How the Brain Works, Curing Blindness & How to Navigate a Career Path | Dr. E.J. Chichilnisky

1. 0:00 – 2:31

   Dr. E.J. Chichilnisky

   1. AHAndrew Huberman0:00

      (uptempo music) Welcome to the Huberman Lab Podcast, where we discuss science and science-based tools for everyday life. I'm Andrew Huberman, and I'm a professor of neurobiology and ophthalmology at Stanford School of Medicine. My guest today is Dr. E.J. Chichilnisky. Dr. E.J. Chichilnisky is a professor of neurosurgery, ophthalmology, and neuroscience at Stanford University. He is one of the world's leading researchers trying to understand how we see the world around us, that is, how visual perception occurs, and then applying that information directly to the design of neural prostheses, literally robotic eyes that can allow blind people to see once again. Today's discussion is a very important one for anyone who wants to understand how their brain works. Indeed, E.J. spells out, in very clear terms, exactly how the world around us is encoded by the neurons, the nerve cells within our brain, in order to create these elaborate visual images that we essentially see within our minds. And with that understanding, he explains how that can be applied to engineer specific robotic, AI, and machine learning devices that can allow human brains not only to see once again, in the blind, but also to perceive things that typical human brains can't, and indeed, for memory to be enhanced and for cognition to be enhanced. This is the direction that neuroscience is going. And in the course of today's discussion, we have the opportunity to learn from the world expert in these topics, where the science is now, and where it is headed. During today's discussion, we also get heavily into the topic of how to select one's professional and personal path. And indeed, you'll learn from Dr. Chichilnisky that he has a somewhat unusual path, both into science and through science. So for those of you that believe that everyone that's highly accomplished in their career always knew exactly what they wanted to do at every stage, you will soon learn that that is absolutely not the case with E.J. He describes wandering through three different graduate programs, taking several years off from school in order to dance. Yes, you heard that correctly, to dance, and how that wandering, and indeed dancing, helped him decide exactly what he wanted to do with his professional life, and exactly what specific problems to try and tackle in the realm of neuroscience and medicine. It's a discussion that I'm certain that everybody, scientist or no, young or old, can benefit from, and can apply the specific tools that E.J. describes in their own life and pursuits.

2. 2:31 – 6:06

   Sponsors: Eight Sleep, ROKA & BetterHelp

   1. AHAndrew Huberman2:31

      Before we begin, I'd like to emphasize that this podcast is separate from my teaching and research roles at Stanford. It is, however, part of my desire and effort to bring zero-cost-to-consumer information about science and science-related tools to the general public. In keeping with that theme, I'd like to thank the sponsors of today's podcast. Our first sponsor is Eight Sleep. Eight Sleep makes smart mattress covers with cooling, heating, and sleep tracking capacity. Now, I've spoken many times before on this and other podcasts about the fact that sleep is the foundation of mental health, physical health, and performance. And one of the key aspects to getting a great night's sleep is to control the temperature of your sleeping environment. And that's because in order to fall and stay deeply asleep, your body temperature actually has to drop by about one to three degrees. And in order to wake up in the morning feeling refreshed, your body temperature actually has to increase by about one to three degrees. Eight Sleep makes it extremely easy to control the temperature of your sleeping environment at the beginning, middle, and throughout the night, and when you wake up in the morning. I've been sleeping on an Eight Sleep mattress cover for nearly three years now, and it has dramatically improved my sleep. If you'd like to try Eight Sleep, you can go to EightSleep.com/huberman to save $150 off their Pod 3 cover. Eight Sleep currently ships to the USA, Canada, UK, select countries in the EU, and Australia. Again, that's EightSleep.com/huberman. Today's episode is also brought to us by ROKA. ROKA makes eyeglasses and sunglasses that are of the absolute highest quality. I've spent a lifetime working on the biology of the visual system, and I can tell you that your visual system has to contend with an enormous number of challenges in order for you to be able to see clearly under different conditions. ROKA understands this, and designed all of their eyeglasses and sunglasses with the biology of the visual system in mind. Now, ROKA eyeglasses and sunglasses were initially developed for use in sport, and as a consequence, you can wear them without them slipping off your face while running or cycling, and they're extremely lightweight. ROKA eyeglasses and sunglasses are also designed with a new technology called Float Fit, which I really like because it makes their eyeglasses and sunglasses fit perfectly, and they don't move around even when I'm active. So if I'm running and I'm wearing my glasses, they stay on my face. Most of the time I don't even remember they're on my face because they're so lightweight. You can also use them while cycling or for other activities. So if you'd like to try ROKA glasses, go to ROKA, that's ROKA.com and enter the code Huberman to save 20% off your first order. Again, that's ROKA.com and enter the code Huberman at checkout. Today's episode is also brought to us by BetterHelp. BetterHelp offers professional therapy with a licensed therapist carried out online. Now, I've been going to therapy for well over 30 years. Initially, I didn't have a choice, it was a condition of being allowed to stay in school, but pretty soon I realized that therapy is extremely valuable. In fact, I consider doing regular therapy just as important as getting regular exercise, including cardiovascular exercise and resistance training, which of course I also do every week. The reason I know therapy is so valuable is that if you can find a therapist with whom you can develop a really good rapport, you not only get terrific support for some of the challenges in your life, but you also can derive tremendous insights from that therapy, insights that can allow you to better not just your emotional life and your relationship life, but of course also the relationship to yourself and to your professional life, to all sorts of career goals. In fact, I see therapy as one of the key components for meshing together all aspects of one's life and being able to really direct one's focus and attention toward what really matters. If you'd like to try BetterHelp, go to BetterHelp.com/huberman to get 10% off your first month. Again, that's BetterHelp.com/huberman. And now for my discussion with Dr. E.J. Chichilnisky.Dr.

3. 6:06 – 11:23

   Vision & Brain; Retina

   1. AHAndrew Huberman6:07

      E.J. Chichilnisky, welcome.

   2. ECDr. E.J. Chichilnisky6:09

      Good to see you.

   3. AHAndrew Huberman6:11

      For the audience, we are friends. We go way back. E.J. has been a few years or more ahead of me in the science game, and the best way to describe you and your work, E.J., is you're an astronaut. You go places no one else has been willing to go before. You develop new technologies in order to do that, all with the bold mission of trying to understand how the nervous system, which, of course, includes the brain, works, and how to make it better with engineering. So, today, we are going to get into all of that, but just to start off and get everybody on the same page, maybe we could just take a moment and talk about the brain and nervous system, and, you know, what it consists of that allows it to do all the sorts of things that we're going to get into, like see things in our environment and respond to those things in our environment. So, at risk of throwing too much at you right out the gate, what's your one to five-minute version of how the brain works?

   4. ECDr. E.J. Chichilnisky7:19

      Oh, I don't have a one to five-minute version of how the brain works, but I can tell you how I think, uh, vision is initiated in the brain, and, um, you and I go back a long way, so we have a lot of common understanding about this. But I'll narrate it from scratch if, if that makes sense. Um, so, vision is initiated in the retina, uh, of the eye, which is a sheet of neural tissue at the rear of the eye that captures the light that is incident on the eye, that comes in through the, through the eye, transforms that light into electrical signals, processes those electrical signals in interesting ways and changes them up, and then sends that visual information to the brain, where it is used to bring about our sense of vision. And, uh, you s- you asked me about the one to five-minute version of how the brain works. I don't know, but I do know that the brain receives all these patterns of electrical activity coming out of these nerve cells in the retina and somehow assembles that into our visual experience, whether that be responding to things coming at us, or our circadian rhythms that , that govern our sleep and behavior, or identifying objects for prey or avoiding predators, or appreciating beauty. And what we know is that the brain receives a fantastically complex set of signals from the retina and puts that all together into our visual experience, and we are very visual creatures, o- obviously. So, I think that's a big part of how the brain works, because so much of what we do revolves around vision, revolves around how the brain puts together these signals coming out of the retina, and I would love to understand how that works. At the moment, I don't. Uh, and what we are trying to do is get a really complete understanding of how that begins in the retina, and then how we can restore it in those who have lost sight.

   5. AHAndrew Huberman9:19

      Why focus on this issue in the retina, this thin set of layers of neurons that line the back of the eye? What-- why, why explore vision there? I mean, obviously, there are centers within the brain that, of course, contain neurons, nerve cells that are involved in vision. If one wants to understand visual perception, and I agree, by the way, that visual perception is one of the most dominant forces in the quality and experience of our life. Why focus on the retina? Why not focus on the visual cortex or the visual thalamus? I mean, what's so special about the retina?

