Emergence of specialized third-party enforcement.

The question of how cooperation evolves and is maintained among nonkin is central to the biological, social, and behavioral sciences. Previous research has focused on explaining how cooperation in social dilemmas can be maintained by direct and indirect reciprocity among the participants of the social dilemma. However, in complex human societies, both modern and ancient, cooperation is frequently maintained by means of specialized third-party enforcement. We provide an evolutionary-game-theoretic model that explains how specialized third-party enforcement of cooperation (specialized reciprocity) can emerge. A population consists of producers and enforcers. First, producers engage in a joint undertaking represented by a prisoner's dilemma. They are paired randomly and receive no information about their partner's history, which precludes direct and indirect reciprocity. Then, enforcers tax producers and may punish their clients. Finally, the enforcers are randomly paired and may try to grab resources from each other. In order to sustain producer cooperation, enforcers must punish defecting producers, but punishing is costly to enforcers. We show that the threat of potential intraenforcer conflict can incentivize enforcers to engage in costly punishment of producers, provided they are sufficiently informed to maintain a reputation system. That is, the "guards" are guarded by the guards themselves. We demonstrate the key mechanisms analytically and corroborate our results with numerical simulations.

The shorter a man is, the more he defends fairness: relative height disadvantage promoting third-party punishment-evidence from inter-brain synchronization.

Third-party punishment occurs in interpersonal interactions to sustain social norms, and is strongly influenced by the characteristics of the interacting individuals. During social interactions, height is the striking physical appearance features first observed, height disadvantage may critically influence men's behavior and mental health. Herein, we explored the influence of height disadvantage on third-party punishment through time-frequency analysis and electroencephalography hyperscanning. Two participants were randomly designated as the recipient and third party after height comparison and instructed to complete third-party punishment task. Compared with when the third party's height is higher than the recipient's height, when the third party's height is lower, the punishment rate and transfer amount were significantly higher. Only for highly unfair offers, the theta power was significantly greater when the third party's height was lower. The inter-brain synchronization between the recipient and the third party was significantly stronger when the third party's height was lower. Compared with the fair and medium unfair offers, the inter-brain synchronization was strongest for highly unfair offers. Our findings indicate that the height disadvantage-induced anger and reputation concern promote third-party punishment and inter-brain synchronization. This study enriches research perspective and expands the application of the theory of Napoleon complex.

Spurious normativity enhances learning of compliance and enforcement behavior in artificial agents.

How do societies learn and maintain social norms? Here we use multiagent reinforcement learning to investigate the learning dynamics of enforcement and compliance behaviors. Artificial agents populate a foraging environment and need to learn to avoid a poisonous berry. Agents learn to avoid eating poisonous berries better when doing so is taboo, meaning the behavior is punished by other agents. The taboo helps overcome a credit assignment problem in discovering delayed health effects. Critically, introducing an additional taboo, which results in punishment for eating a harmless berry, further improves overall returns. This "silly rule" counterintuitively has a positive effect because it gives agents more practice in learning rule enforcement. By probing what individual agents have learned, we demonstrate that normative behavior relies on a sequence of learned skills. Learning rule compliance builds upon prior learning of rule enforcement by other agents. Our results highlight the benefit of employing a multiagent reinforcement learning computational model focused on learning to implement complex actions.

Context Modulates Perceived Fairness in Altruistic Punishment: Neural Signatures from ERPs and EEG Oscillations.

Social norms and altruistic punitive behaviours are both based on the integration of information from multiple contexts. Individual behavioural performance can be altered by loss and gain contexts, which produce different mental states and subjective perceptions. In this study, we used event-related potential and time-frequency techniques to examine performance on a third-party punishment task and to explore the neural mechanisms underlying context-dependent differences in punishment decisions. The results indicated that individuals were more likely to reject unfairness in the context of loss (vs. gain) and to increase punishment as unfairness increased. In contrast, fairness appeared to cause an early increase in cognitive control signal enhancement, as indicated by the P2 amplitude and theta oscillations, and a later increase in emotional and motivational salience during decision-making in gain vs. loss contexts, as indicated by the medial frontal negativity and beta oscillations. In summary, individuals were more willing to sanction violations of social norms in the loss context than in the gain context and rejecting unfair losses induced more equity-related cognitive conflict than accepting unfair gains, highlighting the importance of context (i.e., gain vs. loss) in equity-related social decision-making processes.

