Poster #199 - The Development of Children’s Emotion Understanding: A Meta-Analysis of Results from the Test of Emotion Comprehension

• In Event: 3-053 - Poster Session 10
  In Poster Session: PS 10 Section - Social Cognition

Integrative Statement

Many instruments are currently available for the assessment of emotion understanding from early childhood to adulthood (see Castro et al., 2015 for a review). However, many of these instruments need expertise, take time to administer and score, assess only one or two core components of emotion understanding (often quite basic), can be used either with younger or older children (but not both), are demanding in terms of language production and/or comprehension, and are occasionally expensive (thereby limited accessibility). The Test of Emotion Comprehension– TEC (Pons & Harris, 2000), which can be used with children between 3 to 11 years old, overcomes these limitations. The TEC examines nine components of emotion understanding: (i) recognition of facial expressions of basic emotions; understanding the impact of (ii) external causes, (iii) desires, (iv) memories, and (v) beliefs on emotion; (vi) understanding the control of the expression of emotion; (vii) understanding the regulation of emotion, (viii) understanding mixed emotions, and (ix) understanding moral emotions. The TEC has been available for nearly 20 years and has been translated and adapted so far into more than 25 languages. In the present study, a meta-analysis of studies using the TEC was conducted in order to: (i) identify trends in the overall development of emotion understanding; (ii) explore different developmental patterns for the development of the nine components of emotion understanding; and (iii) examine potential culture differences. Following a literature search, 54 reports or articles including 95 separate age groups were included in the current analysis. Figure 1 shows the relation between sample size, age and mean TEC score. The relation between age and mean TEC score is nonlinear. The quadratic standardized beta is significant (t=8.03, p<0.001), and the R2 of the quadratic model (R2=0.87) is higher than that of the linear model (R2=0.85). Figure 2 shows the relationship between age and mean scores on the nine components (total sample groups = 31), using the curvilinear estimating method. Recognition, External, Desire and Regulation were fully mastered by 11 years of age whereas other components were still being acquired beyond 11 years of age. Children’s data from four countries, China, Germany, Italy, and UK, were compared to assess cultural differences. Multilevel regression was used to test for the effect of country. Age was entered at the first level, and country (set as dummy variables) was entered at the second level. The results showed that, after controlling for age at the first level (R2=0.88，F(1,71)=507.46，p<0.001), the main effect of country was still significant at the second level, ΔR2=0.02，Fchange(3,68)=2.95，p<0.05. Inspection of each pair of countries showed that Chinese children performed better than British (t=2.70, p<0.01) and German children (t=2.30, p<0.05). The mean score of Chinese children on the TEC was 0.9 points higher than British children, and 0.8 points higher than German children. No other differences between countries were found. The analyses reveal a robust development improvement in the understanding of emotion with modest variation in the rate of development across different cultures.

Authors

• Yulong Tang, Zhejiang University of Technology
  Presenting Author

• Paul Harris, Harvard Graduate School of Education
  Non-Presenting Author

• Francisco Pons, Oslo University
  Non-Presenting Author