   6. ECDr. E.J. Chichilnisky9:56

      Well, we have to focus on all of it, because understanding the retina won't give us a full understanding of how all this works, obviously, and if you don't have your visual cortex and visual thalamus, you won't see. But if you don't have your retina, you also won't see. You won't even have a chance to see. So, I focus on the retina, because, um, I enjoy the possibility that we can really understand a piece of the nervous system in my lifetime, in our lifetimes. We can understand it so well that we can build it, replace it, restore its function. That's farther off in the s- in the central regions of the brain. That's gonna be quite a bit harder. Um, I find satisfaction in really understanding something so well that I can write down in a mathematical formula what it's doing, that I can test my hypotheses up and down, and yes, we really get how this little machine works, and that I can engineer devices to replace the function of that circuit when it's lost. That, to me, is just deeply satisfying. But there also has, is a really fundamental role for people who want to go and do more exploratory work in the visual brain, as you mentioned, in the visual cortex, in the thalamus, and other places, 'cause ultimately, those retinal signals won't lead to anything if those areas aren't putting it all together to govern our perception and our, ultimately, our behavior.

4. 11:23 – 18:37

   Retina & Visual Processing

   1. ECDr. E.J. Chichilnisky11:23

   2. AHAndrew Huberman11:23

      So, let's talk about the retina in its full beauty and detail. Three layers of cells that line the back of the eye like a pie crust somehow take light that comes into the eye. The lens focuses that light. If it doesn't do that well, we put lenses in front of our eyes, such as contact lenses or spectacles, and somehow takes that light and transforms it, as you said, into neural signals, and processes that within the retina. So, let's take a deep dive into the retina and do so with the understanding, at least my understanding, is that, in part thanks to your work and the work of others, this is perhaps the best-understood piece of the brain.

   3. ECDr. E.J. Chichilnisky12:06

      Yes. I think it's a solid argument that it's the best-understood piece of the brain, and, uh, we'll turn back to that in a minute. So-Um, the retina begins with a sheet of cells called the photoreceptor cells that are highly specialized. These are cells that essentially don't exist anywhere else in the brain, and what they do is transform light energy into electrical signals in neurons. Very specialized, very, uh, demanding cells. They require a lot of maintenance, and they die relatively easily, which is what gives rise to se- some of the forms of blindness. Those are the... You might call them pixel detectors. They're tiny cells called photoreceptors that each one captures light from a particular location in the world. That sheet of cells has done that initial transduction process, where light is converted into neural signals that the brain can then begin to work with. The second layer is responsible for processing, adjusting, changing, mixing and matching, uh, comparing signals in different neurons, many complex operations that we're still trying to understand, and consists of dozens of distinct cell types that extract features, if you will, of the visual world from the elementary pixels represented in the photoreceptor cells. So, that second layer is receiving the input from that sheet of photoreceptors, and picking stuff out of it. The third layer of cells is the so-called retinal ganglion cells. That's the only, uh, term that I'd like to... And it probably will come up repeatedly in this conversation, uh, so for your, for your viewers and listeners, um, these retinal ganglion cells are the ones who are responsible for taking the signals that are there in the retina, and sending them to the brain so that the process of vision can begin. They are the, the messengers, if you will, from the retina to the brain. The retinal ganglion cells, and there are about 20 different types in humans, um, are, again, feature extractors. They pick out different bits and pieces of the visual scene, and send interesting stuff to the brain, trying to leave out the uninteresting stuff, and the 20 or so cell types all pick out different types of information from the visual scene. You can eli- uh, sort of think of them as Photoshop filters, each cell type in the retina. Um, again, about 20 different ganglion cell types. Each type represents the full scene, the entire visual world, but picks out different features.

   4. AHAndrew Huberman14:32

      Such as?

   5. ECDr. E.J. Chichilnisky14:33

      Some cells pick out spatial detail, tiny little points of light, almost. Some cells pick out and signal information about things that are moving in the visual world. Some cells pick out information that's been captured about different wavelengths from the photoreceptor cells, and there- thereby giving us our sensations of color, and probably more things in those 20 different ganglion cell types that we don't fully understand. The result then is that the retina has this sort of a representation of the visual world, but it has 20 different representations, not one. It's not one picture that comes out of the retina and gets sent to the brain. No, no, no. It's 20 different pictures, and you can think of maybe as 20 different Photoshopped pictures, but one of them has the edges highlighted. One of them has the colors highlighted. One of them has movement, uh, encoded in it, and these, somehow, these filters send the information to many different targets in the brain, and then our brain puts it all together, and then we have a cohesive sense of the visual world, which is the remarkable feature that we really don't understand.

   6. AHAndrew Huberman15:35

      Amazing. Is it fair for those that don't work with Photoshop to think about these, um, different Photoshop filters perhaps as, like, different movies of the visual world? One movie contains the outlines of objects and people and things. Another movie is showing the motion of blobs in the environment, meaning whatever's moving in your environment is kind of just represented as blobs. Another movie is just the color in the environment. Another mo- And then all of those what I'm calling movies are sent into the brain, and then the brain somehow combines those in ways that allow us to see each other, and see cars and objects, and recognize faces. Is that, is that one way to think about it?

   7. ECDr. E.J. Chichilnisky16:16

      That is, that's exactly how I think about it. Maybe it's a better way to say it, yeah.

   8. AHAndrew Huberman16:19

      No, I, I, I like the Photoshop filter, uh, analogy. I just, for those that don't work with Photoshop, um, you know, uh, I just think that the movie analogy might, might be an, a, a decent alternative.

   9. ECDr. E.J. Chichilnisky16:30

      How the retina works is an example, we think, of how all sensory systems work. There's an initial representation in a specialized cell type that is r- it is, that is responsible for and capable of extracting physical features from the world, and then neural circuits in the brain use that information in different ways to grab stuff out of the visual world, and the auditory system, there's, uh, the sound world is represented also in specialized cells that capture sound energy and transduce that into neural signals, and then subsequent pa- uh, stages of processing in the auditory system pick out different features of our auditory world.

   10. AHAndrew Huberman17:12

       Like the frequency, how high or how low a tone is.

   11. ECDr. E.J. Chichilnisky17:15

       Right.

   12. AHAndrew Huberman17:15

       The direction it's coming from.

   13. ECDr. E.J. Chichilnisky17:17

       Right, the movement of it, uh, how loud it is. Different features are extracted. So, the, we think the visual system is just an example of how the external world is represented in our brain, and, of course, in some sense, in, uh, a philosophical approach to the brain is really saying, "Well, there's the sensory world, and then there's the actions we take, and there's almost nothing else that we really know other than those two things, how the sensory world comes in, and then finally it results in our action." That's what our brain is about. Because vision is so important for people, I find it absolutely compelling and fascinating. I mean, as an example, as you know well, many people study rodents, uh, to understand how different aspects of the brain work, and, um, you know, rodents are interesting animals, and do all sorts of really cool things, but they interact with the world differently than we do. They, in, in a lot of ways, they sense by smelling. They, they identify objects by smelling, and they navigate with their whiskers in- to a large extent. We don't do any of that. You don't navigate with your whiskers. At least I don't think you do.You don't recognize me when I walk in the room by my smell. No, you use vision for all that, and we humans use vision, so it's a really fundamental aspect of who we are as biological creatures.

5. 18:37 – 23:01

   Vision in Humans & Other Animals, Color

   1. ECDr. E.J. Chichilnisky18:37

   2. AHAndrew Huberman18:37

      I wonder if, just for sake of entertainment, uh, we could think about how the human retina, and therefore vision in our species, differs a little bit from some extreme examples of vision in other species. Not to make this a comparative, um, or zoological, um, exploration, but just to really illustrate the fact that the specific cell types within our retinas create a visual representation of the outside world that can be, and often is, very different from that of other species. Um, (smacks lips) for instance, or at least my understanding, is that the mantis shrimp sees, I don't know, 60 to 100 different variations of each color that we are essentially blind to, because their photoreceptors can detect very subtle differences in red. For instance, long wavelengths of light, what most people refer to as red. Um, pit vipers can sense heat emissions, essentially with their eyes, but also other organs, um, and on and on. You know, it's, I raise this because I think the, the human neural retina is such an incredible example of extracting features from the visual world that then we recreate, but I think it's also worth reminding everyone, and ourselves, that it's not a complete representation of what's out there. Like, there's a lot in light that we don't see, because our neural retina just can't turn it into electrical signals.

   3. ECDr. E.J. Chichilnisky20:11

      Right.

   4. AHAndrew Huberman20:11

      Do you wanna give some examples of what we can't see? And if any, um, particular examples from the animal kingdom delight you, feel free to sh- throw those out.

