Neurophysiological Correlates of Sensory‐Based Subtypes in Autism

Abstract: Purpose: Children with autism spectrum disorders often present with atypical responses to sensory stimuli in the environment. Additionally, this population has shown differences in autonomic nervous system activity, in both parasympathetic and sympathetic systems, as well as neuroendocrine response during the presentation of sensory challenges. However, findings are mixed and no one consistent responsivity pattern appears to explain these differences within this heterogeneous population. Sensory-based subtypes have been developed to help create more homogenous autism subgroups. One such system, using both type of responsivity (hypersensitivity, hyposensitivity, sensory seeking) and sensory domain shows initial promise. However, differences in nervous system response to sensory input between these sensory-based subtypes have not yet been explored. This study used indices of neuroendocrine (salivary cortisol) and autonomic nervous system activity (skin conductance level and respiratory sinus arrhythmia [RSA]), in order to explore patterns that could differentiate the subtypes. Design: This retrospective study utilized secondary data from two university-based research groups. Data was combined and subjects were assigned to each of the four different sensory-based subtypes. Neurophysiological measurements previously recorded during the Sensory Challenge Protocol were then compared between each of the different subtypes. Baseline measures were also included to allow pre- and post-Sensory Challenge Protocol comparisons. Method: Quantitative analyses included between-group comparisons (repeated-measures ANOVA and MANOVA) and logistic regression to examine predictors of subtype membership. Dependent variables were cortisol, skin conductance and RSA, and the independent variable was subtype membership (four independent subtypes). Results: Results were largely non-significant, with the exception of RSA that was able to differentiate subtypes with typical versus atypical sensory responsivity. Differences were found during baseline RSA, and also during tone, tactile, and movement stimuli (p<.05). Additionally, membership in certain subtypes was predicted by RSA during auditory stimuli and during recovery periods (p<.05). Small sample size from secondary data and measurements available for each subject were substantial limitations for the analyses. Conclusion: The selected subtyping system, based on sensory responsivity, may not best distinguish actual sensory processing differences within the autism population. It is also possible that these neurophysiological markers do not fully reflect the complex response of the nervous system to multi-sensory inputs in the natural environment. Patterns of neurophysiological response, rather than isolated responses to individual sensory stimuli, may better distinguish different subgroups and be more useful for diagnosis and treatment planning. This research provides further evidence that children with sensory processing disorders demonstrate different neurophysiological responses to sensation than typically developing children. Moreover, the results suggest that a new approach to subtyping may be necessary. Additional research is needed to explore the merits of behavior-based sensory subtypes in relation to neurophysiological response patterns.