Sensitivity to Visual and Proprioceptive Error During Motor Adaptation in Children with Autism

Background: When a force field perturbs a reaching movement, the brain adapts the motor commands and improves performance on the subsequent try.  This adaptation is generalized to other movements.  From the coordinate system of this generalization we recently inferred that in children with autism spectrum disorder (ASD) there may be an increased reliance on proprioceptive sensory feedback as compared to typically developing (TD) children. 

Objectives: To directly examine and quantify autism-associated differences in sensitivity to proprioceptive vs. visual feedback during motor adaptation. 

Methods: Children with ASD, ages 8-12, and TD age matched controls participated in the experimental task.  Subjects were seated in front of a robotic manipulandum.  Above their hand and parallel to their lap was a screen in which cursor and target positions were displayed, obstructing their hand from view.  Children were instructed to hold the manipulandum and make ballistic reaching movements to a target 8-cm from their starting position.  On random trials a perturbation was imposed, consisting of two components.  A proprioceptive perturbation was generated by a force field, displacing their hand perpendicularly from the direction of movement.  A visual perturbation was generated by scaling the hand’s trajectory and displaying that with the cursor, thus creating a smaller, equal or greater visual error than proprioceptive error.  By varying the magnitude of both perturbations, and measuring the learning resulting from each error trial, we could determine the sensitivity to both visual and proprioceptive error.

Results: Current results are based on examination of 5 children in each group.  To assess sensitivity to proprioceptive error, fields of three magnitudes were applied in either direction while visual error was clamped with a gain of zero.  Though proprioceptive perturbations were the same, adaptation was greater for the ASD group than for the TD group (two-way ANOVA , main effect of group [F(1,24)=7.06 , p=0.01]).  This indicates a greater sensitivity to proprioceptive error in children with ASD.  Adaptation with respect to visual error was comparable among the two groups.  However, after accounting for the increased adaptation to proprioceptive error, the ASD group showed decreased visual sensitivity.   This is particularly noticeable when considering the relative decrease in adaption from a gain of one, in which there is both visual and proprioceptive error, to a gain of zero, in which there is only proprioceptive error, for a constant field.  The TD group showed decreased learning in response to the decrease in visual error.  The ASD group, however, was less affected by the loss of visual error with a trend towards significance (one-tailed t-test, p=0.095).  This suggests adaptation to visual errors in the ASD group may be largely due to the underlying proprioceptive error.

Conclusions: The results of this study support the hypothesis that during motor adaptation, children with ASD show both an increased sensitivity to proprioceptive error and a decreased sensitivity to visual error when compared to TD children.  These findings have important implications for improvements in therapeutic practices and highlight key brain regions for future research.