Shared representations between seeing and feeling touch in people with high tactile empathy

Abstract

AbstractSimulation theories propose that tactile empathy arises when seeing someone else being touched triggers corresponding representations of being touched. Prior electroencephalography (EEG) findings show that feeling touch while simultaneously seeing touch (versus seeing a no-touch stimulus) modulates early somatosensory responses. Functional Magnetic Resonance Imaging (fMRI) studies have shown that seeing touch increases somatosensory cortical activation. These findings have been taken to suggest that when we see someone being touched, we simulate that touch in our sensory systems. The overlap between somatosensory activation when seeing and feeling touch differs between individuals, potentially underpinning variation in tactile empathy. Increases in amplitude (EEG) or cerebral blood flow response (fMRI), however, are limited in that they cannot test for the information contained in the neural signal: seeing touch may not activate the same information as feeling touch. Here, we use time-resolved multivariate pattern analysis on whole-brain EEG data from people with high and low tactile empathy to test whether seen touch evokes similar representations to the first-hand experience of touch. Participants felt touch to the fingers (tactiletrials) or watched carefully matched videos of touch to another person’s fingers (visualtrials). In both groups, EEG was sufficiently sensitive to allow decoding of touch location (little finger vs. thumb) on tactile trials. Only in individuals who reported feeling touch when watching videos of touch (high tactile empathy) could a classifier trained ontactiletrials distinguish touch location onvisualtrials. This demonstrates that, for these people, there is similarity between the neural pattern when seeing touch and feeling that same touch. The timecourse of this overlap shows that seeing touch evokes similar representations tolater stagesof tactile processing. Therefore, while simulation may underlie tactile empathy, our findings suggest this involves abstract high-level processes rather than early low-level processes.