Is Seeing Believing: Children’s Trust Testimony and Demonstration in Physical and Psychological Contexts

• In Event: Children’s selective trust across domains, agents, and cultures

Abstract

While children learn via the testimony of other people, they can also procure evidence from others’ demonstrations. Recent work suggests that children take both the strength of observational evidence and the reliability of the informant into account when deciding what to believe and what not to believe (Bridgers et al., 2016; Gualtieri et al., 2019). However, little is known about the contextual factors that guide children's choices between demonstrated and verbally communicated information. Here, we examined children’s trust in what they hear from one agent and what they see from another in both physical and psychological domains across both lower- and higher-stakes contexts.
In a sample of 207 5- to 10-year-old children and 69 adults, participants were randomly assigned to two between-subjects conditions: a Neutral condition and a High-stakes condition. In each trial, children heard stories of two agents, one who was described as using action-based demonstrations (e.g., “Sara takes out crayons from the box one by one, shows them to you, and they are yellow”), and the other who was described as making direct statements to the child (e.g., “Katie looks into the box, and then tells you, ‘There are yellow crayons in the box.’”). To explore whether children's distrust for one source would be more pronounced when the stakes of uncovering the truth were higher, a High-stake condition was included in which participants were directly involved in the stories. In addition, children were presented with scenarios in both the physical (e.g., content of a box) and psychological (e.g., one’s food preferences) domains. After each presentation, participants were asked to choose the less trustworthy source of information, and answered a series of questions assessing their subsequent epistemic and social judgments of the demonstrator and the testifier.
We found that across all age groups, participants were significantly more likely to distrust the testifier at above chance levels in the Physical trials. By contrast, in the Psychological trials, participants were more likely to distrust the demonstrator in the Neutral condition, but less so in the High Stakes condition (see Figure 1). With age, children were more likely to differentiate between the two conditions (OR = 1.59, p = 0.001) and the two types of trials (OR = 0.67, p = 0.001) in their distrust responses. Participants’ selective learning choices and social preferences followed similar patterns, in which they were more likely to selectively learn from and affiliate with the testifier in the Neutral condition compared to the High-stakes condition, and more in the Psychological trials compared to the Physical trials.
Taken together, our results suggested that children were flexible when it comes to using demonstration and testimony as sources of information. With age, they become more adult-like in their trust decisions, and understand that while demonstrated information might be more trustworthy when evaluating the physical existence of objects, testimony could be more reliable when it comes to revealing psychological states. Furthermore, when the stakes are higher, children understood that seeing for oneself might be better than believing what is told.

Authors

• Pearl Han Li, Duke Univeristy
  Presenting Author

• Narges Afshordi

• Melissa A. Koenig, University of Minnesota - Twin Cities