Abstract
Direct cortical stimulation (DCS) of primary somatosensory cortex (S1) could help restore sensation and provide task-relevant feedback in a neuroprosthesis. However, the psychophysics of S1 DCS is poorly studied, including any comparison to cutaneous haptic stimulation. We compare the response times to DCS of human hand somatosensory cortex through electrocorticographic grids with response times to haptic stimuli delivered to the hand in four subjects. We found that subjects respond significantly slower to S1 DCS than to natural, haptic stimuli for a range of DCS train durations. Median response times for haptic stimulation varied from 198 ms to 313 ms, while median responses to reliably perceived DCS ranged from 254 ms for one subject, all the way to 528 ms for another. We discern no significant impact of learning or habituation through the analysis of blocked trials, and find no significant impact of cortical stimulation train duration on response times. Our results provide a realistic set of expectations for latencies with somatosensory DCS feedback for future neuroprosthetic applications and motivate the study of neural mechanisms underlying human perception of somatosensation via DCS.
Introduction
For a prosthetic limb to feel like a natural part of the body, it needs to send sensory feedback to the brain. Brain-computer interfaces (BCIs) often achieve this through Direct Cortical Stimulation (DCS)—essentially "writing" sensations directly into the primary somatosensory cortex (S1). Intuitively, one might assume that stimulating the brain directly would be faster than touching the skin, as it bypasses the "long wires" of the peripheral nerves and the spinal cord. However, the timing of this artificial feedback—the latency—is a critical constraint for the control loops of neuroprosthetics. This study investigates whether DCS acts as a shortcut or if the artificial nature of the signal creates a processing bottleneck.
Methods
The researchers worked with four human subjects (N=4) already implanted with electrocorticographic (ECoG) grids for clinical epilepsy monitoring. This provided a rare opportunity to stimulate the surface of the hand-area of the somatosensory cortex with high precision.
The experimental paradigm was a simple reaction-time task:
- Haptic Condition: An experimenter used a digital touch probe to deliver a physical tap to the subject's hand.
- DCS Condition: Researchers delivered bipolar, constant-current electrical stimulation trains (200Hz) to the S1 area corresponding to the same hand location.
- The Task: Subjects were instructed to press a button with their other hand as soon as they felt a sensation.
The team tested various DCS "train durations" (from 100ms to 800ms) and used "off-target" stimulation and "null trials" as controls to ensure subjects weren't responding to general electrical noise.
Results
The "Shortcut" is a Detour
In all four subjects, DCS was significantly slower than haptic touch. Median haptic reaction times ranged from 198 ms to 313 ms. In contrast, median DCS reaction times ranged from 254 ms to 528 ms. This suggests that the delay isn't caused by physical distance, but by the time the brain takes to "decode" the artificial signal.
Invariance to Signal Length and Learning
Increasing the duration of the electrical pulse train (e.g., from 200ms to 800ms) did not significantly speed up reaction times. Once the threshold for perception was met, additional signal duration did not aid processing speed. Furthermore, subjects did not get faster over multiple blocks of trials, implying the delay is a fundamental physiological constraint rather than a lack of familiarity with the sensation.
Discussion
The authors hypothesize that natural touch activates specific pathways with precise timing, whereas DCS excites a messy mix of excitatory and inhibitory neurons simultaneously. This creates a "spatiotemporal smear" that higher-order processing centers struggle to interpret quickly.
While the sample size is small and limited to a clinical population, these results provide a realistic set of expectations for latencies in somatosensory DCS feedback. Future BCI designs may need to be "tuned" to the specific latency of each user and explore stimulation patterns that more closely mimic natural neural firing to reduce these processing delays.
Summary updated