Superior Temporal Sulcus Response in Attributing Social Meaning to Actions in Autism Spectrum Disorder

Background:  Previous studies have shown that individuals with Autism Spectrum Disorder (ASD) often struggle to accurately attribute emotions and intentions to others. Such failures in ASD may result from a limited ability to attribute biological agency. In other words, a reduction in anthropomorphism - the attribution of human characteristics, intentions, or behaviors to non-human animals or objects (Chaminade et al., 2007). Due to its widespread and implicit nature, anthropomorphism is generally considered an innate human tendency (Hutson, 2012). However, children with ASD are less likely than their peers to use anthropomorphic language (Hieder & Simmel, 1944). Furthermore, neuroimaging studies have shown reduced activation in certain mentalizing regions (superior temporal sulcus, STS; medial prefrontal cortex, MPFC; and anterior cingulate cortex, ACC) during social attribution in ASD (Kana et al., 2009) and in response to biological motion (Pelphrey et al., 2005). Relatively less work has looked at the neural correlates of anthropomorphism combining social attribution and biological motion processing in ASD.

Objectives:  To examine the neural correlates of attributing social meanings to human and non-human figures in dynamic anthropomorphic scenarios in ASD.

Methods:  34 age-and-IQ-matched participants (17 ASD, 17 TD) viewed short animations of either human-like stick figures or geometrical shapes (such as triangles) engaged in a series of random or socially meaningful movements (i.e. bullying, helping) during functional MRI scanning. In an event related design, participants were asked to determine via button press whether the character’s movement was social or random. Structural and functional images were acquired in a 3T Siemens Allegra head-only scanner (17 oblique-axial slices; TR=1,000ms; TE=30ms) and processed in SPM12.

Results:  ASD and TD groups were equally accurate at identifying social movement by human and non-human characters [Diagnosis x Movement x Character: F(1,28)=.103,NS]. Within-group analysis of functional activation revealed that observation of socially meaningful movement, regardless of character, activated bilateral posterior STS, bilateral MPFC, and the precuneus (Monte Carlo simulation corrected, p=.005, k=100) in the TD group. Activation in the ASD group was limited to the right pSTS and left MPFC. Furthermore, in the TD group attributing social vs non-social motives to shapes (anthropomorphism) was associated with greater activation of the left pSTS and bilateral MPFC compared to no activation differences in the ASD group. Direct comparison of brain activation between ASD and TD groups revealed a significant decrease in left pSTS activation in the ASD group associated with anthropomorphism (MC corrected, p=.01, k=200).

Conclusions:  This study demonstrated that despite equal accuracy at identifying social movement of non-social objects, high functioning individuals with ASD may be using slightly different neural resources to accomplish the task compared to their TD peers. The findings of this study underscore the role of pSTS in biological and social information processing as it was recruited for human and non-human social movements alike. Greater bilateral recruitment of pSTS by the TD group when observing the social movement of shapes may reflect a stronger neural response toward anthropomorphism, whereas underactivity of this region in ASD may be a neural signature of their altered social functioning.