   5. ECDr. E.J. Chichilnisky20:19

      Well, one thing, uh, you mentioned color. Um, we experience a rich sensation of color when we look at the world, and say, "Wow, I see all these colors." That's immediate, and that's just how we talk about it. But in fact, we have very little information about color. Color is a very high-dimensional, complex thing, or wavelength, I should say. Wavelength information really is about how much energy there is in the light around us at different wavelengths. We only have three sort of snapshots of that in our retinas, with the three different types of photoreceptor cells that are sensitive to different wavelength, different bands of the wavelength spectrum. Three is not a lot. As you just said, other creatures have many more ways of capturing wavelength information, and one way you can verify for yourself that we just have three is to realize that if you look at your TV, there are only three primaries on your TV. There's a red, there's a green, there's a blue. That's it. And from those three primaries, the entire richness of the experience on your TV set is, uh, composed. So, with just those three things, you basically are able to create any human visual sensation. Well, a mantis shrimp would be like, "That's nothing. There's so much more stuff out there that's not represented on this TV," if you could speak to the mantis shrimp. Another thing we, we maybe don't see, another example of a difference in the animal kingdom is... So, um, again, taking rodents as an example, one of the thing, one of the things that rodents have to do is to not be hunted by birds that are coming down toward them. And so, it, it appears that there are cells in the retina that are, seem to be quite sensitive to what, to, uh, looming, to something dark that's getting bigger, uh, like a shadow coming from a bird coming down at you. We don't know for sure that this is exactly what causes animals to avoid being eaten by birds, but there's, there's interesting evidence in that direction. That's not really a big thing for us as, for humans, as far as I know. We're not typically hunted by huge birds, so that's not a thing we need, and I think that's, that's where it comes back to, where you were headed, if I understand right, um, which is, we occupy different biological niches, we and the mantis shrimp and the rodents, and our visual systems reflect that. We have different stuff that we're looking for in our visual environment than other creatures are, and so our eyes are different. And that's one of the reasons that we emphasize work on the human retina, um, as opposed to other, certain other animal species that would be less clearly relevant to the visual experience you and I have.

6. 23:01 – 29:48

   Studying the Human Retina

   1. AHAndrew Huberman23:01

      So, let's talk about these incredible experiments that your laboratory has been doing for several decades now. I've had the, uh, privilege of sitting in on some of these experiments, and they are very involved, um-

   2. ECDr. E.J. Chichilnisky23:14

      (laughs)

   3. AHAndrew Huberman23:15

      ... as, to say the least. If you could just walk us through one of these experiments, I think the audience would appreciate understanding what goes into, quote-unquote, "trying to understand" what's going on in the electrical activity of these specific retinal cell types, the retinal ganglion cells in particular. Um, you know, what does this look like? You know, you're in your laboratory at Stanford, and you get a phone call. Someone says, "I got a retina." What happens next?

   4. ECDr. E.J. Chichilnisky23:43

      We scramble like crazy. We drop everything we're doing, cancel all our appointments, and get ourselves ready for 48 hours of non-stop work down in the lab, getting as much data as we possibly can from the retina. The most exciting exa- example of what you just said is when we get a human retina. When, for example, there's a, a donor who has died, and the retina is available for research, we jump at that opportunity.

   5. AHAndrew Huberman24:06

      How soon after, uh, the person is deceased do you need to get the eye globe, the eyeball, in order to get the retina in a condition that would allow you to record electrical signals from it?

   6. ECDr. E.J. Chichilnisky24:17

      A few minutes. Just a f-

   7. AHAndrew Huberman24:19

      So, you're waiting-

   8. ECDr. E.J. Chichilnisky24:19

      Just a few minutes.

   9. AHAndrew Huberman24:19

      ... in the hospital?

   10. ECDr. E.J. Chichilnisky24:20

       Um, typic- the way we typically get these eyes is from braindead individuals.... so people who are legally and medically dead, but their hearts are still pumping, and therefore, their retinas are still alive and functioning. When, when those individuals, uh, are used, their-- the bodies of those individuals are used for organ donations, we can benefit from that organ do- donation setup that organ distribution centers do to save many people's lives and also to promote research. So, we sometimes get those retinas, and that begins the experiment for us.

   11. AHAndrew Huberman24:56

       Um, I'm gonna ask for a few more details here just to put the picture in peoples' minds, and, uh, not to be gruesome. I just really want people to understand what's involved here.

   12. ECDr. E.J. Chichilnisky25:03

       Yes.

   13. AHAndrew Huberman25:03

       So, you'll get a call. "We've got a, a patient who is soon to be deceased. Um, they've consented to giving their eye globes, their eyeballs-

   14. ECDr. E.J. Chichilnisky25:11

       Yep.

   15. AHAndrew Huberman25:11

       ... for research so that you can study the human retina."

   16. ECDr. E.J. Chichilnisky25:14

       Yeah.

   17. AHAndrew Huberman25:14

       Um, is it you who goes over and takes the eyes out? Does somebody do that or hand them to you, uh, in a bucket of ice? I'm sorry if I'm get- making people queasy at all, but th- this is, folks, how, uh, one goes about trying to understand how the human brain works.

   18. ECDr. E.J. Chichilnisky25:29

       Absolutely, and this is also how you go about donating your heart so that you can save somebody else's life who needs a heart transplant. The same incredible organizations that do the harvesting of the tissue for us, their primary goal is to do that for organ donations to save lives. They save lives every day. These people are incredible. Donor Network West is one of those organizations, the one that we work with. Um, they're really amazing. Um, so their technicians or a retinal surgeon will take the eye out, give it to myself or somebody from my lab, who will bring it back to the lab, and we have a way to keep the eye alive and functioning, just the eye by itself.

   19. AHAndrew Huberman26:05

       Is this always at Stanford, or do you sometimes travel elsewhere?

   20. ECDr. E.J. Chichilnisky26:07

       Local hospitals up to an hour away, so-

   21. AHAndrew Huberman26:09

       Oh, so then you drive them back.

   22. ECDr. E.J. Chichilnisky26:10

       We drive them back.

   23. AHAndrew Huberman26:12

       Okay.

   24. ECDr. E.J. Chichilnisky26:12

       It's the retina express-

   25. AHAndrew Huberman26:13

       (laughs)

   26. ECDr. E.J. Chichilnisky26:14

       ... and when, when we're bringing back the retina express, it's, uh, again, it's all hands on deck in our lab. We are scrambling, setting up all of our equipment, getting everything ready. You've been at these experiments. They're intense, and they really are, really are 48-hour marathons of incredible activity by really dedicated individuals. So, (smacks lips) um, we, we might get those eyes sometimes, again, at 2:00 in the morning. That's common. And from that 2:00 in the morning time, uh, begins the experiment. So, we bring the eyes back. We open them up, and we, we have access to the rear of the eye, which is what the-- where the retina is. It's a thin sheet of neural tissue at the back part of the eye. We hemisect the eye, cut it in half, so that we can see the back. It's like half of a globe, if you will, and then we put in relaxing cuts and lay it out flat so we can see what we're working with. Then, we take little segments of retina out in the subsequent 48 hours, cut them out, maybe a three-by-three millimeter piece of the retina, little chunk of retinal tissue, and bring it into an electrophysiology recording and stimulation apparatus that allow- allows us to interact with it, and we do two types of experiments with that. So, this electrophysiological recording and stimulation apparatus is very custom-built by our physics collaborators, who have developed high-end equipment. It allows us to record and stimulate through 512 channels simultaneously at very high density. This is pretty high-end stuff in terms of, uh, technology for interrogating and manipulating the electrical signals in the retina. That's what we specialize in in my lab.

   27. AHAndrew Huberman27:46

       Could I just ask a question about this device?

   28. ECDr. E.J. Chichilnisky27:48

       Yeah.

   29. AHAndrew Huberman27:49

       Um, I've seen it before. Uh, it's very small. As you mentioned, you're recording from a few millimeters square of the, of the retina, um, from this recently deceased patient. Um, it looks a little bit like a bed of nails, right? Like tiny little microwires all arranged very closely to one another. You got the retina laying down on top of it-

   30. ECDr. E.J. Chichilnisky28:09

       That's right.
7. 29:48 – 31:16

   Sponsor: AG1

   1. AHAndrew Huberman29:49

      I'd like to take a brief moment and thank one of our sponsors, and that's AG1. AG1 is a vitamin mineral probiotic drink that also contains adaptogens. I started taking AG1 way back in 2012. The reason I started taking it, and the reason I still take it every day, is that it ensures that I meet all of my quotas for vitamins and minerals, and it ensures that I get enough prebiotic and probiotic to support gut health. Now, gut health is something that, over the last 10 years, we realized is not just important for the health of our gut, but also for our immune system, and for the production of neurotransmitters and neuromodulators, things like dopamine and serotonin. In other words, gut health is critical for proper brain functioning. Now, of course, I strive to consume healthy whole foods for the majority of my nutritional intake every single day, but there are a number of things in AG1, including specific micronutrients, that are hard to get from whole foods, or at least in sufficient quantities. So, AG1 allows me to get the vitamins and minerals that I need, probiotics, prebiotics, the adaptogens, and critical micronutrients. So, anytime somebody asks me, if they were to take just one supplement, what that supplement should be, I tell them, "AG1," because AG1 supports so many different systems within the body that are involved in mental health, physical health, and performance. To try AG1, go to drinkag1.com/huberman, and you'll get a year's supply of vitamin D3K2, and five free travel packs of AG1. Again, that's drinkag1.com/huberman.

