Autism Spectrum Disorder (ASD) is a developmental disorder characterized by impairments in social cognition, communication, and repetitive behaviors¹. In addition, difficulties with sensory stimuli have been increasingly recognized and are now included as a diagnostic feature of ASD. Multiple lines of evidence suggest that GABA, the main inhibitory neurotransmitter in the brain, plays a role in the pathophysiology of ASD (e.g.²). It is well known that GABA plays a key role in regulating tactile processing^3-5. However, the link between GABA and autism-associated impairments in vibrotactile processing remains unclear. GABA can be measured using edited Magnetic Resonance Spectroscopy (MRS) and we developed a technique to measure tactile sensitivity in children objectively, in tasks linked to inhibition⁶. In this study, in a large cohort, we aimed to investigate whether 1) children with ASD have reduced GABA levels, 2) children with ASD show altered vibrotactile processing, and 3) altered GABA levels are associated with abnormal tactile processing in ASD.Subject and parental consent were obtained under local IRB approval. Eligibility: Children with ASD met the DSM-V criteria for ASD, confirmed using the Autism Diagnostic Observation Schedule-Version 2 (ADOS-2). All Typically Developing Children (TDC) were free of criteria for psychiatric disorders using the Diagnostic Interview for Children and Adolescents. Data were acquired in 37 children with ASD (10.69 ± 1.4 years, 6F) and 35 TDC (10.09 ± 1.25 years; 8F). Children had normal IQ. Imaging: GABA-edited MR spectra were acquired from (3 cm)3 voxels over right primary sensorimotor (Fig 1A&B) and occipital cortices using MEGA‑PRESS, on a 3T Philips ‘Achieva’ scanner (Philips Medical Solutions, Best, the Netherlands). Parameters: TE 68 ms; 14 ms editing pulse at 1.9 (ON) and 7.46 (OFF) ppm; TR 2000ms; 320 transients, 32-channel head coil. A T1w MPRAGE was acquired prior to MRS acquisition for voxel placement and segmentation (TR = 7.99 ms, TE = 3.76 ms, Flip angle = 8°). GABA levels were calculated against the unsuppressed water signal from the same voxel. Data were tissue corrected, normalized to a cohort (ASD + TDC) voxel average grey and white matter ratios as described previously⁷, and implemented within Gannet⁸. Behavioral: Children performed: 1) Static and Dynamic detection tasks (where the increasing (dynamic) sub-threshold stimulus is thought to act through feed-forward inhibition⁹); 2) Amplitude discrimination with- and without an adapting stimulus (linked to lateral inhibition¹⁰); and 3) Frequency discrimination (encoded through GABAergic inhibition^3,11; FD).Sensorimotor GABA levels were significantly reduced in children with ASD compared to TDC (2.20 ± 0.44 i.u. and 2.40 ± 0.25 i.u. respectively, p = 0.016). There were no group differences in occipital GABA levels. For children with ASD, sensorimotor GABA levels correlated positively with dynamic detection threshold, with higher GABA levels indicating a higher detection threshold (Fig 2A). The difference between a static and dynamic detection threshold (Figure 2B) were significantly correlated with GABA levels within the entire cohort (R = 0.39, p < 0.005); this finding was driven by children with ASD (Fig 2B), such that those with lower GABA levels showed a reduced effect of sub-threshold stimulation. Amplitude discrimination performance after single-site adaptation correlated with GABA levels in TDC (R = 0.44, p = 0.019) but not in ASD (R =-0.21, p = 0.3; Fig 2B) and the effect of adaptation was absent in ASD. Tactile FD is correlated with GABA in TDC, which has been previously reported¹¹ (R = -0.41, p =0.025; Fig 2C); this association was not observed in ASD.Sensorimotor GABA levels were reduced in children with ASD compared to TDC, while occipital levels are normal, consistent with previous work¹². Tactile abnormalities in ASD were consistent with previous work, and the absence of the effect of either a sub-threshold stimulus, or an adapting stimulus, in ASD, is consistent with reduced GABA-mediated inhibition. Our correlative results are consistent with the prediction for children with ASD; reduced GABA level was associated with stronger effect of modulating stimuli (sub-threshold and adaptation). Finally, GABA levels in TDC are correlated with FD, which is consistent with published findings in healthy adults. While children with ASD do not show differences in FD performance, they do not show GABA-FD correlation, suggesting that GABA does not act similarly to TDC in terms of frequency encoding.Children with ASD show reduced GABA levels, and are associated with abnormalities in tactile performance. Altered in vivo GABA levels might contribute to abnormal tactile-information processing in ASD. The GABA system may be a future and novel target for therapies in ASD.