Noland Arbo has a computer chip implanted in his skull and an array of electrodes in his brain. But Arbo, an early adopter of Neuralink’s brain-computer interface (BCI), says he wouldn’t have known about the hardware if he didn’t remember having surgery. “If you had told me when I woke up with memory loss that something had been implanted in my brain, I probably wouldn’t have believed you,” says the 30-year-old Arizona resident, who was paralyzed below. In the middle of the neck after a swimming accident in 2016. “I can’t feel it; you wouldn’t know it was there unless someone went and physically pushed it.”

The Neuralink chip may not be physically visible, but Arbo says it has had a huge impact on his life, allowing him to “reconnect with the world.” In January, he underwent robotic surgery to receive Neuralink’s N1 implant, also called “The Link,” in its first approved human trial.

BCIs have been around for decades. However, the company attracted great attention because billionaire technology expert Elon Musk owns Neuralink. This has led to a resurgence of public interest in technology like Arbo that could significantly improve the lives of those living with tetraplegia, as well as those with other disabilities or neurodegenerative diseases.

BCIs record electrical activity in the brain and convert that data into output actions, such as opening and closing a robot arm or clicking a computer mouse. They vary in their design, level of dissemination, and resolution of the information they collect. Some detect the electrical activity of neurons using completely external electroencephalogram (EEG) arrays placed over the subject’s head. Others use electrodes placed on the surface of the brain to monitor neural activity. There are also intracortical devices that use electrodes implanted directly into brain tissue to get as close as possible to target neurons. Neuralink implant belongs to this category.

Capturing neural activity is like recording a conversation between two people in a crowded stadium, says Douglas Weber, a mechanical engineer and neuroscientist at Carnegie Mellon University. To hear more than the roar of the crowd, you need to get closer to the person speaking. “The further away you get from the speaker, the more complex and confusing conversations become,” he explains. Neuralink places electrodes in the brain’s motor cortex, which controls movement, placing “sensors right next to talking neurons.”

Neuralink isn’t the first to do this. A device called the Utah Array (a small rectangular grid of silicone tips) is the standard electrode system for intracortical BCI. It was developed in the 1990s by University of Utah bioengineering professor Richard Norman; In 2004, Matthew Nagle was the first person to use the BCI Utah Array to control a cursor with his thoughts. Neuralink’s design, based on previous microwave research, is also not the first to replace Utah’s rigid array with a network of thin, flexible filaments with electrodes along their length.

But the company Neuralink has combined numerous advances into a single implantable intracortical wireless device. “They took the best of everything I had seen and put it together,” says Jennifer Kollinger, a biomedical engineer and assistant professor at the University of Pittsburgh.

Data is turned into action

Link’s circular electronic core connects to 64 ultrathin filaments containing a total of 1,024 electrodes. This is about 10 times more electrodes than the Utah matrix (although several Utah matrixes are implanted in a person’s brain at the same time). Link transmits compressed neural data from the brain via Bluetooth, and an algorithm tuned to the user’s unique neural patterns translates that data into action.

Arbaugh says he was able to move a digital cursor within a week after implant surgery. She does this in two ways. There is what he describes as an “attempt to act” – or simply a desire to do something that a paralyzed limb can no longer do. By stimulating the muscles in his hand (which he says can still cause a slight wobble) and performing the mental movements of using a mouse with that hand, he can move the cursor across the screen with minimal effort. “It’s very intuitive,” says Arbo.

He also found that he could move around the screen by looking at the cursor and imagining the path he wanted to take. He calls it “phantom movement.” He uses both methods often in combination with each other. The former is a little more physically strenuous, while the latter requires additional mental focus. But both allow simultaneous operation: Arbo can talk or eat with the computer at the same time.

