective reactions to injustice in maintaining reputational information sharing in groups. We conclude by discussing implications for reputational theories of the maintenance of cooperation in human groups.

Competition for Trophies Triggers Male Generosity

BackgroundCooperation is indispensable in human societies, and much progress has been made towards understanding human pro-social decisions. Formal incentives, such as punishment, are suggested as potential effective approaches despite the fact that punishment can crowd out intrinsic motives for cooperation and detrimentally impact efficiency. At the same time, evolutionary biologists have long recognized that cooperation, especially food sharing, is typically efficiently organized in groups living on wild foods, even absent formal economic incentives. Despite its evident importance, the source of this voluntary compliance remains largely uninformed. Drawing on costly signaling theory, and in light of the widely established competitive nature of males, we hypothesize that unique and displayable rewards (trophies) out of competition may trigger male generosity in competitive social environments.Principal FindingsHere, we use a controlled laboratory experiment to show that cooperation is sustained in a generosity competition with trophy rewards, but breaks down in the same environment with equally valuable but non-unique and non-displayable rewards. Further, we find that males' competition for trophies is the driving force behind treatment differences. In contrast, it appears that female competitiveness is not modulated by trophy rewards.SignificanceOur results suggest new approaches to promoting cooperation in human groups that, unlike punishment mechanisms, do not sacrifice efficiency. This could have important implications in any domain where voluntary compliance matters — including relations between spouses, employers and employees, market transactions, and conformity to legal standards.

Beliefs about God, the afterlife and morality support the role of supernatural policing in human cooperation

Reputation monitoring and the punishment of cheats are thought to be crucial to the viability and maintenance of human cooperation in large groups of non-kin. However, since the cost of policing moral norms must fall to those in the group, policing is itself a public good subject to exploitation by free riders. Recently, it has been suggested that belief in supernatural monitoring and punishment may discourage individuals from violating established moral norms and so facilitate human cooperation. Here we use cross-cultural survey data from a global sample of 87 countries to show that beliefs about two related sources of supernatural monitoring and punishment — God and the afterlife — independently predict respondents' assessment of the justifiability of a range of moral transgressions. This relationship holds even after controlling for frequency of religious participation, country of origin, religious denomination and level of education. As well as corroborating experimental work, our findings suggest that, across cultural and religious backgrounds, beliefs about the permissibility of moral transgressions are tied to beliefs about supernatural monitoring and punishment, supporting arguments that these beliefs may be important promoters of cooperation in human groups.

Structural resolutions of collective action problems

This paper concerns itself with the general problem of achieving cooperation in human groups and higher levels of social systems. Several social situations are considered where cooperation is problematic because self-interest contradicts group or collective interest: the prisoners' dilemma game; the commons' problem (Hardin, 1972); the collective action problem, i.e., the failure of memberships of many large interest groups, e.g., consumers and the general public, to cooperate to achieve common goals (Olson, 1968); and the problem of competitive panics, e.g., a crowd in a burning theater. We introduce a structural framework and methodology, using social system level concepts, to characterize and analyze such problems. It is shown that the various cases have a common underlying structure. In the analysis, we focus on the social context of the problematic situations and, in particular, on social processes that structure human interaction and collective behavior. A basic idea guiding the analysis is that actors purposively structure and transform interaction situations or games into situations of greater or less cooperation or conflict, depending upon the social context. We examine specific social control processes that may be found operating in social systems to resolve problems of achieving cooperative action, that is, to deal with contradictions between individual interests and autonomy on the one hand and group interest and need for cooperative action on the other. In particular, we focus on the social structuring and restructuring in groups of perceptions and evaluations, action possibilities, and decision procedures and, therefore, likely interaction patterns of those involved.