Third-party punishers who express emotions are trusted more.

Third party punishment (TPP) is thought to be crucial to the evolution and maintenance of human cooperation. However, this type of punishment is often not rewarded, perhaps because punishers' underlying motives are unclear. We propose that the expression of moral emotions could solve this problem by advertising such motives. In each of three experiments (n = 1711), a third-party punishment game was followed by a trust game. Third parties expressed anger or disgust instead of, or in addition to, financial punishment. Results showed that third parties who expressed these emotions were trusted more than those who didn't express (Experiment 1), and more than those who financially punished (Experiment 2). Moreover, third parties who expressed while financially punishing were trusted more than those who punished without expressing (Experiment 3). Findings suggest that emotion expression might play a role in the evolution and maintenance of cooperation by facilitating TPP.

Association of sleep with moral severity to accidental harm transgressions: A cross-sectional study.

Based on the assumptions that moral judgement activities require cognitive control, a capacity impaired by low sleep quality or a lack of sleep, several studies have explored the association between sleep and moral judgements. However, even if some studies support the association between sleep and both moral awareness and unethical behaviours, others failed to find a robust association between sleep and moral utilitarianism. In the present well-powered preregistered cross-sectional study, we explored the role of sleep in another class of moral judgement, namely third-party punishment (in which people have to assess the morality of an agent who transgressed a moral rule). Specifically, we targeted the association of sleep with judgements of accidental harm transgressions, which are assumed to be especially cognitively costly. Our main analysis showed no association of overall sleep quality during the past month with moral severity in these transgressions. This result was confirmed for other sleep indexes (sleep quantity in the past month, and sleep quantity and quality in the past night). Lastly, we exhaustively explored the associations of all sleep indexes with all classes of moral judgement (accidental, intentional, attempted transgressions and control scenarios). These additional results revealed associations between sleep and moral severity, but none survived correction for multiple testing. Equivalence tests confirmed that the effect sizes of all these associations were relatively low (|r < 0.25|). We ensured that the lack of robust association between natural sleep and third-party punishment could not be explained by a low quality of the data collected.

Children endorse deterrence motivations for third-party punishment but derive higher enjoyment from compensating victims.

Children's punishment behavior may be driven by both retribution and deterrence, but the potential primacy of either motive is unknown. Moreover, children's punishment enjoyment and compensation enjoyment have never been directly contrasted. Here, British, Colombian, and Italian 7- to 11-year-old children (N = 123) operated a Justice System in which they viewed different moral transgressions in Minecraft, a globally popular video game, either face-to-face with an experimenter or over the internet. Children could respond to transgressions by punishing transgressors and compensating victims. The purpose of the system was framed in terms of retribution, deterrence, or compensation between participants. Children's performance, endorsement, and enjoyment of punishment and compensation were measured, along with their endorsement of retribution versus deterrence as punishment justifications, during and/or after justice administration. Children overwhelmingly endorsed deterrence over retribution as their punishment justification irrespective of age. When asked to reproduce the presented frame in their own words, children more reliably reproduced the deterrence frame rather than the retribution frame. Punishment enjoyment decreased while compensation enjoyment increased over time. Despite enjoying compensation more, children preferentially endorsed punishment over compensation, especially with increasing age and transgression severity. Reported deterrent justifications, superior reproduction of deterrence framing, lower enjoyment of punishment than of compensation, and higher endorsement of punishment over compensation together suggest that children felt that they ought to mete out punishment as a means to deter future transgressions. Face-to-face and internet-mediated responses were not distinguishable, supporting a route to social psychology research with primary school-aged children unable to physically visit labs.