8. 31:16 – 36:00

   Cell Types

   1. AHAndrew Huberman31:16

      Let's take this moment to talk a little bit about cell types. So, um, you mentioned there are about 20 different types of these retinal ganglion cells, what we may refer to in brief as RGCs. So, retinal ganglion cells, RGCs, same thing, and as you mentioned, these cover the entire retina, so that if each cell type is extracting a different set of features from the visual world, motion, color, specific colors, et cetera, that essentially no location in the world around us fails to be represented by these cells. Put differently, these cells are looking everywhere. (laughs) Um-

   2. ECDr. E.J. Chichilnisky31:57

      Each cell type is looking everywhere.

   3. AHAndrew Huberman31:58

      Each cell type is looking everywhere. Um, so that if movement occurs in any region of our visual world, we are in a position to detect it. Um, but maybe we could talk a little bit about cell type. Cell types is such an important theme in the field of neuroscience, and indeed, in all of biology, but it's actually not something we t- have talked about very much on this podcast before, either in solo episodes or in, uh, guest episodes. Um, I don't have any specific reason for that. We've talked about brain areas, prefrontal cortex, basal ganglia, anterior insular cortex, and on and on. We've talked about neural circuits. But we've never really talked about cell types. So, the ganglion cells, as we-

   4. ECDr. E.J. Chichilnisky32:35

      Brother, you let me down.

   5. AHAndrew Huberman32:36

      Huh?

   6. ECDr. E.J. Chichilnisky32:36

      No talking about cell types.

   7. AHAndrew Huberman32:37

      Well, but that's why you're here.

   8. ECDr. E.J. Chichilnisky32:38

      That's why I'm here.

   9. AHAndrew Huberman32:39

      That's why you're here.

   10. ECDr. E.J. Chichilnisky32:40

       (laughs)

   11. AHAndrew Huberman32:40

       Um, t- tell us about cell types. How do you figure out if you have a cell type? How do you know if it's a cell type, or, you know, is it the shape? Is it how it responds? Um, how do you know if you have a cell type? What, what, what, what's this about? And I want to just, um, put in the back of this question, um, or rather in the back of people's minds, that, um, this issue of cell types is not just an issue pertinent to the retina. This is an issue that is critical to understanding how the brain works.

   12. ECDr. E.J. Chichilnisky33:09

       Absolutely.

   13. AHAndrew Huberman33:09

       It's critical to understanding consciousness. I know a lot of people are like, "What is consciousness," right? We're not going there just yet.

   14. ECDr. E.J. Chichilnisky33:14

       (laughs)

   15. AHAndrew Huberman33:14

       But, uh, what are cell types? How do you determine if you have a cell type, and why is this so important to understanding how the brain works?

   16. ECDr. E.J. Chichilnisky33:22

       Yeah, I mean, as, as you said, as, as far as we understand, every single brain circuit is full of very distinct cell types. Those cell types are distinguished by their genetic expression, their shapes and sizes, which other cells they do contact and which cells they don't contact, where they send their information to in other parts of the brain, and what they represent. And as far as we know, this is true throughout the brain, and it's true in the retina. The different ganglion cell types, retinal ganglion cell types, about 20 of them, each of which is looking at the whole visual scene, extracts different stuff. This cell type one extracts one thing, cell type two extracts something else, but they all represent the entire visual scene. But those cell types, we know from lots of beautiful work, work that you're closely connected to and some of which you've done, um, those cell types have different morphology, different shapes and sizes, different patterns of gene expression, different targets in the brain. They send their outputs to different places in the brain. So, really, w- to study the retina without understanding cell types, you're kind of lost right away. You have to know what's going on with the cell types. Otherwise, you can't make sense of this retinal signal. The way, we, we identify them in two ways, and they're different, for different purposes. The, the basic way we identify the different cell types is their function, because we study their function. We study how they respond to light images, and we can clearly separate them out, and in fact, it's, it's a simple thing to say, but it's (laughs) really true, our 512-electrode array technology, which you've seen in our lab, and stuff that we developed with collaborators, um, about 20 years ago, um, was crucial for this. Because with that 512-electrode technology, we could see many cells of each type, and we could clearly parse them apart from one another, whereas previous studies working on one cell at a time had great difficulty doing that. So, with our technology, with 512 electrodes, we record hundreds of cells simultaneously. We can say, "Oh, there's 20 of these. There's 50 of those. There's 26 of those, and here they are." And we can just set them in different bins and say, "Okay, this is what's present in this retina," just what the information is they're extracting. There's another purpose, again referring forward to the neuroengineering aspect. We need to identify the cell types not just based on what visual information they carry, but based on their electrical features, properties, electrical properties of the cell. Cells, as you know, neurons are electrical cells. They fundamentally tr- receive and transmit electrical information, and the way that they do that has a distinctive electrical signature. That turns out to be super important for developing devices to restore vision.

9. 36:00 – 43:39

   Determining Cell Function in Retina

   1. AHAndrew Huberman36:00

      Could you explain how you determine what a given cell type does, its electrical properties? Um, let's just draw a mental image for people. Uh, the retina's taken out of this deceased individual, put down on this bed of nails, of electrodes. Those electrodes can detect electrical signals within the ganglion cells. You are able to shine light onto the retina, and see how the retinal ganglion cells respond, meaning what electrical signals they would transmit to the brain, if they were still connected to a brain. They're not connected to a brain in the experiment. They're sitting there, and they're s-

   2. ECDr. E.J. Chichilnisky36:34

      But they're trying.

   3. AHAndrew Huberman36:34

      But they're trying. I could imagine playing those cells a movie of, I don't know, a checkerboard going where every, um, square on the checkerboard goes from white to black to gray, could do that. I could, um, play a cartoon. I could, um, show it, uh, this year's Academy Award winner for Best Picture. But how do you decide what to show the retina? This is a human being's retina, after all. Uh, presumably, it looked at things that are relevant to human beings until that person died. But how do you determine cell type electrical signals if you don't know what specific things to show it? I mean, are you gonna show it, I don't know, Disney movies? Like, what, what, what do you show it?

   4. ECDr. E.J. Chichilnisky37:19

      (laughs)

   5. AHAndrew Huberman37:19

      (laughs)

   6. ECDr. E.J. Chichilnisky37:21

      So, what we show now reflects the fact that we've built up a lot of information, and, and our work stands on the shoulders of many scientists who have studied the retina for decades to figure out what different cell types respond to. And, um, we know that certain cell types respond primarily to increments of light, when light gets brighter than it was, so a change of, from a certain brightness to a higher brightness, this particular cell type fires. Another cell type fires, or sends spikes to the brain, when it gets darker. Some cell types, uh, respond primarily to large targets in the visual world. Other cell types respond better to small targets in the visual world. Some cell types respond to different wavelengths of light that we can identify. There exist certain cell types that are still poorly understood that respond to movement. So, we can tailor visual stimuli to types that we kind of already know about because of much preceding research. That's not actually how we do it in our experiments, for the most part. Instead, we use a very unbiased flickering checkerboard pattern, as it turns out, which is a really efficient, unbiased way to sample many cells simultaneously, so that in a half hour of electrical recording from a retina, we can figure out what all the 512 or so cells are that we're recording, and know all of their types. And the way we do that is to play essentially random garbage TV snow-type image to the retina for a period of time, and determine which bits of brightening or darkening or moving or whatever in that random garbage activated this particular cell, by looking acr- at average across the half-hour recording, and saying, "Oh, it looks like this cell was always firing when it became bright in this region of the screen. That must be an on cell sensitive to light in this region of the screen," and so on. So, we have sophisticated, efficient ways of doing it, but it all comes back to these basic things about what features in the visual world tend to cause a given cell to send a signal to the brain.

   7. AHAndrew Huberman39:28

      Yeah, that makes a lot of sense. So, you take essentially what you called random garbage, um, snow, w- white, black, and gray pixels on a screen (static sound) .

   8. ECDr. E.J. Chichilnisky39:39

      Yep.

   9. AHAndrew Huberman39:39

      The retina views that, and then the cells in the retina will respond every once in a while with an electrical potential. They'll fire, as we say. Spike, it's sometimes called. And then, you take sort of a forensic approach a bit later. You look back in time, and you say, "You know, what was the arrangement of pixels in this (static sound) random garbage right before this cell fired an electrical potential?"

   10. ECDr. E.J. Chichilnisky40:08

       That's right.

   11. AHAndrew Huberman40:08

       "A spike?" And then, from that, you can reconstruct the preferred stimulus.

   12. ECDr. E.J. Chichilnisky40:14

       Yeah.

   13. AHAndrew Huberman40:15

       Right? The... You can say, "Oh, well, this cell, and cells around it, seem to like motion of things going in a particular direction," for instance. And, um, how do you know that the cell doesn't also like a bunch of other stuff that you didn't pick up on using this (static sound) random garbage?

