Do You Like What I Like? Children’s Understanding of the Subjectivity of Preferences

• In Event: Poster Session 09
  In Poster Session: PS 09 Topic - Social Cognition

Abstract

All preferences are subjective, but some preferences are more subjective than others. For example, whereas most people will share your preference for eating apples over eating olives, it’s less certain that most will share your preference for the color blue over red. Building on previous work which suggests that young children understand that preferences are subjective (Repacholi and Gopnik, 1997), we investigated whether young children anticipate that some preferences are more likely to be shared than others.

In 4 experiments, children saw sets of items, and identified theirs and another person’s preferences; see Figure 1. In Experiment 1, 2-year-olds (n=32) chose between images of Oreos and celery in two food trials, and between images of the color red and blue in two color trials. In each trial, children were first asked about their preference (e.g., “Which food do you like?”), and then about the experimenter’s preference (e.g., “Which food do I like?”). A GEE model revealed that children were more likely to predict shared preferences for foods than for colors, p=.001. Although this could suggest that toddlers view food preferences as more shared than color preferences, the finding could stem from differences in the items’ desirability. The Oreos were more desirable than the celery (Oreos preferred in 86% of choices), whereas the colors were similarly appealing (red preferred in 61% of choices). Hence, in the next experiment, both foods were highly appealing, but one color was more attractive.

In Experiment 2, 2-3-year-olds (n=64) chose between Oreos and chocolate-chip cookies, and between bright blue and a shade of greenish-brown, sometimes characterized as the world’s ugliest color (Pantone 448C). Again, children were more likely to predict shared preferences for foods than for colors, p<.001. The next experiments examined whether these findings extend to other colors and foods, and whether they replicate when children were asked about a gender-matched child.

In Experiment 3, 2-3-year-olds (n=64) were shown a gender-matched child named Morgan and identified theirs and Morgan’s preferences. They chose between Oreos and chocolate-chip cookies, apples and bananas, red and blue, and green and yellow. Children were more likely to predict shared preferences for foods than for colors, p=.007. In the final experiment, we also asked children about two other domains.

In Experiment 4, 2-3-year-olds (n=64) identified theirs and Morgan’s preferences for sets of items in different domains. We created two sets of four pairs of items from the domains: food, colors, activities, and animals. Children’s shared preferences between themselves and Morgan depended on the domain, p=.023. Children were more likely to predict shared preferences for foods than for colors, p=.014, and for activities than for colors, p = .011. There were no differences in children’s predictions of shared preferences between the other domains, ps>.134.

Together these findings show that young children recognize that some preferences are more likely to be shared than others. Further, their expectations about these preferences appear to be driven by domains, rather than the relative strengths of preferences.

Authors

• Tiffany Doan, University of Toronto
  Presenting Author

• Ori Friedman, University of Waterloo
  Non-Presenting Author

• Stephanie Denison, University of Waterloo
  Non-Presenting Author