Characterizing Individual Differences in Attention across Autism Spectrum Disorder: A Multiple Object-Tracking Study.

Background: Multiple-object tracking (MOT) tasks have been used to characterize differences in attention between typically and atypically developing populations (i.e. Autism Spectrum Disorder (ASD); Koldewyn et al., 2013). Furthermore, we have recently demonstrated that MOT tasks can characterize individual differences in attention resource capacity across neurotypical adults (Tullo et al., 2018). Attention resource capacity is an ideal descriptor of attention as it identifies cognitive load; the amount of task demands an individual can process. By manipulating the task’s cognitive load, via speed and number of target items, the aforementioned study concluded that analytical reasoning (i.e., fluid intelligence) is predictive of performance at the limits of attentional capacity. Given the attention-related challenges in autism and how they affect various cognitive spheres (Anthsel et al., 2013), the characterization of attention resource capacity for individuals with autism is a priority. Previous attempts at isolating attention in autism using visually-based tasks have been limited by their accuracy in targeting attention, and their accessibility to lower-functioning participants.

Objectives: The aims of the current study are twofold: (i) To examine whether attention resource capacity can characterize individual differences in ASD (ii) whether fluid intelligence contributes to the characterization of individual attentional differences in ASD.

Methods: We recruited 108 participants: 59 neurotypical (NT) children, adolescents, and adults (M_MentalAge=16.61), and 59 children, adolescents, and adults with an ASD diagnosis (M_MentalAge=16.25). All participants completed a MOT task, which required participants to visually track either 1, 2, 3, or 4 target items (separate cognitive load conditions) from among 8 items moving arbitrarily for 8 seconds. MOT performance was defined as the average speed (cm/s) at which a participant tracked all target items across load conditions. Participants were then administered the Wechsler Abbreviated Scale of Intelligence-2nd edition (WASI-II) to obtain a measure of fluid intelligence.

Results: As expected, the MOT average speed score decreased logarithmically as cognitive load increased, replicating the unique trend found in the neurotypical group (y = -118.2ln(x) + 215.09, R² = 0.989). However, the average speed score of the ASD group was lower than that of the neurotypical group across both low (1-2 targets) and high cognitive load (3-4 targets) conditions (t(58)=-2.432, p=.018, t(58)=-2.411, p=.019). Furthermore, fluid intelligence was predicative of MOT performance for only high load condition in the ASD group (b=.562, t(58)=3.508, p=.001, R²=.405, F(2,55)=18.753, p=.000).

Conclusions: Results corroborate the association between fluid intelligence and attention for individuals with ASD. Our findings suggest that MOT capability can be used as an assessment of attention and is generalizable to a sample with ASD. Findings are also clinically relevant since MOT can used as a tool to assess attention resource capacity for atypically developing populations. In the classroom, attention resource capacity is a potential descriptor for how much material a student with ASD can process.