   14. ECDr. E.J. Chichilnisky40:37

       (laughs) Yeah. Two things. First, let me, let me just say f- for the record, we don't record from these cells that signal motion in particular directions. They are an elusive cell type that is best understood in rodents and other creatures, and not well-understood in the primate, as you know, although some people are discovering potential cells of that type now, and have, have recently discovered them. But-

   15. AHAndrew Huberman40:56

       Okay, so let's say cells that respond to-

   16. ECDr. E.J. Chichilnisky40:58

       Small light.

   17. AHAndrew Huberman40:59

       ... like, like spots that are red, um, y- you know, that go from dim red to bright red.

   18. ECDr. E.J. Chichilnisky41:06

       Right.

   19. AHAndrew Huberman41:06

       Yeah.

   20. ECDr. E.J. Chichilnisky41:07

       So, we can go through that colored TV snow, and pick out the cells that responded to a transition of the kind you described, from darker to lighter, or from greener to redder, or something like that. Cells tend to respond to transitions in the visual scene rather than static imagery.

   21. AHAndrew Huberman41:25

       Mm-hmm.

   22. ECDr. E.J. Chichilnisky41:26

       Um, and so we can pick that stuff out. But you asked the question, "Well, gee, is the TV snow gonna capture everything about what these cells are doing?" That's a really important question that I wanna just mention more. Um, quite likely not. That's a scientific instrument. It's an unbiased way to sample a whole bunch of cells in a first cut look at, you know, generally speaking, what are they up to? But that doesn't mean we've really captured their role in natural visual perception, 'cause actually, you don't go through the world perceiving visual snow. You go through the world (laughs) perceiving objects, and meals, and mates, and targets, and all these things, right? So, the study of how the retina responds to more naturalistic visual stimuli...... uh, in my lab and in many la- other labs around the world, is really get- getting off the ground now, and I would say we have limited understanding, um... I would say... We know that our simple laboratory experiments with the TV snow don't capture the whole story. There's more. There are about 20 different cell types in the retina. We have basic characterizations of seven of them, if you count a certain way, um... We know that there are another 15 or so lurking right behind the curtain that we've started to sample. We don't know what naturalistic targets they respond to in the visual life of the animal. That's work that's underway, exciting, interesting work 'cause this... We know that the retina, they, they gotta be there for something. One, one way to think of it, I'm, I'm pretty sure you think of it this way too, is that the retina is a highly evolved organ with a lot of evolutionary pressure for it to be efficient, to have a small optic nerve s- sending to the brain. It's probably the case that there's no accidental stuff sitting around in the retina that's vestigial and sending information to the brain. It's probably the case that those, those signals are all, are all doing something important for our visual behavior, for our wellbeing, for our sleep, all sorts of stuff, and I think... And the field is still trying to figure that out. These, these are the big mysteries, I think, that, uh, in terms of the retina, what are those signals exactly in all those different cell types? What different behaviors and aspects of our life do they control?

10. 43:39 – 49:27

    Retinal Cell Types & Stimuli

    1. AHAndrew Huberman43:39

       What is the wildest cell type you've ever encountered?

    2. ECDr. E.J. Chichilnisky43:43

       (laughs)

    3. AHAndrew Huberman43:45

       Like, what did it do? Like, uh, what did it respond to? That's, uh, that's what I mean when I say wildest. Um, you know, it's cells, retinal ganglion cells, that respond to, you know, increasingly red portions of the visual scene or decreasingly green portions of the visual scene. Like, okay, cool. That seems cool, like, gets some... You know, around the time of Christmas, that, that's useful, and it's, uh, useful on other days of the year as well. But, you know, given that the retina is indeed the best understood piece of the brain, and given that you have 20 cell types, 20 as in 20 million, it's, um... You know, it seems tractable-

    4. ECDr. E.J. Chichilnisky44:22

       Yeah.

    5. AHAndrew Huberman44:22

       ... probably get to u- understanding it in its entirety, or understanding them in their entirety, excuse me. Um, one would like to know, like, what, what, what stuff is... Are we paying attention to at the level of the retina? I mean, are there, like, spiral, cells that like spiral stuff in the environment? Are there cells that like emojis? Like, what, what, what's going on in there?

    6. ECDr. E.J. Chichilnisky44:42

       (laughs)

    7. AHAndrew Huberman44:42

       You spend a lot of time doing this.

    8. ECDr. E.J. Chichilnisky44:44

       We do.

    9. AHAndrew Huberman44:44

       So-

    10. ECDr. E.J. Chichilnisky44:44

        We spend a lot of time on this.

    11. AHAndrew Huberman44:45

        ... you know, you... After all, you give up two nights sleep, which is kind of incredible, by the way. I'll just do a little, uh... Take a moment here and just say, y- you know, for a guy that's been doing this for this long with these sleepless nights, um, you, you look pretty good. You look pretty rested.

    12. ECDr. E.J. Chichilnisky44:57

        (laughs)

    13. AHAndrew Huberman44:57

        You know?

    14. ECDr. E.J. Chichilnisky44:58

        I tend to go home... I go home before the graduate students do. They-

    15. AHAndrew Huberman45:01

        Oh, they stay up.

    16. ECDr. E.J. Chichilnisky45:02

        They stay up. I was doing-

    17. AHAndrew Huberman45:03

        You used to stay up.

    18. ECDr. E.J. Chichilnisky45:03

        I used to stay up until my mid-40s. I was, I was in there doing the all-nighter-type things.

    19. AHAndrew Huberman45:08

        Got it.

    20. ECDr. E.J. Chichilnisky45:08

        And, and, um, maybe you can help me figure out my sleep patterns better so that I can do- (laughs)

    21. AHAndrew Huberman45:11

        Yeah, yeah, we can talk about that this episode. We talk about how to pull all-nighters and still survive.

    22. ECDr. E.J. Chichilnisky45:15

        (laughs) Exactly.

    23. AHAndrew Huberman45:15

        Done plenty of those, um, but yeah, like, what, what's-

    24. ECDr. E.J. Chichilnisky45:19

        Yeah.

    25. AHAndrew Huberman45:19

        Lots at stake here. There's a human retina, uh, uh, you know, meaning a human gave up their eyeballs to... For this experiment after they died, of course.

    26. ECDr. E.J. Chichilnisky45:27

        Yeah, yeah.

    27. AHAndrew Huberman45:27

        Um, you've got many people on this. These are... Uh, these sorts of experiments, um, are very expensive.

    28. ECDr. E.J. Chichilnisky45:34

        Yep.

    29. AHAndrew Huberman45:35

        Um, lot of fancy equipment, lot of salaries to try and figure this stuff out. This is the chief mission of the National Eye Institute. There's a lot of tax dollars, like-

    30. ECDr. E.J. Chichilnisky45:41

        Yep.
11. 49:27 – 1:00:25

    Retinal Prostheses, Implants

    1. ECDr. E.J. Chichilnisky49:27

    2. AHAndrew Huberman49:27

       So, we've been talking a lot about how to understand the signals that the retina is sending the brain, and I know your lab has done, um, incredible work in this arena, and figured out a number of the different signals as you described some of the features that the different cell types are extracting just a moment ago, these blobs of different colors, et cetera. What good is this to, you know, the everyday person, right? Um, what, what... In addition to wanting to understand how we see, you know, what sort o- sorts of medical applications can this provide, in terms of potentially restoring vision to the blind? Um, but perhaps even larger theme is this notion of neuroengineering, right? Taking this information, and creating devices that can help us help our nervous system function better, maybe even function at supraphysiological levels. I know there's a lot of interest in this these days, um, in part due to Neuralink, right? Because Elon's out there, front-facing, very vocal about his vision of the, uh, no pun intended, of, you know, chips being implanted in people's brains that would allow them to be in conversation with, you know, a hundred people at the same time just by hearing those voices in the head, maybe filtering things out so it doesn't sound like a clamor of a hundred different voices. Um, perhaps giving people super memory. I mean, you know, sky's the limit. No, no one really knows where this is all headed.

    3. ECDr. E.J. Chichilnisky50:53

       Yep.

    4. AHAndrew Huberman50:53

       You're working in what we call, uh, a very constrained system, where it has specific properties that you're trying to understand, and once you understand those, you can start to think about real applications of, like, what's possible. Like, could you create a visual system that, um, can extract more color features from the world that no other human can see? Um, can you restore, um, pattern vision to somebody who is essentially blind, and dependent on a cane or a dog, or, you know, God forbid, can't even lea- leave their house because they can't see anything at all. You know, where is this headed? What is the information useful for? And perhaps we should frame that first within the medical rehabilitative context of repairing, uh, or restoring vision rather, and then get into the more kind of, um, uh, sci-fi-type neuroengineering stuff.

