Computational signatures of inequity aversion in children across seven societies.

[Correction Notice: An Erratum for this article was reported online in Journal of Experimental Psychology: General on Aug 10 2023 (see record 2023-96713-001). In the original article, there were affiliation errors for the first and 14th authors. The affiliations for Dorsa Amir are Department of Psychology, University of California, Berkeley; and Department of Psychology, Boston College. The affiliation for Katherine McAuliffe is Department of Psychology, Boston College. All versions of this article have been corrected.] Inequity aversion is an important factor in fairness behavior. Previous work suggests that children show more cross-cultural variation in their willingness to reject allocations that would give them more rewards than their partner-advantageous inequity-as opposed to allocations that would give them less than their partner-disadvantageous inequity. However, as past work has relied solely on children's decisions to accept or reject these offers, the algorithms underlying this pattern of variation remain unclear. Here, we explore the computational signatures of inequity aversion by applying a computational model of decision-making to data from children (N = 807) who played the Inequity Game across seven societies. Specifically, we used drift-diffusion models to formally distinguish evaluative processing (i.e., the computation of the subjective value of accepting or rejecting inequity) from alternative factors such as decision speed and response strategies. Our results suggest that variation in the development of inequity aversion across societies is best accounted for by variation in the drift rate-the direction and strength of the evaluative preference. Our findings underscore the utility of looking beyond decision data to better understand behavioral diversity. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Children's collaboration induces fairness rather than generosity.

Children across diverse societies reject resource allocations that place them at a disadvantage (disadvantageous inequity aversion; DI). In certain societies, older children also reject advantageous allocations (advantageous inequity aversion; AI). Other work demonstrates that after collaboration, children reduce inequity by sharing. However, it is unknown whether collaboration leads to greater sharing because it encourages prosociality (Generosity Hypothesis) or because collaboration elicits stronger equitable tendencies (Equity Hypothesis). Here we use measures of inequity aversion that can disambiguate between these hypotheses. We tested 7- to 13-year-old children from rural India, a population that has shown DI but not AI, and 4- to 10-year-old children from rural Canada, a population that shows both AI and DI. Pairs of children worked either collaboratively or in parallel obtaining candy that was then used in a test of DI and AI. Results showed that in both societies collaboration did not encourage children to accept DI offers, providing evidence against the Generosity Hypothesis. However, in both societies older children demonstrated AI after collaboration but not after parallel work. For children in India AI emerged in ages where it had not been previously observed and children in Canada showed AI during early middle childhood. This suggests that collaboration can induce a willingness to sacrifice an advantage to achieve equity, consistent with the Equity Hypothesis.

Give as I give: Adult influence on children's giving in two cultures.

Adult influence on children's altruistic behavior may differ between cultural communities. We used an experimental approach to assess the influence of adult models on children's altruistic giving in a city in the United States and rural villages in India. Children between 3 and 8 years of age were tested with their parents in the United States (n=163) and India (n=154). Parents modeled either a generous or stingy donation; children then performed a similar task in private. Children in both communities were influenced by the stingy model, but only children in India increased their giving after viewing a generous model. The model's influence also increased with age in India. Results of a questionnaire revealed that parents in both communities believed that children learned sharing behavior from them. We consider these results in light of differences between these societies, including different socialization goals, cultural values, and content biases that may affect altruistic giving.

Modeling referential actions in 6- to 18-month-old infants: a precursor to symbolic understanding.

Social precursors to symbolic understanding of pictures were examined with 100 infants ages 6, 9, 12, 15, and 18 months. Adults demonstrated 1 of 2 stances toward pictures and objects (contemplative or manipulative), and then gave items to infants for exploration. For pictures, older infants (12, 15, and 18 months) emulated the adult's actions following both types of demonstration trials. For objects, infants did not emulate actions following either stance at any age. The findings suggest that infants enlist their imitative learning skills in the context of learning the conventions of action on pictorial symbols. The data are interpreted as pointing to the importance of social learning in developing an understanding of the referential function of pictorial symbols.

Emergence of graphic symbol functioning and the question of domain specificity: a longitudinal training study.

The impact of social scaffolding on the emergence of graphic symbol functioning was explored in a longitudinal training study. Links among graphic, language, and play domains in symbolic development were also investigated. The symbolic functioning of 16 children, who were 28 months at the outset of the study, was assessed in comprehension and production tasks across the three domains at monthly intervals from 28 to 36 months, and again at 42 months. Training was delivered in between monthly assessments during weekly visits. Half of the children received training, which consisted of the experimenter drawing common objects and highlighting the relation between pictures and their referents, for 16 consecutive weeks early in the study (early training, ET). The remaining half received a placebo version of training for these 16 weeks, followed by actual training for 4 weeks in the fifth month (late training, LT). After the first 4 months of training the ET group was found to have accelerated comprehension and production of graphic symbols relative to the LT group. After the fifth month, the LT group reached the same level of graphic symbol performance as the ET group. There were strong positive correlations found among graphic symbol functioning and language and play, and between play and language. These findings support the view that graphic symbolic development can be influenced by cultural scaffolding, that more extensive training is needed early rather than later in development, and that interrelationships exist among symbolic domains.