Poster #24 - Children's Pragmatic Inferences From Generic Statements About Social Categories

• In Event: 1-005 - Poster Session 01
  In Poster Session: PS 01 Section - Cognitive Processes

Integrative Statement

Adults frequently communicate to children information about social groups through broad generalizations using generic language (e.g., "boys play sports”, “girls like pink”). Generic statements make explicit claims about kinds (e.g., the category “boy”), not individuals (Carlson, 1977). However, these kinds of statements may also implicitly convey information about other groups that were not explicitly mentioned. That is, hearing “boys play sports” may lead children to assume that girls do not play sports.

Previous work has shown that children as young as 2 use generic language as a signal for category boundaries (Rhodes, Leslie, Bianchi, & Chalik, 2018), and by age 4, they interpret generics as providing more centrally relevant information about categories than non-generic statements (e.g., Cimpian & Markman, 2009). While previous research speaks to children’s ability to interpret information explicitly conveyed by generics, it has not yet addressed the kinds of implicit information children may draw from generics, and when. We know children as young as 3 can look beyond the literal meaning of language and infer the implied meaning of a speaker’s statement, at least in some contexts (e.g., Stiller, Goodman, & Frank, 2015), but have difficulty in others (e.g., Huang & Snedeker, 2009). Given the frequency children hear generic statements (Gelman, Goetz, Sarneck, & Flukes, 2008) and how early they understand them (Graham, Nayer, & Gelman, 2011), young children may also derive implicit meanings from generic language (e.g., infer that girls do not play sports upon hearing “boys play sports”).

The present study tests this by examining the kinds of inferences that 4- to 6-year-olds (N = 158) and adults (N = 60) make from generic statements about a set of two contrasting novel social groups: zarpies and gorps. Participants were introduced to a town where people were either zarpies or gorps, not both; category membership aligned with wearing green or yellow clothing. Across 4 trials, participants heard statements about a property. Those in the “generic” condition heard properties true of the group (e.g., “zarpies are good at climbing trees”) and those in the “specific” condition heard properties true of the individual (e.g., “this zarpie is good at climbing trees”). Participants were then asked to rate yes or no on whether two additional individuals (one zarpie and one gorp) also possessed the property. Participants completed an additional task akin to that used by Stiller et al. (2015) to measure general ability to infer non-literal meanings.

As expected, our results (estimated using a GEE binomial model) found an interaction between condition and category membership (χ2(1) = 22.4, p < .001): children participants in the “generic” condition responded “yes” another zarpie had the property 82% of the time, but only 21% of the time for a gorp; however, those in the “specific” condition did not make the same strong inference (percentage of “yes” responses: zarpie: 54%; gorp: 45%, see Figure). This provides support that children as young as 4 are able to infer non-literal meanings from generic statements. Additional findings and the implications of these results will be discussed.

Authors

• Kelsey Moty, New York University
  Presenting Author

• Marjorie Rhodes, New York University
  Non-Presenting Author