Poster #71 - Examining Whether Children’s Automatic Encoding of Social Categories is Intersectional

• In Event: Poster Session 05

Abstract

Race and gender emerge as meaningful and salient social categories in early development (Bennett et al., 2000; Bennett & Sani, 2003; Dunham et al., 2015; Kinzler et al., 2010) and guide how children parse the complex social world around them (Bigler & Liben, 2007). The literature on social cognitive development has provided important insights into how children use race and gender as separate dimensions, but recent perspectives suggest that these categories should also be considered intersectionally—that is, how the systematic overlap of these categories may result in unique biases towards those who have multiple subordinated identities (Lei & Rhodes, 2021; Purdie-Vaughns & Eibach, 2008).

Although previous work indicates that children systematically overlap race and gender when categorizing others (Lei et al., 2020, Leshin et al., 2021), whether children account for race and gender simultaneously in memory processes is unclear. To test whether children spontaneously track race and gender, we used a modified version of the “who-said-what” task (Sesko and Biernat, 2010; Weisman et al., 2015).

Children were recruited from community partner sites in and around Philadelphia, (N=124, Mage=6.61 years; 53% female, 46% male, 1% not reported; 60% White, 7% Asian, and 21% Black, 5% Mixed Race, 1% Hispanic, and 6% not reported). In the who-saw-what task, children were shown 2 sets of images, each of 8 target children (2 White girls, 2 White boys, 2 Black girls, and 2 Black boys), paired with animals the children saw at the farm or the zoo. In a later phase, participants were then asked to recall which child saw which animal. Based on their recall of who saw what, we calculated 4 error types: within-race/within-gender, within-race/between-gender, between-race/within-gender, and between-race/between-gender. We also assessed whether children were more likely to make errors regarding Black girls than any other target group.

Results demonstrated that children made significantly more within-gender/within-race errors (M = 4.60, SE = 0.19) than any other error type (gender error type x race error type interaction), b = 1.06, SE = .39, t = 2.71, p = .007, indicating that their categorical encoding is intersectional. Although children in general improved in performance score over age, b = -.41, SE = .11, t = -3.73, p < .001, participant age did not moderate the main results, p > .10.

This work provides valuable insight into how children encode race and gender simultaneously into their minds at a young age, informing their perception and memory retrieval capabilities.

Authors

• Ravenel Davis, Society for Research in Child Development (SRCD)
  Presenting Author

• Aaron Cohen, Columbia University

• Ryan Lei, Haverford College