Biological Motion Perception in Autism and Unaffected Siblings

Background: Recent work by Klin et al. (2009) suggests that preferential attention to biological motion, an early-emerging and evolutionarily well-conserved mechanism, is disrupted in autism. Dysfunction in the posterior superior temporal sulcus (pSTS), a region implicated in biological motion perception, has been shown in adults with autism (Pelphrey et al., 2005, Brain). While pSTS abnormalities may underlie biological motion perception deficits in autism, it is possible that these brain abnormalities result from a lifetime of disruption in preferential attention to biological motion. Little is known of the developmental trajectory of these neural processes in children with an autism spectrum disorder (ASD) or unaffected siblings (UAS). Given the high heritability of autism, UAS children may be susceptible to similar neural deficits.

Objectives: We sought to characterize the neural mechanisms involved in biological motion perception in 6-18-year-old children comprising the following groups: ASD (n= 16), UAS (n = 16), and typically-developing (TD)  (n= 11).

Methods: During an fMRI scan, participants viewed point-light displays of dynamic human biological motion and randomly moving point-light animations. Random motion animations were created with equivalent amount of motion. Each condition was presented five times in a block design, with each block lasting 24 seconds.

Results: We defined structural regions of interest including the right pSTS, a middle temporal/occipital cortical region sensitive to viewed motion (MT), the right fusiform gyrus (FFG), the right and left amygdalae, and the right ventrolateral prefrontal cortex (VLPFC). Children with an ASD exhibited hypoactivation (biological > nonbiological motion contrast) relative to the TD and UAS children in the right pSTS, and right and left amygdalae. Both the ASD and UAS children exhibited hypoactivation in the right FFG and right VLPFC relative to TD children. In contrast, we observed equivalent levels of activation in MT for all three groups.

Conclusions: In agreement with prior studies, we observed pSTS dysfunction in children with an ASD relative to TD children. We extend the prior results by revealing continuity in this aspect of brain dysfunction in ASD from childhood to adulthood. In addition, consistent with previous neuroimaging studies of individuals with ASD, we found reduced activation to socially meaningful stimuli in the amygdala. Hypoactivation in the pSTS and amygdala in response to biological motion was specific to individuals exhibiting the autism behavioral phenotype – no such dysfunction was observed in the UAS group. Our results from area MT indicate that neither generalized cortical hypoactivation or global deficits in the processing of motion are part of the neuroendophenotype of ASD. Strikingly, both the ASD and UAS groups showed reduced biological > non-biological activation in the FFG and the VLPFC, relative to TD children. This finding raises the intriguing possibility that hypoactivation in these regions represents a brain mechanism for vulnerability to developing an ASD, even in the absence of the behavioral phenotype.