Dynamic Eyetracking As a Measure of Treatment Response for an ASD Social Skills Intervention

Background: The development and evaluation of novel ASD treatments currently relies on clinical observations and parent-report questionnaires to examine response to treatment. However, most commonly used measures of ASD symptoms lack sensitivity to change. Therefore, unbiased ASD treatment outcome measures are urgently needed to accurately evaluate novel ASD interventions. Eyetracking-based measures of social attention have the potential to be a reliable measure of response to ASD treatments that target social communication and social cognition skills.

Objectives: This preliminary study aimed to examine the extent to which time spent looking at faces in a dynamic social eyetracking paradigm reflected treatment-related changes in social communication skills due to a social skills intervention for adolescents and young adults with ASD. Changes in clinical report measures of social communication were also evaluated across time points.

Methods: Fourteen adolescents and young adults with ASD (Age: M=16.28 years, SD=2.19) completed a dynamic eyetracking task prior to and immediately following Social Cognitive and Interaction Training for Autism (SCIT-A; Turner-Brown et al., 2008) an 8-week social skills group intervention. Data were collected using a Tobii TX300, which sampled at 300Hz. Participants viewed a silent 8-minute video eyetracking paradigm, which showed children participating in both joint and parallel play (Chevallier et al., 2015). Stimuli include 22 semi-naturalistic, 15-second silent video clips, in which two school-age siblings play together (joint condition) or independently (parallel condition). Tobii software produced a Total Fixation Duration (TFD) outcome measure, which represents the summed length of all fixations within each area-of-interest (AOI). This TFD measure for faces was used as an index of the degree to which participants prioritized faces in joint conditions relative to parallel conditions. The Social Responsiveness Scale, Second Edition (SRS-2) was also collected before and after the intervention.

Results: Paired t-tests were used to examine the change in visual preference for social stimuli and SRS-2 scores before and after the social skills intervention. Results from the eyetracking task showed an increase in visual prioritization of social stimuli following the intervention, t=2.09, p=0.057. This change in visual preference reflects a large effect size, d=0.82. Additionally, SRS-2 scores declined significantly following the social skills intervention, t=6.46, p=<.0001, reflecting a large effect size, d=2.53.

Conclusions: Eyetracking and SRS-2 measures of social communication showed improvement in visual social preference and social communication skills, respectively, in adolescents and young adults with ASD following an 8-week social skills group intervention. These preliminary results suggest that this eyetracking-based measure of social attention may be sensitive to change in the context of an ASD intervention. Further research is needed to examine the reliability and generalizability of these findings to other interventions with appropriate controls for practice effects.