Sensory over-Responsivity and Amygdala Response to Emotional Faces in Youth with and without ASD

Background: Sensory over-responsivity (SOR) is extremely common in individuals with ASD (Baranek et al., 2006; Ben-Sasson et al., 2007) and is now included in the DSM-V diagnostic criteria. Previous studies in our lab have shown that in children with ASD, higher SOR symptoms are associated with amygdala hyperactivation in response to mildly aversive visual, auditory, and tactile stimuli (Green et al., 2014; under review). Individuals with SOR may over-attribute salience to sensory stimuli resulting in an overactive amygdala response. Additionally, sensory stimuli may trigger an anxiety response, given the overlap between SOR and anxiety (Green & Ben-Sasson, 2010). It has also been shown that individuals with ASD have an exaggerated amygdala response to emotional faces, when directed to look at the eyes (Dalton et al., 2005). However, it is not known if the overreactive amygdala response in individuals with SOR is specific to basic sensory stimuli, or if they have more generalized amygdala hyperactivity to emotional faces, such as in individuals with ASD. We were interested in fearful faces in particular, given that children with anxiety disorders display amygdala hyperactivity to fearful faces (e.g., Thomas et al., 2001), and SOR is so highly correlated with anxiety.

Objectives: To understand if SOR symptoms are associated with amygdala hyperactivity in response to emotional faces, and to fearful faces specifically.

Methods: Participants were 20 children and adolescents with ASD and 20 matched TD controls, ages 8-16. While undergoing fMRI, participants passively observed faces displaying different emotions (fearful, happy, sad, angry, and neutral) in the scanner. We used a jittered event-related design, where faces were presented every 3 sec (for 2 sec each) according to an optimized random sequence. Parents rated children’s symptoms of SOR with the Short Sensory Profile (Dunn, 1999) and with the Sensory Over-Responsivity Inventory (SensOR; McIntosh, 1999). Scores from subscales reflecting over-responsivity were standardized and combined to create a sensory composite score. Parents also rated their children’s anxiety symptoms using the CBCL Anxiety subscale.

Results: FSL was used to run subject-level and group-level analyses with the following contrasts: All faces > Fixation and Fear > Neutral. Within- and between-group analyses were thresholded at Z > 1.7. Anxiety scores and SOR composite scores were included as regressors in a second group-level analysis to examine the relationship between SOR symptoms and amygdala response, while controlling for anxiety. The ASD group had greater amygdala activation in the All > Fixation condition only (whole-brain correction at p<.05); there were no group differences in the Fear > Neutral condition.  SOR and anxiety scores were not correlated with amygdala response for either group in either condition, based on a small volume correction at p<.05 within the right and left amygdalae.

Conclusions: Our findings suggest that for individuals with SOR, hyperactivity of the amygdala is not a generalized response, but is instead specific to mildly aversive sensory stimuli. Clinical implications of this finding include the possibility that for individuals with SOR, anxiety and distraction can be significantly decreased by limiting exposure to extraneous aversive environmental stimuli.