The BCI Race
As was the case a year ago, when Elon Musk first unveiled Neuralink’s technology to the public [NBR Jul19 p1], his webcast update earlier this month caused a bit of a stir in the brain-computer interface community. New at this year’s talk was a description of a disc-shaped implant called the “Link,” which would sit flush with the skull, replacing the bone that the firm’s neurosurgical robot would extrude. The wirelessly charged Link features 1000 channels at this stage, and interacts with the thin wires implanted by the Neuralink robot.
During the demo Musk trotted out three live pigs, two of which had been implanted with the Link, and one of which was explanted. While Musk’s purpose in showing the three pigs was to demonstrate the apparent safety of the device and its ability to capture brain signals in a behaving animal, an article in Technology Review referred to it as “neuroscience theater.”
Many in the BCI community were quick to applaud the elegance of the wireless interface. But neural interface pioneer Doug Weber, now at Carnegie Mellon, was more impressed with another BCI development. “While @neuralink‘s Gertrude was ‘hogging’ everyone’s attention yesterday, our friends at Synchron had big neurotech news to reveal,” he posted on Twitter. As we detail in our vendor profile on page 7 of this issue, Synchron has made steady progress with a brain implant that will likely be less invasive than Neuralink’s. Longtime readers of this publication will appreciate this approach, since the Stentrode device fulfills some of the prognostications we made in this space 17 years ago [NBR Aug03 p2].
While the publicity that Elon Musk brings to the neurotech industry is probably a net positive, there is also the danger that it sucks away attention from other efforts to construct a viable BCI or neuroprosthetic implant. Aside from ventures like Synchron, other research teams are working to improve upon the original BrainGate implant, which Musk refers to derisively as a “box on your head.”
As an example, Jaimie Henderson’s team at Stanford recently published an article in Nature Biomedical Engineering that demonstrates how to generate wireless signals from the brain using 90 percent less power, and thereby spare much of the tissue damage that might otherwise result. The Stanford team validated the hypothesis that a wireless interface could accurately control an individual’s motion by recording a subset of action-specific brain signals, rather than acting like the wired device and collecting brain signals in bulk.
While the perfect BCI has yet to be created, it’s in the field’s interest to view the race to build the best brain interface as a marathon and not a sprint.
Editor and Publisher