Procedural Learning Abnormalities in First-Degree Relatives of Individuals with Autism

Background: Procedural learning involves the implicit learning of simple behavioral response patterns through repeated practice.  Individuals with autism have impaired performance on complex serial response learning tasks involving manual responses, and an altered ability to time predictive responses during an oculomotor serial reaction time task. 

Objectives: To investigate procedural learning during a predictive saccade task of procedural learning in first-degree relatives of individuals with autism. 

Methods: Fifty-one first-degree relatives (parents and siblings) of individuals with autism and 37 age- and IQ-matched healthy control individuals between 8-54 years of age performed an oculomotor predictive saccade task and a sensorimotor control task. During the predictive saccade task, participants looked at a light that shifted position back and forth between locations 6 degrees of visual angle to the left and right of central fixation every 750 ms seconds for 40 trials.  A visually guided saccade (VGS) control task was administered in which saccades to unpredictable targets were examined. The latency of saccades was examined in each eye movement task.

Results: The timing of saccades to unpredictable targets was not altered in the unaffected family members.  In the predictive saccade task, family members’ rate of learning to speed their responses to predictable target appearance was slower than that of healthy controls.  Further, their response latencies remained slow after learning asymptote had been achieved. 

Conclusions: First-degree relatives of individuals with autism show decreased rates of procedural learning on an oculomotor serial reaction time task, and they are not able to respond as rapidly as controls once they achieve peak performance.  This pattern of findings likely reflects alterations in frontostriatal systems.  The demonstration of reduced procedural learning capacity in unaffected family members of individuals with autism suggests that this deficit may serve as a useful endophenotype for family genetic research.