The Role of the Corpus Callosum in Sensory Hyporesponsiveness of Children with Autism Spectrum Disorder

Background: Atypical experience of sensory stimuli is commonly reported in individuals with Autism Spectrum Disorder (ASD) (Baranek et al., 2006) and in those without ASD who have higher autism-like traits (Robertson & Simmons, 2013). However the neurobiological basis of these sensory features is unclear. The white matter integrity of the corpus callosum (CC) is a viable candidate for the neurobiological basis of atypical sensory experience as it has common reports of group differences related to ASD (Travers et al., 2012), and has been shown to have an atypical developmental trajectory in children with ASD (Travers et al., 2015). Importantly, the corpus callosum is critical for the facilitation of hemispheric synchrony, which has demonstrated a role in typical sensory experience (Ouimet et al., 2010). Therefore, the white matter integrity of the CC may potentially correspond to individual differences in sensory features both within and beyond an ASD group.

Objectives: To characterize the relationship between CC white matter integrity and degree of sensory hyper- and hypo-responsiveness in children with ASD, children with typical development (TD), and children at higher genetic risk for ASD (ASD-Related).

Methods: Seventy-four children (6-10 years old) completed a diffusion-weighted imaging (DWI) scan and sensory measures (32 children with ASD, 26 children with TD, and 16 ASD-Related children). The ASD-Related group included children with an ADHD diagnosis and children with a first-degree relative with ASD, major depressive disorder, bipolar disorder, or schizophrenia. Groups were well-matched on age (p =.30). Hyper- and hypo-responsiveness were assessed using the parent-reported Sensory Experience Questionnaire Version 3 (Baranek, David, Poe, Stone, & Watson, 2006). Multi-shell DWI was performed on a 3T GE Scanner with a 32-channel head coil with protocol: 63 encoding directions: b=0 (6); b=350 (9); b=800 (18); b=2000 (36) s/mm²; isotropic 1.8mm voxels: 3.6 mm thick slices with 1.8 mm overlap. Median fractional anisotropy (FA) in each of ten CC sub-regions (Genu: G1, G2, G3; Body: B1, B2, B3; Isthmus; Splenium: S1, S2, S3) was used for correlational analyses. False discovery rate (fdr) was used for multiple comparisons.

Results: Hyporesponsivity was significantly correlated in the Autism group with FA in G2 (r=.46, p=.01 [fdr]), G3 (r=.47, p=.01 [fdr]), B1 (r=.61, p<0.001 [fdr]) (Figure 1). These CC sub-regions primarily connect to the frontal lobe. In contrast, hyperresponsivity was not correlated with any sub-region in any diagnostic group.

Conclusions: More severe hyporesponsivity symptoms were associated with greater white matter integrity in the CC genu and body, although no area of the CC was associated with hyperresponsivity. These results suggest that potential overconnectivity between the left and right frontal lobes may be associated with more severe hyporesponsivity. However, follow-up analyses will examine other DWI metrics to see if these hyporesponsivity associations are consistent with greater axonal packing, myelination differences, or both. Intriguingly, the present data suggest that hypo and hyper-responsivity may have different neurobiological substrates, although this will need to be corroborated by future research.