To help or not: negative aging stereotypes held by younger adults could promote helping behaviors toward older adults.

Aging stereotypes affect older adults' behaviors, however, it is unclear whether and how (negative) aging stereotypes influence younger adults' behaviors toward older adults. Two possibilities arose, such that aging stereotypes would reduce helping behaviors according to TMT and SIT; while based on the BIAS map, we would expect the opposite. The present study aimed to further compare the two possibilities by examining the effect of negative aging stereotypes on younger adults' helping behaviors, and testing which theory would fit the data better. In a cross-sectional study (Study 1), 112 Chinese younger adults (M = 22.67, SD = 2.56) were recruited. Aging stereotypes were measured by the Ambivalent Ageism Scale and the abbreviated ageism questionnaire. And their prosocial behaviors were measured by the modified third-party punishment task. The results revealed that high benevolent ageism would increase helping behaviors toward older adults. In the following experiment with aging stereotype priming (positive, neutral vs. negative) among 130 Chinese younger adults (M = 26.82, SD = 3.70), we confirmed the influence of negative aging stereotypes on prosocial behaviors measured by both third-party punishment and Social Value Orientation tasks. Study 2 further demonstrated that pity might mediate the association between negative aging stereotypes and behaviors. Our results indicated that younger adults' negative aging stereotypes could increase their prosociality toward older adults through pity in line with BIAS maps. It also had significant theoretical and practical implications for future research. For example, with more education and intergenerational contact in younger generation which could evoke pity feelings for older adults, could help to build harmonious intergenerational relations. Supplementary Information: The online version contains supplementary material available at 10.1007/s12144-023-04371-0.

The squeaky wheel gets the grease: Recipients' responses influence children's costly third-party punishment of unfairness.

Third-party punishment promotes cooperation by deterring opportunistic behaviors. Even children are willing to pay a cost to implement third-party punishment of unfair behavior. Whether in judicial practice or in daily third-party punishment, people take recipients' feelings into account out of restorative motives. Restorative motives pay attention to both the offenders and the victims and are committed to best repairing harm. This work examined whether children adopt restorative motives by considering recipients' responses when punishing unfair dividers. Participants (N = 128) were 6-, 8-, and 10-year-old Chinese children. Children were shown allocations proposed among a divider and a recipient with response (positive vs. negative) or without response and were asked to accept or pay a cost to reject the allocation. Two experiments indicated that costly third-party punishment increased with age. Furthermore, children took recipients' responses into consideration, with negative responses prompting children to punish more. These findings show that children adopted a restorative view when implementing costly third-party punishment.

Ire and punishment: Incidental anger and costly punishment in children, adolescents, and adults.

Why do children, adolescents, and adults engage in costly punishment to sanction fairness violations? Two studies investigated the differential impact of incidental anger on the costly punishment of 8-year-olds, 13-year-olds, and adults. Focusing on experimentally manipulated incidental anger allows for a causal investigation as to whether and how anger affects costly punishment in these age groups in addition to other motives such as inequity aversion. Study 1 (N = 210) assessed the effect of incidental anger (vs. a neutral emotion) on second-party punishment, where punishers were direct victims of fairness violations. Study 2 (N = 208) examined third-party punishment, where the punisher was an observer unaffected by the violation. Across ages, incidental anger increased the second-party punishment of unequal offers but not equal offers. Thus, anger seems to play a causal role in the punishment of unfairness when fairness violations are self-relevant. As predicted, adults' third-party punishment of unequal offers was higher in the incidental anger condition than in the neutral emotion condition. Children's third-party punishment of unfairness was not affected by the emotion condition, but incidental anger increased adolescents' third-party punishment across offers. Overall, our data suggest that the association between anger and costly punishment is based on the self-relevance of the violation. In third-party situations, where unfairness does not affect the self, social-cognitive processes that develop well into adulthood, such as emotional appraisals, might be necessary for third parties to engage in costly punishment.

