Infants’ Generalization Across Moral Subdomains

• In Event: Poster Session 04
  In Poster Session: PS 04 Topic - Moral Development

Abstract

Evidence suggests that adults use information about an individual’s moral character to make inferences about how they will behave across contexts (Siegel, Crockett & Dolan, 2017). Moreover, past work suggests that roots of moral sensitivity can be traced back to infancy. Preverbal infants prefer agents who help others over those who hinder others in their goal pursuit (Hamlin & Wynn, 2011; Hamlin, Wynn & Bloom, 2010), and by the start of the second year of life, infants showed pronounced attention to unequal resource distributions (Schmidt & Sommerville, 2011), prefer individuals that distribute resources equally (Geraci & Surian, 2011; Burns & Sommerville, 2014) and perceive fair individuals to be morally praiseworthy (Deschamps, Eason & Sommerville, 2018). While it is relatively unknown whether infants generalize behavior from one moral subdomain to another, existing evidence suggests that they may: 15-month-old infants expect agents who previously helped another agent to distribute resources equally to recipients (Surian, Ueno, Itakura & Meristo, 2016). We investigated whether infants expect agents who are fair to subsequently be helpful, and for agents that are unfair to subsequently be unhelpful.
Twelve- to 24-month old infants (n = 34; in progress) infants saw a protagonist distributing cookies between two individuals either fairly or unfairly during familiarization trials. During test trials, the protagonist either helped a new agent achieve her goal (by opening a box to allow her to retrieve her toy) or hindered the new agent from achieving her goal (by closing the box lid). Preliminary results show that infants in both conditions look longer to the helping event than the hindering event: unfair condition, t(15) = 2.507, p = .024; fair condition, t(17) = 2.373, p = .030. A prior control experiment established that, after receiving neutral information about a protagonist, infants look equally to the helping and hindering actions, (t(32) = .46, p = .65, r = .65), suggesting that our findings are unlikely to stem from a baseline preference for the helping event.
One interpretation of our findings is that infants may be focusing on the fact that the protagonist is giving away resources thereby construe both distributions as generous on this basis. This possibility is supported by the fact that in our experiment, the faces of the recipients are not featured (infants only see the backs of their heads), whereas in prior work, recipients’ side profiles are visible (Lucca, Pospisil & Sommerville, 2018). Thus, Experiment 3 will investigate whether the current findings extend to situations in which the resource recipients’ faces are visible. We predict that when the recipient faces are visible, infants may perceive the distributions as fair/unfair thereby influencing their expectations of the distributor in help/harm scenarios (i.e., that the actor will act in a consistent manner). If our prediction holds true, our findings would suggest that infants can generalize behavior across moral domains, at least under some circumstances. Together, the results of these studies will shed light on not only the scope of infants’ moral inferences, but also the contextual factors that influence them.

Group Authors

University of Toronto, Department of Psychology

Authors

• Jessica A Sommerville, University of Toronto
  Presenting Author

• Inderpreet Kaur Gill, University of Toronto
  Non-Presenting Author