Drift-Diffusion Model Parameters Underlying Cognitive Mechanism and Perceptual Learning in Autism Spectrum Disorder

Abstract

Previous studies have analyzed the performances of autism spectrum disorder (ASD) individuals in two-alternative forced-choice tasks (TAFC) and reported the group differences on the basis of drift diffusion model (DDM) parameters: boundary separation, drift rate, and non-decision time. This model investigates the cognitive ability, perceptual judgment, and other decisive mechanisms underlying the responses of an individual. The present paper has implied DDM to analyze cognitive and learning mechanism in ASD individuals while performing a decision-oriented task. Three different cases of DDM were considered relative to the variations in its parameters: decision process, non-decision, and inter-trial variability. The ASD and neuro-typical (NT) individuals were provided with two stimuli (risk-free and risk-containing images) and instructed to classify the images in their corresponding category. The experiment has been conducted over two consecutive sessions (Day 1 and Day 2). The results showed that means response time and accuracy values show an improvement in perceptual learning in ASD, although the values were poor in comparison to NT. The results suggest the role of drift rate and boundary separation parameters in modeling the cognitive mechanism and perceptual learning in ASD.