Symptoms of Sensory Sensitivity and Anxiety As Predictors of Amygdala and Hippocampus Activation to Sensory Stimuli in Youth with and without ASD

Background: Children with ASD often exhibit sensory over-responsivity (SOR), which may cause them to react negatively to sensory stimuli such as noisy or visually stimulating environments (Liss et al., 2006).  Rates of SOR are over five times higher in children with ASD than in typically developing (TD) children (e.g., Baranek et al., 2006; Ben-Sasson et al., 2007) and SOR is associated with increased functional impairment in children with ASD (e.g., Liss et al., 2006; Pfeiffer et al., 2005).  SOR has been shown to frequently co-occur with anxiety in children with ASD (e.g., Ben-Sasson et al., 2008). The neural bases for SOR are still unknown, but the high co-occurrence of SOR and anxiety suggests that underlying limbic system abnormalities may put individuals with ASD at risk for both conditions (Green & Ben-Sasson, 2010; Hitoglou et al., 2010; Waterhouse et al., 1996).  However, no functional MRI (fMRI) studies have investigated the neural bases of SOR in children with ASD.

Objectives: The purpose of this study was to examine the relationship of parent-reported SOR and anxiety symptoms with brain responses to mildly aversive sensory stimuli in youth with and without ASD.

Methods: Participants were 24 children and adolescents with ASD and 24 typically-developing (TD) controls, between 8-17 years.  During fMRI, participants were presented with mildly aversive auditory (white noise) and visual (a continually rotating color wheel) stimuli.  Each stimulus trial was 3 seconds long and consisted of either the auditory stimulus, visual stimulus, or both. Each trial type was presented 12 times. Participants’ parents rated their symptoms of SOR with the Sensory Profile (Dunn, 1999) and Sensory Over-Responsivity Inventory (Schoen et al., 2008). Scores on the relevant subscales (auditory and visual sensitivity) of these measures were standardized and combined to create a sensory composite score. Parents also rated their children’s anxiety symptoms using the Child Behavior Checklist (CBCL).

Results: Amygdala and hippocampus activation across groups during all three conditions was used to create functional masks, and parameter estimates were extracted from these masks for each participant during each sensory condition (auditory, visual, or both) as compared to baseline. Hierarchical regression was used to predict amygdala and hippocampus activation during each condition. Status (ASD vs. TD), a sensory composite score, CBCL anxiety score, and a status by sensory composite interaction term were entered as predictors in four separate steps. Higher sensory composite scores significantly predicted greater amygdala and hippocampus activation in the auditory and joint conditions after accounting for anxiety (DR^s ranged from .09-.14, p<.05). The interaction term was not significant.

Conclusions: Findings suggest that SOR is related to hyperactivation of the amygdala and hippocampus and these results cannot simply be accounted for by higher anxiety in children with SOR.  The relationship between SOR and amygdala/hippocampus activation appears to be similar across children with and without ASD.