The Neuralink Show&Tell from last night had a welcome announcement: they are moving into prostheses for the blind, and is full of interesting tidbits. If they had any real competition they'd want to keep those details confidential but they have no competition: the other companies that make human brain implants use technology from 30+ years ago. It's a fun watch. Presentation starts exactly 1 hour from the beginning of the video, and I suggest x1.5 speed for most of it: https://youtu.be/NhAU8p985-4
Among the things I liked in the Neuralink presentation was the fact that they now go through the dura mater while using fluorescence to see blood vessels, the effort to make realistic fake brains to practice insertion strategies, the effort to study durability of implants, and the new ability to provide stimulation, which is crucial for sensory prosthetics. (Musk speaks only for the first few minutes, easily skipped.)
@MatteoCarandini I agree that it's an impressive technical development. Many of the techniques could benefit scientific research for sure. I'm curious if they're considering that they might need to undo much of the functional reorganization that happens in blind V1 to gain functionality. Would this work? If artificial cochlear implants have told us anything, these types of prosthesis will need to be done quite early. Or is the shear number of electrodes they're proposing to use going to be the solution?
@hkl I'm not sure there is any reorganization!
@MatteoCarandini It seems they are ignoring how hard it is to gain plasticity in the adult brain, especially with congenital blind cases where we know V1 undergo crossmodal recruitment. You'd think the first step is to either undo those crossmodal changes or engage in adult cortical plasticity to make V1 "learn" these new signals.
@MatteoCarandini
I agree that late-onset blind could benefit from such implant if there is large enough coverage of the visual field as they propose to do. The book sounds interesting!
@hkl Ah I see, you're mentioning congenital blindness. In the presentation they only talk about patients who had sight at some point. With those, there is good evidence that the basics like retinotopy are preserved though more complex mechanisms like motion processing may be impaired. I met a patient who paid $100k decades ago to have a rudimentary implant by Dobelle, which lasted only a short time, and he said it was to have those few phosphenes. (He wrote a book too, I can dig it up)