Validating electrical stimulation through the somatosensory system: a perspective on mismatch negativity and auditory...

Natural touch engages specialized mechanoreceptors, while electrotactile stimulation bypasses them, directly depolarizing a mixed and non-uniform population of afferent axons. Whether this difference in peripheral encoding affects the brain predictive processing remains an open empirical question. The mismatch negativity (MMN), an established electroencephalography (EEG) marker of automatic change detection, provides a powerful tool to address this question without perceptual confounds. Somatosensory studies comparing MMN responses to natural versus electrical stimulation within the same healthy individuals can identify candidate mechanisms for testing in the auditory domain, where such designs are inherently not feasible. In particular, they can help disentangle device effects from deafness-related confounds in cochlear implant (CI) users. Existing somatosensory evidence in healthy subjects suggests that electrotactile stimulation elicits mismatch responses with less bilateral integration than natural touch, though this evidence remains limited to a single magnetoencephalography (MEG) study using an oddball paradigm. CI studies are more extensive and show feature-specific MMN deficits. However, they also rely on oddball or multi-feature paradigms that conflate prediction error with sensory adaptation. This Perspective therefore encourages the adoption of alternative MMN paradigms (roving, omission, or local-global) to determine whether electrical stimulation truly engages the brain's predictive machinery. With emerging multimodal stimulators, the somatosensory system offers a valuable platform to test whether electrical stimulation reproduces natural predictive processing-insights that can improve both tactile neuroprostheses and CI devices.