Training Attention in Students with ASD Using a Multiple Object Tracking Paradigm: A Pilot Study

Background: The ability to pay attention is a primary predictor of academic achievement (Duncan et al., 2007), while inattention is associated with problem behaviors and poor academic performance (Barriga et al., 2002). A deficiency in the subcomponents of selective and sustained attention are characteristic of atypically developing populations; specifically, in children with Autism Spectrum Disorder (ASD; Koldewyn, et al., 2013). Consequently, several cognitive-based training approaches have been developed to reduce problems of attention by targeting the subcomponent processes (Sonuga-Barke et al., 2014). Cognitive training is commonly achieved through implicit practice via repeated performance on tasks (Willis & Schaie, 2009). Here, we used Multiple Object Tracking (MOT) paradigm as a method of cognitive training. The reasons for this are threefold: first, MOT is considered to be the best empirical measure of object-based visual attention (Scholl, 2009); second, it is non-verbal in nature; and third, it is accessible to children of different ages and levels of cognitive functioning.

Objectives: We aimed to assess the efficacy of a training program using a MOT paradigm in adolescents with ASD. We asked the following questions: (i) will the performance of participants with ASD improve significantly with training, and (ii) will increased performance on the MOT paradigm transfer to another test of attention (i.e., near-transfer; Redick et al., 2014)?

Methods: Nine high-school students diagnosed with ASD (aged 12-17), and 15 students diagnosed with a neurodevelopmental condition other than ASD (learning disorder) were pre-assessed on the Wechsler Abbreviated Scale of Intelligence (WASI-II) and the Conners Continuous Performance Task (CPT-3). An experimental group (ASD: n = 6 & non-ASD: n = 6) received the MOT cognitive training program, while an active control group (ASD: n = 4 & non-ASD = 8) was trained on 2048 (a puzzle-like, math game). Both groups trained three times a week, over a period of five weeks, for a total of fifteen MOT or 2048 training sessions. Performance on the MOT task was defined as the average speed at which the participant could track three of eight target items. Post-training performance on the CPT-3 was compared to pre-training scores to measure an effect of transfer; groups switched training tasks and followed the same routine and post-test schedules.

Results: MOT performance doubled from the first to the last (fifteenth) training session for both the ASD and non-ASD group. This increase was similar to that found in typically-developing adolescents (Tullo et al., 2015). Preliminary analyses revealed that the improvement on the MOT task did not transfer to the CPT-3 task; this was not surprising given our small group size.

Conclusions: Results demonstrate that our MOT training is accessible to adolescents with ASD of different ages and levels of cognitive functioning. Our findings illustrate that adolescents with ASD are able to train attention on a non-verbal task similarly to those without ASD and typically-developing adolescents. We are presently analyzing whether the outcome measure of attention, and verbal- and non-verbal cognitive functioning of the participants with ASD is related to the effect of near transfer.