Brain implants create a realistic touch in paralyzed hands

His eyes closed, electrodes connected his brain to the tablet, and a quadriplegic man described the warm, silky fur of a cat under his fingertips. Another participant reported feeling the smooth, cool surface of the apple and the rough texture of the towel-a hand that hadn’t felt anything in years.
This outstanding touch recovery achievement comes from a collaborative study published today by the University of Pittsburgh and the University of Chicago researchers, published today in the field of natural communication, where scientists put participants in unprecedented control over their artificial feelings.
Previous brain implantation systems might have created some touch, but according to researchers, these sensations often feel like a blurry buzz or tingling sensation and have no changes between objects. The key innovation of this study is to enable BCI users to control the details of electrical stimulation that produces tactile sensations, rather than letting scientists make these decisions themselves.
“Touch is an important part of nonverbal social communication; it’s a personal feeling and has a lot of meaning,” said lead author Ceci Verbaarschot, who conducted research as a postdoctoral fellow at the University of Pittsburgh before joining the Texas-Western Side. “Designing your own feelings allows BCI users to make interactions with objects more realistic and meaningful, bringing us closer to creating a neural prosthesis that is pleasing and intuitive to use.”
Participants rarely had obfuscated objects with distinct tactile properties, but more frequently mixed objects with similar characteristics. For example, they are unlikely to confuse cats as keys, but they may mistake soft towels for cats.
The research team and three men who lost their feelings due to spinal cord injuries. Each participant had a tiny array of electrodes implanted in the somatosensory cortex—a brain region that processed information. These electrodes pass small, precisely controlled electrical pulses to specific brain cells.
Unlike previous researchers’ scheduled stimulation patterns, the study introduced a new approach: participants themselves adjusted stimulation parameters when they interacted with virtual objects on a tablet screen.
The results reveal a very vivid and appropriate tactile experience. When creating a feeling for the apple, one participant reported that it was “light, smooth, curved and slightly moist and moist”. For one cat, another describes “a very light touch, like stroking a cat. It’s smooth and smooth at your fingertips. The cat’s resistance has this oily feeling. It even has a warm feeling.”
To test whether these sensations really represent different objects, participants later received stimulation without seeing any images and had to identify the sensations represented. Two of the three participants performed significantly above the opportunity level, correctly identifying about 35% of the time.
What is more convincing is the wrong pattern. Participants rarely had obfuscated objects with distinct tactile properties, but more frequently mixed objects with similar characteristics. For example, they are unlikely to confuse cats as keys, but they may mistake soft towels for cats.
The study found that when selecting stimulation mode, participants always chose different parameter combinations of objects with different compliance (softness versus hardness) and temperature (warm versus coolness). This suggests that the brain can interpret complex stimulation patterns as consistent object properties.
Although the study represents substantial advancement, the challenge remains until a completely intuitive touch recovery becomes a reality. Two participants showed a consistent ability to identify subjects from their senses, but the third participant spent less time exploring different stimulation patterns.
“We designed this study to shoot for the moon and go into orbit,” said Robert Gaunt, an associate professor at the University of Pittsburgh. “The participants’ task was very difficult to distinguish objects by simply tactile sensations and were tactile. Even if they made mistakes, these mistakes were predictable: Because both were soft, it was difficult to distinguish between cats and towels, but they were unlikely to confuse cats as keys.”
These findings open up new possibilities for brain computer interfaces that may eventually restore natural touched prosthetics. Scientists hope this user-guided sensory creation approach can lead to more personalized and intuitive neural prosthesis systems.
The research team plans to explore whether similar techniques can be used to recover other aspects beyond touch, including proprioception – sense of body position – which is crucial for natural movement.
Related
Discover more from Neuroweed
Subscribe to send the latest posts to your email.