Computers on the Mind

it's ours. I don't know what we'll do with it or why it's forward thinking, but we may do an episode on the future of axes and ax text technology. That's good. That's good of of plastics. Yeah, plastic ac is in particularly quite different future than that's true. Well, that's not what we're talking about today though. Today we're talking about brain computer interfaces. Uh what now, you looking at me like what I want to talk about axes? No,

absolutely not. I have no idea what we talked about that. Okay. Brain computer interfaces is what we're going to concentrate on today. And uh, and really it's exactly what it sounds like. It's an interface that allows you to interact with a computer using just your brain power. That sounds like science fiction. Well at one point it was just science fiction, but now it's science fact and in fact there we've made

a lot of progress in this field. There. Essentially, the way this works, you have to understand the way your brain works. You know, you have neurons in your brain, and the neurons communicate, They pass along messages and commands to the various systems in your body using a combination of electric signals and chemical processes. Right, and all these neurons in your brain are connected by things axons and dendrites, and they and they create complex patterns of electrical charges

that surge back and forth. And this is what manifests as thought right and uh, and also as a command, So like when you lift your arm, you know, or

as Buck Rubonds would say, throw a harpoon. Um. So yeah, I mean the the neurons, because they are communicating through partially through electricity, not solely through It means that if we had found a way, if we could find a way to detect electrical activity and to assign particular types of electric electrical activity to particular tasks, we could interpret that electrical activity and translate that into something that a

computer could understand like a command. Right. Yeah. Well, the interesting point this goes all the way back to was the first time that someone noticed that the brain makes this electrical current, a British physician named Richard Cayton, I believe, And uh yeah, so so this this has been this is not a new idea, no, no, no, it's certainly a fantastic one, right, Yeah, The whole idea of being

able to harness computer power is relatively new. But the fact that we have we've learned about the electrical activity in the brain, We've not about that for a while, right. And so essentially, what what your brain is doing is it's producing meaningful action based on electrical impulses, which is actually the same thing a computer does, right, yeah, yeah,

electrical impulses manifest as meaningful action. Yeah. In that case, the electrical impulses tend to be on very tiny microprocessors and going through extremely tiny channels on these microprocesses we're talking about on the nano scale. Uh. And so if we were able to to detect and interpret these these electrical patterns that go on in our brains in a meaningful way, we could then create an interface with a computer and pass commands through thought and in fact, yeah,

pretty crazy, but it's awesome. But how do we read that? I mean, how do how do you figure out what's going on inside someone's skull? Well, it's actually simpler than you might think. It's it's it's difficult, and it's it's simpler. At the same time, it's um So your brain is producing electrical impulses and in a way, it's not all that different than the data you'd create by punching keys on a keyboard or moving around the mouse. When your

brain thinks a thought, something physical happens. If you have a way of detecting what that physical thing is that happens, then you have a way of creating input. All you have to have is some kind of sensor or machine or way of gathering the input from the brain. That's the part that's hard. Now, that's one part that's hard is actually exactly teaching teaching the computer or the or

the person or the robotic limb. You have to teach how too, how to interpret a thought into an action, and that there's a part of a challenge of that is that two people could go into the same lab to have their their thoughts analyzed Essentially, you can get an m R I machine, for example, and look at the brain activity and and you tell the person inside the machine, all right, I need you to think about

lifting your left arm. And then you watch the parts of the brain that light up, and you say, all right, well, those are the parts of the brain that I want to concentrate on, because I'm designing the system so that, uh, the person will be able to control a robotic arm that would be in the place of their left arm. So you think, but then you get a second person in there and they think about it, and it's gonna be slightly different places, so it's not like one size

fits all. Well, let's make this concrete. Let's talk about what we're actually talking about doing in terms of the technology. Okay, um, so what are some of the ways that people think they can actually detect electrical impulses in the brain and and do something useful with them. Well, the the least invasive and the least expensive, and the uh probably the easiest would be the e G. Right. Well, first of all, you already mentioned m R I, and m R I can do this, but it can't do it on the

go right. No, right machines are very large and not magnets, and then m r I is really better used to again determine the best places to put an electrode to detect these electrical impulses. So you kind of use it in the testing phase, not in the execution phase. Right, The m R I U seems very useful for UM figuring out what you need to do, but not for

real time control UM. But so yeah, you mentioned the e g. The electro and cephalogram um and this this goes back by the way, a German neurology names Hans Burger found the way to read current by electronic placement, honestly, so they put electrodes on the scalp itself, which detect any electrical activity. Wait a minute, though, you'd have to think that if you're just putting electrodes all over your head, that's got to be a really high noise to signal ratio.

