Elon Musk’s declaration that Blindsight, a cortical implant to restore vision, “may ultimately exceed normal human vision” has been denounced as “unrealistic at best” by US neuroscientists.
Writing in Nature’s open access journal Scientific Reports, University of Washington (UW) Professors Ione Fine and Geoffrey Boynton said Musk’s projection for the latest Neuralink project rests on the flawed premise that implanting millions of tiny electrodes into the visual cortex will result in high-resolution vision.
Reporting the results of a UW study, which used a computational model to simulate the experience of a wide range of human cortical studies, including an extremely high-resolution implant like Blindsight, the researchers said a single electrode doesn’t represent a pixel, but stimulates, at best, a single neuron. One of their simulations showed a movie of a cat at a resolution of 45,000 pixels is crystal-clear, but a movie simulating 45,000 electrodes implanted in the visual cortex shows the patient would perceive the cat as barely recognisable, they said.
Engineers often think of electrodes as producing pixels, but that is simply not how biology works, said Prof Fine. “We hope that our simulations based on a simple model of the visual system can give insight into how these implants are going to perform. These simulations are very different from the intuition an engineer might have if they are thinking in terms of a pixels on a computer screen.”
Musk is making important strides in the engineering challenge of visual implants, but once the electrodes are implanted and stimulating single cells, you still need to recreate a neural code that creates good vision, she said. “Even to get to typical human vision, you would not only have to align an electrode to each cell in the visual cortex, but you’d also have to stimulate it with the appropriate code.” Since each cell has its own code, it would take years to map out each one, said Prof Fine.
