1 00:00:05,120 --> 00:00:09,039 Speaker 1: Why is it so hard to reverse engineer the brain? 2 00:00:09,240 --> 00:00:11,480 Speaker 1: Can't we just measure the signals and all of the 3 00:00:11,520 --> 00:00:15,040 Speaker 1: brain cells and then figure out the neural code. And 4 00:00:15,080 --> 00:00:18,040 Speaker 1: if not, why not? And what does this have to 5 00:00:18,079 --> 00:00:23,160 Speaker 1: do with solving vision loss and eavesdropping on the activity 6 00:00:23,160 --> 00:00:29,319 Speaker 1: of cells using other cells and communication between brains using 7 00:00:29,360 --> 00:00:35,479 Speaker 1: something other than conversation or observing and understanding and maybe 8 00:00:35,600 --> 00:00:43,000 Speaker 1: changing our own experience of the world. Welcome to Intercosmos 9 00:00:43,080 --> 00:00:46,239 Speaker 1: with me David Eagleman. I'm a neuroscientist and author at 10 00:00:46,240 --> 00:00:49,800 Speaker 1: Stanford and in these episodes, we sail deeply into our 11 00:00:49,880 --> 00:00:55,720 Speaker 1: three pounds universe to understand the mysterious creatures inside the 12 00:00:55,760 --> 00:01:00,160 Speaker 1: eighty six billion neurons that are chattering along with tiny 13 00:01:00,440 --> 00:01:08,479 Speaker 1: electrical and chemical signals producing our experience. Now, today's question 14 00:01:08,640 --> 00:01:12,480 Speaker 1: is how do you actually get inside the brain to 15 00:01:12,560 --> 00:01:16,160 Speaker 1: study it? After all, we know that the brain is 16 00:01:16,280 --> 00:01:19,480 Speaker 1: the root of all of our thoughts and hopes and 17 00:01:19,640 --> 00:01:23,679 Speaker 1: dreams and aspirations and our consciousness. And the reason we 18 00:01:23,800 --> 00:01:28,080 Speaker 1: know this is because even very small bits of damage 19 00:01:28,080 --> 00:01:31,880 Speaker 1: to the brain change who you are and how you 20 00:01:32,000 --> 00:01:36,039 Speaker 1: think and whether you're conscious. Note that other parts of 21 00:01:36,080 --> 00:01:40,320 Speaker 1: your body, like your heart, can get completely replaced by 22 00:01:40,400 --> 00:01:43,839 Speaker 1: a machine and you are no different. Or you can 23 00:01:44,280 --> 00:01:47,000 Speaker 1: lose your arms and your legs and you can still 24 00:01:47,040 --> 00:01:50,560 Speaker 1: be conscious, or you can get a kidney replacement and 25 00:01:50,800 --> 00:01:53,200 Speaker 1: you're still thinking about your life and your family and 26 00:01:53,200 --> 00:01:56,800 Speaker 1: what you need to do tomorrow. But even a tiny 27 00:01:56,880 --> 00:02:00,800 Speaker 1: bit of damage to the brain caused by let's say 28 00:02:00,800 --> 00:02:03,800 Speaker 1: a stroke or a tumor or a traumatic brain injury, 29 00:02:03,840 --> 00:02:08,480 Speaker 1: even a small bit of damage can change you entirely. 30 00:02:09,000 --> 00:02:11,440 Speaker 1: Even if you don't lose your consciousness, you might lose 31 00:02:11,480 --> 00:02:16,000 Speaker 1: your ability to think clearly, or to speak, or to move, 32 00:02:16,760 --> 00:02:21,280 Speaker 1: or to recognize animals or understand music, or understand the 33 00:02:21,320 --> 00:02:24,520 Speaker 1: concept of a mirror, or a thousand other things that 34 00:02:24,560 --> 00:02:29,040 Speaker 1: have taught us over the centuries about the complex landscape 35 00:02:29,440 --> 00:02:33,240 Speaker 1: of this three pound inner cosmos. So we know the 36 00:02:33,240 --> 00:02:37,240 Speaker 1: brain is necessary for our cognition and experience, but we 37 00:02:37,360 --> 00:02:41,720 Speaker 1: didn't get to that understanding through detailed studies of the 38 00:02:41,760 --> 00:02:47,279 Speaker 1: intricate circuitry, but instead mostly through observations of crude damage. 