Abstract Sensory and motor difficulties are common in autism. Altered excitation-inhibition (E-I) balance is a putative framework for understanding atypical sensory and motor function. We investigated whether sensory differences of autism mediate motor difficulties of autism via differences in E-I balance. 106 children were included in the study (Autism n = 44, Typical development children (TDC) n = 62, age 10.32 ± 1.49). E-I balance was assessed through magnetic resonance spectroscopy (MRS), quantifying Glutamate and Glutamine (Glx) and Gamma-Aminobutyric Acid (GABA) in primary sensorimotor cortex (SM1) and thalamus (Thal). Sensory function was evaluated using both objective vibrotactile perceptual sensitivity assessments and subjective parent ratings via the Sensory Experience Questionnaire (SEQ). Motor ability was assessed objectively through the Movement Assessment Battery for Children-second edition (MABC-2) and the Physical and Neurological Examination for Subtle Signs (PANESS). Our findings reveal that lower sensory reactivity and lower tactile thresholds are both predictive of better motor ability (R sig range between 0.32 and 0.57) with higher sensory scores reflecting poorer sensory filtering predicting worse motor function (R sig range −0.22 and −0.63). We identified significant associations between MRS-measured Glx and GABA+ levels and sensory reactivity (p < 0.001). Importantly, sensory reactivity sub-scores were found to fully mediate E-I balance to motor associations in domain-specific patterns: Hyper-reactivity mediated the impact of SM1 Glx levels, while hypo-reactivity mediated the impact of SM1 GABA levels. Additionally, sensory seeking mediated the impact of Thalamic GABA levels with all indirect paths ab p < 0.01. These results propose a model where regional metabolite-specific markers of E-I balance explain patterns of autism-associated sensory and motor difficulties, and where subsequently, distinct sensory phenotypes differentially mediate metabolite-motor associations (see Graphical Abstract for detail).
Shi et al. (Fri,) studied this question.