3-D Social Attention Training for Young Children with ASD

Friday, May 13, 2016: 10:00 AM-1:30 PM
Hall A (Baltimore Convention Center)
Z. Zheng, Z. Warren, H. Zhao, Q. Fu, A. S. Weitlauf, A. Swanson and N. Sarkar, Vanderbilt University, Nashville, TN
Background:  As we unravel ASD’s complex neurogenetic origins and environmental influences, there will be increasing need for technological tools that help us better understand symptom profiles for detection and potentially for intervention/prevention.  A poor response to one’s name being called, in children with adequate hearing has consistently been identified as an early red-flag for ASD in infancy and toddlerhood and is included as part of standardized, widespread screening checklists and diagnostic instruments.  We therefore designed and tested a novel technological system that tracks children’s visual attention and responds independently in order to measure and scaffold an exemplary social attention task for young children, namely, response to their names being called. 

Objectives:  The ultimate objective of this study was to empirically test the feasibility and usability of an adaptive technological learning environment, capable of intelligently administering early social orienting prompts and adaptively responding based on within system measurements of performance. 

Methods:  This system was comprised of name prompting and attention tracking sub-systems, an animated attractor (a bouncing ball), and a reinforcing feedback mechanism.  Eight toddlers with ASD and eight infants and toddlers without developmental concerns participated in a pilot study.  Each participant completed a single experimental session consisting of 10 trials of response to name training. During the trials, when a child did not respond to his or her name being called on the target monitor, a bouncing ball appeared on the monitor closest to the child’s current gaze.  This ball then bounced across monitors to direct the child’s gaze toward the target monitor.  If the child looked away or failed to follow the attractor, then additional motion and sound effects were added to the bouncing ball to heighten the attractor’s effect.  If the child looked at the target screen (target hit) in response to his/her name, the attention tracking sub-system recognized the success and promptly delivered praise with a firework animation. 

Results:  All participants tolerated the system well.  No sessions were terminated due to participant distress or engagement challenges.  All participants in the ASD group eventually hit the target (turned toward their name) across all trials.  On average, children in the ASD group averaged 3.07 seconds between initial prompt and success.  TD children, an average of 10 months younger than ASD children, also performed with a high level of accuracy.  The average time required for TD children to hit the target was 1.19 seconds.

Conclusions:  We studied the application of an innovative closed-loop adaptive technological system with potential relevance to core areas of vulnerability related to ASD in infancy and toddlerhood.  The proposed computer system was extremely well tolerated by the children.  Toddlers with ASD and typically developing infants and younger toddlers were able to ultimately respond accurately to prompts delivered by the technological system within the standardized protocol.  Further, the system was capable of attracting and pushing toward correct performance autonomously without user intervention.