Recognizing emotions in music through a computerized method: A novel way of evaluating social maturity

Hyunchan Hwang¹Jin Hyung Lee²Juri Yun³Doug Hyun Han^1*

• ¹Chung-Ang University, Seoul, Republic of Korea
• ²Chulalongkorn University, Bangkok, Thailand
• ³Ewha Womans University, Seodaemun-gu, Republic of Korea

This study used music emotion recognition to evaluate emotion perception and aimed to explore the relationship between social maturity and musical emotion recognition in both autism spectrum disorder (ASD) and controls groups. 84 people with ASD and 50 controls were included in the study. Participants were evaluated using the Wechsler Intelligence Scale of corresponding age (IQ), social maturity scale (SQ), childhood autism rating scale 2 (CARS). Emotion perception was evaluated using two tools. The emotion perception test (EPT) shows multiple facial expressions and asks to find the face with a different emotion than the rest. The music emotion recognition test (MEPT) asks the user to choose the emotion they feel when they hear the given music. The control group had significantly higher IQ, SQ, EPT scores, MEPT scores, and significantly lower CARS scores. Since IQ differences were substantial between the two groups, individuals with IQ lower than 70 were removed from the main analysis, and are not interpreted causally, but only presented as associations. SQ was positively correlated with EPT scores in all participants and ASD group, but not in the control group. Correlation between MEPT scores and SQ were also similar to that of EPT scores and SQ. The MEPT shows that social intelligence as music emotion recognition are correlated. Further research may prove this tool to be useful for people with ASD.