Journey to a Better Brain SXSW 2021

Show video

Hi, Vivienne. Thanks for joining me. As always fun. I think we're gonna have a great chat today. Pretty exciting topic. Something that you and I have talked about a couple of times in the last couple of years, too. Welcome to everyone who's joining us.

This is one of those recorded panels a bit of a chat between Vivienne and I, about neurotechnology. For those who don't know neurotechnology is technology that interfaces directly with the human brain. And it can be very direct. It can be implanted in the skull and require brain surgery or it can be quite a lot more indirect stuff that's indirectly interfering from our interactions with existing technology like our phones, what's going on even. There's a broad definition for you to kind of understand what we're talking about.

I think the stuff that we'll be talking about today will be a bit more brain focused than the broader stuff but we're also not going to be talking about the stuff that's being implanted in your brain by a brain surgery. We're gonna be talking about some of the more noninvasive stuff. Similar to what my company Humm does and also something that Vivienne's very interested in. Vivienne, I would love to just have you introduce yourself and everyone to find out a bit about your background.

Absolutely. It's probably worth saying for the purpose of this discussion that long, long ago when I was interviewing to go to grad school I would ask the people next to me, young undergrads trying to get into a neuroscience program. "What do you wanna do research?" And they would say something like, "Well, there's this protein cascade in yeast." And of course in my head, I'm thinking "What the hell does that have to do with neuroscience?" But for any of you who don't know to a molecular neuroscientist we are all approximately yeast.

Really, the differences are pretty minimal. It's like a spherical cow situation. And they say, "What do you wanna research?" And I would say, "Cyborgs." And they would literally scoot away from me for fear that my crazy would rub off and they wouldn't get in either. But, as you were suggesting this is not so crazy an idea anymore. And while we may not be talking about certainly cyborgs like the Borg or Robocop there is huge work and advances that have been made ranging from deep brain stimulation and motor neuroprosthetics, robotic limbs driven by our own motor cortex.

Those would be invasive chronic implants. All the way through to work I've been able to do looking at brain stimulation, looking at predicting manic episodes, trying to improve face recognition in autistic children. There's this huge realm of things that so many of us take for granted the ability to speak for a paralyzed person or to reach for a cup of coffee that so many of our fellows on this planet just...

They don't get to take it for granted. And we live in a science fictional enough universe where cookie people like us actually get to build this stuff not just as home brew products and projects but for direct medical applications. And in many cases just about improving our own wellbeing. And it was really exciting to live in the world that when I read "The Diamond Age" and other inspirational pieces of science fiction and I think, "God, I wanna build something like that." When I was a kid or even not as a kid like this is a thing I get to do as a job.

I've started a number of companies. I've been an academic researcher at Stanford and Cal and nowadays I largely do philosophical... Philanthropic research and a real focal area is this work on neuroprosthetics, neurotechnologies, sitting on a number of boards and advising a number of companies. Awesome. Thanks for the introduction. I'll give a tiny bit of an introduction of me and then I'll explain how I came to meet Vivienne and the journey we're on together together a little bit too.

My name is Iain. I am the CEO and one of the co-founders of a company, based in Berkeley, but we started as you may be able to tell from my accent in a place far, far away in Western Australia, around some, about 2012 there was some research going on that my co-founder was involved with. His name is Dr. Fiori he's a medical doctor in Australia that they were using powerful magnets on people's brains not putting an inside the brains, but putting a, basically like a hat on and then putting a an electric signal through a magnet that would then cause the electrical signals in the brain to change something called TMS. And I was studying physics and law.

I was just a normal young person on his way to a career. I've been interested in entrepreneurship and stuff like that, but I've never been on the path to start a neuroscience company. I was in the middle of college and at that time to discover that there was this technology which could do the things that I discovered in that study in that research that I participated in was mind blowing. And it really opened the window onto this huge field of the stuff that Vivienne was just talking about which is these technologies that can directly influence normal people, but also, really sick people, like Vivienne was talking about, people who have difficulty stemming from how their brains work and make a big difference in all of these people's lives. And it's technology that's been in the lab for a while. We've had some really interesting and exciting things happen, but what's really fascinating and what we're here to talk about today is it's parked out of the lab and into the real world and some of the interesting issues and challenges and exciting opportunities involved in that.

