About this paper symposium

---

Panel information
Panel 13. Moral Development

Paper #1
Does the emergence of third-party punishment co-vary with community size?
Author information	Role
Dr. Julia Marshall, Ph.D., Boston College, United States	Presenting author
Abigail McLaughlin, Boston College, United States	Non-presenting author
Chang (Amy) Lu, Boston College, United States	Non-presenting author
Shalini Gautaum, Boston College, United States	Non-presenting author
Shashwati Shubhadarshini, Mystery of Mothers, India	Non-presenting author
Patrick Tusiime, Kibale Forest Schools Program, Uganda	Non-presenting author
Emily Otali, Kibale Forest Schools Program, Uganda	Non-presenting author
Katherine McAuliffe, Boston College, United States	Non-presenting author
Abstract
A willingness to punish on behalf of others—third-party punishment (TPP)—is one of many ways in which humans help to maintain large-scale cooperative societies and has been the long-standing focus of developmental inquiry. Indeed, many studies from the past decade reveal that children around age six engage in TPP in response to a variety of transgressions, although there is considerable variability across countries (Marshall & McAuliffe, 2022). Despite these findings, we still have yet to understand the community-based factors that support the development of TPP. One factor that plays a prominent role in predicting TPP in adults is the number of people living within a particular community, so-called community size (Marlowe et al., 2014). In particular, the size of a community positively predicts the extent to which adults engage in TPP. The rationale for this relationship is that, within smaller societies, many mechanisms, such as reputational ones (i.e., gossip), can help sustain cooperation in the absence of direct TPP. Within large-scale societies, however, reputational mechanisms are not as effective because information does not propagate throughout large communities as efficiently, making TPP more valuable in such circumstances. Given the role of community size in adults’ TPP, a question remains about when in development community size also shapes children’s TPP. Here, we address this by using a developmental approach. Specifically, we tested the prediction that community size should interact with age to predict greater TPP at earlier ages in larger communities. To test this, we implemented a well-established TPP task (McAuliffe et al, 2015) in four different countries (all pre-registered)—China (n = 334), India (n = 349), Peru (n = 255), and Uganda (n = 226)—with children between the ages of six and twelve. Children heard stories about mean (e.g., withholding resources) and neutral actions (e.g., sharing resources) and decided whether such individuals should keep valued resources (non-intervention) or whether they should be taken away (punishment). Within each country, we tested children in an urban and rural location (as a proxy for community size). Preliminary analyses point toward trends in the predicted direction, but they do not fully support our hypotheses. Even though children across different countries punish antisocial actions relative to neutral ones, we do not find that this tendency is particularly prominent amongst younger children in urban societies where there are presumably stronger norms about engaging in TPP, χ2(1) = 1.20, p = .273. Instead, children appear generally more inclined to punish with age, r = .13, p < .001, and this tendency is not statistically different between rural, r = .17, p < .001, and urban communities, r = .09, p = .011 (Figure 1). These results show that children punish transgressions but also indicate that community size plays less of a role in the development of TPP than would be anticipated from the adult literature.

