Smart or Nice?: The Role of Trait Understanding in Preschoolers' Face-based Inferences

• In Event: Poster Session 12
  In Poster Session: PS 12 Topic - Social Cognition

Abstract

Five-year-olds are able to make inferences about others’ future behavior using face-trait information, whereas 4-year-olds’ inferences are less consistent (Charlesworth et al., 2019). When Palmquist and DeAngelis (under review) presented children with two positively-valenced faces that depicted different traits, only 5-year-olds consistently associated competent-looking faces with knowledge and trustworthy-looking faces with social skills. The present study was designed to explore two explanations for this developmental shift. A low-level explanation: 5-year-olds are better than 4-year-olds at perceiving subtle differences in others’ facial features. And a trait-based explanation: 5-year-olds are better able to use traits to make inferences about others.

Here, 28 4- and 5-year-olds participated in two tasks. The first measured children’s ability to match competent- and trustworthy-looking faces with other similarly competent- and trustworthy-looking faces to determine whether they could perceive subtle differences in their features (Todorov, 2015; Figure 1.). The second measured children’s trait understanding by presenting them with pairs of faces, one competent- and one trustworthy-looking. These faces were also labeled with traits: the competent-looking faces as smart and the trustworthy-looking faces as nice. Next, children were asked to decide which of the two faces would know more about unfamiliar objects or would be a better sharer.

On face perception, 4-year-olds matched the target face to its corresponding group of faces 80.5% of the time. Similarly, five-year-olds matched the target face to its corresponding group of faces 80.2% of the time. A chi-square test showed that both 4- and 5-year-olds did not differ significantly from chance on this task (p’s < 0.001). The age groups also did not differ from one another (p = 1.00). The COVID-19 pandemic has delayed our coding of these face perception data to look for differences in performance by face type (competent and trustworthy).

On trait understanding, 4-year-olds chose the competent-looking face on 59.4% of object-function and 36.7% of sharing questions. On the other hand, 5-year-olds chose the competent-looking face on 72.9% of object-function and 20.8% of sharing questions. A chi-square test indicated that 4-year-olds’ choices did not differ significantly from chance on object-function (p = 0.13) questions, but did differ on sharing questions (p = 0.032). Five-year-olds’ choices did differ significantly from chance on object-function (p = 0.0011) and sharing (p = 0.000024) questions.

We estimated the probability of choosing the competent-looking face as a function of age (4-year-olds vs. 5-year-olds) and question type (object-function and sharing). A binary logistic regression indicated a significant interaction such that 5-year-olds responded differently to object-function (p = 0.001) and sharing (p = 0.001) questions than did 4-year-olds. In other words, 5-year-olds, but not 4-year-olds, were more likely to consistently choose competent-looking faces on object-function questions and trustworthy-looking faces on sharing questions. Therefore, although both age groups are similarly able to perceive differences in facial features (as indicated by the face perception task), 5-year-olds are able to reliably use trait information to select different kinds of faces for different kinds of tasks, whereas 4-year-olds’ use of trait information is less clear.

Authors

• Julia P Ruggiero, Amherst College
  Presenting Author

• Carolyn M Palmquist, Amherst College
  Non-Presenting Author