We're here today to have just a conversation about some of the emerging trends, how they impact people's lives and also just talk about some of the challenges in the neuroscience field and neurotech field. We're gonna spend the next about 45 minutes just diving into a couple of questions. And Vivienne, I'll start you off with a question about what you see as the big trends that you're seeing emerge in these technologies, that interface with the brain right now? A bit of a broad question to start you off. Yeah. That is pretty broad. And yet, you can look at trends in terms of what are the... What's the cutting edge of the bioengineering research.

What is sort of the cutting edge in trends in actual application and even just what are the problems people are working on. And I think what's interesting is that last one it's the same ones that I was interested in 20 years ago. On my website for decades, I had the just this little inexplicabe line, unless you were an efficient auto of scientific publications.

It said, "What if it was 20 (plus or minus two)?" And it was a reference to this, one of the most influential and famous scientific papers of all time, certainly psychology papers of all time, where the title was "The Magic Number Seven Plus or Minus Two". And it was talking about the earliest work, establishing this concept of working memory how much things people can keep in memory at any given time. And it relates to attention. It relates to the complexity of an argument you can follow. And it turns out it's interesting more than just as a passing scientific curiosity.

A paper, just in the domain of COVID-19, people with larger working memory spans were more likely to wear masks. They show fewer and lower infection rates. People with larger working memories, a roughly a half of a standard deviation of increased in working memory span.

I already gave you that spread seven plus or minus two we'll skip defining what those numbers mean but let's just say, you're eight or nine, turns out that ties into about a 30% increase in wages and earnings. These are real world variables. And 20 years ago, when I was going around telling people I wanna build cyborgs that was actually what I meant. I've done work on cochlear implants which is probably the neuroprosthetic most people are most familiar with, a hearing aid. That's not what this is, but a hearing aid but one that actually wires into the eighth cranial nerve into the auditory nerve and restores hearing to people that are profoundly deaf.

To know that the same questions are what we're still trying to address, memory deficits, emotional instability, things like anxiety and suicidal ideation, paralysis, Parkinson's disease. Like those haven't changed. That's still the trend, is solve these core problems. We're just so much closer to it now because of changes in those other domains.

I think one of the big trends is if you are a "Cyberpunk" fan that you were probably raised thinking of brain-computer interfaces as things you literally like plug into your head like you'd plug in a printer or the amplifier for your guitar or something certainly I've played "Shadow Rider", "Cyberpunk 2020", God, it is disturbing to think that we're now in 2021 then that was exactly what you did to log into the net in those role-playing games. Well, now brain-computer interfaces are noninvasive systems. One of the more exciting trends that one of the companies advice, focuses in on is mashing up the ability to read brain activity in this case, using EEG. Looking at the electrical activity that bleeds through to the scalp, combine that with machine learning couldn't have done this even five years ago this would have been absolute bleeding edge but now advances in deep neural networks mean you can take this smeared useless electrical activity on the scalp, useless from a single person in a single instance. And suddenly you can tell what's going on.

And then linking that back into a closed loop for an augmented reality system. In this particular case you might be profoundly paralyzed, unable even to speak and yet by tracking eye movements with cameras and looking with these systems that are picking up on attentional cues the system can actually figure out what it is you're trying to identify and allow these people that might otherwise be locked in to their bodies, to interact with the rest of the world activate things in their immediate environment. And again, sounds like science fiction but the chance to be a part of actually building this, right now, give upwards of 100,000 people in the United States with advanced cerebral palsy the chance to interact with their world. That is an exciting transformation.

That's one matching up old school EEG with machine learning and augmented reality. Super exciting. Another big trend is actually closing the loop of the whole system. Using EEG to do noninvasive readout but feeding it back in using things like ultrasound or steerable magneto-inductive systems. And of course as Humm is intimately familiar with using technologies like transcranial alternating current stimulation. You brought up the magnets early on.

I mean, magnets, the results, there were really interesting but what's freaky about all of this is we didn't know what the biophysics was going on. It's just, wow. If you put a magnet really close to this part of a rat then this really interesting thing happens. If you put it in the wrong part you zap the ponds and it stops breathing. But if you did this one part of their their barrel fields that control their whiskers you see this crazy behavior, no one knew that what you were doing was causing the poles of all these individual neurons in your brains to like line up, sync up and then go crazy. We could have this transient little stroke and figure out what that part of the brain was doing.