39 00:02:47,720 --> 00:02:52,160 Speaker 1: So there's still an enormous amount that we don't understand 40 00:02:52,240 --> 00:02:55,800 Speaker 1: about how the whole system works. We only have a 41 00:02:55,840 --> 00:02:58,880 Speaker 1: sense of how it breaks. It would be like if 42 00:02:58,919 --> 00:03:02,160 Speaker 1: you were a space salien and you looked at cell 43 00:03:02,240 --> 00:03:05,200 Speaker 1: phones and discovered that if you zap the phone with 44 00:03:05,280 --> 00:03:09,760 Speaker 1: your laser, then it doesn't make calls anymore. Okay, that's important, 45 00:03:09,760 --> 00:03:14,120 Speaker 1: but it doesn't tell you how telecommunication works in terms 46 00:03:14,200 --> 00:03:17,960 Speaker 1: of base stations and frequency bands and compression and sim 47 00:03:18,000 --> 00:03:21,400 Speaker 1: cards and everything else. For that, you would need to 48 00:03:21,480 --> 00:03:24,640 Speaker 1: take off the cover of the cell phone to figure 49 00:03:24,680 --> 00:03:29,640 Speaker 1: out what the billions of transistors are actually doing. And 50 00:03:29,680 --> 00:03:34,480 Speaker 1: that's really our modern challenge in neuroscience to study this 51 00:03:34,639 --> 00:03:41,840 Speaker 1: incredibly detailed system more directly. So why is progress still 52 00:03:42,160 --> 00:03:45,680 Speaker 1: so slow on that front? Well, it turns out it's 53 00:03:45,760 --> 00:03:51,320 Speaker 1: very hard to study the brain's trillions of neurons directly, 54 00:03:51,760 --> 00:03:57,520 Speaker 1: this pink, magical computational material that mother nature has refined 55 00:03:57,640 --> 00:04:02,920 Speaker 1: through hundreds of millions of years evolution. Why because this 56 00:04:02,960 --> 00:04:06,640 Speaker 1: is the computational core, and so mother nature has protected 57 00:04:06,640 --> 00:04:10,320 Speaker 1: it in armored bunker plating. So that's the first challenge. 58 00:04:10,360 --> 00:04:15,160 Speaker 1: The brain is tightly protected inside the prison of our skull. 59 00:04:15,920 --> 00:04:17,680 Speaker 1: But that's only part of the challenge, and that can 60 00:04:17,720 --> 00:04:22,520 Speaker 1: be addressed by careful neurosurgery. The bigger difficulty is that 61 00:04:22,720 --> 00:04:25,520 Speaker 1: even when we can get in there by drilling a 62 00:04:25,520 --> 00:04:28,560 Speaker 1: little hole in the skull. What we find is an 63 00:04:28,560 --> 00:04:35,080 Speaker 1: incredibly densely packed device made of very sophisticated units that 64 00:04:35,160 --> 00:04:39,799 Speaker 1: are microscopically small, and there are almost one hundred billion 65 00:04:39,880 --> 00:04:43,040 Speaker 1: of them, which is about twelve times more than there 66 00:04:43,080 --> 00:04:46,920 Speaker 1: are people on the planet. And each one of these 67 00:04:47,000 --> 00:04:52,640 Speaker 1: neurons is sending very tiny electrical signals tens or hundreds 68 00:04:52,680 --> 00:04:56,600 Speaker 1: of times per second, and these signals zoomed down axons 69 00:04:56,920 --> 00:05:02,000 Speaker 1: and cause chemicals neurotransmitters to be released. And it's not 70 00:05:02,400 --> 00:05:07,760 Speaker 1: generally clear how to read this insanely dense circuitry to 71 00:05:07,880 --> 00:05:13,960 Speaker 1: understand how these trillions of incredibly small signals racing around 72 00:05:13,960 --> 00:05:18,040 Speaker 1: in there lead to a particular outcome at the scale 73 00:05:18,240 --> 00:05:21,880 Speaker 1: of a