That yeah, because your skull actually does block some of the electrical signals and also distorts some of them, which means that if you want to talk in terms of resolution, you would have low resolution signals, meaning that uh that you would get you'd be able to detect activity, but you would your accuracy would be fairly low. So be like taking directions from somebody who's on a cell phone inside a bomb shelter underwater. Yeah, yes, it's exactly like that.

Or as I was going to say, maybe like if you were to put the electrodes on your scalp and the the let's say that the hardware and software that you're working with allows you to move a cursor by concentrating on it on a screen. If you're wearing the electrodes on your scalp, then it may not be that accurate, and it may require you to to spend a lot more time concentrating and trying to move that cursor than it would if you were to have a more invasive procedure. Right.

So here we come to the other and perhaps most promising technology, right is the the surgically implanted electrode. Right. So this in this case, you have you have gone and you've actually removed part of the skull and gone into the brain. Or you may just put the electrode on the surface of the brain, but there are also procedures where in order to get to specific parts of the brain im you actually implant it directly into the brain.

This also has some problems. Obviously, one being invasive surgery. There's always worst there and and it's you know, that's a big decision. Uh. The second problem is, or potential problem, is that it can cause over time, scar tissue can form over the electrode, which will create that sort of insulating problem that the e G has where you're going

to get signals blocked or distorted because of the scar tissue. Uh. And then also until recently anyway, it also meant that you were you had to be tethered to or Yeah, you would have essentially a connection point that would extend out from your skull and you would have to connect a wire to that, and that wire would be connected on the other end to whatever device it was that you were controlling, whether it was a computer or a robotic arm or whatever. Though this problem has recently been

somewhat conquered. Um, we can talk about that in a minute. So okay, so we're talking about neural implants. An electrode a little piece of metal. It's a chip that goes in your brain and it takes those electrical impulses and it sends them to a computer. But we have to think that uh, it can't be that easy, right, It

can't just send the electrical impulse to the computer. There has to be really advanced software r right, it understands how to deal with these impulses, or based on an extremely uh complicated set of probabilities, can figure out that that little blippy block probably means this thing and as opposed to that little blippity blob, which generally speaking, the way that I understand that this works is that it requires a long session or multiple sessions with scientists and

doctors to go through the process of saying, all right, uh, we want let's say, let's say given, we'll give an EXAMPLETS say it's a robotic arm and uh and they say, well, we want you to uh to lift your left arm. And let's say it's a person who is fully capable of doing this, And so the person lifts their left arm and they interpret those signals and they say, all right, these are the signals we want. These are the signals that say lift the left arm for the robotic arm.

And you repeat that test many, many, many, many many times to get as accurate a picture as possible of the neural activity that's going on when you lift your left arm. And then you eventually say this represents the

input for the robotic arm to raise up. So then the next time you lift your left arm and you're hooked up to this system, the robot arm also lifts it arm it's arm, and you have to do the same thing for all the different commands, so things like gripping or twisting your wrist unless you've actually designed the

robotic arm to have some of these features automatically. I saw one video where the robotic arm would automatically grip something if it if the palm came into contact with an object, so that you didn't have to go that extra step to think, make the fingers close at a certain point. Yeah, the sensor is easier to produce than the brain transfer. That makes it hard to do like an open hand slap though, Yeah, you just griped the person to challengee instead pinching their cheek. Yeah yeah, so

so so don't future problems duel of the future. I would say that if my cheek were pinched by robotic arm, I would consider myself chagrined. Yeah, but anyway, the the would defend your honor and you'd have to I would demand satisfaction. Yes, Okay, I'd have to go and grab

our mystical acts. UM, so they bringing it back. So the getting back to this, if you're talking about someone, obviously, one of the big uh potential you is for this technology is to help people who are severely disabled, who are perhaps paralyzed or quadriplegic and cannot cannot do a

lot of things for themselves. And so for them it's a learning experience because they may not know, they may lack the capability of doing whatever physical action they need to do to send the command, So it ends up being more of a learning process both for the person and for the machine. Yeah, and luckily, our brands are incredibly plastic, and UM are are very good at still continuing into adulthood developing new neural paths in order to

figure out these kinds of processes. Right write plastic as in changeable, not actually right, correct that they are not made of vinyl. So uh yeah. So you talked about UM helping people who have severe body disabilities, and I've actually read about a few of these cases and they're kind of amazing stories there. UM. In the end of two thousand twelve, there was a woman who is quadruple