But now we can do something so much richer which is drive synchrony across brain regions or look at phase locking within certain regions and induce with a great deal of precision. And again, with the potential to marry these two things together the EEG readout, the phasic stimulation plus machine learning. Again, you wouldn't have been able to do this, five, 10 years ago, and yet this is a big new trend in the space with wide applications. Again, the exciting things have been moving away from the idea that the only progress here is going to be drilling a hole in someone's skull, jamming a bunch of wires in there, looking like you're a member of the Borg, to systems, you could literally just take off and put on as you need them whether conceptually or literally that's really exciting.

I will, even though it's not a focal point of our discussion I'll also say there's really interesting trends in chronic implantation. Building, looking at the biophysics of systems that are chronically inside your brain which allows a different level of control and finesse. And knowing that this might be something that is a part of you for the rest of your life, building something that's a safe part of that.

Lots of advances there in areas like neural dust and expandable nets, injectable nets, these are exciting spaces. And again, if you were me and it was theoretical neuroscience and science fiction and that's how people saw this field. And now you have the chance to take someone with a memory deficit or someone with cerebral palsy and make a difference in their life. I get to live in science fiction. It's really cool.

Yeah. A couple of just fascinating trends there. I think obviously as science develops and we get a better understanding of the brain the tools develop as well. And we're better able to just increase the quality of that technological interface where we can collect more data. Some of the things that Vivienne was just talking about EEG is basically just a little headset with a couple of little, either sticky electrodes or like pieces of metal just pull electricity. They just conduct electricity that goes through the outside the brain, through the skull, the skin.

And yet the data that you can get out of something like that is able to tell you how someone is thinking and feeling. And some of the trends that we're seeing is as the quality of that data and the ability to collect it gets easier. And it gets put together with other technologies like what Vivienne was mentioning with the transcranial stimulation stuff.

It's just opening entirely new doors for what's possible for us using your technology to enhance our brains and choose how we think and feel. Yeah, really cool. I'm super fascinated, Vivienne, by what you've seen in maybe the last 10 years, these different technologies not necessarily neurotechnologies even adding together and becoming more than the sum of their parts. I know you do some really fascinating research using AI.

Could you tell us a little bit about how your experience in machine learning is crossing over with your experience in neurotechnology? Yeah, well, I mean as someone that wanted to build cyborgs, but for certain experimental domains it wasn't really a career plan. I studied this field called theoretical neuroscience which was really using machine learning to study the brain and sort of vice versa. And a lot of the work being there, ended up being foundational for the development of deep neural networks, this new class of AI that really has been responsible for this the recent revolution in the field where suddenly these systems are comparable to and in some cases better than the best human performers in their space, AlphaGo, those are really fun things building systems that learn how to play soccer games or first person shooters better than humans can play them.

And seeing those things come out, left me with the sense of what if we did this in education? What if we did this to combat bias in workforce? And again, for me, this was very early days, but in a couple of cases, this tied right back to neurotechnologies at least with a slightly broader definition. For example, one time someone said to me, "Hey, you know I have this wonderful startup idea, but I would drop it all if you could help me figure out how to do bipolar disorder." And it just...

It put a bug in my head and I just... I couldn't drop it. I spent a whole night reading papers about bipolar... I mean, I'm a neuroscientist I know what it is, but I'm not a neuropsychologist. I don't know the fine details of the research in the space where it didn't at the time.

And reading these papers, I just convinced myself, "I bet we could predict manic episodes and bipolar sufferers just using mobile phones." And then the craziest thing happened the very next day, I joined the board of a small startup that was building technologies to track people's emotional lives on their phones. Like it was the perfect moment.

And then to make it even better, one of the employees... I said, "I'll join the board if they let me build a system to protect these manic episodes as a side project." And one of their employees said to me, in confidence at the time, "Hey, I have bipolar. I overheard you guys talking. You can use all of my data."

And sure enough we were able to use without a direct interface at all just based on GPS, gyroscope, accelerometer, data some other things, but like movement data turned out to be really informative. We could see evidence of an emerging manic episode long before they were aware of it in some cases three weeks ahead of time. We can really think about invasive neurotechnologies or what I call a funny hat, the non-invasive but you still gotta wear this big thing, but in fact we can affect who we are just using the data that exists out there in the world. Anyone who's thinking right now can't you do bad things with that data as well. Hell yes. Yeah, of course you can. And in fact, that was part of the heartbreaking part of that research is we could see people in our dataset, clearly having manic episodes and we couldn't do anything about it because we...

