An unexpected difference in reaction times to artificial and natural touch

Wayne Gillam

A new research finding contributes to the development of neuroprosthetics (assistive devices that supplant or supplement the input and output of the nervous system) by revealing the need to close a performance gap between artificial and natural touch.

David Caldwell, a fourth-year doctoral bioengineering student and member of the University of Washington Medical Scientist Training program, is exploring new ways to make a difference for people who are impacted with neurological damage from conditions such as stroke or paralysis. Neural pathways are often badly damaged or completely severed in these disorders, so sensory feedback from the peripheral nervous system (such as a person’s sense of touch) can no longer make its way back to the brain. In the search for engineering-based solutions, Caldwell and other researchers at the Center for Neurotechnology (CNT), under the supervision of CNT Co-Director, Rajesh Rao, and CNT members, Dr. Jeffrey Ojemann and Dr. Andrew Ko, are investigating ways tactile feedback could be artificially induced through direct electrical stimulation on the brain’s surface.

The research team works with patients under Dr. Ojemann’s and Dr. Ko’s care at Harborview Medical Center who are undergoing surgery with the goal of eliminating or reducing the frequency of their epileptic seizures. Prior to the surgery, patients have an electrocorticographic (ECoG) grid implanted on the surface of their brain to help precisely locate the origin of seizures and prevent inadvertent brain damage while operating. By participating in CNT research, these patients provide the team with a unique opportunity and access point to study sensory feedback and motor control in the human brain.

Caldwell is the lead author of a new study that examines how patients would respond to artificially-induced hand sensations through direct electrical stimulation of the brain’s cortex. The study builds on the team’s previous research work using brain surface stimulation to provide artificial touch feedback capable of directing hand movement. In this study, Caldwell and his team wanted to discover the difference in reaction times between an artificially-induced sense of touch and natural touch.

In the study’s experimental setup, electrical brain stimulation gives patients the artificial sensation of a slight touch on their hand. When they feel this sensation, the patient pushes a button. This gives researchers a measurement of how long it takes people to notice and react to artificially-induced “touch.” This data was then compared to how long patients took to notice and react to a natural touch on the same spot on their hand via an instrument that recorded the time when touch occurred.

The results of the study were the opposite of what the team initially expected. In natural touch, nerve impulses have a long way to go, traveling from the hand to the arm, all the way up to the brain. Caldwell and the other researchers hypothesized that because direct brain stimulation was much closer to where the brain processes sensory feedback, reaction times for the artificial feedback would be quicker. Surprisingly, the team discovered that natural touch reaction times were significantly faster than artificially-induced touch, sometimes close to half of the best time direct brain stimulation could muster.

According to Caldwell, there are many possible reasons for this counterintuitive result, and he talked about three plausible explanations.

“Normally, when you stimulate something on your hand or anywhere else, you have this amplification process that happens. The [nerve] signal goes through these different regions of the spinal cord and the brain, increasing the ‘volume’ of the signal. You’re missing all of that when you stimulate the brain directly,” Caldwell said. “Another possible reason is that with electrical stimulation, you activate both inhibitory and excitatory cells non-discriminately, making everything ‘talk’ at once and creating noise along with the signal. One more possibility is that we are stimulating the surface of the brain to induce a sense of touch, but in natural processes, much of the primary sense of touch occurs deeper within folds in the brain.”

Impact on neuroprosthetic development

This study has implications for the development of neuroprosthetics, devices that could assist and empower those impacted by stroke, spinal cord injury and other neurological disorders. A key takeaway from Caldwell’s study is the need for more research that accounts for the difference in reaction times between artificial and natural touch. According to Caldwell, these findings will impact neuroprosthetic development in at least a couple of different ways.

“One is that the study illustrates more work can be done in figuring out what the best ways to stimulate sensory parts of the brain are to make the response times faster,” Caldwell said. “The other is that this points to the fact that you can’t assume that if you have someone’s brain interact with a computer, using brain stimulation, each part would function like it normally would. By using an artificial system, you may have timings that are not natural, and that’s important to consider for any kind of neural prosthetic device.”

Next steps for the research team include experimenting with different brain stimulation patterns and waveforms to try and reduce the performance gap between artificially induced and natural touch, as well as studying the brain closely to see if any of its other functions are impaired by direct cortical stimulation. This build-up of knowledge and data over time will help scientists better “map” the brain’s sensory feedback and motor control systems. The map will then aid engineers in designing neuroprosthetics capable of sending useful sensory feedback. For example, a sensory-enabled artificial limb could send valuable data to a user such as “How should I adjust my grip to hold this glass of water?” or “How delicately do I need to hold this egg without crushing it?” in a timely and natural way.

Caldwell said he was inspired by the generosity and selflessness of patients participating in CNT research. In some ways, it was almost as unexpected as the findings of his recent study.

“What’s always surprising to me is how willing people are to work with us when they think they have the opportunity to make a difference in someone else’s life down the road. They still know that by working with us it doesn’t make them any better any faster,” Caldwell said. “It doesn’t have any impact on their clinical care, but they are willing to let us come in for three or four hours a day over a couple days on a weekend while they’re sitting with electrodes in their brain getting ready for a second brain surgery. I think the patients are amazing to work with.”

For more information about this CNT research study contact David Caldwell.