Typing Telepathically

The concept of brain-computer-interfaces has been floating around for a while now and has become more prevalent in the media, but the idea can still seem a bit ‘sci-fi’. Today in When Science Fiction Becomes Reality, I’ll be exploring a recent publication that came out in Nature that may offer hope to paralysed people.

In this work, the researchers have measured neural activity/patterns while the paraplegic test subject imagined hand-writing letters. They analysed the neural patterns associated with each letter and used this as a kind of ‘translator’ so that when the test subject ‘wrote’ (imaginary) sentences, the computer program compared their neural patterns to those in the ‘translator’ and spat out on-screen what the subject was ‘writing’.

This is pretty incredible when you think about it. There is so much activity going on in our brains at any one time that the ability to hone in on the specific signals made by imagining to write a specific letter is mind-boggling. They’ve filtered out all the other stray signals and created a program that had a pretty incredible accuracy rate (higher than 90%) and speed as well (90 characters per minute). This is much faster than previous systems which have relied on the test subject moving a cursor on the screen (40 characters per minute). To me it suggests that fine motor-inputs that are distinct may offer better/faster methods of brain-computer interfacial communication.

Obviously this kind of technology would be a huge boon to paralysed people. Just having the ability to ask for the kind of drink you want! But as always when looking at the science-fiction side of scientific breakthroughs we have to ask ourselves, what are the downsides? The obvious is unauthorized mind-reading. Right now, the technology relies on physical electrodes, but surely in the future there may be a way to do it wirelessly. Then you could monitor anyone without them knowing it. And sure, right now monitoring what they’re thinking about hand-writing might not be so useful (unless it’s a password…) but what if other patterns could be identified, like anti-government sentiment? It’s a long way in the future, but still a pretty amazing thing to consider.

How can you see this technology being used? What’s the next big breakthrough you’re looking forward to?

If you’d like to read the article in full, you can find it here: https://www.nature.com/articles/s41586-021-03506-2

For more science-related articles, go here: https://saffronbryant.com/category/science-fiction-becomes-reality/