Doing so would, for example, might disclose to their employer that they are bipolar. And we didn't know that wasn't why they were using the app. They would...

This was not a medical app for them to get a diagnosis. We felt trapped like we couldn't do anything about it. We've launched... Relaunched that project as a purely philanthropic one now, just to look at what we can do but that broadens our horizons on what's possible in a space like this. Another domain we're looking at now with a non-invasive neurotechnology more similar to the scope of Humm and others is a project looking at phasic light therapy for Alzheimer's dementia and possibly even just depression.

One could look inside the brain using an EEG and see that particularly when you think about a lot of very human complex type behaviors, like li... A lot of language and a lot of planning and memory things. You see a lot of activity in a certain frequency range particularly around like 30 to 40 Hertz which sometimes called a gamma band. And it turns out this group at MIT were just kicking around this idea what if we flashed really high intensity light at that same frequency range paired it with a sound in the exact same phase would this do anything with rats that effectively have Alzheimer's disease? And sure enough, they saw this big change and I'll be perfectly honest.

If you'd asked me a few years ago, "Could you go in and like, forget with a light, even with direct electrical stimulation, not just create a change in the brain, but it would be a meaningful one. I mean something that might actually someday treat Alzheimer's or improve memory performance?" I just said, "Come on. It's all out of phenomenal. There's no effect there." Turns out I was wrong.

There is some very solid evidence that this stuff at least has transiently positive effects. And if you're staring at an Alzheimer's diagnosis even that is a kind of hope that that hasn't existed before. But again, this isn't...

It's not plugging into your brain. We're not putting electrodes directly on your skull. The company on whose board I sit I'm not gonna name anyone in any of these cases. They're just building a little box that flashes a light with some really cool advances in understanding the nature of color and the idea that something like that could prolong someone's life or at least quality of life. That's huge. That's amazing.

And it, again doesn't require you to go in for brain surgery. I don't know when the date comes when elective brain surgery becomes a normal part like cosmetic brain surgery just because, "I feel like my wit isn't quite what I'd like it to be. I wanna up that." But we... That's been one of the exciting spaces is the viability of noninvasive strategies for making real change in people's lives.

This is just a couple of examples of spaces where artificial intelligence or other sort of contactless technologies have actually shown some real positive effects in people's lives. Fascinating. I think it doesn't really make its way yet into the public consciousness, from what I can see just how fast some of this stuff is moving. And some of the like incredible stuff that we're seeing in situations like the companies that you're involved in.

You're involved in five or six really fascinating companies that in some way relate to what's going on in the brain. You know what's funny? You mentioned earlier how the science was driving technology. Let's also be honest. A lot of it's been the other way around. We all have this big cranial nerve called the vagus nerve. And it goes from your central deep of your brain down into an interface, your gut and other parts down into your body stem.

And you... Many people probably don't know this but in humans, once we get away from certain things like the cortex you've got a whole separate nervous system down in your gut called the enteric nervous system. There are more nerve cells down there than are up in the older parts of your brain. And so there's a whole lot going on.

What people discovered is wow, some of these signals in people's guts and coursing on the vagus nerve seem to have some relationship to things like depression and mood disorders. And so someone thought, "Well what if we stimulated the vagus nerve?" And in a funny sense with no idea why it was going on they found, "Oh, this actually seems to help with depression." And then they started following it up and up and up. And it got in to what, I guess we're gonna call deep parts of the brain, because what we are talking about is called deep brain stimulation. Getting up into the basal ganglia and these other deeper brain structures. And they stimulate there.

And not only was it addressing depression, but things like Parkinson's, which seems completely unrelated and seemed like it would treat everything. And we didn't know why what a crazy thing to let technology, this thing, this electrical stimulation and people try phasic stimulation and they just try direct current and all the different things. And it helped. And people didn't... It's only recently that we've just begun to get ideas of what's really going on in the biophysics of that process much less than neurophysiology, more broadly. And so knowing, and being honest about it like in some of the cases here, the tail's wagging the dog. I guess our dog has like a lab coat on and a degree from Caltech and the tail is industry or at least medical application.

'Cause if it makes someone's better, just follow it. It's certainly not unique to neurotechnologies. We see this a lot in neuropharmacology as well.

There are treatments for schizophrenia and depression C-SSRS and others. For the longest time we did it 'cause it works not because we knew why. And maybe one of the best examples is actually bipolar disorder. Like why did lithium...

