Children's Emotion Understanding from Vocal Bursts

• In Event: Young Children's Emotion Expression and Understanding

Abstract

To be able to navigate in the world and maintain social connection, people need to understand other’s and their own emotions starting from very early ages. People convey emotional cues in different modalities and one of them is through voice. In fact, those emotional cues contain important, even more accurate, information about the intentions and feelings of others (Simon-Thomas et al., 2008). Thus, we reasoned understanding other’s emotions from more subtle cues instead of explicit one may give us a better insight on one’s emotion competencies. Plus, while the previous literature heavily relies on children’s understanding of facial signals of emotions, we still know very little about the developmental trajectory of interpreting affective cues in the voice. To address that we recruited 162 5- to 8-year-old children: 44 5-year-olds (23 boys, 21 girls), 426-year-olds (22 boys, 20 girls), 41 7-year-olds (20 boys, 21 girls), and 358-year-olds (20 boys, 15 girls). To understand children’s ability to understand other’s non-verbal and more subtle emotional messages, participants completed a forced-choice categorization tasks consisting of 24 emotional sounds (Sauter, 2006) including laughs, sighs, and grunts. Ekman’s basic emotions approach was adapted to determine the selection of emotions (Happiness, Anger, Disgust, Surprise, Sad,Fear).

Consistent with previous work, multiple regression models revealed that recognizing emotions improved with age and there was a significant age and gender effect (β = .17, F(3,152) = 12.94, p < .001). At earlier ages, girls were better than boys at recognizing emotions accurately from subtle vocal cues; however, this gender difference was eliminated by age 8.Furthermore, there was no significant change in accuracy of recognizing emotions for girls across age, whereas boys became more accurate with age, effectively catching up to their female peers. We also looked at the recognition rate per emotion that was tested. Multilevel analysis yielded significant mean differences in the order of recognizing emotions: aligned with the previous research (Widen and Russell,2013) happiness was the most recognized emotions for both young (5-6) and old (7-8)children and across girls and boys while fear was the least recognized emotions from vocal cues. We did not observe any gender interaction for happiness and fear recognition, however, girls had advantages over boys recognizing anger, disgust, surprise, and sadness. On the other hand, older kids performed better recognizing anger, surprise, sadness, and fear but not happiness and disgust. Our study provides a more nuanced way of measuring children’s affect recognition competencies which we believe have more predictive power for socio-emotional outcomes, such as empathic accuracy (Kraus, 2017). The findings also will help curriculum creators, teachers, and parents to design more effective tools with the purpose of increasing emotion competencies at early ages so that children can grow up with the required skill sets. Future work should expand those findings beyond basic emotions to further explore children’s age and gender effects one motion recognition. 

Author

• ozge ugurlu, University of California - Berkeley
  Presenting Author