Toddlers and infants expect individuals to refrain from helping an ingroup victim's aggressor.

Adults and older children are more likely to punish a wrongdoer for a moral transgression when the victim belongs to their group. Building on these results, in violation-of-expectation experiments (n = 198), we examined whether 2.5-year-old toddlers (Exps. 1 and 2) and 1-year-old infants (Exps. 3 and 4) would selectively expect an individual in a minimal group to engage in third-party punishment (TPP) for harm to an ingroup victim. We focused on an indirect form of TPP, the withholding of help. To start, children saw a wrongdoer steal a toy from a victim while a bystander watched. Next, the wrongdoer needed assistance with a task, and the bystander either helped or hindered her. The group memberships of the wrongdoer and the victim were varied relative to that of the bystander and were marked with either novel labels (Exps. 1 and 2) or novel outfits (Exps. 3 and 4). When the victim belonged to the same group as the bystander, children expected TPP: At both ages, they detected a violation when the bystander chose to help the wrongdoer. Across experiments, this effect held whether the wrongdoer belonged to the same group as the bystander and the victim or to a different group; it was eliminated when the victim belonged to a different group than the bystander, when groups were not marked, and when either no theft occurred or the wrongdoer was unaware of the theft. Toddlers and infants thus expect individuals to refrain from helping an ingroup victim's aggressor, providing further evidence for an early-emerging expectation of ingroup support.

Infants expect leaders to right wrongs.

Anthropological and psychological research on direct third-party punishment suggests that adults expect the leaders of social groups to intervene in within-group transgressions. Here, we explored the developmental roots of this expectation. In violation-of-expectation experiments, we asked whether 17-mo-old infants (n = 120) would expect a leader to intervene when observing a within-group fairness transgression but would hold no particular expectation for intervention when a nonleader observed the same transgression. Infants watched a group of 3 bear puppets who served as the protagonist, wrongdoer, and victim. The protagonist brought in 2 toys for the other bears to share, but the wrongdoer seized both toys, leaving none for the victim. The protagonist then either took 1 toy away from the wrongdoer and gave it to the victim (intervention event) or approached each bear in turn without redistributing a toy (nonintervention event). Across conditions, the protagonist was either a leader (leader condition) or a nonleader equal in rank to the other bears (nonleader condition); across experiments, leadership was marked by either behavioral or physical cues. In both experiments, infants in the leader condition looked significantly longer if shown the nonintervention as opposed to the intervention event, suggesting that they expected the leader to intervene and rectify the wrongdoer's transgression. In contrast, infants in the nonleader condition looked equally at the events, suggesting that they held no particular expectation for intervention from the nonleader. By the second year of life, infants thus already ascribe unique responsibilities to leaders, including that of righting wrongs.

Who initiates punishment, who joins punishment? Disentangling types of third-party punishers by neural traits.

The act of punishing unfair behavior by unaffected observers (i.e., third-party punishment) is a crucial factor in the functioning of human societies. In everyday life, we see different types of individuals who punish. While some individuals initiate costly punishment against an unfair person independently of what other observers do (independent punishers), others condition their punishment engagement on the presence of another person who punishes (conditional punishers). Still others do not want to partake in any sort of punishment (nonpunishers). Although these distinct behavioral types have a divergent impact on human society, the sources of heterogeneity are poorly understood. We present novel laboratory evidence on the existence of these three types. We use anatomical brain characteristics in combination with stated motives to characterize these types. Findings revealed that independent punishers have larger gray matter volume in the right temporo-parietal junction compared to conditional punishers and nonpunishers, an area involved in social cognition. Conditional punishers are characterized by larger gray matter volume in the right dorsolateral prefrontal cortex, a brain area known to be involved in behavioral control and strategic reasoning, compared to independent punishers and nonpunishers. Finally, both independent punishers and nonpunishers are characterized by larger gray matter volume in an area involved in the processing of social and monetary rewards, that is, the bilateral caudate. By using a neural trait approach, we were able to differentiate these three types clearly based on their neural signatures, allowing us to shed light on the underlying psychological mechanisms.