human like you move your arm, or you 74 00:05:21,960 --> 00:05:26,200 Speaker 1: have a craving for pistachios, or suddenly you're reminded of 75 00:05:26,240 --> 00:05:31,080 Speaker 1: the poem osomandias or whatever. What is the relationship between 76 00:05:31,400 --> 00:05:34,839 Speaker 1: this small scale and the large scale? So how do 77 00:05:34,960 --> 00:05:40,080 Speaker 1: neuroscientists try to decode this incredible complexity. The answer is 78 00:05:40,640 --> 00:05:45,240 Speaker 1: by marrying the technology that we have like computers, directly 79 00:05:45,320 --> 00:05:48,200 Speaker 1: to the cells of the brain. And this is what 80 00:05:48,240 --> 00:05:53,479 Speaker 1: we generally call a brain computer interface or BCEI. We 81 00:05:53,640 --> 00:05:56,799 Speaker 1: use that term to refer to essentially anything that allows 82 00:05:56,920 --> 00:06:01,200 Speaker 1: direct communication between the brain and to an external device. 83 00:06:01,240 --> 00:06:05,960 Speaker 1: So people use these to control wheelchairs or robotic arms, 84 00:06:06,360 --> 00:06:09,600 Speaker 1: or type directly onto a screen or speak through a 85 00:06:09,640 --> 00:06:14,360 Speaker 1: synthetic voice. The idea is to use BCIs to restore 86 00:06:14,520 --> 00:06:19,160 Speaker 1: functions in people who have lost them, like paralysis or blindness, 87 00:06:19,360 --> 00:06:24,159 Speaker 1: and someday perhaps to enhance the capabilities of healthy people. Now, 88 00:06:24,360 --> 00:06:28,680 Speaker 1: how does a BCI actually work. People sometimes think about 89 00:06:28,760 --> 00:06:32,360 Speaker 1: BCIs as measuring electrical activity on the scalp with an 90 00:06:32,400 --> 00:06:36,320 Speaker 1: EEG electroncephalogram, and that counts, but you don't get very 91 00:06:36,400 --> 00:06:39,359 Speaker 1: much detail from the outside of the skull. So the 92 00:06:39,440 --> 00:06:45,599 Speaker 1: more sophisticated forms of BCIs involves measuring brain activity directly 93 00:06:45,800 --> 00:06:48,720 Speaker 1: from the cells. And the main way to do this 94 00:06:49,120 --> 00:06:52,800 Speaker 1: is with small metal electrodes that you insert into the 95 00:06:52,839 --> 00:06:57,000 Speaker 1: brain tissue. And with these electrodes you can send little 96 00:06:57,200 --> 00:07:00,160 Speaker 1: electrical zapps to stimulate the neurons, and you can can 97 00:07:00,200 --> 00:07:04,599 Speaker 1: also listen to hear when the neurons themselves are giving 98 00:07:04,640 --> 00:07:08,480 Speaker 1: off small electrical signals. Now, this has been a technology 99 00:07:08,480 --> 00:07:12,600 Speaker 1: that researchers and neurosurgeons have used for many decades, but 100 00:07:12,760 --> 00:07:15,320 Speaker 1: it's still a challenge because you have to drill a 101 00:07:15,320 --> 00:07:18,480 Speaker 1: hole in the skull and these little, tiny metal electrodes. 102 00:07:18,520 --> 00:07:21,640 Speaker 1: Although they're tiny, they're actually pretty big from the point 103 00:07:21,640 --> 00:07:24,120 Speaker 1: of view of neurons. From the point of view of 104 00:07:24,160 --> 00:07:27,440 Speaker 1: the neurons, it's like inserting a tree trunk. It damages 105 00:07:27,520 --> 00:07:31,320 Speaker 1: the tissue. Now you've probably heard of companies like Neurlink. 106 00:07:32,000 --> 00:07:36,040 Speaker 1: They're still inserting electrodes just like neurosurgeons have done for decades, 107 00:07:36,240 --> 00:07:39,480 Speaker 1: but they're working to make them smaller and finer and 108 00:07:39,640 --> 00:07:43,960 Speaker 1: robotically inserted, and also wireless in their communications so the 109 00:07:44,000 --> 00:07:47,520 Speaker 1: information can go back and forth without having a cable there. 