    5. ECDr. E.J. Chichilnisky51:40

       Absolutely. Yeah, this, this really is my passion these days, turning that corner. Continuing to figure out the mysteries in the retina, but also saying, "Wait a second, we actually know quite a lot about this. Shouldn't this be the first place that we can solve problems like restoring vision, restoring function, or augmenting our function?" I think it should be. The concept of how to do this is straightforward, and not invented by us in any way, and that is the following. One of the major sources of blindness in, uh, the western world is, uh, loss of the photoreceptor cells that capture light. Macular degeneration and retinitis pigmentosa are two well-known, uh, ailments that give rise to vision loss, and the vision loss is because the cells that capture light in the first place, that we talked about earlier, die off, so you're no longer sensitive to light, and then you're blind. The concept is that you may be able to bypass those early sections of the retina that capture the light and process the signals, and instead, build an electronic implant that connects up directly to the retinal ganglion cells, and this electronic implant would h- would do the following. It would capture the light using a camera, which is relatively easy. It would process the visual information in a manner similar to the way that the retina normally does, as similar as possible, and then it would electrically activate the retinal ganglion cells by passing current, and causing the ganglion cells to fire spikes, and send those spikes to the brain. And if we do this really well, we can essentially replace those first two layers of the retina, and piggyback onto the third layer, and say, "Okay, we're just jumping right into that third layer. We're gonna force those ganglion cells to send reasonable visual signals to the brain, and then the brain is gonna think it got a natural visual signal, and proceed accordingly." That's the concept. This is not our idea. People have had this concept for decades, and some people have even started to make it work in, in human patients, and what I mean by that is implanting electrodes on the retina, stimulating and causing people who have been profoundly blind for decades to see visual sensation, blobs and streaks of light in their visual world that they, that are reproducible.

    6. AHAndrew Huberman53:56

       So, that's happening now.

    7. ECDr. E.J. Chichilnisky53:57

       That's happened. That's-

    8. AHAndrew Huberman53:58

       People who were, at once, blind-

    9. ECDr. E.J. Chichilnisky54:00

       Yep.

    10. AHAndrew Huberman54:01

        ... are able to see objects.

    11. ECDr. E.J. Chichilnisky54:03

        C- Are able to see crude blobs and flashes of light.

    12. AHAndrew Huberman54:07

        In ways that allow them to navigate their world better?

    13. ECDr. E.J. Chichilnisky54:09

        A little bit. A little bit, and-

    14. AHAndrew Huberman54:10

        Avoid a coffee table.

    15. ECDr. E.J. Chichilnisky54:12

        ... maybe, or at least see a bright window in a dark wall and be able to point to that bright window or the bright doorway in a dark wall, something like that. So, it's a glass half full, half empty story that I wanna turn... that I'd like to turn attention to in, in this conversation because I think it's very exciting. Um, yes, you can see stuff by artificially, electrically stimulating the ganglion cells, and you can see stuff that actually helps you interact with your world a tiny, little bit. So, great. That's the glass half full. The glass half empty is it's nothing remotely resembling what we understand as naturalistic vision, where we see s- fine, spatial detail, and color, and objects, and can navigate complex environments, and all that stuff. N- no chance, you can't do anything remotely like that. You can see that there's a bright doorway over that way and turn toward it, which is a helpful step in your human experience, but there's a long way to go. So, the question is, why does this existing technology fail to give us high-quality vision? What- what's needed to give us high-quality vision? And this is the piece I'm really passionate about. It turns out that the- the devices that have been implanted in humans so far by pioneering bioengineers who did really hard stuff were p- were fairly simple devices that treated the retina as if it's a camera that is just a grid of pixels, and they put a grid of electrodes down there, and they stimulated according to the pixels in the visual world, and thought, "Well, hopefully, that will cause the retina to do the right thing and send a nice visual... piece of visual information to the brain and initiate vision." Unfortunately, they left the science on the table, and this is actually what I'm dedicating the next phase of my career to: bringing the science that we know, that we talked about, to the table for vision engineering, and in particular, the fact that there are, there really are 20 or so distinct cell types, and they send different types of visual information to the different targets in the brain. I like to think of them a little bit as, uh, an orchestra playing a symphony. Each different cell type has its particular score. The violins do one thing. The oboes do something else. It's a very organized signal coming out of the retina, presenting to the brain this complex pattern of electrical activity that the brain assembles into our visual experience. Well, current retinal implants, unfortunately, are too crude to do anything like that. The, the conductor has just scattered the sheet music everywhere, and people are playing whatever. It's cacophony, okay? You can maybe recognize a tune in there a little bit sometimes, navigate toward a doorway, but you're not gonna get the full richness of the experience by ignoring the different cell types. And I'm so passionate about this, in part, for reasons that a little bit are similar to y- your reasons for doing this kind of work that you do, which I think is great, w- which is I, I feel we have a mission to give back as scientists, to take all this stuff we've been talking about because we find it so interesting and cool, and to deliver something to the society that has allowed us to explore these fascinating areas. And in our case, it's... in my... in the case of my lab and what we've done is to turn around and say, "Wait a second, we understand that there's these different cell types. We understand a lot about what they do. None of this information appears in current epiretn- retinal implants. Can't we do better by using the science to restore vision in a way that respects the circuitry of the retina?" That's what we're trying to do, and the mismatch is intense. Um, I, I told you when we were chatting before that nothing that we have learned about the retina since the founding of the National Eye Institute in 1968 is incorporated into the existing retinal implants. Nothing. We've learned a ton about the retina. Your research was funded by the National Eye Institute. My research is funded by National Eye Institute, a fantastic organization that allows us to learn all these things. How is this showing up in the neuroengineering to restore vision to people? Well, currently, it's not, and so we're trying to do that. Now, doing that turns out to be hard, and maybe we'll talk about that. It's a, it's a technological feat that's really challenging. You have to build a device that you can implant in a human that can recognize the distinct cell types, see where they are, stimulate them separately from one another, and conduct this orchestra to create a musical score that reasonably closely resembles the natural one. That's what we're all about doing, and as it turns out, and maybe we'll talk about that separately, that mission of being able to restore the patterns of activity that the retina normally creates also has extremely exciting spin-offs in three directions. One of them is understanding better how the brain puts the signals together. That's research for the brain. The second one is augmenting vision, creating novel types of visual sensations that weren't possible before, and maybe doing something along the Elon Musk lines of delivering more v- visual information than was ever possible. And third, figuring out how to interface to the brain more broadly, because as you and I know, the structure of the retina is very much representative of the structure of many brain areas, and if we're gonna figure out how to interface to the visual cortex, we darn well better fig- figure out how to interface to the retina first. That's what we're all about doing in my lab these days, is figuring out how to do that well. That's a mixed science and engineering effort. We've done about 15 years of basic science on that. How do we stimulate cells? How do we recognize cells? How are we gonna build a device that does all this and talks to the cells in this way? And I can go into lots of gory detail about it, um, but that's what we've been doing the basic research on, and the last few years, we've worked at Stanford with, um, fantastic engineers from various disciplines, electrical engineers, material scientists, others, to figure out how to put together the pieces and build an implant that can do this in a living human.

12. 1:00:25 – 1:06:05

    Artificial Retina, Augmenting Vision

    1. ECDr. E.J. Chichilnisky1:00:25

    2. AHAndrew Huberman1:00:25

       So, is the idea to build a robotic retina?

    3. ECDr. E.J. Chichilnisky1:00:30

       Yeah.

    4. AHAndrew Huberman1:00:30

       ... to build a, a, essentially an artificial retina that could be put into the eye of a blind person, or even put into the eye of a sighted person, that would fundamentally change their ability to see, or in the latter case, uh, enhance their ability to extract things from the visual world that, uh, they would otherwise not be able to see, like, like seeing twice as far, um, getting, you know, hawk-like resolution of the visual world. Um-

    5. ECDr. E.J. Chichilnisky1:00:55

       Yep.

    6. AHAndrew Huberman1:00:56

       ... or, which, that would be cool.

    7. ECDr. E.J. Chichilnisky1:00:57

       Yep.

    8. AHAndrew Huberman1:00:58

       Could be distracting.

    9. ECDr. E.J. Chichilnisky1:00:59

       Yep.

    10. AHAndrew Huberman1:01:00

        Like, I'm not sure I wanna see the, the fine movements of a piece of paper in somebody's notebook from across a café as they flutter the pages. Um-

    11. ECDr. E.J. Chichilnisky1:01:08

        But you might want to for a moment. There might be a moment when you want to, and if you have an electronic device that you can control, that you can dial in to sense different aspects of the visual world, and, you know, by, by your choice, you might be like, "Yeah, that's pretty cool. I wanna be able to do that right now." There's an example I like to give which I think maybe is, is helpful for interpreting what we're talking about when we say, "Being able to do more things with the optic nerve." You gave the example of many voices and stuff. Here's an example that I like. We know that we can drive down the road and have a phone call hands-free, and do that quite safely, pretty safely, right? And you're-- be-- why? Because you're tapping in, uh, you've got two types of signals coming into your brain, your visual signal of the cars on the freeway, any one of which could kill you in an instant, so it's important, and the sound coming into your ears which carries the voice of your girlfriend that's telling, who's telling you something that you're interested in hearing. And these are different parts of the brain processing this information, and so you can do both of these things at the same time 'cause there's no interference. One part of the brain's working doing one thing, another part's doing something else, you're good. What you can't do is to read your texts and drive down the freeway at the same time. That's not good, because now that visual system of yours that needs to be detecting these fast-moving dangerous objects is being distracted by looking at the text, and you might die, and some other people might die with you. So-

    12. AHAndrew Huberman1:02:31

        I see a lot of people texting and driving.

