Belief-Based Evaluations of Helping in 15-Month-Old Infants

• In Event: Mental State Understanding and Moral Development

Abstract

Adults’ moral evaluations depend on an analysis of not only whether an individual has caused a positive or a negative outcome, but what mental states drive an individual’s actions. Theory of mind, the ability to represent and understand others’ mental states, is critical to moral cognition in adults. In two experiments, we explored whether mentalistic belief-based moral evaluation may have origins in infancy. Despite early findings that infants can represent others’ false beliefs in nonverbal tasks, there have since been notable failures to replicate these early findings (Poulin-Dubois et al., 2018). It therefore remains a largely open question whether infants can represent others’ false beliefs, and even if they can, whether they can incorporate others’ false beliefs in their social evaluations. We build on a body of research that has demonstrated that infants evaluate others on the basis of their social actions, preferring individuals who help others over individuals who hinder others (see Margoni & Surian, 2018). Importantly, by late in the first year, infants not only positively evaluate helpers over hinderers, but take into account and privilege others’ mental states (e.g., intention, ignorance) when making their evaluations (Hamlin et al., 2013; Woo et al., 2017). These findings suggest that infants may form robust inferences about others’ mental states when viewing agents in social interaction with one another. Could infants also incorporate others’ false beliefs in their social evaluations? In two preregistered experiments, fifteen-month-old infants (N = 94) viewed a protagonist who sought to obtain one of two objects, each inside a different box, and two helpers alternately opened the box containing that object. Then the two objects switched boxes, either in the helpers’ presence or absence, and infants saw one helper open the new box, affording access to the desired object, and the other helper open the original box, affording access to the forsaken object. When both helpers had witnessed the change in object locations, infants preferentially reached for and looked at the former helper, who acted to make the desired object available in its new location. In contrast, when neither helper had witnessed the change in object locations, infants preferentially reached for and looked at the helper who opened the original box where the two helpers had last seen the desired object. These findings provide evidence that infants inferred the beliefs of the helpers from the events they did or did not witness, and that infants evaluated the helpers in accord with their inferred beliefs. Belief-based social evaluation thus occurs early in the second year. In light of recent, mixed evidence for implicit false-belief understanding in infancy, the present experiments raise the possibility that belief inferences occur more readily when infants are challenged to infer the beliefs of a character whose intentions are social: to help another character fulfill its desires. Future research should systematically test whether more richly social contexts elicit more robust mental state inferences in infants than do the scenarios used in many implicit tests of false-belief understanding, in which infants only see a single agent.

Authors

• Brandon Matthew Woo, Harvard University
  Presenting Author

• Elizabeth S. Spelke, Harvard University
  Non-Presenting Author