The Influence of Linguistic Form and Causal Explanations on the Development of Social Essentialism

• In Event: What's in a group? New Directions for Children's Essentialist Perceptions

Abstract

Overview: Generic claims about categories (e.g., “boys have short hair”) are widespread and are known to experimentally elicit essentialist beliefs (e.g., leading children to think that boys and girls are naturally and inevitably different from one another; Rhodes et al., 2012; Leshin et al., 2020). Given that generics are used to describe a wide-range of category-related features, it has been unclear whether generics lead to essentialism particularly when they are used to describe biological properties (e.g., “girls have ovaries”) or more generally—including when they are used to describe socially-determined features (e.g., “girls wear pink”). This question is critical for understanding how strongly these effects of language shape children’s understanding of the social world. To address this issue, the present study examined whether the effects of generic language vary based on whether it describes biologically- or socially-determined properties.

Methods: In this pre-registered study, 4.5 to 7-year-old children (N = 201; Mage= 6.07; 55% female; 71% White; 12% Hispanic) participated remotely from their home computers via an online platform designed for unmoderated remote research with children. Participants included children from across the United States and several other countries; by April 2021, we will complete analyses of how variability in our findings relates to variation in children’s communities and neighborhoods. Children were shown a storybook describing 16 properties about a novel group named “Zarpies.” Using a 2x2 between-subjects design, we varied the language form (generic vs. specific) used to describe the properties (e.g., “Zarpies have striped hair…” vs. “This Zarpie has striped hair…”) and the causal explanation provided (biological vs. social) for the property (e.g., “because they are born with stripes in their hair” vs. “because they are taught to paint stripes in their hair”). Children then completed a series of measures probing various components of their essentialist beliefs, including beliefs that Zarpie features are caused by intrinsic vs. extrinsic mechanisms, inflexible, inherited, and generalize across all members of the group.

Results: Overall, children who heard generic descriptions subsequently endorsed more essentialist beliefs about the category than children who heard specific descriptions (ß = -2.75, SE = .75, z = -3.68, p < .001; see Figure 1), with no effects of the causal explanation provided. Effects were generally consistent across each component of essentialism, and increased in magnitude across age for children’s endorsement of intrinsic, category-based causal mechanisms for category features (ß = -0.91, SE = .39, z = -2.33, p = .02), beliefs that category properties are inflexible (ß = -0.91, SE = .36, z = -2.54, p = .01), and beliefs that category features generalize to all members of the kind (ß = -0.71, SE = .33, z = -2.19, p = .03). These findings suggest that generic language elicits various components of essentialist beliefs regardless of whether they describe biologically- or socially-determined properties, consistent with the possibility that generics can broadly shape how children think about the world.

Group Authors

Josie Benitez, New York University, siebenitez@nyu.edu

Rachel Leshin, New York University, rachel.leshin@nyu.edu

Marjorie Rhodes, New York University, marjorie.rhodes@nyu.edu

Authors

• Josie Benitez, New York University
  Presenting Author

• Rachel Leshin, New York University
  Non-Presenting Author

• Marjorie Rhodes, New York University
  Non-Presenting Author