110 00:07:47,720 --> 00:07:51,120 Speaker 1: So it's a better version of the same idea of 111 00:07:51,160 --> 00:07:55,600 Speaker 1: sticking electronics into the brain. But are there new ideas 112 00:07:55,680 --> 00:07:58,840 Speaker 1: about how to read and write to brain cells, about 113 00:07:58,880 --> 00:08:01,560 Speaker 1: how to interface with the brain. Today, we're going to 114 00:08:01,640 --> 00:08:04,240 Speaker 1: talk about what is at the cutting edge, and so 115 00:08:04,440 --> 00:08:07,160 Speaker 1: for that I called a colleague of mine who is 116 00:08:07,520 --> 00:08:12,440 Speaker 1: shaping the future of BCI technology, Max Hodak. Max is 117 00:08:12,600 --> 00:08:17,720 Speaker 1: an unusually brave thinker. He started studying brain machine interfaces 118 00:08:18,080 --> 00:08:21,120 Speaker 1: as an undergraduate, and while most people would be thrilled 119 00:08:21,120 --> 00:08:23,840 Speaker 1: to simply be a part of that. He was already 120 00:08:23,880 --> 00:08:26,840 Speaker 1: thinking about the ways that parts of the science were 121 00:08:27,200 --> 00:08:30,400 Speaker 1: inefficient and could be improved. Some years later, he went 122 00:08:30,480 --> 00:08:32,800 Speaker 1: on to be a part of the co founding team 123 00:08:33,080 --> 00:08:36,320 Speaker 1: at Neuralink and he became the president, and then four 124 00:08:36,400 --> 00:08:40,920 Speaker 1: years ago he left to found his own company, Science Corporation. 125 00:08:41,320 --> 00:08:44,480 Speaker 1: When I visited him at Science Corporation recently, many of 126 00:08:44,480 --> 00:08:47,880 Speaker 1: the things I saw there would have seemed like science 127 00:08:47,880 --> 00:08:52,080 Speaker 1: fiction fantasy just a few years ago. So here's my 128 00:08:52,160 --> 00:09:00,679 Speaker 1: interview with Max Hodak. You started a company called Science Corp, 129 00:09:00,840 --> 00:09:03,640 Speaker 1: which will refer to as Science and tell us about 130 00:09:03,640 --> 00:09:05,560 Speaker 1: science because it's so exciting what you're doing there. 131 00:09:05,760 --> 00:09:08,839 Speaker 2: Our main focus at Sciences is restoring vision to people 132 00:09:08,840 --> 00:09:10,800 Speaker 2: that have gone blind because they've lost the rods and 133 00:09:10,840 --> 00:09:15,040 Speaker 2: cones in the retina. And I this was not something 134 00:09:15,080 --> 00:09:17,280 Speaker 2: I'd not worked on the retina before, but I had 135 00:09:17,280 --> 00:09:19,720 Speaker 2: this thesis that that the technology was there that this 136 00:09:19,760 --> 00:09:23,880 Speaker 2: would be possible. There's I think two different ways to 137 00:09:23,960 --> 00:09:26,559 Speaker 2: do this that people have been thinking about in the retina. 138 00:09:26,679 --> 00:09:30,560 Speaker 2: There's a technical optogenetics, where you use a gene therapy 139 00:09:30,679 --> 00:09:33,080 Speaker 2: to deliver a little bit of DNA to the cells 140 00:09:33,080 --> 00:09:35,760 Speaker 2: of the optic nerve to make them light sensitive. That 141 00:09:35,760 --> 00:09:37,920 Speaker 2: then you could activate with a laser, or you could 142 00:09:37,960 --> 00:09:40,480 Speaker 2: put an electrical stimulator under the retina and drive the 143 00:09:40,520 --> 00:09:42,719 Speaker 2: remaining the cells that are still there electrically. 144 00:09:43,320 --> 00:09:45,040 Speaker 1: And the mean for just one saying, the retina is 145 00:09:45,080 --> 00:09:47,479 Speaker 1: the lawn of cells at the back of the eyeballer 146 00:09:47,600 --> 00:09:50,440 Speaker 1: catching the photons that are coming in through the front. 