Neural Mechanisms of Habituation to Sensory Stimuli and Generalization of Response in Youth with and without ASD

Background: Sensory over-responsivity (SOR) is an impairing condition that is extremely common in children with ASD, and is manifested as negative reactions to sensory stimuli (e.g., noisy environments, scratchy clothing; Liss et al., 2006; Ben-Sasson et al., 2007). Prior results from our lab showed that SOR in ASD was related to greater brain activity and reduced habituation to auditory and tactile stimuli in amygdala and somatosensory cortex (Green et al, 2015), suggesting that treatment of SOR needs to take into account fundamental differences in neural mechanisms of response to sensory information, including initial arousal, habituation, and generalization of response. In this study, we extend our findings on neural responses to sensory stimuli by examining neural mechanisms of generalization in children with and without ASD and differing levels of SOR.

Objectives: To examine (1) differences in brain responses to auditory and tactile stimuli in youth with and without ASD; (2) the extent to which ASD and TD youth generalize responses to new but similar stimuli; and (3) brain responses and generalization as a function of SOR severity.

Methods: Participants were 38 children and adolescents with ASD and 22 TD matched controls, aged 8-17 years. During fMRI, participants were presented with mildly aversive auditory (pulsing white noise) and tactile (scratchy sponges) stimulation. Stimuli were presented first in a habituation phase (6 blocks of 15-sec trials each of auditory and tactile stimuli together). Subsequently, different but similar stimuli matched for aversiveness (i.e. different frequencies of white noise; different sponge material) were presented in a generalization phase (4 blocks of 15-sec trials each of the stimuli). Participants’ parents rated their symptoms of SOR with the Short Sensory Profile (Dunn, 1999) and the Sensory Over-Responsivity Inventory (Schoen et al. 2008). Scores on the auditory and tactile sensitivity subscales were standardized and combined to create a sensory composite score.

Results: Within- and between-group analyses were conducted using FSL and corrected for multiple comparisons (p<.05). In the first (Habituation) phase, the ASD group showed significantly greater activation in widespread regions including limbic areas, sensory cortices, and medial prefrontal cortex (Figure 1). Activity in these regions was positively correlated with parental report of SOR severity within the ASD group. In the second (Generalization) phase, the ASD group showed significantly greater activation only in sensory-motor cortex (Figure 2). However, in this phase, SOR was correlated with greater activity in right amygala and orbital and medial prefrontal cortex but not with activity in auditory or somatosensory cortex.

Conclusions: We replicate prior findings that SOR in ASD is related to hyperactivation in response to sensory stimuli in sensory and limbic regions, and further highlight potential mechanisms for deficits in generalizing responses to sensory stimuli. Specifically, ASD youth may have a deficit in generalizing primary sensory processing of tactile stimuli, and those with SOR may have a distinct deficit in modulating amygdala response to sensory stimuli. Results suggest that ASD youth with SOR may respond to new but similar stimuli as if they are novel and salient, with implications for intervention.