Results of a randomized controlled trial support the efficacy of a wearable digital intervention in improving socialization among children with autism spectrum disorder (ASD), according to a study published in JAMA Pediatrics.
Investigators conducted a randomized clinical trial to assess the effectiveness of Superpower Glass (SG), an artificial intelligence-driven wearable behavioral intervention. SG is deployed via Google Glass, a wearable set of eyeglasses, and an Android smartphone application. The SG system uses the glasses’ outward-facing camera to capture facial image data from the wearer’s social partner. Facial data are then transmitted to the smartphone, which interprets facial expressions and transmits a corresponding emoji to the glasses’ peripheral monitor. A total of 8 emojis were available to guide the child in interpreting 8 emotions. Children aged 6 to 12 years with a formal ASD diagnosis who were currently receiving applied behavioral analysis therapy were selected for inclusion (n=71). Children were randomly assigned to receive either the SG intervention plus applied behavioral analysis therapy (n=40) or applied behavioral analysis therapy only (n=31). As primary outcomes, 4 socialization measures were assessed via family-administered surveys over the course of 6 weeks. Assessments were conducted 4 times a week.
The final study cohort comprised 71 children with a formal ASD diagnosis, among whom 63 (89%) were boys. The mean (SD) age of all participants was 8.38 (2.46) years. Children receiving the intervention displayed significant improvements on the socialization dimension of the Vineland Adaptive Behaviors Scale compared with treatment-as-usual control children (mean [standard deviation] treatment impact, 4.58 [1.62]; P =.005). Positive mean treatment effects were also found for the other 3 primary measures, although not to a significance threshold (all P =.0125). Specifically, greater improvements were observed on the Social Responsiveness Scale, Developmental Neuropsychological Assessment, and Emotion Guessing Game in intervention participants compared with control participants.
Limitations included low participant adherence to the recommended treatment dosing and the varying levels of social skills developed by participants before the study.
A mean gain of 4.58 points on the Vineland Adaptive Behaviors Scale socialization subscale is comparable to improvements observed in standard of care therapy. These data thus demonstrate efficacy of the SG intervention in improving social behavior in children with ASD. Further research to explore the effect of facial engagement and emotion recognition on social behaviors is necessary to refine such therapies.
Voss C, Schwartz J, Daniels J, et al. Effects of wearable digital intervention for improving socialization in children with autism spectrum disorder: a randomized clinical trial [published online March 25, 2019]. JAMA Pediatr. doi:10.1001/jamapediatrics.2019.0285