147 00:09:50,760 --> 00:09:53,560 Speaker 1: And so if you've got a problem where let's say 148 00:09:53,559 --> 00:09:56,240 Speaker 1: those cells have died for whatever reason, lots of reasons, 149 00:09:56,679 --> 00:09:58,760 Speaker 1: then what you're talking about is how do you how 150 00:09:58,800 --> 00:10:01,520 Speaker 1: do you get those cells to catch the photons and 151 00:10:01,640 --> 00:10:03,600 Speaker 1: send their signals back along the optic nerve. 152 00:10:03,840 --> 00:10:05,720 Speaker 2: Yeah, so I think you know to take a step back. 153 00:10:05,760 --> 00:10:08,000 Speaker 2: If you're thinking about getting vision into the brain, there's 154 00:10:08,000 --> 00:10:09,880 Speaker 2: a couple of different places you could think to do it. 155 00:10:09,920 --> 00:10:11,960 Speaker 2: The first is the retina. So the back of the 156 00:10:12,000 --> 00:10:15,079 Speaker 2: eye is the retina, which is this really nice two 157 00:10:15,160 --> 00:10:17,800 Speaker 2: D sheet of neurons and a big cable going into 158 00:10:17,840 --> 00:10:20,960 Speaker 2: the brain. So in some ways this is like a 159 00:10:21,000 --> 00:10:24,440 Speaker 2: really ideal interface to the brain. Evolution has done this 160 00:10:25,480 --> 00:10:28,439 Speaker 2: to give us vision. The first stop of the optic 161 00:10:28,480 --> 00:10:31,240 Speaker 2: nerve out of the eye is a structure in the 162 00:10:31,360 --> 00:10:34,360 Speaker 2: thalamus called the lateral janiculate nucleus, which is a very 163 00:10:34,360 --> 00:10:37,839 Speaker 2: deep structure in the brain. It's very old evolutionarily, and 164 00:10:38,440 --> 00:10:41,360 Speaker 2: there's about one point five million cells in the optic nerve. 165 00:10:41,520 --> 00:10:45,000 Speaker 2: There's about about the same number of cells in the thalamus. 166 00:10:45,240 --> 00:10:47,920 Speaker 2: And then from there you go out to a much 167 00:10:47,960 --> 00:10:51,400 Speaker 2: larger number of neurons in cortex called primary visual cortex. 168 00:10:51,760 --> 00:10:54,760 Speaker 2: And so if you want to supply vision to the brain, 169 00:10:55,080 --> 00:10:58,880 Speaker 2: in some sense synthetically your choices are really in the retina, 170 00:10:59,000 --> 00:11:02,920 Speaker 2: in the in the LGN, or in V one, and 171 00:11:03,720 --> 00:11:07,920 Speaker 2: everywhere past the optic nerve gets much much harder. Nobody 172 00:11:07,920 --> 00:11:10,440 Speaker 2: has ever really shown the restoration of form vision by 173 00:11:10,480 --> 00:11:13,800 Speaker 2: directly stimulating either the LGN or V one. I mean, 174 00:11:13,800 --> 00:11:16,000 Speaker 2: people haven't even really shown the restoration of form vision 175 00:11:16,000 --> 00:11:19,200 Speaker 2: simulating the optic nerve. The device that we're bringing to 176 00:11:19,240 --> 00:11:21,600 Speaker 2: market now that just recently finished to face three clinical 177 00:11:21,600 --> 00:11:24,720 Speaker 2: trial sits under the retina and stimulates a layer of 178 00:11:24,720 --> 00:11:27,679 Speaker 2: cells called the retinal bipolar cells, which are the first 179 00:11:27,720 --> 00:11:30,680 Speaker 2: cells past the rods and cones. And so this is 180 00:11:30,679 --> 00:11:33,400 Speaker 2: really in many ways the first opportunity to get a 181 00:11:33,480 --> 00:11:36,400 Speaker 2: visual signal back into the signaling pathways into the brain. 