Brain-computer interface gives lock-in sufferers a way to communicate


#1

[Permalink]


#2

This is one of the most uplifting things I’ve read all week.


#3

What about Morse code? A single-bit signal can be then extracted from the brain signals, and used for output.


#4

Do they say anything except “Kill me, kill me, kill me”? Because that’s sure as hell what I’d be saying. Is there a way do to a living will that protects you against being kept alive if there’s even an outside chance of lock-in?


#5

Yes. In the USA, at least. After a year and a day (I’m a traditionalist) there will be no more intravenous feeding or any other non-consensual measures to keep me alive. And no harvesting me for body parts, either, there are already far too many people on this rock.


#6

“What’s wrong with Facilitated Communication”, he asked mischievously…


#7

Indeed, I would have thought that they’d been using that for some time already. Brainwave-monitoring neurofeedback techniques have been around for quite some time, after all.


#8

That would require learning morse code, but it still beats the “one beep for yes, two beeps for no” approach.


#9

Yes, this exact scene led me to that idea years ago. If you have one bit wide comm channel, you can say whatever you want. It may take a time but that’s you usually have plenty of in such situation.

If you can see or hear, you can learn simple code over few days. (I learned NATO spelling alphabet from a card during walking between station and office, over several such trips - it’s a must-have when dictating passwords or commands. I later taught it my techs, when I got them. Prevents a lot of miscommunication.) Either seeing a cheat sheet or listening to a tape in a loop should help.

And a computer can be taught Morse code fairly easily - so keyboard input without cumbersome cursor shifting around the screen should be much faster. I believe there’s even some ATmega-based Morse USB-HID keyboard…? The issue is getting the timing autodetection working right. When I was younger, I had a proof-of-concept that transcribed whistled Morse to text - it was crude (no FFT, just amplitude from the soundcard as beep-on detection, and fixed brackets for timing) but functional enough to show it can work.


#10

This topic was automatically closed after 5 days. New replies are no longer allowed.