Intrinsic functional connectivity of the frontoparietal network predicts inter-individual differences in the propensity for costly third-party punishment.

Humans are motivated to give norm violators their just deserts through costly punishment even as unaffected third parties (i.e., third-party punishment, TPP). A great deal of individual variability exists in costly punishment; however, how this variability particularly in TPP is represented by the brain's intrinsic network architecture remains elusive. Here, we examined whether inter-individual differences in the propensity for TPP can be predicted based on resting-state functional connectivity (RSFC) combining an economic TPP game with task-free functional neuroimaging and a multivariate prediction framework. Our behavioral results revealed that TPP punishment increased with the severity of unfairness for offers. People with higher TPP propensity punished more harshly across norm-violating scenarios. Our neuroimaging findings showed RSFC within the frontoparietal network predicted individual differences in TPP propensity. Our findings contribute to understanding the neural fingerprint for an individual's propensity to costly punish strangers, and shed some light on how social norm enforcement behaviors are represented by the brain's intrinsic network architecture.

Acute psychosocial stress increases third-party helping but not punishing behavior.

Despite extensive research on the effects of stress on the brain and behaviors, there is a debate whether stress promotes prosocial behaviors, especially acute stress due to intricate costly punishment in the ultimatum game. Therefore, the present study introduced an irrelevant third party to examine how acute stress and the triggered cortisol influence third parties' punishing and helping behaviors as more convincing altruistic behaviors. The 65 participants were exposed to a psychosocial stressor (n = 33) or a control condition (n = 32). Afterwards, two third-party intervention tasks (a token allocation task and criminal scenario judgment task) were completed, during which the participants, as an "irrelevant" third party, could choose whether to sacrifice their own interests to help the victim or punish the transgressor. Participants' affective states, heart rate, and salivary cortisol were repeatedly measured throughout the experiment. Results showed that acute stress can lead to more third-party helping behaviors but not more punishing behaviors. Specifically, participants under stress tended to transfer more monetary units to the victim in the token allocation task than the control-group participants, and they tended to help the victim in the scenario task. In contrast, there was no significant difference in punishing behavior between the stressed and control participants. These findings reveal that acute psychosocial stress triggers the "tend and befriend" response, which might reflect the prosocial intuition under acute stress.

Empathy-based tolerance towards poor norm violators in third-party punishment.

Third-party punishment (TPP) plays an important role in fairness norm enforcement. This study investigated how the economic status of proposers could modulate third parties' behavioural and neural responses to unfairness. Participants played a TPP game as third parties deciding whether to punish proposers after observing the offers from proposers while behavioural and electroencephalogram (EEG) data were recorded. The proposers were of either high economic status or low economic status, and the recipients were middle class. The behavioural results indicated that participants reported decreased punishment for poor-proposed unfair offers compared to rich-proposed unfair offers, and this effect was stronger for highly unfair offers. Neurally, greater P200, a component involved in empathy processing, was observed in response to highly unfair offers (i.e. 90:10 and 80:20) proposed by the poor, suggesting that when the targets of severe punishments were poor proposers, participants showed greater empathy for poor norm violators in highly unfair trials. Taken together, these findings help to elucidate that the third-parties tend to tolerate the norm-violating behaviours conducted by the poor and provided further neuroscience evidence for the influence of economic status of proposers on TPP.

Social status affects how third parties assess unfairly shared losses and unfairly shared gains.

