Typing By Brain Arrives: No Surgery Necessary (wired.com) 93
mirandakatz writes: 2017 has been a coming-out year of sorts for the brain-machine interface. But the main barrier to adoption is the potentially invasive nature of a BMI: Not many people are going to want to get surgery to have a chip implanted in their brains. A New York company may have found a solution to that. It's created a BMI that works just by an armband -- and it works now, not in some far-off future.
Steven Levy describes a recent demo by the CEO of CTRL-Labs: After [typing] a few lines of text, he pushes the keyboard away... He resumes typing. Only this time he is typing on...nothing. Just the flat tabletop. Yet the result is the same: The words he taps out appear on the monitor... The text on the screen is being generated not by his fingertips, but rather by the signals his brain is sending to his fingers. The armband is intercepting those signals, interpreting them correctly, and relaying the output to the computer, just as a keyboard would have...
CTRL-Labs, which comes with both tech bona fides and an all-star neuroscience advisory board, bypasses the incredibly complicated tangle of connections inside the cranium and dispenses with the necessity of breaking the skin or the skull to insert a chip -- the Big Ask of BMI. Instead, the company is concentrating on the rich set of signals controlling movement that travel through the spinal column, which is the nervous system's low-hanging fruit. Reardon and his colleagues at CTRL-Labs are using these signals as a powerful API between all of our machines and the brain itself.
Steven Levy describes a recent demo by the CEO of CTRL-Labs: After [typing] a few lines of text, he pushes the keyboard away... He resumes typing. Only this time he is typing on...nothing. Just the flat tabletop. Yet the result is the same: The words he taps out appear on the monitor... The text on the screen is being generated not by his fingertips, but rather by the signals his brain is sending to his fingers. The armband is intercepting those signals, interpreting them correctly, and relaying the output to the computer, just as a keyboard would have...
CTRL-Labs, which comes with both tech bona fides and an all-star neuroscience advisory board, bypasses the incredibly complicated tangle of connections inside the cranium and dispenses with the necessity of breaking the skin or the skull to insert a chip -- the Big Ask of BMI. Instead, the company is concentrating on the rich set of signals controlling movement that travel through the spinal column, which is the nervous system's low-hanging fruit. Reardon and his colleagues at CTRL-Labs are using these signals as a powerful API between all of our machines and the brain itself.
Misleading title (Score:5, Insightful)
That's not typing by brain. That's typing by muscles. It won't work for paralyzed people like Stephen Hawking.
Re: (Score:2)
Re: (Score:1)
The F-35 isn't capable of mach 2+. It can't even super-cruise.
In fact there's a heck of a lot the F-35 still can't do, and the effects of vibrations on the (custom for each pilot) helmet's internal display is known to be a major ongoing development problem.
Re: Misleading title (Score:1)
Not necessarily. Without the biofeedback from actually moving your fingers you might have difficulty generating the right signals consistently.
If this thing REALLY worked as advertised, then they could already be hooking this up to a prosthetic robitic hand and using it to move the fingers. It has some promise, but it's still mostly a novelty research item.
Even Worse (Score:2, Funny)
Only this time he is typing on...nothing. Just the flat tabletop.
Since the tabletop does not give, he gets carpal tunnel syndrome even faster!
Re:Misleading title (Score:4, Insightful)
That's not typing by brain. That's typing by muscles. It won't work for paralyzed people like Stephen Hawking.
Why not? I would think that even if my hand was paralyzed or amputated I could imagine typing and the brain would send the signals, it just wouldn't arrive at the muscles. I suppose if you've been that way since birth it might be different, but I'm not so certain. I mean even if it's not working you'd think the brain is still wired to try. Don't some advanced prosthetics work like that, they're not just mechanical but they actually read the nerve impulses to recognize what you're trying to do. If you could pair this up with a VR/AR headset maybe you could learn to use virtual hands.
The downside is of course that there's no physical truth to compare to, but if I could see the computer's interpretation of it we could work on that, like this is me counting on my fingers one-two-three-four-five, this is me bending my index finger forward, this is me curling my index finger, this is me bending it sideways, this me doing V for victory, a fist, giving you the finger, the horns, the Spock greeting, the okay sign etc. and then I could virtually type on a virtual keyboard. It could go a helluva lot faster than eye tracking.
Re: (Score:3)
Why not? I would think that even if my hand was paralyzed or amputated I could imagine typing and the brain would send the signals, it just wouldn't arrive at the muscles
I can 'imagine typing' and that's not the same thing as 'typing'. Can you really 'type' without a hand to type with? Maybe yes? Maybe no. I don't know. I know there are cases where you are injured and you *are* trying to move your hand it it doesn't move, and those cases this should work, but after you've been injured for a while (years) can you still even send the signals to move your hand or whatever, or do you forget how?