Why does it work? And why does it only work on half of people? It's amazing that this has been a standard treatment for major neurodisorder, maybe the biggest one. And we didn't understand it until essentially the last decade what was going on. There's, I guess I used to joke when neurotechnologies go full in industry and you have to have a customer service department then we'll start to really understand how this stuff works.

I mean, I love science, but it can be a little plotting sometimes. Things are accelerating. Fascinating. When you were talking about maybe the dog and the lab coat one of the things that I was thinking was there was this funny product a couple of years ago.

I don't know if you saw it. It was an EEG, one of those brain sensing headsets for dogs so that you could put it on when your dog was barking and you could combine that with like their bark data and find out what they were thinking and feeling. Yeah. What a time to be alive. Finally, the human condition solved or at least the canine condition solved. And I bet we can really probably brainstorm the three things, "Feed me, pet me... We don't need to speculate further on what it is doggy wants, but, "I gotta go walkies."

I guess even to segue from the dog example, I mean the dog's brain is so much less complicated than the human brain, but even then that was a really challenging problem. And it was a hard product to build, partly because we don't know a lot about the dog brain, but to talk about the human brain, I think it's clear like listening to the things that you're talking about as we learn more we're having some incredible capabilities pop up for things that we can do helping people with Alzheimer's and the other things you mentioned, but just as exciting as the fact that we know so little still about the brain and there's so much to learn and so much more to discover. And so I really think that a lot of people don't realize how this century I think is going to be something of an awakening about how we experience our lives and use our brains because we're just learning so much and having so many cool discoveries.

Yeah, I... You'll often come across people saying, "We know nothing about how the brain works." Or others that think "We've nearly figured it all out." It's one of those classic cases where my rule of thumb is however complex you think it is, it's more complex than that but that does mean we're learning, The advances have been huge. It just turns out, Oh my goodness there is so much more nuance here. There's transcriptomics and large-scale brain networks and synchronous activity and everything from the molecular scale to the entire body and even distributed beyond that, recognizing one of the things we've done is recognize how big and complex the system is that is our brains.

That's us really when it boils down to it. And that's why there are so many ways that you can tap into this. When I was younger, the... I can think of a couple of the domains in which people were putting the most effort into known prosthetics. I mentioned one already cochlear implants.

The other that was sort of hot at the time were motor prosthetics, an exploratory space but the first human work had been done. And there's a good reason why those two were so far ahead of everywhere else because inside your ear, there is essentially like the keys of a piano, all laid out by frequency from low to high, and then it's kind of all wrapped up. And so you could just stick a bunch of wires in there and stimulate these different frequency places or what's called tone atopy in the cochlea, this inner ear part, which is technically, a part of your brain or at least directly innovates a part of your brain. And so in that sense it's simple, relative to the rest of the brain.

It's simple, turns out the motor system had something kind of similar. We have the stretch of primary motor cortex, M1 you might call it. And in it basically, there's like a whole map of our body including little sections that are hands and tongues and arms. And so if I just go in and I record from that part so for this ear, it's sensory we're stimulating individual parts of that tone atopy but for the motor prosthetics I wanna read out. I just put all these wires in and they read out these different parts of the brain and you find a cell that seems to really like it, like it.

They don't care about anything. They're just little cells but bear with my anthropomorphizing them. Then like reaching in this direction and another cell really prefers reaching in this direction.

And they found that all they had to do was record all this activity and then average it together. And if predominantly the cells were reaching here that's how the robot limb reach also. It required so little understanding of the actual brain function to make progress that we were able to make progress. Well, Alzheimer's isn't like that. Memory, whether it's long-term memory or working memory is so much more complicated than that. The HIPAA campus guide, if anyone's stuck wires in there like that, that's a crime you should do nothing like that.

Making progress in spaces like that seemed impossible. We didn't have a good theoretical models of what was happening so you couldn't make any progress on the technology and engineering, but now we can recognize things like, well maybe the micro structure of the dorsolateral prefrontal cortex is still a little complicated, but we can see that some of what it's doing is essentially modulating the activity in other parts of our brain that are farther back, more posterior pridal and so forth. Well, what if we just increased the strength of that relationship allow it to take a little bit more influence over these posterior regions. Turns out if you do that, then working memory goes up and you do similar things more immediately in the frontal lobe that might be involved.

