Young Children’s Close Counterfactual Reasoning and Happiness Inferences

• In Event: Poster Session 05
  In Poster Session: PS 05 Topic - Social Cognition

Abstract

Our emotions depend on close counterfactual alternatives – events that came close to happening, but did not (Kahneman & Varey, 1990). Adults consider close counterfactual alternatives when inferring others’ emotions (e.g., Kahneman & Tversky, 1982; Mellers et al., 1997). However, it is unknown whether children do. Here we examined whether children consider close counterfactual alternatives when inferring other people’s happiness.

In Experiment 1 (n=180), 4-6-year-olds saw a story about a girl who could win either a regular or special balloon from an array of ten closed boxes; see Figure 1. The girl chose a box and received a regular balloon. Next, she was shown the balloons in the remaining boxes. In the Mostly Special condition, there were many special and few regular balloons. In the Mostly Regular condition, the distribution was reversed. Children rated the girl’s happiness about getting a regular balloon and judged whether she could have easily gotten a special balloon, with responses recorded as “yes” or “no”. For happiness ratings, a GEE ordinal logistic regression revealed a significant main effect of age, Wald X2(1)= 9.51, p<.001, and a significant age-by-distribution interaction, Wald X2(1)=12.18, p<.001; see Figure 2. Six-year-olds rated the girl as happier in the Mostly Regular condition than in the Mostly Special condition, p<.001. Five-year-olds responded similarly across conditions, p=.218, and 4-year-olds gave higher ratings in the Mostly Special condition, p=.034. However, children’s close counterfactual responses did not significantly differ between the two conditions, ps>.147.

These findings suggest that by age 6, children can use probability to infer emotions, but may not do so through close counterfactual reasoning. Is this disconnect between the two judgments due to children not considering close counterfactuals or is it because our close counterfactual question was inadequate at assessing counterfactual reasoning? In Experiment 2 (n=120), 5-6-year-olds saw the same stories as Experiment 1, however, the closeness of the counterfactual alternative was more salient; see Figure 1. In the Mostly Special condition, in addition to having a high chance of receiving a special balloon, the girl was also physically close to one. In the Mostly Regular condition, she was physically far from one. Children made the same happiness judgment as Experiment 1. To assess close counterfactual reasoning, children either judged whether the girl could have easily or almost got a special balloon. For happiness ratings, there was a significant main effect of distribution, Wald X2(1)=12.23, p<.001, where children rated the girl as happier in the Mostly Regular condition than in the Mostly Special condition; see Figure 2. Further, 6-year-olds were more likely to say she was close to getting a special balloon (regardless of which counterfactual question was asked) in the Mostly Special condition than in the Mostly Regular condition, p<.001. Five-year-olds did not respond differently between the two conditions, p=.161.

Together, these findings suggest that by age 5, when a counterfactual alternative is especially close, children can use it to infer people’s happiness. However, it is not until age 6 that they can explicitly report that the alternative is close.

Authors

• Tiffany Doan, University of Toronto
  Presenting Author

• Ori Friedman, University of Waterloo
  Non-Presenting Author

• Stephanie Denison, University of Waterloo
  Non-Presenting Author