Effects of Rhythm and Robotic Interventions on the Affective States of Children with Autism Spectrum Disorders

Background: Children with ASDs also demonstrate several behavioral and affective impairments that may indirectly influence their social interactions with others. For example, children with ASDs have a significant incidence of neutral or negative affect, behavioral problems, as well as difficulty with social connections and empathetic behaviors to others (Mazefsky et al., 2012).  We are currently conducting two randomized controlled trials to evaluate the effects of two novel, movement-based interventions on the affective states of children with ASDs.  Children with ASDs find musical activities enjoyable and motivating (Srinivasan & Bhat, 2013). Similarly, there is some anecdotal evidence for the use of robot-based interventions to promote social interaction skills in children with autism (Kozima et al., 2007).

Objectives: In the current project we compared the effects of robotic and rhythm interventions on the affective states of children with ASDs, specifically, positive, interested, neutral, and negative affect.

Methods: 36 children diagnosed with ASDs will be matched on age and level of functioning and randomly assigned to three training groups - a rhythm group or a robot group or a control academic group.  Children received training sessions across a total of 8 weeks (2 sessions per week from experts and 2 more sessions from caregivers).  The rhythm group engaged in music-based activities involving singing, joint rhythmic action, and whole-body coordination with an expert trainer and an adult model. The robot group engaged in imitation games involving whole body movements, drumming, and walking with a humanoid robot, Nao and an adult model. The academic group engaged in stationary, fine motor and academic activities including reading, building, and arts and crafts. We are coding for affective changes in all three groups across the early, mid, and late training sessions. Specifically, we code for positive, interested, negative, and neutral affect during the entire session.

Results: We have preliminary data from two low-functioning children with autism, one from the rhythm group and one from the robot group. Both children demonstrated an increase in positive and interested affect across weeks of training. The child in the rhythm group showed a 33% increase with more time spent in positive and interested affect with training. Similarly, the child in the robot group demonstrated a 23% increase with time spent in more positive and interested affect with training. We hypothesize that overall, children in the rhythm and robot groups will enjoy the training activities more than the children in the academic group.  We are continuing to code the remainder of the dataset.

Conclusions: Based on the preliminary results, it appears that novel movement-based interventions (robotic or rhythm) may be an effective training tool to enhance positive affective states in children with ASDs.  If our hypotheses are upheld, we will be providing objective evidence supporting the efficacy of embodied movement interventions for addressing the behavioral and affective impairments of children with autism.