Egypt and UM. From what I've read, it looks like UM neurobiologists at the University of Pittsburgh Medical Center they enabled her to through a neural implant as so an electrode in her brain control a robotic arm with which she could feed herself. And all it took her was a couple of days of practice before she could move this robotic arm with her brain to feed herself a

piece of chocolate. So yeah, the the the good thing is that people are very are very good at adapting to this, and that we are getting better and better at building the hardware and software that can act as the other half of this system. Right. And I saw an interesting, uh exoskeleton thing which I thought was pretty cool. It actually used an e G cap, so uh, you know, it didn't it was noninvasive, you know. The person This was for people who who could not walk for themselves.

They would get into this exoskeleton and the cap would go on top of their heads. It was called a mind walker, and then they would by by concentrating. I don't I don't you know, I hope not from one of those lost years. It's not sleepwalkers. I'll tell you that's terrible, terrible movie. Don't tell my dad it's like his favorite movie. Wow. Okay, I have no response to that. The uh. But at any rate, So there are other

applications as well. But but the main focus has really been on giving people who otherwise would have difficulty interacting with their environments more ability to do so through these interfaces. Uh. And you were alluding earlier, Joe, to the idea of the wireless approach, because one of the big drawbacks to this was the fact that you would be tethered to a machine, and so you had even more limited mobility

because of that. But um, yeah, there looks like, um, there were resear cheers at Brown University who they came up with a you just close it off. It's a it's a wireless neural implant. So it's it's doing the same job that these other ones we've talked about, where um that you know, it's reading the electrical impulses in the brain and it's sending a signal, except this one is sending it wirelessly. It's got a wireless transmitter, lithium

ion batteries, rechargeable. It's its own little unit and it transmits wirelessly to the computer that controls UM that controls the the operate the device right right, whether it's a computer or whether it's a robotic arm. And yeah, it's uh it can I think according to what I was reading, it sounds like right now it has very limited operating range about a meter away. So you're still that's I mean,

that's still way better than having a court. Yeah, definitely, And and uh, right now, no humans have had this implant uh implanted in their heads. They don't know. It's only been uh used on test animals monkeys and pigs so far, so no humans have had this procedure done. And uh, I don't know when that might happen, if when that might change, but it it's an interesting advance

in the field. So uh, speaking of monkeys, I mean, there have been some really amazing studies done with monkeys and these neural implants, and uh, when I was reading about it, showed that, uh, with some software tweaks, UM engineers had gotten it to where monkeys could and and keep in mind, these are monkeys, you know, they're not people who can understand complex instructions given by researchers and

can really respond to training like this. The monkeys figured out how to move a computer cursor into a target zone with I can't remember exactly the percentage now, it was a significant percentage of what they could do with their arms, you know, just moving it by hand. Well, there's there's also idoya. Did you hear about idea? Oh yeah,

the runner. Yeah yeah, So we're talking about a twelve pound in monkey that, through thought alone, caused a two hundred pound, five foot tall humanoid robot to walk on a treadmill. That's awesome. And the monkey was in North Carolina. The robot was in Japan. Wow. So so cyborg monkey armies storming the walls of our cities. Excellent, remotely controlled, remotely controlled by their by their true monkey overlords in the you know, far away so power chambers. In the future,

there will be monkeys. Yeah. Well, I mean so we've been talking about how these in the near future, neural implants and other brand computer interfaces UM could be very helpful to people who suffer from, say, quadriplegia, or any condition that that's debilitating physically. You could create a robotic arm or you could you know, mess with the computer without being able to move your arms to move a mouse, right you could communicate in ways that you might not

have been able to before. What happens when these things get good enough that, as we've talked about before, you might want them even if you don't need them. Yeah, it'll be interesting to see. I mean, I would imagine that unless unless the state of the art is so amazing for the invasive forms, I can't imagine those ever really taking off, at least in the foreseeable future. Depends on what they can do. It would take. I'm to

talk a lot about it. I mean, you know, you hear all the racket about what is it a homo cyberneticus, and uh, yeah, I'm thinking, oh, you're talking about like a species. Yeah, homo space cyber cyberneticus. Yeah, that kind of thing. So I think I think that we're at least fifty years away from that personally, I mean, the it's I think that's pretty safe. It's good number. It's a good numb at least from but but I mean, honestly, like, I can't to me, it's hard for now. I say that.

