Implicit Sequence Learning in Persons with ASD

Background:

Recent studies have found impairments in implicit concept learning in persons with autism spectrum disorder (ASD).  However, it is unclear whether persons with ASD also show impairments in implicit motor learning.  The serial reaction time (SRT) task is the most common measure of implicit motor learning.  Previous research with the SRT in persons with ASD has yielded contradictory findings.  One study showed implicit motor learning impairments (Mostofsky et al., 2000), but two studies did not show impairments (Gordon & Stark, 2007; Müller et al., 2000).  

Objectives:

The present study sought to test whether persons with ASD show impairments in the SRT task.   This study aimed to decrease the role of explicit learning processes in the SRT task so that it more strongly tapped implicit learning processes. 

Methods:
Fourteen high-functioning adolescents and young adults with ASD and 15 age and verbal-ability matched controls completed a SRT task.  Participants responded with a keypress indicating the location of a racecar on the computer. Unknown to participants, the location of the racecar followed a repeating 12-step sequence. Participants completed four blocks of 120 sequenced trials followed by a block of 120 random location trials and another sequenced block.

Results:

For this task, learning is seen as the difference in reaction time to the final blocks of sequenced trials compared to the block of random trials.  Both groups of participants responded faster to the sequenced trials than the random trials (control group difference=+35.03ms; ASD difference=+43.24ms).  Differences between groups were not significant. 

Conclusions:

It appears that persons with ASD demonstrated implicit motor learning equivalent to a matched control group.  This result suggests that not all implicit learning is impaired in persons with ASD.  Differences may exist between implicit learning of concepts and motor actions.