Sorting out emotions: How labels influence categorization of emotions

• In Event: Poster Session 12
  In Poster Session: PS 12 Topic - Attention, Learning, Memory

Abstract

Recent literature on emotion development claims that emotion categories are constructed with the help of language (Barrett, 2017; for a review, see Lindquist & Gendron, 2013). In the present study we examined whether the language provided during an emotion face sorting task affected children’s ability to categorize emotional faces. We hypothesized that emotion labels given during the emotion categorization task would help children correctly categorize the faces more than neutral labels. We also hypothesized that this effect would be more pronounced for 2-year-old children than 3-year-old children because these language labels would act as a support for 2-year-olds that the older children would no longer need.
Forty-seven children (goal: 64) participated in this study, 15 were 2-year-old children (M=2.73, SD=0.23), and 32 were 3-year-old children (M=3.55, SD=0.28). Preliminary data are presented below, but data collection is ongoing in an online format due to the COVID19 pandemic. The data presented here were collected in a quiet room in the children’s preschool.
Children were first given a pre-test to ensure they understood the task and it included three images of animals (i.e., a horse, a dog, and a cat). There were three buckets with one of the images on each bucket. An experimenter labeled each bucket with the phrase “_____ pictures go in this bucket”, naming each animal in turn. Then the experimenter showed the child one of the animal pictures and asked, “where does this one go?” If they got one wrong, they were given a second chance. Children only continued to the experiment if they could get all three right on the first or second try.
The main experiment followed the same basic structure as the pretest, except children were given emotional faces. The buckets had one emoji to represent the three emotions: happy, sad, and angry. In the neutral condition, the buckets were labeled with the phrase “These kinds of faces go in this bucket” while the experimenter pointed to the emoji. In the emotion label condition, the buckets were labeled with the phrase “________ faces go in this bucket” for each emotion word. The bucket labeling procedure was given at the beginning and after every 3rd picture. The children were shown pictures of faces that displayed one of the three emotions and asked, “Where does this one go?” These pictures were given one at a time until all 30 pictures were sorted.
Preliminary analysis shows a significant difference in number of pictures correctly sorted between 2- (M = 18.93, SD = 8.40) and 3-year-olds (M = 25.13, SD = 4.52), (F = 11.934, p = .001). However, preliminary findings do not show a main effect for condition (F = 2.763, p = .104), nor is there an interaction between age and condition (F( = 2.129, p = .152). Further data collection is ongoing and will provide valuable insights into how labels may, or may not, influence their emotion categorization.

Authors

• Gwendolyn Price, University of California - Los Angeles
  Presenting Author

• Marissa Ogren, Rutgers University - Newark
  Non-Presenting Author

• Catherine Sandhofer, University of California - Los Angeles
  Non-Presenting Author