I can also say that I need a keyboard to type on. I need the feedback. The home key
Re: (Score:3)
Paralysis is usually due to the signals not getting to the muscles for some reason, like a broken wire. Obviously if that is the case then putting the sensor on the muscle isn't going to work, you need to attach it at the brain end. Unfortunately the brain is rather complex and has a large amount of I/O, so it's much harder to do it that way.
Re: (Score:1)
There's already prosthetics that work like this. Attempting to move the limb, even though it no longer exists does still generate the appropriate nerve signals. However, the nervous system uses a lot of trained feedback loops to provide accurate, consistent movements. Feedback from the muscles doesn't come back, so precision is severely impacted.
This demonstration won't show that, because he still has his hands. Try it with someone with a trans-radial (mid-forearm) amputation or even a wrist amputation (whe
Re: (Score:2)
If they are doing an EMG, they are reading muscle electrical activity, not nerve electrical activity. So besides requiring a working nervous system, it also requires muscles that work to some degree.
Re: (Score:2)
Why not? I would think that even if my hand was paralyzed or amputated I could
If your hand was amputated, sure. If it was paralyzed, unlikely. They are measuring the signals down close to the appendages. If they got there in the first place you're unlikely to be paralised.
If this worked on the source (brain), or on the spine then it may be different. But the people who need this most won't benefit from this design.
A half-way solution that nobody wanted (Score:2)
Re: (Score:2)
Well, the existing ways that involve surgery also require lots of effort, which people imagine they won't. They're also slower than typing on a keyboard, and have other defects.
I doubt that we'll come up with a way similar to what people imagine when they think of typing via the brain in the coming decade, and if we do it will involve reading the tensions of muscles around the larynx. With direct brain feed it would probably be simpler to do visualization transmission...but do you have any idea how poorly
Re: (Score:2)
Of corse it avoids it.
You are not actually using your fingers or muscles.
Re: (Score:1)
DVORAK (Score:1)
Hope he doesn't use DVORAK or AZERTY. Or pick-and-peck. This type of thing will be quickly rendered obsolete/redundant with overlaid interfaces rendered on an AR device, with a LeapMotion type solution. Could be useful for those with disfigured hands trying to type, though.
Re: (Score:2)
Re: (Score:2)
This is merely interpreting the muscles/nerve signals, not the brain signals...
Excuse me, but your brain consists entirely of nerves.
Re: (Score:2)
While I agree it's not as good as a "real" BMI, it's better than the laser keyboard in a few ways:
1) It would work with a prosthetic limb if your hand had to be amputated. Allow these signals to control it, and even if your brain doesn't send it the right signals initially, the human brain is actually really good at adapting to changes like that.
2) TFA: "You could be blasting a hundred words a minute on your smart phone with your hands in your pockets." I'm not sure how accurate that is, but it still sounds
Re: fMRI is the only non-invasive way for BMI. (Score:2)
No, it is EEG.
I want (Score:1)
I've been wanting this for nearly 20 years. I knew it was only a matter of time before it happened.
Does this mean that typing speeds won't be relevant anymore in jobs?
"How fast can you type?" "Well, I can type by brain at over 1000 words per minute..."
Re: (Score:2)
Read the article. You still have to type with your hands, so no.
Re: (Score:2)
Did they test it? I would assume there are a number of possible issues like muscular atrophy in such cases.
Re: (Score:2)
I can write with my brain at over 2000 words per minute with only about 15% typaus and 5% gramatiical errors.
Well done (Score:2, Insightful)
You've just won the prize for the most disingenuous headline of the day.
You still have to type with your hands, so its basically fucking pointless because you could just use a keyboard.
Re: (Score:2)
Well, of course. It's a recursive acronym.
BSD means BSD is Still Dead.
Prior Art (Score:2)
https://youtu.be/oSyivQLQfQw?t... [youtu.be]
But the main barrier ... (Score:2)
Wrong. The main barrier is that nobody has found a way to connect electronic circuits to neural tissue in a sustainable way. The body rejects that shit sooner or later. Without that, there can never be a useful computer brain interface. This is your chance for a Nobel prize- make it happen!
Re: (Score:3)
The main barrier is that nobody has found a way to connect electronic circuits to neural tissue in a sustainable way. The body rejects that shit sooner or later.
Couldn't be bothered to do your research [wikipedia.org] before posting?
Re: (Score:2)
Longevity is not the main barrier, getting the signals right is.
Re: (Score:2)
You should read the link you posted before displaying your ignorance. Among other things it says:
"Much research is also being done on the surface chemistry of neural implants in effort to design products which minimize all negative effects that an active implant can have on the brain, and that the body can have on the function of the implant."
