Poster #28 - Effect of Generic Language on the Categorization Strategies for Conventional and Privileged Information

• In Event: 3-003 - Poster Session 09
  In Poster Session: PS 09 Section - Cognitive Processes

Integrative Statement

We have yet to determine children’s categorization strategies for different types of information; that is, do children consider information to be an objective, natural category or one that is artifactual and human-made. Here, we investigated the strategies children and adults use to categorize conventional information, known by others and shareable to all, and privileged information, not known by others and intentionally concealed (Girgis & Behrend, 2015). Children may categorize information types by its category membership (similar to natural kinds) or by its use, i.e., whether it is shared or not (similar to artifacts). In addition, we examined the effect of generic labels versus specific ones on categorization strategies. Generic nouns convey properties that are associated with the category as a whole rather than a specific instance (Gelman, Ware & Kleinberg, 2010). Because of the properties inherent in privileged information, generic labels may have a larger effect on it as compared to conventional information.

In two studies, we tested a total of 49 preschoolers, 53 kindergarten and 48 adults. In both studies, participants were shown two test pictures: one depicting conventional information (e.g., animal) and the other privileged information (e.g., secret). Following this, a target picture was presented. In consistent trials, the target picture’s information type matched its use (privileged, not told). In the conflict privileged (target is privileged information) and conflict conventional (target is conventional information), the target picture’s information type did not match its use (e.g., privileged/told, conventional/not told). The test question was which test picture was more like the target. In Study 1, labels were specific (e.g., this animal) while in Study 2, labels were generic (e.g., these animals).

We conducted separate 3 (age) x 3 (trial type) ANOVAs for category membership responses for each study. For Study 1, there was a main effect of trial type, F(2, 130) = 26, p < .001, qualified by an age by trial type interaction, F(4,130) = 7, p < .001. Follow up tests revealed preschoolers and adults categorized conventional information by its category membership more than kindergartners, ps < .04, see Figure 1. For Study 2, there was a main effect of information type, F(2, 158) = 30, p < .001 and of age F(1,79) = 11, p < .001. Follow up tests revealed more category membership responses for conflict privileged trials than conflict conventional, p < .001, and that adults used category membership to identify information types more often than the younger age groups, ps < .001, see Figure 2.

Similar to previous research findings on the effects of generics on categorization strategies, we found adults used category membership to identify both information types and did so more than children did. In addition, all age groups used category membership responses to identify privileged information as compared to conventional. In this first attempt to examine the strategies we use to categorize information, it appears differences in the criteria used to categorize information shift with age and type of information.

Authors

• Helana Girgis, Hartwick College
  Non-Presenting Author

• Douglas A Behrend, University of Arkansas
  Presenting Author