Individual Submission Summary
Share...

Direct link:

Poster #196 - Children’s Beliefs about the Controllability of Thoughts, Actions, and Feelings

Fri, March 22, 2:30 to 3:45pm, Baltimore Convention Center, Floor: Level 1, Exhibit Hall B

Integrative Statement

BACKGROUND: During the elementary school period, children form beliefs about agency and action control (Kushnir et al., 2015; Nichols, 2004). For example, 6-year-olds (but not 4-year-olds) believe that a person can desire one thing (to eat cake) but do something different (eat broccoli). The current study expands on this work by examining children’s beliefs about how much control a person can have over another’s actions and mental states. Moreover, we ask whether children treat control of their own mental states and actions differently from those of others.

METHOD: Three- to 6-year-olds (N=77), 36 boys and 41 girls, heard stories in which one character (the agent) attempted to control another (the patient). Each story featured 1 of 4 trial types using a 2 (agent: child versus other) x 2 (patient: child versus other) design. Additionally, each story featured 1 of 3 domains of control: thinking, feeling, and doing. Following each story, participants were asked whether the agent could control the patient within the featured domain [e.g., Can X make Y think (or feel, or do)?] Table 1 provides a script for the “think” stories for each of the four trial types. The scripts for the other domains were structured similarly.

RESULTS: Responses to the make question were analyzed via a General Estimating Equation assuming a binary logistic response. Trial type and domain were analyzed as within-subject factors; gender and age (in months) were between-subject variables. This model was then rerun, removing nonsignificant interactions, which resulted in a better fit to the data (as measured by QICC = 1125.49 vs. 1182.34). The final model revealed a significant effect of trial type, Wald χ2(3) = 13.83, p = .003, a significant effect of age, Wald χ2(1) = 5.31, p = .03, and a significant interaction between trial type and age, Wald χ2(3) = 15.80, p = .001. Overall, older children were more likely to say that the agent could not control the patient; this effect was driven by the two trials in which a stranger was the recipient of the action. Independent of age, when children were agents, they were more likely to respond that they had control than when another person was the agent, b = -5.59, SE = 1.71, Wald χ2(1) = 10.72, p = .001. Finally, comparisons to chance showed that children were likely to say “yes” to the make question. This was true overall, by domain, and by trial type (ps < .01), see Figure 1.

CONCLUSIONS: These findings suggest children believe that people can be controlled, and that this can happen across a variety of domains including those touching on a person’s physical and epistemic functioning. This was particularly true for trials in which the child was the agent (i.e., self-other and self-self trials). On these, children were more likely to answer “yes” when asked if they could “make” another person do, think or feel something. Findings provide important insights into children’s beliefs about when and how they can affect, or be affected by, the world around them.

Authors