Integrating Behavioral and Electrophysiological Assessments of Sensory Reactivity in ASD

Background: Sensory reactivity is now recognized as a symptom of autism spectrum disorder (ASD) in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). Previous literature indicates that sensory symptoms are present across the lifespan and affect individuals at all levels of functioning. In addition, greater sensory symptoms have been reported in more severely affected children. Obtaining quantifiable measures of sensory reactivity is particularly important for individuals who are unable to describe their own sensory experiences. While there is a clear need to objectively measure sensory reactivity, these symptoms have historically been described on the basis of caregiver report. The current study combines a direct observation and corresponding caregiver interview with an electrophysiological assessment in an effort to develop outcome measures that may be sensitive to change in response to treatment, while gaining insight on the underlying neurobiology of sensory symptoms in ASD.

Objectives: To examine whether observed and reported sensory reactivity is correlated with neural measures of sensory processing.

Methods: This study included minimally verbal children between the ages of 2 and 7 with a diagnosis of ASD. Diagnoses were established using DSM-5 criteria, the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2), and the Autism Diagnostic Interview-Revised (ADI-R). All participants received the Seaver Sensory Assessment for Neurodevelopmental Disorders (SSAND), which consists of a standardized clinician-administered observation and corresponding caregiver interview that characterizes sensory reactivity symptoms based on DSM-5 criteria for ASD. The SSAND provides an overall total score (observed + reported), scores by sensory modality (visual, auditory, tactile), and scores by DSM-5 symptom domain (hyper-, hypo-reactivity, seeking). Visual evoked potentials (VEP) were collected using single-channel EEG recording over the occipital cortex. A contrast revering checkerboard pattern consisting of 32 x 32 checks (check size = 18.75 minarc) contrast reversed with a 1-Hz square-wave signal (100% contrast) was used to elicit transient VEPs.

Results: Results to date indicate a relationship between sensory reactivity symptoms and neural responses. Specifically, P100 amplitudes were negatively correlated with both SSAND total scores and SSAND visual domain scores. P100 reflects inhibitory (GABAergic) activity and therefore the smaller the inhibitory response, the greater were the number of behavioral sensory symptoms. Importantly, there were no correlations between P100 amplitude and auditory or tactile domain scores, confirming the specificity of this finding. Activity in several frequency bands was also correlated with overall SSAND scores and visual domain scores (observed and reported), indicating that greater sensory reactivity at the behavioral level was associated with weaker VEP signal power.

Conclusions: Our results suggest that there is a relationship between observed and reported visual reactivity symptoms and electrophysiological responses. Future directions include examining whether auditory evoked potentials are correlated with auditory domain scores on the SSAND. Ultimately, the goal is to develop outcome measures that combine information from behavioral and neural testing and to examine whether these measures are sensitive to change in the context of clinical trials.