Why do you suppose these people are wasting time and money in this research when all the answers have been found (according to you)?
would be nice if you could hang or rest your arms (Score:1)
The standard keyboard position plays havoc with the wrists, the forearms, the teres, the neck muscles, the back.
Being able to sit comfortably with arms folded and "type" would help with many of those issues.
Re: (Score:1)
RSI doesn't affect everyone.
But I work on folks who your approach fails for.
Keep doing your regular breaks. Most the people I work on do not take breaks.
What the hell am I doing then? (Score:1)
Even the amoeba's I call my relatives have to use their brains for typing. Now, using ones brain to think about what one is typing is a totally different question.
--
"I'm just a cat with upgraded parts" - Jim Carrey
What could go wrong.... (Score:5, Funny)
Okay, while this may be pretty handy in some cases, but I am hoping people are taking into account that IF it is brain activity that simulates a keyboard based on what your thinking..... Get where this is going. No more thinking about last night with your wife, no more cubicle walks ups that will interrupt your thought process, and holy cow, do we need to reference the micro-managing boss scenario? I see benefits and the opposite with technology like this. I just hope someone does think about this.
It would be kind of bad if a co-worker drops by and "Yea, I would tap that" shows up on your screen.
Tip of the iceberg (Score:1)
I toldja this was coming. Eventually everyone in the work-place will have to have one to compete with countries that make it wide-spread.
And soon after, we'll even skip the screen and hook into the optic nerve. (All your damned JS libraries will be obsolete yet again. Lobe.js will be in :-)
Permissions Granted (Score:2)
Show of hands: Given what we know about what apps do with our personal information, who wants to install the app for typing with your brain?
Poor timing (Score:4, Funny)
Re: (Score:2)
myoelectric prosthetics (Score:2)
Sounds like they are actually just implementing myoelectric prosthetics [ottobockus.com]. Note that that is measuring muscle signals, not nerve signals. Similar technology has been used for subvocalized speech recognition [youtube.com]
There have been prosthetics based on measuring signals from the spinal column or peripheral nerves, but they usually still use implanted electrodes because nerve signals are much weaker than myoelectric signals.
BMI (Score:2)
I'm certainly no health fanatic but I doubt I'm not the only one who reads a medical article and immediately thinks BMI means Body Mass Index.
Solving a non-problem. (Score:3)
And it doesn't help most of the people who would need a 'typing by brain' interface - because the reason they need it is usually that the signals their brain is trying to send don't get anywhere near their muscles because parts of the brain or the spinal cord are not working as they should.
Re: (Score:3)
YET despite not alleviating the difficulties of every possible disability it is still helping and much better than some.
I can imagine there is a numerically large enough part of the worldwide population that can benefit from this.
It's very much like curing one specific rare type of cancer. It still solved a serious issue and it is progress.
Not that I wish it on anyone but imagine if you have a disability that prevents you from using your fingers/hands but your forearms are fine. -will you say it's a "
Not a brain-machine-interface. (Score:2)
Possibly. They shouldn't be calling a "brain-machine-interface" then, because it is not. If they're really doing an EMG, it's not even an interface to a part of the nervous system, because they're measuring electrical activity of muscle cells.
Ironically, I probably won't be able to use this toy, due to an SCI that generates spurious muscle contractions, among other symptoms.
Re: (Score:2)
FTFA: "The innovation lies in picking up EMG more precisely—including getting signals from individual neurons—than the previously existing technology, and, even more important, figuring out the relationship between the electrode activity and the muscles so that CTRL-Labs can translate EMG into instructions that can control computer devices."
Re: (Score:2)
EMG doesn't pick up signals from neurons. It picks up electric activity of muscle cells (the M stands for 'myo'). If anything, their method gets signals from individual muscle fibers.
Picking up an electrophysiological signal from unprepared skin and without an adhesive electrode sounds interesting, though.
Re: (Score:2)
This.
It's a muscle-machine-interface. Just electrical instead of mechanical.
Finally! (Score:2)
Finally, someone [else] has realised that the brain is not necessary for a computer interface:
https://slashdot.org/comments.... [slashdot.org]
https://slashdot.org/comments.... [slashdot.org] *
https://slashdot.org/comments.... [slashdot.org] ... although my guess is that these people are still trying to teach computers how our brain works, rather than the other way round.
* Probably the closest slashdot comment I've written to TFA.
This explains ... (Score:2)
all the 'y0u fAiL' posts.
(Damn, she has a nice pair of ....) where the hell is the backspace?
Body Mass Index???? (Score:1)
BMI
I get sick of acronyms without an explanation. But, what do you expect in an article about a company whose name is an abbreviation from back in the 8 bit computer days. Hint, modern computers can handle actually using whole words.