For example, in emotion or emotion regulation you can actually see benefits there as well. It turned out, we didn't have to understand exactly what the brain was doing at a micro-architectural level to be able to play some improved role. For example, in autism It really looks like there are connectivity deficits. This may not be the whole of what autism is but different parts of the cortex are not talking to each other as robustly as what we would call a neuro-typical person. Well, it seems pretty clear that we could actually help that. We could help interplay that transcortical communication.

Why would that be interesting? Because so much of how I imagine you as an audience member, think of who you are. If you got in trouble did you feel like you needed to lie to get out of trouble? Or would you never do that sort of thing? Would you begin sweating if someone told you you have to get up on stage? Or do you just roll with these sorts of things and keep it together. Understanding how to produce that in the brain is incredibly complicated but understanding how to modulate it is something we now know how to do.

And seeing these kinds of technologies come into play is really exciting. So exciting. I mean, one of the numbers that people often throw out to explain just how complicated the brain is that there is many connections between neurons in the brain. Something like a hundred trillion as there are stars in the Milky Way galaxy.

It's a pretty complicated system, but at the same time it's also just incredible how much we can do even today at a very early stage of our development of this industry with these simple techniques, like what you're talking about. And on that note, we do have about 15 minutes left and I would love to talk a little bit about how this industry is gonna develop and some of the challenges. To start us off.

I think something that I often say to people, Vivienne and I'd love to get your thoughts is that working in your technology right now, I kind of feel like I imagine someone might've been feeling about working in the internet industry in the late '80s, early '90s. We are having an effect like we are able to already do really cool things. Some of the stuff that you've mentioned is game-changing for people, but we also can feel that it's not even really yet begun we're putting these technologies together and starting to see even more incredible discoveries and results from the things we do. We're really yet to see a lot of massive commercial successes outside of perhaps the medical industry. We're really coming to grips with the fact that this technology could actually in the way that maybe AI and a couple other technologies like that will in the 21st century really define what it is to be human.

But of course when the internet became more and more adopted it didn't necessarily benefit everyone at the start. And even with things like social media, more recently we've started... We've seen that technology doesn't always help people homogeneously and in a democratic way. And one of the things that I love about your work Vivienne, is that it's really focused on making sure we do good here and we learn from the way that those technologies are developed in the past. I wanted to ask you, if you could tell the audience a little bit about your perspective on just in maybe the short and then the longer term who are gonna be the people that are affected and benefited and what are some of the challenges that we have as an industry to make sure that we're really helping people and everyone too. We have really short memory spans when it comes to new technologies.

The iPhone was the first smartphone. Well, that's funny 'cause before the iPhone came along I owned an HP Windows CE phone, and I had a Handspring before that and a Phillips Nino before that in the '90s and, and had color screens and ran programs. I actually loved my little Nino.

I'm sure now it would feel like I'm carrying around a whole computer with me with no power whatsoever but still it existed long before the iPhone, 10 years before the iPhone came out, the first genuine smartphones existed, nobody remembers them anymore. Steve Jobs invented the smartphone. AI, AI Geoff Hinton's lab, a famous AI researcher. They won an image recognition contest in like 2006 using the very earliest kinds of deep neural networks. And within a few years, that technology really took off. And now it's everywhere.

And in some ways, ways that we don't appreciate like I love, particularly my wife loves, who has hearing loss that she can hop in a video chat and in real time a conversation which will never take place again with no human doing transcription can get usable real-time transcription at the bottom and that's modern artificial intelligence. It essentially entered our lives in 2012. Well, no, the entire internet economy is based on AI. All those bidding wars for your attention involving ads that's all artificial intelligence.

We just don't think of it that way. We've sort of forgotten where we started well, we're here in a similar place with neurotechnologies. Again, cochlear implants have been around for very, a long time like decades. And motor neuroprosthetics the first work in that space was just after the turn of the century, 2001, 2002 in humans.

I mean, people with chronic implants, driving robotic limbs or controlling cursors, there's this great video of this guy who was in a motorcycle accident been for like the last six years, profoundly paralyzed. He rolls out in a wheelchair and he's got this giant thing like literally glued to his head, it's dental cement, that's in a hole and it's sitting on top of his head and they say, "You see that screen. You see the screen." This is gonna be deep cut for anyone who isn't 150 million years old but it looks like a Commodore Vic 20 screen or something.