Paper #2
How Do Children Behave Towards Transgressions? -the Choice between Third-Party Punishment, Forgiveness, and Non-Intervention
Author information	Role
Rizu Toda, Osaka Uniersity, Japan	Presenting author
Nazu Toda, Osaka University, Japan	Non-presenting author
Hiromichi Hagihara, Osaka University, Japan	Non-presenting author
Yasuhiro Kanakogi, Osaka University, Japan	Non-presenting author
Abstract
When norms for maintaining cooperative relationships are violated, humans try to cope in various ways (e.g., Fehr & Fischbacher, 2004). Previous studies have independently explored two important responses to norm violations—third-party punishment and forgiveness. Third-party punishment is punishment for an offense that does not directly harm oneself (Fehr & Gachter, 2000). Children begin to engage in third-party punishment around the age of four, even when it incurs a cost (e.g., Yang et al., 2018). Forgiveness, that is, prosocial change toward the transgressor, is another potential response to norm violations (McCullough et al., 2000). Children exhibit forgiving behaviors in early childhood (e.g., Oostenbroek & Vaish, 2019). Previous research paradigms have equated “not punishing” with “forgiving” and “not forgiving with “punishing.” However, in real-life situations, an individual who refrains from punishing does not necessarily have prosocial intentions toward the transgressor, and one who refrains from forgiving does not always expect negative results on the transgressor. Therefore, the conventional view of punishment and forgiveness as two sides of the same coin is inaccurate. In this study, we developed a new experimental paradigm that includes three responses to norm violations (Punishment, Forgiveness, and No Intervention). Additionally, we examined how these behaviors change in response to violations depending on the transgressor’s apology (Experiment 1) and the cost of intervention (Experiment 2). In Experiment 1, 41 6-year-olds in Japan were asked to choose one of three actions toward a transgressor who did not apologize (No-Apology Condition) or apologized (Apology Condition). The children could choose to take the transgressor’s toy (Punishment), give the transgressor a new attractive toy (Forgiveness), or neither take nor give a toy (No intervention). Only when choosing Punishment did the participants incur the cost of giving up their playtime with the toy. The results showed that the proportion of children who chose Punishment was the lowest among the three options in either condition (No Apology: 17.1%; Apology: 19.5%; Figure 1). In the Apology Condition, children were likelier to choose Forgiveness (Apology: 48.8%; No Apology: 29.2%) and less likely to choose No Intervention (Apology: 34.1%; No Apology: 51.2%). In Experiment 2, another 41 6-year-olds in Japan chose between Punishment, Forgiveness, or No Intervention toward a transgressor who did not apologize in two conditions—a Cost Condition, identical to Experiment 1’s No-Apology Condition, where the punishment involved a cost, and a No-Cost Condition, where there was no such cost. Significantly more children chose Punishment in the No-Cost Condition (43.2%) than in the Cost Condition (13.6%, Figure 2). However, the presence or absence of costs did not significantly affect the choice of Forgiveness (No Cost, 25.0%; Cost, 31.8%) or No Intervention (No Cost, 31.8%; Cost, 54.5%). In summary, when presented with options besides punishment, such as forgiveness and non-intervention, the proportion of children choosing to punish a transgressor was extremely lower than observed in prior work. The findings demonstrate how children respond to transgressors in contexts closer to real-life situations and how their responses vary depending on social information.

Paper #3
Getting away with it: Does it feel good not to be punished?
Author information	Role
Dr. Katarzyna Myslinska Szarek, University of Michigan, University SWPS in Poland, United States, Poland	Presenting author
Felix Warneken, University of Michigan, United States	Non-presenting author
Abstract
Fairness and justice are integral to human social life, guiding how individuals and societies function. Research has shown that children are attuned to fairness from an early age and understand that harmful actions should result in punishment (Ziv& Sommerville, 2016; Li et al., 2016; Kanakogi et al., 2022). However, less is known about how children respond when punishment is applied unequally towards two transgressors for the same norm violation: (1) How do children perceive and react to unequal punishment for the same transgression? (2) What emotions do they attribute to both the punished and unpunished individuals? In two experiments, children aged 6–9 from the US and Poland were presented with scenarios involving two characters who both broke the rules, but a teacher punished only one of them and let the other one get away with it. Specifically, the first character violated a school rule and was either punished or not. Subsequently, a second character committed the same act but received the opposite consequence than the first character (Figure 1). Children rated the characters' behavior, predicted their emotions before and after the teacher's decision, and suggested how the teacher should resolve the inequity caused by punishing only one individual. We hypothesized that: (i) children would endorse the same punishment for identical transgressions; (ii) children would ascribe mostly negative emotions to the punished character and more positive emotions to the unpunished one; (iii) as children grow older, they would be more likely to attribute mixed emotions to both the punished and unpunished characters. In Study 1, N= 91 children (Mage= 7.49, SD = 1.12) expected consistent consequences for similar transgressions. The results showed that children endorsed punishment or non-punishment depending on whether the first character had been punished or not. This initial teacher’s decision strongly shaped children’s judgments, χ2 (2,91) = 53.74, p < .001, setting a clear precedent for children’s solutions to unequal treatment (Figure 2). When asked to describe the characters' emotions, children attributed mixed emotions to both. The first character's emotions were directly influenced by the teacher's decision—feeling good when spared and bad when punished—while the second character’s emotions, mainly negative, reflected the unfairness of unequal treatment. Study 2, involving N=98 children (Mage = 7.42, SD = 1.10), aimed to replicate these findings using an improved method to measure mixed-emotion attributions by asking children to choose between different emotions by displaying emoticons. The results confirmed that children consistently expected equal treatment and attributed mixed emotions to the characters after the punishment decisions. The punished character was often seen as feeling both relief and sadness, while the unpunished character was associated with feelings of guilt alongside happiness. Our results highlight children's nuanced understanding of fairness in punishment and suggest that our sample children had a preference for individuals to receive the same treatment, rather than a preference for transgressors being punished or not. We further discuss the possibility children attribute mixed rather than just simply negative emotions since being punished might also alleviate a transgressor's guilt.

