Ask people living with chronic paralysis what they would open their skull for, and the answer is not the thing the neurotechnology industry keeps demonstrating on stage.
In a survey of 71 adults with chronic stroke or spinal cord injury, 98% said they would be very or moderately likely to accept surgical implantation of a brain-computer interface if it could restore control of urination and defecation. Ninety percent said the same for the ability to feel a full bladder. Walking at various speeds with the ability to turn drew 92%. Basic hand grasp and release, the function most often showcased in laboratory demonstrations, drew 70%, the lowest figure in the survey.
The findings, from a team at the University of California, Irvine, were published in Frontiers in Neuroscience in March 2026.
The Function Nobody Puts in the Press Release
Bowel and bladder dysfunction is one of the most common and least discussed consequences of spinal cord injury and severe stroke. It drives urinary tract infections, hospital admissions, and loss of independence, and treatment options remain limited.
The survey team, led by Derrick Lin and An H. Do, gave every participant a written and visual description of a hypothetical device based on electrocorticography, a grid of electrodes laid on the surface of the brain rather than pushed into cortical tissue. Participants then rated how likely they would be to undergo surgery for each of a series of specific functional gains.
Autonomic function topped the list. That tracks with two decades of rehabilitation priorities research in spinal cord injury, in which people with high injuries have repeatedly ranked bladder and bowel control at or above walking. What is new is that the preference held up when the question involved brain surgery.
The authors note that commercial programs, including Neuralink and Synchron, have largely focused on restoring communication for people with amyotrophic lateral sclerosis or severe speech impairment. Their data point to a different, possibly larger group of candidate users whose priorities lie elsewhere.
Willingness That Did Not Track with Benefit
The more unsettling finding is what did not predict willingness.
Offering fine finger control on top of basic grasp, or turning on top of walking, produced no statistically significant increase in stated willingness. Severity of disability did not predict willingness either. Neither did how important a participant said that function was to them. Prior knowledge of the technology made no difference except in one subgroup: stroke survivors considering upper limb restoration.
Interest was high and roughly flat. People said yes to the idea of an implant more readily than to any particular benefit it might deliver.
The researchers treat that pattern as a warning rather than a market signal. They write that patients with severe disabilities may be especially susceptible to optimistic portrayals of emerging neurotechnologies, and that the disconnect between willingness and expected benefit raises questions about whether respondents were making fully informed assessments of risk and benefit. They call for robust consent frameworks before invasive devices reach wider clinical use, citing the neuroethics priorities published in Nature.
What People Were Actually Worried About
Participants were not naive about risk. Asked to name concerns about a prospective implantation, 77% cited surgical risks and complications, including infection, bleeding, blood clots and reactions to anesthesia. Sixty-two percent increased the device's long-term durability. Others flagged cost, the prospect of additional surgeries if the hardware failed, training time, and the possibility that the system would simply not meet expectations. Free-text answers included worries about "unknown side effects and/or long-term effects of the device" and "brain damage from surgery."
Asked what else they would want restored beyond the listed functions, six participants named sexual function, the most frequently endorsed write-in category. Vision, facial movement, trunk movement, cognition, and speech followed.
One demographic association reached statistical significance. Higher household income was linked to a greater willingness to undergo surgery specifically for sensory restoration, with relative risks ranging from 1.36 to 1.46 compared with lower-income respondents. Age, gender and education showed no significant associations.
Why a Hypothetical Survey Still Matters
This is a small, single-region study with real limits, and the authors say so. Participants came through UC Irvine clinics, rehabilitation centers, and support networks, which likely selected for people actively engaged in their own recovery. Subgroup sizes were too small to provide sufficient statistical power. Sixty-four percent did not answer a question about mobility aids, for reasons that remain unclear.
Most importantly, nobody was actually offered surgery. The authors point to the well-documented gap between stated health intentions and real behavior, and say they would expect far lower uptake in practice than these numbers imply.
Even so, the survey answers a question device developers have largely been answering for themselves. Long-term outcome data on implanted systems remains thin. A five-year follow-up of one implanted system tracked a single spinal cord injury patient using an electrode array at home, and a landmark brain-spine interface report described one participant walking again. The field is still measured in individuals.
Work on pairing neural interfaces with sacral nerve stimulation for bladder control remains in its early stages and largely unexplored. What this survey adds is a preference map drawn by the people who would carry the hardware, and it points somewhere the demonstrations have not.
Anyone considering enrollment in a neural implant trial should discuss risks, realistic expectations and long-term device support with a qualified clinician.
Key Questions Answered
What did the study actually measure?
Stated willingness to undergo surgery for a hypothetical brain implant, along with rehabilitation priorities and concerns. No participant received a device.
Why is the bowel and bladder result surprising?
Most public attention to brain-computer interfaces focuses on movement, cursors and communication. Autonomic control showed the highest willingness among the functions tested.
Does this mean an implant can restore bladder control?
No. No approved device does this. Research combining neural interfaces with sacral nerve stimulation is in its early stages and largely unproven.
How reliable are these numbers?
The sample was 71 people recruited from one university system and its support networks, which limits generalizability. Several subgroup analyses were underpowered.
What worried participants most?
Surgical risk, cited by 77%, followed by concerns about how long an implanted device would keep working, cited by 62%.
Who funded the work?
Partial support came from the National Science Foundation and internal funds at the University of California, Irvine. The authors reported no commercial conflicts of interest.
