Atypical Premotor, Parietal, and Cerebellar Functioning Underlies Sensorimotor Impairment In Autism

Background: The brain system abnormalities underlying sensorimotor deficits in autism are not well established.

Objectives: Investigate brain alterations underlying visuomotor control impairments in autism.

Methods: Using fMRI, we studied 16 individuals with autism and 17 age-matched healthy control individuals performing a sustained precision grip force task.  All subjects were right-handed.  Subjects gripped a transducer while viewing a white force bar that moved upwards with increased force toward a fixed target bar.  They were instructed to maintain a constant force level so that the white force bar remained at the level of the target bar set to 15% of each individual’s maximum force contraction.  The gain of the feedback, defined as the vertical distance the force bar moved per Newton of force applied to the transducer, was compared in low and high conditions each presented during 26 sec trials.  When gain was low, the white bar moved a smaller distance for every Newton of force applied compared to when gain was high.  Subjects also performed trials in which no visual feedback was given.

Results: Subjects with autism and controls did not differ in their mean force level.  The error of sustained force was greater for subjects with autism relative to controls, P=.01, indicating a difficulty using sensorimotor feedback to sustain a constant force output.  When the gain setting was low, and thus visual feedback less precise, and when no visual feedback was provided, subjects with autism showed increased force error compared to controls, P’s<.05.  During the low gain condition, subjects with autism showed reduced activation in left motor and premotor cortices, bilateral superior parietal lobule, and bilateral anterior (lobules III/IV) and posterior cerebellum (lobules V/VI), and increased activation in right MT/V5 and left posterior cerebellum (Crus I).  During the high gain condition, subjects with autism showed reduced activation in left middle frontal gyrus, right inferior parietal lobule and bilateral V3, and increased activation relative to controls in bilateral supplementary motor area (SMA), left motor cortex, right MT/V5, bilateral putamen and posterior cerebellum (Crus II).  During the no visual feedback condition, subjects with autism showed reduced activation within bilateral dorsolateral prefrontal cortex, medial frontal gyrus, and ipsilateral cerebellar lobules V/VI.

Conclusions: Individuals with autism showed greater isometric force error, suggesting impaired utilization of visual feedback to guide motor performance.  These impairments implicate dysfunction within lateral cerebellar circuitry that typically modulates the accuracy of motor performance by transforming sensory inputs about performance error into corrective afferent motor commands.  This system is not able to guide accurate motor performance in autism when the precision of visual input is attenuated.  Increased activity within the SMA, putamen and posterior cerebellum in the context of intact performance during a condition of increased sensory feedback indicates that subjects with autism may recruit alternate brain systems typically involved in complex motor planning.  These findings provide evidence that cerebellar dysfunctions and their interaction with neorcortical systems may underlie dyspraxia in autism, and that even intact sensorimotor performance in autism may involve atypical recruitment of compensatory brain systems.