Paper #4
Children learn what is right or wrong selectively from a legitimate authority’s punishment
Author information	Role
Young-eun Lee, MIT, United States	Presenting author
Setayesh Radkani, Massachusetts Institute of Technology, United States	Non-presenting author
Rebecca Saxe, Massachusetts Institute of Technology, United States	Non-presenting author
Abstract
Children must learn the norms of their society. One source of information is to see which behaviors parents, teachers, and other authorities punish. Adults deliver and receive punishment as communication about socio-moral norms (Sarin et al., 2021). Yet, adult learners make a critical distinction. Socio-moral norms are learned and internalized from punishment imposed by an authority perceived as impartial, unbiased, and procedurally just (i.e., legitimate; Tyler, 2006). By contrast, adults distrust a biased or self-serving (thus illegitimate) person’s punishment (Mulder, 2018). Do children similarly learn socio-moral norms from punishment, and do they do so selectively from legitimate authorities? Here, using within-subject experiments, we test (i) whether young children learn that a punished act is wrong, (ii) whether they do so selectively from legitimate punishment, and (iii) if so, by what age. To ensure that children had no prior knowledge of the relevant norms, we used hypothetical vignettes set in a faraway society. In three pre-registered studies, 6- to 11-year-old children (total n=196) heard vignettes, in which one character engaged in an unknown target action (e.g., daxing). Critically, the authority was described as legitimate or illegitimate (see Fig.1C). Then, the authority chose to do nothing or punish the target by making them go to a time-out or by taking a cookie. Our dependent measure was children’s inferences on how right or wrong the target action (e.g., daxing) was, ranging from 0 (“very wrong”) to 5 (“very right”). In Study 1, 9- to 11-year-olds inferred that e.g., daxing is more wrong when it is punished by legitimate authorities (versus illegitimate authorities) across motive and designation cues (Fig.2B), β=-0.36, SE=0.09, p<.001. However, this tendency was not observed in 6- to 8-year-olds, β=-0.10, SE=0.13, p=.433. In Study 2, 9- to 11-year-olds inferred that e.g., daxing is more wrong when legitimate authorities punished it than when they did nothing (β=1.24, SE=0.28, p<.001), while their inference did not change depending on illegitimate authorities’ punishment decisions (β=-0.05, SE=0.28, p=.865). In contrast, 6- to 8-year-olds inferred that daxing is more wrong when it is punished than when it is not (β=-0.32, SE=0.10, p=.003) regardless of legitimacy. These results suggest that 9- to 11-year-olds learn new norms from legitimate authorities selectively. However, Studies 1 and 2 may have underestimated younger children’s sensitivity to legitimacy, by requiring them to make inferences in unfamiliar vignettes with alien-like characters (Fig.1A). In Study 3, we presented 6- to 8-year-olds with scenarios involving more familiar authorities (i.e., teacher; see Fig.1B) in a classroom setting where punishment by time-out frequently occurs. In Study 3, 6- to 8-year-olds learned selectively from legitimate teachers (β=-0.23, SE=0.10, p=.027), showing responses similar to older children. This work delineates a key cognitive mechanism by which children learn the norms of their society from authorities like parents and teachers.

⇦ Back to session