And so it's like this giant blocky screen. And they say, "Just imagine you're moving the cursor." and is like, "What am I supposed to do?" And they say, "Well just imagine you're moving your hand in a circle." Which he obviously he can't do is profoundly paralyzed. And suddenly the cursor starts to shake and then it starts to move and he starts screaming, "Holy shiet! holy shiet!" Like this is the first thing he's moved on his body other than like his face in his eyes in six years.

And now he's moving a cursor on a screen. And again, that was nearly 20 years ago. We've seen this progress coming but it's been a lot of scientific curiosities and very particular medical applications. Now, we're on the border of something like an internet revolution. And yeah, even there I remember when I was a kid putting the phone on the hook and having the dial tones and having a tape drive over here, not that engaging unless you were in a phenomenal nerd that really wanted to get on early message boards. But that was the precursor of everything we know today.

And you can really see that early concept there, but frankly, it was allowed to evolve in a way that didn't, people didn't pay a lot of attention to. So much of what we hate about the internet today came about just because we let people move fast and break things. Yeah, that's a little bit of a dig but let me ask you quite honestly I don't think of Mark Zuckerberg as a villain. I don't love Facebook, but I don't think he's a bad guy but do we really want Mark and Elon Musk and a handful of other billionaires that have been first movers in this space? They all have neurotechnology companies. Do we want them deciding the industry norms for our own brains? Do you want Facebook in your head? It's an...

It'll be another domain to gather information about who you are except here, better than you know yourself like deep information. And this is a field that already exists. Neuromarketing has had many successful startups actually huge acquisitions have happened in this space. Understanding how our brains are responding in ways we can't even control.

That's terrifying and justifiably terrifying. I'm not worried about the Borg. I'm worried about neurotechnologies being used to invade our lives, being used for... By autocracies, being used to predict our behavior in ways that we can't essentially avoid. There's some things that I think are really important that we put forward here at its core principles.

And many of these others have brought up as well. Like I have a right to decide me, who I am and what I'm willing to share. Other people shouldn't be able to influence that but we have already talked about technologies here that can influence my memories, that can influence my preferences, my honesty, my creativity, emotion regulation no one should have those sorts of decisions, powers over me.

But there's another domain here, which is... And it comes up a lot when people hear this one aspect of what I do with my time, which is... But what happens when in a sense we're not good at anything anymore. It's all the machines, making all the decisions for us.

Kind of like we're Eloy except from the outside, we still look normal. We're all just little kids that never had to grow up because we have our own built-in parent, making all of our decisions for us. Here's something I believe deeply.

It's not just about neurotechnologies but I actually developed this philosophy working in neurotechnologies. I should not only be better when I'm using the technology. I should be better than when I started when I turn it off again. If I got an implant, a chronic implant if I'm using a non-invasive technology or even these sort of broader, more diffused technologies like the bipolar system it's express purpose should be to make that intrinsic me better. The things I truly have. In some places obviously with major spinal injury or advanced neurodegenerative disease or sensory deprivation and even there that's can be really controversial in the deaf community.

There are places where there's nothing... A replacement is needed and a permanent augmentation is called for but everywhere else, like our own memory, the starting point of a memory prosthetic should be that it actually isn't replacing our memory. It's making our memory better. It's not replacing our emotion regulation. It is improving our brain's own intrinsic ability to regulate itself. For example, improving the coupling between medial frontal activity and a medullar activity that defines so much, for example of emotion, regulation, fear, and even social skills really getting mediated through these sorts of broader level circuits.

I wanna build better people. And the here's another thing, my son has autism, he has diabetes, he's a cyborg. I got to turn my son into a cyborg and it's the kind of cyborg I want. Yes, I tried to develop a system.

I don't want diabetes, I want it to go away but I don't want the autism to go away. My son has a superpower. Everyone... If someone walks into a room and I'm looking at a sofa through a window here. It looks black to me but I know it's blue, but I'm an authority figure. I found a bunch of companies. I'm really important.

My employees start walking into the room and I'm saying that black sofa in there, I guarantee you they all agree with me that it's black. Here's a cool finding. They will actually perceive it as more black simply because I planted that idea in their head. Our brains are so wired into social signals we will change the way we see the world to fit in. My son won't.

People with autism are essentially resistant to those sorts of social cues. Now that can pose problems, but it's also a super power. They see the truth at least more vertically than we do. I didn't wanna build something that cured autism.