But then as a kid, I never would have imagined that I could carry something that would have access to pretty much the sum total of all human knowledge in my pocket. I didn't think that was gonna happen either, and it totally did, So I could be wrong about this. I think it's more likely that we will see and we already have seen things that are that are essentially using the e G method to to act as some sort of control device or um. You know, a toy

like I've seen the toys that you wear. It has a little helmet that you wear, and through concentration, you're supposed to be able to control the uh, the path

of an object on a board. There's there's one I saw at CS a few years ago, and the way it worked was it had a a little a little air vent that blew air straight up and you put a ping pong ball and it would suspend in the airflow right and that by concentrating, you could either increase the airflow and make the ping pong fly higher or decrease and make it fly lower, and a rotating obstacle course would go around and you would have to try and maneuver the ping pong ball so it could fit

through hoops or pathways or whatever. So um, you're introducing the possibility of telepathic gaming, telekinetic game. That's the idea. I don't know how well these things really were well right now right now? Yeah, yeah, but I'm saying that we already are seeing this sort of stuff. Yeah, and they're still doing research about more, um, you know, noninvasive ways to figure this kind of thing out and speech

recognition technology. NASA way back in two four actually was was doing a program that they were attaching sensors under the chin into the throat because they found out that when you just think about talking your your muscles still have some kind of electrical activity that can be picked up. And so, you know, stuff like that I think holds some amount of promise. So could what you're saying that by thinking a word, sensors on the muscles in your

throat could could follow commands even without your speaking. Yeah. Well, and on top of that, there are other alternatives that that might uh take some of the load off of the brain computer interface UM train, I guess because you could you could look at things like eye tracking software, uh, and a lot of eye tracking software out there is getting pretty sophisticated to the point where, um, if you

are capable of moving your eyes, not everyone is. There's some people that you know locked in syndrome who are incapable of really making any sort of motion. Um, But if you're able to move your eyes, then you're able to control elements on a computer screen. And I've seen some pretty cool implementations of that, and it all it's really just using cameras to track your eye movements and and plot where you are looking on a screen, and

then interpreting the a as a command. So there are other interfaces out there that we're seeing big advances in that could either compliment or in some cases, depending upon the use case scenario replace brain computer interfaces. All depends on what it's being used for, and you know what you want the outcome to be. Obviously, it wouldn't work for every implementation. So can you imagine any situation in which brain computer interface would be really useful to a

person who has otherwise full control of their body? Spies? Spies immediately, spies, are you kidding me? A spy? Because you if you are able to interact with a computer system without any overt sign that you are doing so, I would say that it would be very useful for a spy. So if you want to like start start your tape recorder, without touching it, or say, or you're sending a message saying all right, I'm about to I'm about to be eradicated, so you might want to burn

my apartment, that kind of thing. I mean, I watched a lot of born identity, you know, even in our own industry. You know, if, if, if our producer Noel could send us a text message just by thinking about it, without having to make the noise of typing on keys or you know, wave his arms wildly or however else he gets messages across to us, it would be filthy. Maybe a filthy message. Noel's just smirking at me right now and saluting me. You're not getting them. I choose

to ignore them. Well anyway, that that's kind of wraps up our our discussion about brain computer interfaces. It's a really interesting field and I'm curious to see how that develops over time. Uh and guys, if you have any suggestions for topics we should cover and forward thinking, please get in touch with us. Our email address is FW

thinking at discovery dot com. Please go to the fw thinking dot com website because that's where we have all our videos, are blogs, the podcasts, all our social media stuff. Is there if we look forward to hearing from you, and we will talk to you again really soon. Please get down the ask. For more on this topic and the future of technology, visit forward thinking dot Com, brought to you by Toyota. Let's Go Places,