Effects of Music Education on Social Skills and Executive Functioning of Children with Autism Spectrum Disorder

Background: There is growing interest in studying the impact of music for children with autism spectrum disorders (ASD). Individual music therapy for ASD increases peer interactions and turn taking (Kim et al., 2009; LaGasse, 2014). However, less is know about the effects of group music making on social skills of children with ASD. Individual and group music making can improve executive functioning of children with typical development (Moreno et al., 2011) and may also have the same effect for children with ASD who show executive impairments.

Objectives: We aim to assess whether group music making will have a positive impact on social skills and executive functioning of children with ASD in a naturalistic group setting.

Methods: Six entire classes comprised of children with ASD and neurodevelopmental disorders 8 to 13 years old completed a music intervention program during music class to maintain a naturalistic intervention setting. Three of the six classes practiced percussions in a structured drum circle (experimental group, N=21) and the other three classes explored different instruments without explicit practice (comparison group, N=18). Music classes targeted by the study occurred once per week for 30 minutes over 10 weeks. Homeroom teachers completed the Social Skills Improvement System: Socio-emotional Learning (SSIS-SEL) questionnaire, as a measure of social skills, and participants completed the NIH-Toolbox Dimensional Change Card Sort task, as a measure of executive functioning, pre and post music intervention. Groups were matched on IQ (p=.18) as measured with the Wechsler Scale of Intelligence-II (FSIQ range: 46-98).

Results: For the SSIS-SEL, repeated measures ANOVA showed no main effect of change over time (p=.43) or group (p=.40) but a significant interaction of time and group (p<.01), such that pre to post intervention scores increased for the comparison group but decreased for the experimental group. For the Card Sort task, repeated measures ANOVA with a sub-sample of the study (N=15) showed a significant increase in performance over time from pre to post intervention for both groups (p=.03), and no significant effect of group (p=.35) or interaction of group and time (p=.34). Although the interaction effect was not significant, visual inspection of graphed data shows that scores seem to have increased more for the experimental than comparison group pre to post intervention.

Conclusions: Contrary to our predictions, results suggests that social skills improved more for the comparison than experimental group, bringing into question whether longer interventions are needed to impact social behaviour outside the music class setting. Group rather than individual measures may best capture positive changes in group dynamics anecdotally reported by teachers. Our findings related to executive functioning, although not significant in a small sub-sample, were in the predicted direction suggesting that longer interventions and frequent practice (e.g. daily) may be needed for skill transfer to non-musical tasks. Our findings show that it is feasible to study the impact of music education within the context of special education while minimally altering the regular school curriculum.