Mechanonociceptive tactile feedback enhances prosthetic grasp control during repeated human-in-the-loop interaction

Upper-limb prosthetic systems continue to face a fundamental challenge in providing sensory feedback that remains functionally effective during repeated interaction. This study investigates innocuous painful feedback (IPF), a controlled mechanonociceptive tactile modality intended to provide salient error-related information during human-in-the-loop prosthetic interaction. A custom experimental platform was developed to emulate prosthetic grasping conditions in healthy participants, enabling systematic evaluation of IPF across feedback modality, stimulation frequency, temporal modulation strategy, and spatial configuration. IPF was associated with lower grasp-position errors than visual-only and visual-plus-force feedback, supporting its potential as a corrective sensory feedback modality for boundary-constrained grasp control. Within the tested frequency range, high-frequency stimulation produced the lowest error levels, demonstrating that the functional effectiveness of IPF depends on stimulation frequency. Progressive frequency modulation generally maintained lower error levels than fixed-frequency stimulation across repeated interaction, while sequential dual-site stimulation yielded lower overall participant-level errors than repeated single-site stimulation, although the magnitude of this difference varied across the trial sequence. Collectively, these findings indicate that the behavioral effectiveness of mechanonociceptive feedback depends not only on feedback modality, but also on its temporal and spatial organization. As a proof-of-concept study conducted in neurologically intact participants using a virtual prosthetic-like grasping paradigm, these findings provide a basis for subsequent validation with individuals with upper-limb amputation and physical prosthetic devices.