The present paper focused on how a third party's social status affects third-party intervention to maintain social fairness. Study 1 adopted a quasi-experimental design. Selecting high and low-status members of real social groups, we observed that high-status individuals intervened more forcibly and more frequently when assessing the fairness of players' behavior in a dictator game (DG). The manifestation of social status is generally divided into power and economic capital. In Study 2a, using the same DG punishment-compensation paradigm we randomly assigned the third party in the lab to high, medium, and low impact conditions, where their actions had relative multiplier effects on the resources retained by dictators and recipients. This tested whether the power to influence the situation would systematically affect third party's behavior. We found that greater influence predicted increased interventions. Study 2b investigated the influence of economic capital or intrinsic wealth on a third party's altruistic behavior by varying how much capital the third party had at their disposal to spend on punishment or recompensing. Having high capital predicted an increase in the size of penalty inflicted or compensation offered, but resource abundance had no effect on the likelihood that the third party would intervene. In conclusion, the paper showed that the social status of the third party truly does affect their altruistic interventions and the impact of social status on altruistic behaviors for maintaining fairness by the third party occurred primarily through the third parties' perception of their power. Furthermore, the influence of gain and loss contexts and social status on third-party punishment and compensation were independent of each other. This paper provided a new perspective for third-party altruistic behaviors and help to clarify the view of social fairness.

Social norms and cultural diversity in the development of third-party punishment.

Human cooperation is probably supported by our tendency to punish selfishness in others. Social norms play an important role in motivating third-party punishment (TPP), and also in explaining societal differences in prosocial behaviour. However, there has been little work directly linking social norms to the development of TPP across societies. In this study, we explored the impact of normative information on the development of TPP in 603 children aged 4-14, across six diverse societies. Children began to perform TPP during middle childhood, and the developmental trajectories of this behaviour were similar across societies. We also found that social norms began to influence the likelihood of performing TPP during middle childhood in some of these societies. Norms specifying the punishment of selfishness were generally more influential than norms specifying the punishment of prosocial behaviour. These findings support the view that TPP of selfishness is important in all societies, and its development is shaped by a shared psychology for responding to normative information. Yet, the results also highlight the important role that children's prior knowledge of local norms may play in explaining societal variation in the development of both TPP and prosociality.

Why do children punish? Fair outcomes matter more than intent in children's second- and third-party punishment.

Humans punish fairness violations both as victims and as impartial third parties, which can maintain cooperative behavior. However, it is unknown whether similar motivations underlie punishment of unfairness in these two contexts. Here we approached this question by focusing on how both types of punishment develop in children, asking the question: What motivates young children to punish in response to fairness norm violations? We explored two potential factors: the direct experience of unfair outcomes and a partner's fair versus unfair intentions. The participants, 5- and 7-year-olds, were given the chance to engage in both second- and third-party punishment in response to either intended or unintended fairness norm violations in a single paradigm. In both age-groups, children were more likely to punish when they were directly affected by the allocation (second-party punishment) than when they were an uninvolved third party (third-party punishment). Reliable third-party punishment was shown only in the older age-group. Moreover, children's punishment was driven by outcome rather than intent, with equal rates of punishment when unequal outcomes were either the result of chance or the intentional act of another child. These findings suggest that younger children may be mainly motivated to create equal outcomes between themselves and others, whereas older children are motivated to enforce fairness norms as a general principle.

Gratitude increases third-party punishment.

Third-party punishment occurs when a perpetrator of a transgression is punished by another person who was not directly affected by the transgression (i.e. a third-party). Given gratitude's demonstrated ability to enhance both cooperation and the value people place on future-rewards, its capacity to increase third-party punishment - a phenomenon theorised to increase future cooperative behaviour - was investigated. In two experiments, participants were randomly assigned to experience one of three emotional states (i.e. gratitude, happiness, or neutrality) prior to making decisions about how much of a previous financial endowment they would spend to punish a person who transgressed against another at differing degrees within the context of a dictator game. As expected, punishment expenditures decreased for all participants as a dictator's decision became fairer. Of primary interest, however, participants who felt grateful, as compared to those who felt neutral or happy, engaged in significantly more third-party punishment across dictator splits that were not altruistic in nature.