Elevated Motion Coherence Thresholds for Slow Stimuli in Individuals with Autism

Background:  While some studies have reported elevated motion coherence thresholds in autism, others report motion coherence thresholds comparable to those of typically developing individuals. This discrepancy may, at least in part, be attributable to differences in task and stimulus parameters across studies. Some authors have suggested that individuals with autism process speed information differently to typically developing individuals. Such atypicalities in speed processing may therefore have an effect on coherent motion perception.

Objectives:  We sought to investigate whether the speed of coherent motion was a contributing factor in determining the extent of differences between autistic and typically developing individuals.

Methods:  We measured motion coherence thresholds under slow (1.5 deg/sec) and fast (6 deg/sec) speed conditions in children with an autism spectrum condition aged 7 to 14 years (n = 23), and age- and ability-matched typically developing children (n = 25). Stimuli were pairs of limited lifetime random dot stimuli presented sequentially: one stimulus had dots moving in random directions, and one stimulus had a proportion of dots moving coherently leftwards or rightwards.  Participants were asked to select the interval containing coherent motion.

Results:  Overall, as expected, children were more sensitive to coherent motion in the fast condition than the slow condition. Children with autism showed elevated motion coherence thresholds relative to typically developing children but only in the slow condition.

Conclusions:  Rather than children with autism having pervasive difficulties in motion processing, these results suggest that they have a selective difficulty in extracting coherent motion information specifically at slow speeds.  This pattern might be attributable to increased neural noise in children with autism due to undersampling of local motion signals, which may have a particularly pronounced effect on coherent motion processing in ‘slow’ motion channels. Understanding the effects of stimulus parameters such as stimulus speed will be important for resolving discrepancies between previous studies that have found elevated motion coherence thresholds in autism and those that have not and also for refining theoretical models of altered perception in autism.