Yay! Yuck! Toddlers use incongruent emotions to reason about hidden objects

• In Event: Poster Session 09
  In Poster Session: PS 09 Topic - Cognitive Processes

Abstract

Infants show impressive sensitivity to others’ emotions from early on – attending to and discriminating different facial emotions, using emotions to decide what to approach or avoid, and recognizing that certain objects and events are likely to produce particular emotional responses. But do infants and toddlers also recognize more abstract features of emotions—features that are not tied to any one emotion in particular? Here we examined the development of the higher order expectation that emotions are more or less mutually exclusive, asking whether young children recognize that people generally do not express two conflicting emotions towards a single stimulus using a manual search task.

Across four experiments (n = 64), toddlers saw an agent look inside a box and produce two emotional utterances and corresponding facial expressions, and then reach in and retrieve an object from the box while keeping its identity hidden. The question was whether children then continued to each the box for any remaining object(s). Experiment 1 asked whether 26-month olds recognize that incongruent emotional expressions are unlikely to refer to a single object. After observing the experimenter peer into the box and say “Yay! Yuck!” (incongruent emotions) they persisted in searching the box longer than if they had heard “Yay! Wow!” (congruent emotions), t(15) = 3.58, p = .003. Experiment 2 showed that this inference relied on the incongruent emotions being produced by a single agent. When two different agents produced the two incongruent emotional responses, children did not assume that two objects must be present, searching equally on incongruent and congruent emotion trials, t(15) = 0.21, p = .83. Finally, we examined the developmental trajectory of children’s ability to reason about objects using emotions. In Experiment 3 we found that younger, 20-month olds failed to use incongruent emotions to individuate objects, t(15) = 0.64, p = .53. This was not due to an inability to succeed at the task in general, as they readily searched the box longer after hearing the experimenter produce two incongruent novel labels (“A blick! A fep!”) compared to a single label (“A dax! A dax!”), t(15) = 2.35, p = .03 (Experiment 4).

Together, our results suggest that by 2-years of age, children use higher order knowledge of emotions to make inferences about the world around them, and that this ability undergoes early development.

Authors

• Alexis Sierra Smith-Flores, University of California - San Diego
  Presenting Author

• Lisa Feigenson, Johns Hopkins University
  Non-Presenting Author