Young children’s ability to rationally revise and maintain others’ beliefs about the self

• In Event: Children’s Construction of the Self Across Contexts, People, and Time

Abstract

We are deeply curious about the self: Who am I and what am I good at? To answer these questions, we often go beyond our own experiences and look to what others know and believe about us. However, despite decades of work on children’s belief-reasoning capacities (e.g., Wellman et al., 2001), little is known about the extent to which young children can infer others’ beliefs about them (the self). Here we investigate the cognitive capacities that support young children’s abilities to infer others’ beliefs about the self and subsequently communicate to manage such beliefs.

Recent work has suggested that 3- and 4-year-old children can use others’ prior observations of their performance outcomes (i.e., their own failures and successes) to not only infer what others believe about their competence, but also actively communicate to revise these beliefs (Authors, 2018, under revision). Here, children showed their success to an observer who only previously saw their failures, effectively changing the observer’s beliefs to be both more accurate and desirable (children could truly make the toy go, and being competent is more desirable than being incompetent). Building on this work, we ask whether children may be motivated to maintain others’ inflated beliefs about the self––that is, beliefs that are inaccurate, but desirable.

In Experiment 1 (preregistered; Age: 3-4; N=43), an observer came in to look at (what she thought was) the participant’s drawing of a tree. Later, children found out that the observer was actually looking at a drawing either much better than their own (i.e., the observer had an inaccurate, but desirable belief about the child’s drawing abilities; Better Condition) or worse than their own (i.e., an inaccurate and desirable belief; Worse Condition). When provided with the opportunity to draw or play with blocks with the observer, participants were more likely to choose blocks in the Better Condition (78.3%) than in the Worse Condition (45%; p = .032, Fisher’s Exact Test).

Even though the observer held a “false” belief in both conditions, these results suggest that children were more likely to avoid doing more drawing in the Better Condition, suggesting that they wanted to maintain the observer’s inflated belief.

In an ongoing conceptual replication (conducted online over Zoom), 4-year-old participants will make a tracing and then send their tracing to two observers. One observer will accidentally receive a worse tracing (as in Study 1 Worse Condition), and the other will receive a better tracing (as in Study 1 Better Condition). When asked which observer to send their actual tracing to, we predict that participants will select the observer who received the worse tracing.

So far, our work suggests that children can reason about the desirability of others’ beliefs and may even avoid certain activities to maintain others’ inflated beliefs. These findings have deep implications for children’s learning decisions in social contexts, as well as their mental health and well-being (e.g., imposter syndrome).

Authors

• Mika Asaba, Yale University
  Presenting Author

• Hyowon Gweon, Stanford University
  Non-Presenting Author