Background Abnormal responses to tactile stimuli are a common feature of Autism Spectrum Disorder (ASD). Several lines of evidence suggest that GABAergic function, which has a crucial role in tactile processing, is altered in ASD. In this study, we determine whether in vivo GABA levels are altered in children with ASD, and whether alterations in GABA levels are associated with abnormal tactile function in these children. Methods GABA-edited MRS was acquired in 37 children with Autism and 35 Typically Developing Children from voxels over primary sensorimotor and occipital cortices. Children performed tactile tasks previously shown to be altered in ASD, linked to inhibitory mechanisms. Detection threshold was measured with- and without the presence of a slowly increasing sub-threshold stimulus. Amplitude discrimination was measured with- and without the presence of an adapting stimulus, and frequency discrimination was measured. Results Sensorimotor GABA levels were significantly reduced in children with autism compared to healthy controls. Occipital GABA levels were normal. Sensorimotor GABA levels correlated with dynamic detection threshold as well as with the effect of sub-threshold stimulation. Sensorimotor GABA levels also correlated with amplitude discrimination after adaptation (an effect absent in autism) and frequency discrimination in controls, but not in children with autism. Conclusions GABA levels correlate with behavioral measures of inhibition. Children with autism have reduced GABA, associated with abnormalities in tactile performance. We show here that altered in vivo GABA levels might predict abnormal tactile information processing in ASD and that the GABA system may be a future target for therapies.

Impaired responses to tactile stimulation are a commonly reported symptom among children with autism spectrum disorder (ASD). Furthermore, impairments in filtering or habituation to tactile input have been described in ASD. This study measured different aspects of tactile processing to investigate atypical touch sensitivity in children with ASD, methodology that has not been previously used in this population. Sixty-seven typically developing children (TDC) and 32 children with ASD (ages 8-12) completed vibrotactile tasks assessing: reaction time (RT); static and dynamic detection threshold (DT); amplitude discrimination with and without single-site adaptation; frequency discrimination; and temporal order judgment (TOJ) with and without concurrent stimulation. Children with ASD showed raised static detection thresholds and an absence of the effect of a dynamically increasing subthreshold stimulus on static detection threshold. Children with ASD also showed poorer amplitude discrimination than TDC, as well as decreased adaptation. There were no significant differences in frequency discrimination or TOJ performance between the groups. Differences in the effect of dynamic stimulation on detection threshold suggest impaired feed-forward inhibition in autism, which may be linked to poor sensory filtering. Increased baseline amplitude discrimination thresholds in ASD suggest that lateral inhibitory connections are weaker in ASD, and an absence of the effect of adaptation suggests impaired modulation of lateral inhibitory connections in ASD, which may relate to aberrant habituation. These results suggest a functional deficit in the somatosensory inhibitory system in autism. Understanding the specific mechanisms underlying sensory symptoms in autism may allow for more specific therapeutic or drug targeting in the near future.

The intensity and submodality of pain are widely attributed to stimulus encoding by peripheral and subcortical spinal/trigeminal portions of the somatosensory nervous system. Consistent with this interpretation are studies of surgically anesthetized animals, showing that relationships between nociceptive stimulation and activation of neurons are similar at subcortical levels of somatosensory projection and within the primary somatosensory cortex (in cytoarchitectural areas 3b and 1 of SI). Such findings have led to characterizations of SI as a network which preserves, rather than transforms, the excitatory drive it receives from subcortical levels. Inconsistent with this perspective are images and neurophysiological recordings of SI neurons in lightly anesthetized primates. These studies show that an extreme anterior position within SI (area 3a) receives input originating predominantly from unmyelinated nociceptors, distinguishing it from posterior SI (areas 3b and 1), long recognized as receiving input predominantly from myelinated afferents, including nociceptors. Of particular importance, interactions between these subregions during maintained nociceptive stimulation are accompanied by an altered SI response to myelinated and unmyelinated nociceptors. A revised view of pain coding within SI cortex is discussed, and potentially significant clinical implications are emphasized.

Functional connectivity is abnormal in autism, but the nature of these abnormalities remains elusive. Different studies, mostly using functional magnetic resonance imaging, have found increased, decreased, or even mixed pattern functional connectivity abnormalities in autism, but no unifying framework has emerged to date. We measured functional connectivity in individuals with autism and in controls using magnetoencephalography, which allowed us to resolve both the directionality (feedforward versus feedback) and spatial scale (local or long-range) of functional connectivity. Specifically, we measured the cortical response and functional connectivity during a passive 25-Hz vibrotactile stimulation in the somatosensory cortex of 20 typically developing individuals and 15 individuals with autism, all males and right-handed, aged 8-18, and the mu-rhythm during resting state in a subset of these participants (12 per group, same age range). Two major significant group differences emerged in the response to the vibrotactile stimulus. First, the 50-Hz phase locking component of the cortical response, generated locally in the primary (S1) and secondary (S2) somatosensory cortex, was reduced in the autism group (P < 0.003, corrected). Second, feedforward functional connectivity between S1 and S2 was increased in the autism group (P < 0.004, corrected). During resting state, there was no group difference in the mu-α rhythm. In contrast, the mu-β rhythm, which has been associated with feedback connectivity, was significantly reduced in the autism group (P < 0.04, corrected). Furthermore, the strength of the mu-β was correlated to the relative strength of 50 Hz component of the response to the vibrotactile stimulus (r = 0.78, P < 0.00005), indicating a shared aetiology for these seemingly unrelated abnormalities. These magnetoencephalography-derived measures were correlated with two different behavioural sensory processing scores (P < 0.01 and P < 0.02 for the autism group, P < 0.01 and P < 0.0001 for the typical group), with autism severity (P < 0.03), and with diagnosis (89% accuracy). A biophysically realistic computational model using data driven feedforward and feedback parameters replicated the magnetoencephalography data faithfully. The direct observation of both abnormally increased and abnormally decreased functional connectivity in autism occurring simultaneously in different functional connectivity streams, offers a potential unifying framework for the unexplained discrepancies in current findings. Given that cortical feedback, whether local or long-range, is intrinsically non-linear, while cortical feedforward is generally linear relative to the stimulus, the present results suggest decreased non-linearity alongside an increased veridical component of the cortical response in autism.

A recent study (Tannan et al., 2006) showed that pre-exposure of a skin region to a 5 sec 25 Hz flutter stimulus ("adaptation") results in an approximately 2-fold improvement in the ability of neurologically healthy human adults to localize mechanical stimulation delivered to the same skin region that received the adapting stimulation. Tannan et al. (Tannan et al., 2006) proposed that tactile spatial discriminative performance is improved following adaptation because adaptation is accompanied by an increase in the spatial contrast in the response of contralateral primary somatosensory cortex (SI) to mechanical skin stimulation -an effect identified in previous imaging studies of SI cortex in anesthetized non-human primates (e.g., Simons et al., 2005;Tommerdahl et al., 2002;Whitsel et al., 1989).In the experiments described in this report, a paradigm identical to that employed previously by Tannan et al. (2006) was used to study adults with autism. The results demonstrate that although cutaneous localization performance of adults with autism is significantly better than the performance of control subjects when the period of adapting stimulation is short (i.e., 0.5 sec), tactile spatial discriminative capacity remained unaltered in the same subjects when the duration of adapting stimulation was increased (to 5 sec). Both the failure of prior history of tactile stimulation to alter tactile spatial localization in adults with autism, and the better-than-normal tactile localization performance of adults with autism when the period of adaptation is short are concluded to be attributable to the deficient cerebral cortical GABAergic inhibitory neurotransmission characteristic of this disorder.