182 00:11:36,679 --> 00:11:38,920 Speaker 1: So let's back up. So how does your device work. 183 00:11:39,559 --> 00:11:42,520 Speaker 2: So the device is called Prima. It's a pretty cool idea. 184 00:11:42,600 --> 00:11:45,800 Speaker 2: So it's a tiny little solar panel chip about two 185 00:11:45,840 --> 00:11:49,360 Speaker 2: millimeters by two millimeters, so it's really very small, and 186 00:11:49,960 --> 00:11:51,679 Speaker 2: there's if you look. If you look at it, you'll 187 00:11:51,679 --> 00:11:53,920 Speaker 2: see all these little hex grids on it, these little 188 00:11:53,920 --> 00:11:56,040 Speaker 2: hex tiles. Each one of those TXT tiles is a 189 00:11:56,200 --> 00:11:59,080 Speaker 2: photodiode and an electrode. So what we do is you 190 00:11:59,120 --> 00:12:01,000 Speaker 2: implant this under the retina in the back of the 191 00:12:01,040 --> 00:12:03,440 Speaker 2: eye where the rods and concept degenerated, and the patient 192 00:12:03,480 --> 00:12:06,880 Speaker 2: wears glasses that have a laser projector on them, and 193 00:12:07,440 --> 00:12:11,120 Speaker 2: the laser projector projects the scene with laser energy onto 194 00:12:11,160 --> 00:12:13,800 Speaker 2: the implant in the back of the eye, and wherever 195 00:12:13,880 --> 00:12:17,920 Speaker 2: the laser energy is absorbed, it stimulates, and wherever there's 196 00:12:18,080 --> 00:12:20,560 Speaker 2: darkness in the scene that it doesn't. And so this 197 00:12:20,640 --> 00:12:23,000 Speaker 2: is a cool idea because there's no implanted battery, there's 198 00:12:23,040 --> 00:12:26,800 Speaker 2: no wires, there's no PCBs, there's no electronics other than 199 00:12:26,840 --> 00:12:29,720 Speaker 2: this tiny little chip. Because you send it both energy 200 00:12:29,840 --> 00:12:34,079 Speaker 2: and information simultaneously in the laser pulse and so this 201 00:12:34,160 --> 00:12:35,840 Speaker 2: is like, it's tough to imagine how you would do 202 00:12:35,880 --> 00:12:39,240 Speaker 2: this more simply than this. And when you look at 203 00:12:39,520 --> 00:12:42,000 Speaker 2: past devices, so like a little over a decade ago, 204 00:12:42,000 --> 00:12:45,040 Speaker 2: there was a company called Second Site that had a 205 00:12:45,080 --> 00:12:48,680 Speaker 2: retinal stimulator that is probably what people would be most 206 00:12:48,679 --> 00:12:52,160 Speaker 2: famous when people think about retinal prustcs. So it worked 207 00:12:52,280 --> 00:12:56,679 Speaker 2: very differently than than the science prima implant. First of all, 208 00:12:56,679 --> 00:12:58,480 Speaker 2: it targeted a different layer of cells. It targeted the 209 00:12:58,520 --> 00:13:02,000 Speaker 2: optic nerve rather than the bip our cells, which are 210 00:13:02,040 --> 00:13:05,840 Speaker 2: just much harder to stimulate naturalistically in this way. And 211 00:13:05,920 --> 00:13:09,199 Speaker 2: the second is because it wasn't it was a conventional 212 00:13:09,280 --> 00:13:12,679 Speaker 2: electrical implant. You had this big titanium box attached to 213 00:13:12,720 --> 00:13:14,760 Speaker 2: the side of the eye. You had cables going in 214 00:13:14,920 --> 00:13:18,439 Speaker 2: through through the eyeball to power it. This was a 215 00:13:18,440 --> 00:13:21,400 Speaker 2: four and a half hour surgery. Being able to just 216 00:13:21,480 --> 00:13:25,280 Speaker 2: put this little two by two millimeter chip of silicon 217 00:13:25,800 --> 00:13:28,760 Speaker 2: fully wirelessly under the eye with a little inserted tool 218 00:13:28,880 --> 00:13:31,480 Speaker 2: is a totally different game and it's and the clinical 219 00:13:31,480 --> 00:13:33,720 Speaker 2: trial results I think really speak for themselves. The first 220 00:13:33,800 --> 00:13:36,000 Speaker 2: time ever in the history of the world, as far 221 00:13:36,040 --> 00:13:37,839 Speaker 2: as we know, that blind patients have been able to 222 00:13:37,880 --> 00:13:38,320 Speaker 2: read again. 