COMBINING NON-INVASIVE SPINAL AND CORTICAL RECORDINGS FOR MULTIMODAL SENSORIMOTOR DECODING

The spinal cord plays a central role in integrating sensory inputs and generating motor outputs, yet non-invasive methods to monitor its activity remain limited. Electrospinography (ESG) has recently emerged as a promising approach for recording spinal electrical signals from the body surface; however, the lack of standardized recording systems and systematic signal characterization has constrained its wider adoption. In this work, we introduce a standardized lumbar ESG recording system and assess its capacity to characterize and decode motor- and sensory-related spinal activity. We hypothesized that ESG captures spinal motor processing that is complementary to cortical activity measured with electroencephalography (EEG). To test this hypothesis, we recorded ESG and EEG simultaneously in healthy participants during controlled peripheral motor and sensory stimulation. We analyzed spatial and temporal signal features and performed decoding analyses to compare ESG and EEG discrimination performance across tasks. Both modalities exhibited distinct, stimulus-dependent spatiotemporal patterns. Decoding results showed that ESG achieved higher classification accuracy for stimulation above the motor threshold, consistent with engagement of spinal motor circuits, whereas EEG performed better during sensory stimulation, reflecting dominant cortical processing. These findings demonstrate that ESG and EEG provide complementary information and highlight the value of their combined use. Overall, this study establishes a reproducible framework for lumbar ESG and demonstrates that spinal activity can be selectively decoded during motor engagement. Although validated here in healthy participants, this approach holds strong translational potential for monitoring residual spinal function in individuals with incomplete spinal cord injury and for informing closed-loop rehabilitation strategies.