    13. ECDr. E.J. Chichilnisky1:02:33

        Yeah. That's why I like to point out this example, to remind people you can't do it well. (laughs) It's like, it's like-

    14. AHAndrew Huberman1:02:38

        You can't do it well. It used to be-

    15. ECDr. E.J. Chichilnisky1:02:39

        ... we probably talk m- (laughs)

    16. AHAndrew Huberman1:02:40

        Yeah, it used to be, you know, I, uh, will just take a brief, uh, tangent here into this topic. Um, a few years back, there were, there were a lot of news articles, a lot of discussion about texting and driving, a lot of attention to getting people to stop texting and driving. I've seen a few people pulled over for texting and driving before, but I would say texting and driving is rampant. Reading what's on one's phone-

    17. ECDr. E.J. Chichilnisky1:03:04

        Yep.

    18. AHAndrew Huberman1:03:04

        ... while driving is rampant.

    19. ECDr. E.J. Chichilnisky1:03:06

        Yep.

    20. AHAndrew Huberman1:03:07

        All you have to do is be on the freeway here in Los Angeles, look in the car next to you-

    21. ECDr. E.J. Chichilnisky1:03:10

        Yeah, look where they're looking.

    22. AHAndrew Huberman1:03:11

        ... um, (laughs) and people are, like, reading and texting while driving. Presumably, when they're doing that, they're just using their peripheral vision to detect any kind of motion, and, um, no doubt this has caused the deaths of many people.

    23. ECDr. E.J. Chichilnisky1:03:23

        Yeah. Change lanes, get away from them.

    24. AHAndrew Huberman1:03:25

        Yeah.

    25. ECDr. E.J. Chichilnisky1:03:25

        And that's, you know, just, just like that other driver. So, so here's the thing, um, and this is, this is, uh, I say this a bit tongue-in-cheek, but it's, it's sort of a real example. It may be that if we harness the different cell types in the retina to deliver different visual information to different cell types, like the image of the text on your screen to a certain cell type that you know, the so-called midget cells, or the motion of the objects in the visual scene, the cars, to a different cell type that you know-

    26. AHAndrew Huberman1:04:00

        Called midget cells, by the way, folks, because they're very, very small.

    27. ECDr. E.J. Chichilnisky1:04:03

        Yes. And by named by anatomists decades ago.

    28. AHAndrew Huberman1:04:06

        Right.

    29. ECDr. E.J. Chichilnisky1:04:06

        So we carry that nomenclature forward.

    30. AHAndrew Huberman1:04:09

        Sure.
13. 1:06:05 – 1:07:12

    Sponsor: InsideTracker

    1. ECDr. E.J. Chichilnisky1:06:05

    2. AHAndrew Huberman1:06:05

       I'd like to just take a quick break and acknowledge one of our sponsors, InsideTracker. InsideTracker is a personalized nutrition platform that analyzes data from your blood and DNA to help you better understand your body and help you reach your health goals. Now, I've long been a believer in getting regular blood work done for the simple reason that many of the factors that impact our immediate and long-term health can only be addressed, that is, can only be measured with a quality blood test-Now, one issue with many blood tests out there is that you get information back about lipid levels, hormone levels, metabolic factors, et cetera, but you don't know what to do with that information. With InsideTracker, they make it very easy to understand your levels and what they mean, and specific actionable items that you can undertake in order to bring those levels into the ranges that are optimal for you. InsideTracker also offers InsideTracker Pro, which enables coaches and health professionals to provide premium and personalized services by leveraging InsideTracker's analysis and recommendations with their clients. If you'd like to try InsideTracker, you can go to insidetracker.com/huberman, and you'll get 20% off any of InsideTracker's plans. Again, that's insidetracker.com/huberman.

14. 1:07:12 – 1:17:01

    Neuroengineering, Neuroaugmentation & Specificity

    1. AHAndrew Huberman1:07:12

       So, to just, um, summarize a little bit of the, the linear flow here of w- of what you've done, um, you started off with the understanding that the neural retina is perhaps the best place to try and understand how the brain works because of its arrangement, the cell types, et cetera. You spend a number of decades doing these wild experiments, um, on human retinas and other retinas, um, recording the different cell types with these high density, what I call bed of nails-

    2. ECDr. E.J. Chichilnisky1:07:40

       Two and a half decades, I'm not that old. (laughs)

    3. AHAndrew Huberman1:07:41

       Two and a half decades. Um, it's your, uh, robustness that matters, EJ.

    4. ECDr. E.J. Chichilnisky1:07:48

       (laughs)

    5. AHAndrew Huberman1:07:48

       Um, and, and you have plenty of it. Um, you've figured out what the cell types are, so then you gained an understanding of how light is transformed into different types of electrical signals that encode different features in the visual scene. Then comes the challenge of developing neuroengineering tools to try and stimulate the specific cell types in a way that more or less mimics their normal patterns of activation. Like, not, um, activating a huge piece of retina, so that, you know, the cells that like increases in redness are also being stimulated with the cells that like, you know, edges in a way that would create some shmooey, like, crazy representation of the outside world.

    6. ECDr. E.J. Chichilnisky1:08:26

       That's right.

    7. AHAndrew Huberman1:08:26

       No, you want the pre- same precision that light stimulation of these cells in the intact human eye provides in this explanted retina, this retina i- on this bed of nails. But then, a device that essentially can mimic what the retina does, and you needed to do all that earlier work, understanding, like, what does the normal retina do, what does the healthy retina do, in order to try and develop this prosthetic device to either restore vision, or because it puts you in the position of being able to stimulate cells however you want. In theory, you could create a situation in a human where the cells that respond to, um, I don't know, outlines of objects are, are hyperactive, so that, you know, the person could, um, effectively see objects in the, one's environment better than anyone else.

    8. ECDr. E.J. Chichilnisky1:09:16

       Yeah.

    9. AHAndrew Huberman1:09:16

       Could perceive... I know motion is a tricky one, but motion better than anyone else. Could see detail in the visual world that no one else could detect. We're not talking about turning people into mantis shrimp, um, but the analogy works, because mantis shrimp can see things that we can't, and vice versa, and so what you're talking about here is neural augmentation through the use of engineering.

    10. ECDr. E.J. Chichilnisky1:09:38

        Yup, and we often do talk about it as sci-fi, because the sci-fi writers have been talk- have been, you know, writing about this for decades. It's not sci-fi anymore. It's sci- it's straight up sci right now. It's really we just need to build the instrumentation, and start working with those experiments to figure out how to make it work. I think we have a responsibility to do this, because this is the way we take this kind of information, all that's been learned about the visual system by the National Institute s- since 1968, and all the people that it's funded to do this research, and turn the corner and make a difference for humanity with it. And I, I assume and think that humanity will be leveraging nervous system knowledge to build all sorts of devices that we can interface to the world with. I think, y- you know, I don't know Elon Musk, but I think he's right about that, that that's where we're headed. It should be done well. It's important to do it well. Um, we will hopefully be more connected to truth in the world if we're able to build devices that give us better sensations, more acute understanding of what's going on out there, better abilities to make decisions, and all that, let alone just see stuff. So, that frontier of developing technologies to allow our brains and our visual systems initially, and then other parts of our brains, to do things better is unbelievably exciting. It is sci-fi, but I would just want to emphasize I think it's the responsible way to go to think about how to do that well. All technologies that we develop can be used for good or for ill, and I'm sure some of your listeners, who are a lo- lot of very passionate thinking people out there, thinking about neuroscience and the implications, worry, "What does this mean? We're gonna be introducing electronic circuits in our brains to do stuff?" Yeah, well, we will. We- it's pretty much clear that humanity will do that. And so, in, in any technology development, you have to think about, well, how do you do it well? How do you do it for good? There are, there are popular movies right now about technology development, such as understanding the structure of the atom, and that technology development can be used for good or for ill, to blow up cities, or to save civilizations. How's it gonna go? Well, I think, I think, um, as scientists, we're responsible for advancing that in a thoughtful and meaningful way. I think we can do this in the retina. It is the place to start. It's the pl- and you know, y- you... I'm curious what you think, actually-

    11. AHAndrew Huberman1:12:11

        Mm-hmm.

    12. ECDr. E.J. Chichilnisky1:12:11

        ... as a scientist, 'cause your background is in this field, or a very close field, to mine. I know you speak with all sorts of scientists on this podcast, um, but this is pretty much your field, or very close. Not the neuroengineering part, but understanding the retina-

    13. AHAndrew Huberman1:12:24

        Hmm.

    14. ECDr. E.J. Chichilnisky1:12:24

        ... and I'm curious if you agree that this is the place to start doing this stuff.

