A recent interesting article in the Wall Street Journal explores the science of where we currently are with neuroprosthetics and what it could mean for brain or sensory augmentation in the future. To an avid sci-fi reader and sometime writer, this kind of notion is awesome to contemplate.
What are the main limitations right now? Mainly power (as in, we’d prefer implants driven by the body’s energy instead of a battery), size, device lifespan, and issues affecting precise placement or the brain-machine interface. Quite a lot of these issues are already being solved a piece at a time. I was doing some research (no really!) recently on cochlear implants and discovered the current devices are really quite sucky in terms of fidelity; I wonder what will happen when this technology leaps forward. Retinal implants are just in their infancy, but apparently the human fovea has a resolution of only about 7 megapixels and the rest of the retina has about 1, putting future developments well within the capabilities of modern CCD cameras. These things are all jumping forward at an astonishing rate.
So as a sci-fi buff, my question isn’t so much when to expect all of this as what will be possible when the obvious goals are met. What can implants do, and to what extent?
Well, I don’t think we’ll ever see implants to make your memory perfect. It’s just never going to happen. But there may well be ways to enhance it, even to provide some solid long-term storage of important info. And hand in hand with this, there may be a way to add knowledge to the brain. Maybe. William Gibson envisioned people sticking hardware sockets into slots in their heads that would offer them instant knowledge; is such a thing possible? My thinking is that if you have an implant with a broad enough interface that the brain can learn to get names, sounds, numbers, and maybe images out of it, as well as accept such input from the brain, you could eventually be trained to use whatever information the socket provided. However, I don’t see that his notion of gaining fluency in a language this way would ever work.
And this is a point that we sci-fi writers seldom appreciate: Short of some kind of technique to speed adaptation, it’ll take time to learn the new implant. Here’s an example closer to reality right now. In 2007, Wired Magazine ran a piece on a belt that provided a constant sense of direction; using 13 vibrating pads, it could indicate the direction of magnetic north at all times. The people who wore it regularly discovered that thanks to the brain’s incredible plasticity, they could come to rely on it. They’d never feel lost, had a much better intuitive understanding of distances, and generally felt like they’d gained an incredibly useful new sense. The corollary is that losing access to the belt required a similar adaptation, causing some disorientation.
Knowing that the brain can adapt to stimuli, though, raises all kinds of interesting possibilities. A better built-in sense of direction would be an enormous benefit to many people, making such a thing a prime candidate for a neural implant.
So the question is if we could have new senses or abilities, what would they be? Assume that stuff like eye augmentation is off the table, except in the matter of total replacement; you wouldn’t want to necessarily alter an existing organ. Or then again you might; scientists have managed to insert genes correcting red-green colorblindness into primates and observe that the monkeys could in fact see the distinction between colors some time afterward, which means the technique is likely already close to human trials. If we can correct colorblindness—and I would argue that we should, also on a genetic level so that stupid gene dies out—then we could, theoretically, inject an eye with codons for infrared- or ultraviolet-sensitive cells that would expand vision further. No one’s tried that and I wouldn’t want to volunteer; plus it’d probably mess with your ability to watch TV. But for some people it could be useful.
But back in the possibility space of neural interfaces, while the knowledge implant possibility intrigues me, I can think of a much better one: a math coprocessor. Human brains aren’t wired to grasp abstract mathematics or discrete numbers very intuitively. A coprocessor could change all of that. After a period of adaptation, imagine that any computation you think of on the fly can give you back a good answer, maybe even exact decimal places if you know how to “ask” for them; if there’s a way to ask, the brain will learn it relatively quickly. Depending on the sophistication of the interface, a math coprocessor could also handle more complex work, like calculus, though it might take some understanding of the basic concepts of calculus to use it. Moreover, a math coprocessor could give you a much better understanding of statistics, to the point of possibly overcoming cognitive biases that make us suck so badly at statistics.
The idea of uncovering brand new senses or abilities is a really cool one to me. I doubt I’ll ever be on the receiving end of the coolest ideas I can come up with, but it sure is exciting to think about. Imagine cybernetics as a mature field, a couple of centuries from now; imagine genetic manipulation growing alongside it. The future’s going to be an awfully interesting place.