223 00:13:38,679 --> 00:13:42,160 Speaker 1: Oh that's so amazing. So all of the electronics and 224 00:13:42,200 --> 00:13:44,520 Speaker 1: all that stuff is in the glasses themselves, which are 225 00:13:44,559 --> 00:13:47,480 Speaker 1: capturing the scene like a camera and zapping it back 226 00:13:47,520 --> 00:13:49,520 Speaker 1: with a laser to the chip. 227 00:13:49,640 --> 00:13:52,520 Speaker 2: Yeah yeah, powering it, Yeah, basically like a solar cell. 228 00:13:52,679 --> 00:13:56,080 Speaker 1: Congratulations on all your progress with that. It's an incredible device. 229 00:13:56,400 --> 00:13:56,600 Speaker 3: Yeah. 230 00:13:56,800 --> 00:13:59,600 Speaker 2: And also I should say we didn't develop this from 231 00:13:59,640 --> 00:14:03,120 Speaker 2: a scratch ourselves. We acquired this from another company called Pixium, 232 00:14:03,320 --> 00:14:08,800 Speaker 2: which was based in Paris and had done has started 233 00:14:08,840 --> 00:14:11,560 Speaker 2: the clinical trial. Originally the technology came from a I 234 00:14:11,640 --> 00:14:14,800 Speaker 2: love at Stanford scientist Daniel Planker in the Electrical Engineering 235 00:14:14,800 --> 00:14:17,400 Speaker 2: department who came up with the idea, did the early 236 00:14:17,400 --> 00:14:20,280 Speaker 2: work at Stanford that was licensed by Pixium. They started 237 00:14:20,280 --> 00:14:23,120 Speaker 2: the clinical trial which we acquired and have finished and 238 00:14:23,200 --> 00:14:25,160 Speaker 2: finished the clinical trial and to bring in to market. 239 00:14:25,080 --> 00:14:27,960 Speaker 1: Right, I mean, I'm so I'm so jazzed that you 240 00:14:28,000 --> 00:14:30,480 Speaker 1: guys are doing that or bringing it to market and 241 00:14:30,520 --> 00:14:33,040 Speaker 1: making making this across the finish line. So that's what 242 00:14:33,080 --> 00:14:36,440 Speaker 1: you're doing in the retina for people who have lost vision. 243 00:14:37,680 --> 00:14:42,720 Speaker 1: Tell me what you're doing with uh reading from neurons? 244 00:14:42,880 --> 00:14:46,520 Speaker 1: So before just before we get there. So the challenge 245 00:14:46,840 --> 00:14:50,400 Speaker 1: with brain computer interfaces has always been, well several One 246 00:14:50,440 --> 00:14:52,520 Speaker 1: of them is that you know, mother nature has wrapped 247 00:14:52,560 --> 00:14:55,480 Speaker 1: the brain in this armored bunker plating, so it's hard 248 00:14:55,480 --> 00:14:57,360 Speaker 1: to get to. But then when you get in there, 249 00:14:57,760 --> 00:15:00,600 Speaker 1: you've got eighty six billion neurons and you have to 250 00:15:00,600 --> 00:15:03,840 Speaker 1: figure out who's saying what. And the traditional way to 251 00:15:03,880 --> 00:15:07,280 Speaker 1: do this is to dunk an electrode in there, which 252 00:15:07,680 --> 00:15:10,640 Speaker 1: really damages the tissue. So obviously people have been trying 253 00:15:10,680 --> 00:15:13,240 Speaker 1: to make electrodes thinner and thinner. But you've got an 254 00:15:13,280 --> 00:15:16,600 Speaker 1: idea that you're working on which is amazing. Tell us 255 00:15:16,640 --> 00:15:17,040 Speaker 1: about that. 256 00:15:17,440 --> 00:15:19,800 Speaker 2: Yeah, so that I can like, there's no free space 257 00:15:19,800 --> 00:15:22,240 Speaker 2: in the brain. The brain is wet, it squished together. 