    15. AHAndrew Huberman1:12:28

        You're the first guest to ever ask me a question on this podcast during a guest interview. Um, I think this would be a fun place to both answer and riff on this a little bit, because, um...... first of all, I think the retina is absolutely the place to start, because we understand so much of what it does, what the cell types are. But maybe by comparison, a different brain region, the hippocampus, which is involved in the formation of memories and other stuff, but formation of memories about what one did the previous day, what one did many years ago, et cetera, is an area that I think any time the conversation about neural prosthetics, or neuroengineering, or neural augmentation comes up people think, "Oh, wouldn't it be cool to have, like, a little stimulating device in the hippocampus? And then if I want to remember a bunch of information from a page or from a, a lecture, I just stimulate, and then voila, all the information is batched in there." Um, while that's an attractive idea, I think it's worth pointing out right now that sure, there is a pretty decent understanding of the different cell types in terms of their shapes, some of their electrical properties of the hippocampus, but there is in no way, shape, or form the depth of understanding about the hippocampus, and what the individual cell types do, and what the different layers of it do that one has for the neural retina. So, what we're really saying is if you stimulate the hippocampus, you'll likely get an effect, but it's unclear what the effect is, and it's not clear how to stimulate. That's, to me, the best reason to focus on the retina, because you know what the cell types are, thanks to the work of your laboratory and many other laboratories. You know what sorts of stimulation matter, and it provides the perfect test bed for this whole business of neural augmentation, neuro- neuroengineering. I think there's also a bigger discussion to just frame this in, which is so much of what we discuss on this podcast with guests and in solo episodes relates to things like dopamine, neuro- neuromodulators, serotonin. Everyone is interested in these things because they can profoundly change the way that we perceive and interact with the world. But one only has to look to the various pharmaceuticals that increase or decrease these neuromodulators and know that indeed those pharmaceuticals can be immensely beneficial to certain individuals. I wanna be clear about that. But that whatever, quote unquote, "side effects" one sees, or lack of effect over time, is because those receptors are, like, everywhere over the, around the brain, so you can't just increase dopamine in the brain and expect to only get one specific desired effect.

    16. ECDr. E.J. Chichilnisky1:14:54

        Yes.

    17. AHAndrew Huberman1:14:55

        So, the reason you're here today, um, is not because we both worked on the retina, and it's not because we happen to also be friends. It's because, uh, to my mind, your laboratory represents the, um, apex of precision in terms of trying to figure out what a given piece of the brain, in your case the neural retina, does, parse all its different components, and then use that knowledge to create a real-world technology that can actually tickle and probe and, um, stimulate that piece of the brain in a way that's meaningful, right? Not just, like, sending electrical activity in. And that, to me, is so important. I think if we were gonna think about levels of specificity for manipulating the human brain to get e- an effect, you would say, okay, crudeness would be drugs. Take a drug, it increases serotonin, which might bind to particular receptors. Let's take the drug psilocybin. A lot of excitement about psilocybin. We know that can lead to increased connectivity between different brain regions. At rest, there's probably, there is some demonstrated clinical benefit. There's also some potential hazards. But it's very broad scale. We don't know what's happening when the person is thinking about a, a, you know, a piece of moss expanding into an image and a memory of their childhood, or w- it's, like, a million different things are happening there. And then at the other far extreme is the kind of experiment that you're talking about, stimulating one known type of retinal cell, seeing what that means for visual processing, or modeling what that means for visual processing, and then building a device that can do exactly that, and then maybe ramp it up 20%, 50%. Uh, 'cause I think that represents the first step into, okay, how would you stimulate the hippocampus to create a super memory? Would you stimulate a particular cell type in a particular way? And to my knowledge, there's, you know, despite the immense excitement about the hippocampus, and understandably so, there just isn't that precision and, of understanding yet. So, uh, forgive the long answer, but, you know, you ask me a question on, on this podcast, I'm going to give you the long answer.

15. 1:17:01 – 1:20:02

    Building a Smart Device, AI

    1. AHAndrew Huberman1:17:01

    2. ECDr. E.J. Chichilnisky1:17:01

       Love that answer. Yeah, and, and specificity is what you're talking about, and we need technology to do that, to, to modulate neural circuits in a highly specific way. We've got to start with the piece, the neural circuit we understand best and we have best access to. That's the retina. It's sitting right there on the surface. We can get right into it, and install devices. We know so much about it. That's the place to start, the place that we understand. Build electronics that is, that is adaptive, that senses what circuit it's embedded in, and responds appropriately. It's not just electronics you stick in there and it does something and that's it. No. It, it first figures out who it's talking to-

    3. AHAndrew Huberman1:17:38

       Mm-hmm.

    4. ECDr. E.J. Chichilnisky1:17:38

       ... and then learns to speak the language of the nearby neural circuitry.

    5. AHAndrew Huberman1:17:42

       So, a smart device.

    6. ECDr. E.J. Chichilnisky1:17:43

       Smart device, yep.

    7. AHAndrew Huberman1:17:44

       Let, let's, let's push on that a little bit. So, um, put a little chip onto the retina that can stimulate specific cell types. Is there a way that it can use AI, machine learning-

    8. ECDr. E.J. Chichilnisky1:17:56

       Yeah.

    9. AHAndrew Huberman1:17:56

       ... that it can learn something about the tissue it happens to be in contact with?

    10. ECDr. E.J. Chichilnisky1:17:59

        Absolutely. In the simplest possible way, the device works in three simple steps. Step number one, record electrical activity, which is what we do in our lab in a room full of equipment, but this is a two-millimeter-sized chip implanted in your eye. Record the electrical activity. Just recognize what cells are there, when they're firing, what electrical properties they have, to identify the cells and cell types in this specific circuit in this human. That's step one. Step two, stimulate and record. So, you figure out, oh, this electrode activates cell number 14 with 50% probability. This electrode activates cell number 12 if we pass a microamp of current with this probability, and so on. You make a big table that tells you how these electrodes talk to these cells in your circuit. That's step two, calibrate the stimulation by stimulation and recording.... then, finally, when you have a visual image, and you want to represent it in the pattern of activity of these cells, you say, "Okay. I know from decades of basic science what these cells ought to be doing with this image that's coming in. I know exactly what they ought to be doing. That's what the science has been telling us. We've been studying the neural code for decades to understand this. I know what they should do. Use my device with my calibration, where I know where the cells are, I know how the electrodes talk to them, and bing, bing, bing, bing, bing, activate them in the correct sequence." That's what I think of as a smart device, a device that records, stimulates and records, and then finally stimulates. Yes, AI is part of that. Of course it is, because this is a very complex, uh, transformation, if you will, from the external visual world to the patterns of activity of these cells. It's not easy to write down in just a few lines of code, or a, uh, few equations. It's complicated, and AI is really helpful for that, and learning by stimulating and recording and aggregating information quickly so that you can then use the device. That's absolutely a part of the engineering. It... Let me be clear. The AI doesn't help us to understand. It's just an engineering tool that helps us to capture what this thing normally does, and then go ahead and execute, and make it do the thing it should normally do.

16. 1:20:02 – 1:25:21

    Neural Prosthesis, Paralysis; Specificity

    1. ECDr. E.J. Chichilnisky1:20:02

    2. AHAndrew Huberman1:20:03

       I hope people, um, will appreciate this example. Uh, perhaps not. You know, not but, goodness, I don't know, 40, 50 years ago? But still today, uh, one treatment for depression is electric shock therapy. Uh, a very, uh, you know, on the face of it, barbaric, um, treatment, but effective in certain conditions. It's still used for a reason, um, but it can appear barbaric, right? You know, people are... have, like, a bite, you know, a bite device, you know, so they don't, um, bite through their tongue or their lips. They're... You know, they're strapped down, and they stimulate (imitates electric current) 　 just, like, stimulate all neurons in the brain, essentially. There's a huge dump of neurotransmitters and neuromodulators, and-

    3. ECDr. E.J. Chichilnisky1:20:42

       It's like drugs. It's completely non-specific stimulation, effectively, right?

    4. AHAndrew Huberman1:20:46

       Right. Probably even less specific than drugs.

    5. ECDr. E.J. Chichilnisky1:20:48

       Maybe.

    6. AHAndrew Huberman1:20:49

       And yet, the clinical outcomes from electric shock therapy, uh, in some cases, are pretty impressive. Like people-

    7. ECDr. E.J. Chichilnisky1:20:56

       Yeah.

    8. AHAndrew Huberman1:20:56

       Um, the brain is, quote-unquote, "reset." They still remember who they are, um, but presumably, through the, the, the release of neuromodulators like dopamine, serotonin, acetylcholine, in a very non-specific way, there has been some symptom relief in some cases. What you're talking about is really the opposite extreme. Um, you know, before I said pharmaceuticals that tickle a particular neuromodulator pathway would be the opposite extreme. I think electric shock therapy is probably-

Episode duration: 2:00:06