Autistic Traits and Visual Sensitivity to Human, Animal, and Object Motion

Background: Autism Spectrum Disorder (ASD) is associated with difficulties in social behavior and gesture comprehension-two processes that depend upon the perception of other people’s movements. This raises the question of whether ASD is associated with decrements in visual sensitivity to human movement. Past research indicates that observers with fewer autistic traits demonstrate greater visual sensitivity to human movement than to object movement while observers with more autistic traits demonstrate equal visual sensitivity to human and object movement (Kaiser, Fermano, & Shiffrar, 2008).  Such evidence is consistent with the hypothesis that ASD involves selective deficits in the visual analysis of socially relevant information.  However, the movements of people and objects differ significantly along many non-social parameters including rigidity, frequency, predictability and degrees of freedom.  Do previous results reflect low-level differences in stimulus complexity or high-level differences in social relevance? A new psychophysical study addressed this question by examining visual sensitivity to animal motion.  
The presence of autistic traits varies continuously across clinical and non-clinical populations (Baron-Cohen et al., 2001).  The Autism-Spectrum Quotient (AQ) measures the extent to which typically developing individuals exhibit autistic traits (Baron- Cohen et al., 2001). We compared visual sensitivity to human, animal, and object motion by typical observers and investigated the relationships between each participant’s perceptual sensitivities and the magnitude of that participant’s autistic traits as measured by the AQ.
Objectives: The goal of this work is to determine whether increases in the magnitude of autistic traits is associated with selective decreases in the ability to perceive other people’s movements per se.
Methods: Fifty-four typical adult participants completed the AQ and a classic psychophysical measure of visual sensitivity to complex motion. Point light displays of a walking person, walking dog, and moving tractor were constructed from motion capture data. In a blocked design, these stimuli were presented coherently (present) or scrambled (absent) and hidden within dynamic point-light masks that inhibited the use of local motion processes. On each trial, participants reported with a button press whether the target stimulus was present or absent. No feedback was provided.
Results: Correlation analysis revealed that AQ score was most strongly associated with visual sensitivity to human motion (r = -.30, p = .01), less strongly associated with sensitivity to dog movement (r = -.21, p = .06) and not significantly related to visual sensitivity to object motion (r = -.06, p = .33). Thus, increases in autistic traits are associated with decreases in visual sensitivity to human motion in specific and biomechanical motion generally.
Conclusions: There is a significant relationship between the presence of autistic traits in typical observers, as measured by the AQ, and changes in visual sensitivity to other people’s actions and, to a lesser extent, changes in visual sensitivity to the actions of animals, at least when they are performing actions that humans can also produce (e.g., walking).  This supports the hypothesis that visual motion processes are meaningfully related to social abilities.