Before the implant, if Arbo wanted to use the computer, he did so with a voice command or by moving the mouth wand on the touch screen (having someone help him position it). But Arbo says it can do more, faster, independently and more comfortably, thanks to its BCI. Using the best BCIs “should be as natural as functional, voluntary movement,” says Lee Hochberg, a neurointensive care physician and neuroscientist at Brown University, Massachusetts General Hospital, Harvard Medical School, and Virginia Providence Health System. He has conducted numerous human BCI studies in his work, including some research done for Neuralink. Hochberg says that sometimes you judge how well a device works by how little the product can define the user experience. “If our participants can’t tell us exactly how they did something,” he says, “we know we’re on the right track.”

Neuralink claims that Arbaugh broke the record for BCI cursor control, achieving eight bits per second, a metric that includes both speed and accuracy. (If you want to compare your ability to Arbo’s, Neuralink has released a cursor control test that’s a square task.) Arbo says it uses its device for hours surfing the web, sending text messages, browsing social media. media, browsing apps, and perhaps most importantly, playing video games. Online chess and Civilization VI’s world-building strategy were his favorites.

According to him, the device has one unavoidable drawback: it needs to be charged regularly, which interrupts gaming sessions. Arbo wears a hat with a built-in wireless charger to charge his implant; This is a big change from the BCI plug-ins still used in many studies. He says Link has been otherwise mostly trouble-free, except for a period in February when it almost stopped working.

Indention

About a month after surgery, Arbo lost significant functionality of his implant. At first he thought it was a software bug, but the Neuralink team soon told him it was a hardware problem. Arbo said Neuralink’s analysis of electrode signals showed that 85 percent of the filaments in its implant were “retracted,” or dislodged. Neuralink first publicly disclosed the issue in a blog post on May 8, several months after discovering the glitch. (Neuralink did not respond to questions scientific american About the withdrawal of flow.)

“It was really hard to accept,” Arbo says. “I just dug it with my teeth. I reached this high place. And in a month [відчувалося, що це] Everything will fall.”

The potential for such frustration and anxiety is one of the “biggest risks” in human BCI research, Weber notes. “Imagine the stress of experiencing a spinal cord injury for the first time. “Now imagine having to go through this again,” he says.

Arbo says Neuralink has managed to restore much of the implant’s functionality by modifying the system’s algorithm to respond to electrodes that are still transmitting data. Since then, he’s shown off his hovering skills in demo videos and says he’s back to breaking speed records. But some fixes required creative solutions. Neuralink engineers created a system where Arbo makes selections on the screen by hovering the cursor over a location for 0.3 seconds, rather than clicking. “We plan to go back to the one-click that I started,” he says. But this has not happened yet.

The company also has not released a formal scientific report on Arbo’s experience. This limits how much can currently be understood about the technology, says George Malliaras, an engineer who directs the Bioelectronics Laboratory at the University of Cambridge. Malliaras noted that it is unclear why or how much the filaments retract, whether their position continues to change, or whether the remaining filaments become fixed. “We have to wait for the data sheets to be published,” he says.

Meanwhile, the US Food and Drug Administration approved Neuralink’s plans to continue the clinical trial and implant a second device in another person. The company will attempt to address the retraction problem by implanting the N1 threads deeper than they are placed in the Arbo case (eight millimeters versus three to five millimeters), as first reported by Wall. Street Diary. “Assuming it doesn’t change the security profile, it’s a strategy worth testing,” says Weber. “They wouldn’t do this if the FDA didn’t think it was right, so they must have something already approved in their protocol. Hopefully that will solve the problem.”

However, Arbo was not deterred by this setback. According to him, everything he goes through has a purpose: to develop technology for others. “The whole point of this study was to find out what works and what doesn’t,” he says. Every bit of information Neuralink collects contributes to a data pool that could one day lead to some of BCI researchers’ most ambitious goals: Restoring mobility to paralyzed limbs or sight to the blind. “I try to keep my expectations pretty reasonable,” he says. But the situation in the BCI field seems to be changing rapidly. He is happy to be one of the first and happy that the next person will get something even better.