Poster #196 - Looking Smart: Preschoolers Judge Others' Knowledge Based on Their Facial Appearance

• In Event: 1-169 - Poster Session 04
  In Poster Session: PS 04 Section - Social Cognition

Integrative Statement

Adults judge others using facial appearance (Todorov, 2017). However, little is known about the origins of this bias. In Cogsdill, Todorov, Spelke, and Banaji (2014), children matched faces varying in trustworthiness with the labels ‘nice’ and ‘mean’, but were less likely to match faces varying in competence with ‘smart’ and ‘not smart’. Here, we investigated whether children would be more likely to associate knowledge with positively-valenced (trustworthy) faces, given that children identify trustworthiness from infancy (Jessen & Grossman, 2016). We also simplified the task by asking children to associate faces with behaviors, not trait labels.

Sixty-four preschoolers (M = 57 months, SD = 6.96 months; 30 girls) participated in a competence or trustworthiness condition; both featured two computer-generated puppets with faces that appeared more or less competent or trustworthy (Figure 1.). First, the puppets’ faces were covered with identical masks. The experimenter explained that the masked puppets would take turns identifying different objects and children needed to guess which puppet had taken a turn.

Test trials always began with both puppets out of view. The experimenter then brought out a third “dummy” puppet. This “dummy” puppet had the same mask glued to its face (preventing children from seeing its features and discriminating between it and the other puppets). This puppet was asked about the functions of familiar objects (e.g., “Which one unlocks doors?”). During knowledgeable trials (4), the puppet indicated a correct object (e.g., keys) and on unknowledgeable trials (4), the puppet indicated an incorrect object (e.g., comb). Afterward, the “dummy” puppet was switched for the test puppets, their masks were removed, and children guessed which puppet they had seen.

Although children in the competence condition did not choose the ‘expected’ puppet more on knowledgeable or unknowledgeable trials, children in the trustworthiness condition did (Figure 2.).

Children in the trustworthiness condition chose the ‘expected’ puppet more often than children in the competence condition, t(62) = -1.82, p = 0.074, d = -0.46, (Figure 2). Children in the trustworthiness condition associated the negatively-valenced puppet with unknowledgeable behavior more often than children in the competence condition, t(62) = -2.39, p = 0.019, d = -0.61, while children did not differ in their associations of the positively-valenced puppets with knowledgeable actions across the two conditions, t(62) = -1.26, p = 0.214.

Children associate particular faces with more or less knowledgeable behaviors. However, preschoolers’ inferences about others’ knowledge appear to be based on broadly positive traits (trustworthiness) rather than on specific traits typically associated with knowledge (competence).

Authors

• Erika DeAngelis, Amherst College
  Presenting Author

• Carolyn Palmquist, Amherst College
  Non-Presenting Author

• Erik W Cheries, University of Massachusets at Amherst
  Non-Presenting Author