Disrupted Brain Mechanisms for Processing Affectionate Touch In Children with ASD

Background: Recent discoveries of the neural mechanisms underlying social dysfunction in Autism Spectrum Disorders (ASD) have highlighted visual social perception (e.g., faces, biological motion, and social scenes). Despite the critical role of proximity seeking and touch hedonics in social development, and known associated deficits in ASD, the neural basis of touch processing in ASD has not been investigated. A prime candidate for examination is a system of slow-conducting unmyelinated C-tactile (CT) afferent fibers that exist only in hairy skin and respond preferentially to light and pleasant touch. This type of touch is particularly reminiscent of the evolutionary conserved system of “affectionate touch” amongst bonded individuals (such as mother and infant or romantic partners). An elegant series of functional magnetic resonance imaging (fMRI) studies has illustrated that somatosensory and insular cortical regions are activated by affectionate touch. The insular activation is particularly interesting given its involvement in social-emotional processing. The available evidence strongly suggests that the CT system is involved in processing the caress-like, skin-to-skin contact, a key part of the bio-behavioral processes of the mammalian affiliative system.

Objectives: We sought to examine brain responses to affectionate touch in children with ASD using fMRI. We hypothesized that children with ASD would exhibit decreased responses to affective touch in the insula but not the somatosensory cortex compared to age- and IQ-matched typically developing (TD) children.

Methods: Samples of children with and without ASD (matched on age and IQ), ranging from 4-17 years of age, participated in the study. Participants received continuous brushing to the palm or forearm in a block design procedure. There were 2 runs of each condition (forearm, palm) which included 8 repetitions of 6-second blocks of touch followed by 12 seconds of rest (no touch). Tactile stimuli were slow strokes (8cm/s) with a 7cm wide watercolor brush administered by a trained experimenter.

Results: Both groups rated the brush strokes as pleasant. Supporting our initial hypothesis regarding the CT afferent system, both groups exhibited activation in the somatosensory cortex during arm touch (relative to baseline). However, several brain regions revealed distinct responses to affective touch in the two experimental groups. In TD children, the bilateral posterior, and middle, insular cortex were robustly engaged whereas this activation was left lateralized and significantly weaker in the children with ASD. Further, TD children revealed unique activation in several additional regions including the medial prefrontal cortex, right posterior superior temporal sulcus, and ventral striatum. Children with ASD revealed amygdala activation that was not present in the TD children.

Conclusions: Children with ASD do not process affectionate touch with the same brain mechanisms utilized by their TD peers. While somatosensory activation in both groups indicates that children with ASD feel the touch, the social brain is not engaged during slow soft touch which is associated with activation of the CT afferent system. These results suggest the disruption of social perception in ASD extends beyond the visual domain. This work has implications for derailed social engagement in ASD and provides potential target for early intervention.