Eww, you want that one? Toddlers use emotion to infer intentionality from non-random sampling events

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

Abstract

Young children use statistical information in their environment, as well as others’ emotional communication, to understand the intentions of social partners. In particular, a noteworthy study found that young children can infer agent preferences from non-random sampling events (Kushnir et al., 2010). However, this study did not consider the influence of emotional communication, by controlling for affect. The role of discrete emotions in attributing intentionality from non-random sampling events has not been studied. This study assessed 2-year-olds’ attributions of intentionality from non-random sampling events and subsequent discrete emotion reactions.

Forty-eight 2-year-olds (M = 27.57 months, SD = 3.14) observed an experimenter remove five objects from a sample one at a time and express joy, disgust, or sadness after each selection. Twenty-four toddlers were assigned to the non-random minority (18%) sampling condition and 24 were assigned to the random majority (82%) sampling condition. After viewing all 5 selections, toddlers were prompted to offer the experimenter the object type that she had previously selected (i.e., target) or had not previously selected (i.e., alternate). Children received the same selection criteria in a given condition (i.e., always viewing the experimenter select either the minority or majority object across trials) and participated in each of the 3 emotion trials.

We hypothesized that toddlers would infer agent intentionality by offering the target object more often in the minority sampling condition after the experimenter expressed joy, replicating previous research (Kushnir et al., 2010). We also hypothesized that this same pattern would exist when she expressed disgust after selecting minority objects since children at this age understand the notion of intentionally removing decontaminated objects (see Brown & Harris, 2012). Lastly, we hypothesized that children would not infer intentionality in sadness trials since sadness communicates goal failure and would be an irrational response after selecting minority objects when the experimenter could have more easily selected a majority object.

Chi-square analyses were conducted separately for each emotion to determine if toddlers offered the target or alternate object more often in the minority or majority condition. In line with our hypotheses, children’s first offers significantly differed between distribution conditions when the experimenter expressed joy, X2 (2, 46) = 7.77, p = .005, and disgust, X2 (2, 46) = 4.26, p = .04, but not sadness, X2 (2, 46) = 0.09, p = .76. Specifically, toddlers in the minority sampling condition offered the target more often when joy (75%) and disgust (65%) were expressed than toddlers in the majority condition viewing joy (42%) and disgust (35%) reactions, but toddlers offered the target at chance in both conditions when sadness was expressed (see Figure 1).

This study replicates findings that young children can infer intentionality from non-random sampling events and extends these findings to discrete negative emotions. In fact, to our knowledge this is the first study to find that toddlers understand that people may intentionally select undesired objects out of a sample. Implications for the development of inferring intentionality from statistical sampling events and discrete emotion communication will be discussed.

Authors

• Lukas D Lopez, University of Utah
  Presenting Author

• Eric A Walle, University of California - Merced
  Non-Presenting Author