Temporo-Cerebellar Interactions Predict Mentalizing Ability in Youth with Autism: Fmri and DTI Evidence

Background: Posterior superior temporal sulcus is specialized for interpreting perceived human actions, and disruptions to this site have been documented in autism spectrum disorder (ASD). Previous work with typically developing adults has indicated that right posterior superior temporal sulcus (RpSTS) may be supported in its role in biological motion processing by interactions with left Crus I (LCrusI) of neocerebellum (Jack et al., 2011; Sokolov et al. 2010; Sokolov et al. 2012). We hypothesized that disordered RpSTS-LCrusI interactions could predict social deficits in ASD.

Objectives: Investigate relations among individual differences in 1) functional interactions between RpSTS and LCrusI, 2) temporal and cerebellar white matter, and 3) mentalizing ability among adolescents with ASD and typically developing comparison youth.

Methods: 15 high functioning adolescents (12-17 yr) with ASD and 15 same-age comparison youth participated in a simple manual imitation task conducted in the scanner. The format of this task, which involved both observation and imitation of a human hand pressing a randomized sequence of buttons, as well as execution of identical sequences of button presses using visuospatial cues in the absence of a human model, allowed for isolation of brain activity related to observation and use of human motion cues. White matter integrity for these youth was assessed using diffusion tensor imaging (DTI), and behavioral ratings of mentalizing ability were collected via parent report and lab-based assessment. We predicted that stronger interactions between RpSTS and LCrusI during processing and use of biological motion information, as assessed via psychophysiological interaction analysis (PPI), would be associated with better mentalizing ability. We also hypothesized that analysis of DTI data using Tract-Based Spatial Statistics would indicate that better mentalizing ability would be related to higher white matter integrity in cerebellar regions and in tracts theoretically relevant to RpSTS function (i.e. inferior and superior longitudinal fasciculi [ILF; SLF]).

Results: Our sample of high functioning adolescents with ASD and comparison youth did not differ on IQ, head motion, response time, or response accuracy. Consistent with our hypotheses, stronger RpSTS-LCrusI interactions were associated with better mentalizing outcomes among teens with ASD. Lowered white matter integrity was observed in temporal ILF/SLF among youth with ASD; furthermore, individual differences in white matter integrity in this region were related to parent-reported mentalizing ability across the full sample, and these differences in temporal ILF/SLF also helped to predict the strength of RpSTS-LCrusI interactions during biological motion processing.

Conclusions: These results indicate that variability in neocerebellar interactions with key cortical social brain sites, and white matter integrity in temporal regions, may help explain individual differences in social perceptual outcomes in ASD.