The Effect of Interactive Music Therapy on Joint Attention Skills in Preschool Children with Autism Spectrum Disorder

Abstract

The purpose of this study was to investigate the effect of interactive music sessions on joint attention behaviors in preschool children with Autism Spectrum Disorders (ASD). Joint attention, the ability to share attention to a stimulus with another person, is a key deficit in children with ASD. Lack of joint attention behaviors contributes to the limited social and verbal skills that characterize ASD; joint attention behaviors are the primary component of the early screening for ASD advocated by the American Academy of Pediatrics. Participants (N=6; 5 male, 1 female) were between 36 and 64 months old at the time of the study, and were recruited from the child development program at a large Midwestern university. All children were enrolled in classrooms with curricula designed specifically for children with ASD. A multiple treatment (within-subject) design was used, with three treatment conditions: interactive music therapy, non-music interactive play, and independent play. Participants experienced each condition six times for a total of 18 ten-minute sessions over a five-week period. Session order was randomized to control for order effect. Behavioral observation of videotaped sessions was used to determine both interaction (responding to a bid for joint attention) and requesting behavior (initiating joint attention). Visual analysis of data graphs and statistical analysis were used to determine treatment effect. Interaction behaviors were most frequent in the interactive music therapy sessions, with less interaction in non-music interactive play sessions, and much less interaction during independent play. Although the difference between was less significant for the two children with the best interaction skills prior to the study, overall, the between-subject ANOVA revealed a significant difference in interaction among all three conditions (F [2, 105] = 62.028, p < 0.001; Bonferroni p < 0.01 between all conditions). Requesting behavior was highly variable across sessions, regardless of treatment condition, although requesting was generally higher in the interactive conditions than in the independent play sessions. Implications, limitations, and opportunities for further research are discussed.