Does social categorization affect toddlers' play preferences? : an experimental test

The purpose of this study was to examine the effect of learning gender and age labels on toddlers' toy preferences. Many researchers (e.g., Arthur, Bigler, Liben, Ruble, & Gelman, 2008; Martin & Halverson, 1981) have suggested that the acquisition of such labels should cause increased in-group preferences, and thus, increased interest in same-category toy preferences. In this study, we used feminine and masculine toys (e.g., a purse, a hammer) to test gender typed preferences and adult and child objects (e.g., toy keys, real keys) to test age typed preferences. Forty 18-20 month old children (M = 19.1 months) from primarily upper-middle class families participated in the study. After taking pre-test measures of their children's social label understanding and toy preferences, parents were randomly assigned with their children, to one of two conditions: the age condition (in which they were asked to practice "grown-up" and "kid") and the gender condition (in which they were asked to practice "boy" and "girl"). Parents practiced the labels with their children for approximately two weeks and returned to the lab. In the post-test session, children's social label understanding and toy preferences were again assessed. Additionally, parents were asked about their attitudes about other-gender toys and behaviors and about the toys their children had at home. The results indicated an interaction between test time and comprehension of the gender and age labels. Children in the age condition's performance on the age labels in the target word comprehension task improved from pre- to post-test. Children in the gender condition's performance on the gender labels also improved from pre- to post-test. However, there were no main effects of condition at post-test. As predicted, comprehension of labels was related to toy play in the age condition. Children who were more successful on the age labels on the word comprehension task also played more with toy items than real items at post-test. Comprehension of labels was not related to toy play in the gender condition. However, at pre-test, girls in the gender condition already exhibited strong sex typed behavior, so it is possible that the manipulation was not enough to change these extant preferences. These data partially support Bigler and Liben's (2006) Developmental Intergroup Theory and work by Martin and Halverson (1981) on role of cognitive processes in the formation of social stereotypes, preferences, and prejudice.