Disentangling response initiation difficulties from response inhibition in autism spectrum disorder: A sentence-completion task

It has been proposed that individuals with autism spectrum disorder (ASD) struggle both with response initiation and with response inhibition, both of which are functions of the executive system. Experimental tasks are unlikely pure measures of a single cognitive domain, and in this study we aim at understanding the contributions of response initiation difficulties to possible deficits in inhibitory control in autism. A sample of adults diagnosed with ASD and a control sample participated in this study. To participants it was asked to perform a sentence-completion task with two different condition: Part A—targeting response initiation and Part B—engaging inhibitory processes. Importantly, we have analyzed the B-A latencies that have been proposed for the removal of the response initiation confound effect. Results show that no differences between the groups were found in accuracy measures, either in Part A (ASD: M = 0.78; Controls: M = 0.90) nor Part B (ASD: M = 0.03; Controls: M = 0.02). However, in both conditions autistic participants were significantly slower to respond than the group of participants with typical development (Part A—ASD: M = 2432.5 ms; Controls M = 1078.5 ms; Part B—ASD M = 6758.3 ms; Controls M = 3283.9 ms). Critically, we show that when subtracting the response times of Part A from Part B (B-A latencies) no group differences attributable to inhibitory processes remained (ASD: M = 4325.76; Controls: M = 2205.46). With this study we corroborate the existence of difficulties with response initiation in autism and we question the existence of troubles in inhibition per se.