I wanted to build something that preserved what was special and unique about him but also allowed him to interact with the world. I wanted to build a translation system that allowed my son and everyone else to interact with a little less friction than would normally be there. Given the queue of all the secret language of faces that the rest of us get for free. That's my last call for neurotechnologies, build technologies that actually augment what make us us unique and different not technologies that just arbitrarily make us all the same. We're all be these smart well-regulated and largely industry, indistinguishable people.

And let me tell you, I am waiting especially now that all these kids are from home someone is going be the first kid to cheat on a test using neurotechnologies by boosting their working memory boosting their emotion regulation 'cause they wanna perform better under the stress of the test, whatever it might be, this is gonna happen like 2001, maybe even already happened in 2020. This is gonna happen if it hasn't already. If someone gets that advantage everyone has to have that same advantage or at least the right to choose for themselves. It needs to be like a vaccine something that is everyone's right, not just a civil right, a human right, which you can refuse but one which is available. 'Cause otherwise you said it we are staring at a future in which the definition of what it means to be human is going to change.

It will change. Not maybe, not that there's this cool story I read. The definition of what it means to be human is going to change. And the fundamental question is, for whom? It's fascinating.

And I mean, people may not realize but companies like Humm where we're making this little patch that you put on your forehead and it improves your working memory temporarily. We're confronting these issues today. These are not science fiction anymore.

These are not issues that we can just laugh at and say that's the kind of thing I'd see in a movie or something like that that these challenges that Vivienne's just briefly mentioned in the short time we have of encouraging her university and making sure this isn't the domain of billionaires and making sure that those billionaires and the big tech companies, they run don't break our brains in the process of moving fast or even just make our lives objectively and subjectively worse. And in the way that some technologies in the last 10 years debatably have. Vivienne, we're coming towards the end of our time.

I did have one little story just before we wrap up which is just that I'm not sure if I ever told you but I think I may have been one of the first people in the world to have ever had my brain hacked. (laughs) I, one day I was sitting in the lab we had just completed a new prototype hooked up to the software and I was there late playing a computer game on my office computer. Actually, one of the engineers logged into the system from home, didn't realize that I was connected to the device and started electrically stimulating me remotely and I didn't realize it. All of a sudden I got better at the computer game. (laughs) That's the funny thing is... I think a lot of people's vision again from science fiction and I love science fiction.

If you went in and I stimulated parts of your insular cortex using your existing rig without you realizing it and I made you... Increased your subjective preference for the thing you're looking at which happens to be my product, my movie, the the ad that I'm showing you. I hack it, I really subtly, whenever you see my ad you actually are like, "Yeah, that looks really great." When normally you would never feel that way yet that's actually your judgment. You do feel that way and you wouldn't even realize it was going on.

That sort of thing I think is legitimately scary and is a legitimate concern. We need to treat it as a real problem now, because again if we wait too long, we're going to end up with everything that we have come to love and loathe about the internet, except it will be us. And we need to collectively make those decisions right now not first mover entrepreneurs or crazy mad scientists all of us need to be making these decisions. Well, yeah, we've run out of time but just ended on that note. I think that your work is incredibly important and underappreciated in the public consciousness, how important it will be to continue thinking in the way that Vivienne does and the work that she does and supporting the companies that she does. But Vivienne I'm told there'll be lots of people would be really interested to follow you ongoing.

And if they heard some cool things for you today they'd like to find out more specifics, how could they get in touch and find out more about what you're doing? Absolutely. If you visit socos.org, SOCOS you'll find my philanthropic lab. People bring us incredibly challenging problems.

And if I think my team can make a difference I pay for everything. And if we come up with a solution we just give it away for free. So at socos.org, you can find some of the stories of the projects we worked on including as you may be able to see some of our work in neuroprosthetics. And some of the stories of hacking my own son's body, although, in my own defense, it is truly terrifying when your son has a life-threatening disease. This is let's call it hacking for good.

You're a super hero Vivienne, and I'm so grateful not just for your time today, but for your continuing help in this industry and in my company too. Thank you everyone for joining us. If anyone's interested in learning a little bit about Humm, the company that I run just go to thinkhumm.com

or anywhere on social media, I think. It's been fantastic to talk to you, Vivienne and I'm looking forward to doing it again at Essex next year, will talk to you soon. We'll make it happen. It was lovely meeting everyone from afar. Look forward to all being back in person soon. Thanks everyone.

2021-05-12

Show video