258 00:15:23,040 --> 00:15:28,680 Speaker 2: Evolution has really compressed as much as much as it 259 00:15:28,680 --> 00:15:30,960 Speaker 2: can into as small a space and an energy budget 260 00:15:30,960 --> 00:15:34,400 Speaker 2: as it possibly can, and so there's there has not 261 00:15:34,520 --> 00:15:37,040 Speaker 2: really left holes that we can take advantage of in there. 262 00:15:37,320 --> 00:15:40,760 Speaker 2: Evolution is extremely good at its job, and there's limits 263 00:15:40,760 --> 00:15:42,720 Speaker 2: how small you can make an electrode. You can't make 264 00:15:42,840 --> 00:15:48,360 Speaker 2: a like one nanometer wire because as a wire, just 265 00:15:48,400 --> 00:15:52,240 Speaker 2: any electrical wire gets smaller, the resistance increases there's just 266 00:15:52,280 --> 00:15:55,880 Speaker 2: real limits how small you can make a recording electrode 267 00:15:56,040 --> 00:15:59,320 Speaker 2: before you lose the ability to distinguish the signal that 268 00:15:59,520 --> 00:16:03,520 Speaker 2: you care the biological activity from the background noise. And 269 00:16:03,560 --> 00:16:07,040 Speaker 2: then on the stimulation side, this is actually worse because 270 00:16:07,080 --> 00:16:09,240 Speaker 2: there's real limits how small you can make a stimulating 271 00:16:09,240 --> 00:16:12,120 Speaker 2: electrode before you start splitting water in the brain and 272 00:16:12,120 --> 00:16:14,720 Speaker 2: producing hydrogen and oxygen, and like, you really don't want 273 00:16:14,720 --> 00:16:17,520 Speaker 2: to be doing this, And so we think about, like, 274 00:16:17,600 --> 00:16:20,080 Speaker 2: what does an ideal neural interface look like. I think 275 00:16:20,120 --> 00:16:23,080 Speaker 2: one of the high level intuitions that I started with was, Yeah, 276 00:16:23,080 --> 00:16:26,200 Speaker 2: the brain is encased in this dark vault of a skull, 277 00:16:26,720 --> 00:16:28,640 Speaker 2: but it has to communicate with the world. 278 00:16:28,960 --> 00:16:30,720 Speaker 3: There's like you, the. 279 00:16:30,720 --> 00:16:34,040 Speaker 2: Brain is not telepathically connected to the outside world. I mean, 280 00:16:34,040 --> 00:16:36,240 Speaker 2: it's also important to realize that you're not seeing the 281 00:16:36,280 --> 00:16:39,400 Speaker 2: world out there, right, You're only ever seeing in perceiving 282 00:16:39,440 --> 00:16:40,960 Speaker 2: information that's arrived at the brain. 283 00:16:41,360 --> 00:16:42,400 Speaker 3: And so how does it get there. 284 00:16:43,400 --> 00:16:45,280 Speaker 2: All of the information that flows in or out of 285 00:16:45,280 --> 00:16:47,760 Speaker 2: the brain flows through a relatively small number of cables. 286 00:16:48,200 --> 00:16:52,160 Speaker 2: There's twelve cranial nerves and thirty one spinal nerves. The 287 00:16:52,200 --> 00:16:55,960 Speaker 2: optic nerve is cranial nerve two. The vestibular cochlear nerve 288 00:16:55,960 --> 00:16:58,520 Speaker 2: that carries hearing in balance is also called nerve eight. 289 00:17:00,080 --> 00:17:03,920 Speaker 2: And kind of thinking about you've got this relatively small 290 00:17:03,960 --> 00:17:06,200 Speaker 2: number of wires, we can think about attaching to those 291 00:17:06,280 --> 00:17:09,320 Speaker 2: like we do for getting vision into the brain through 292 00:17:09,760 --> 00:17:13,560 Speaker 2: the remnants of nerve too. But this also kind of 293 00:17:13,720 --> 00:17:16,120 Speaker 2: got in the back of my mind going this idea 294 00:17:16,119 --> 00:17:18,560 Speaker 2: of can we grow a thirteenth cranial nerve that really 295 00:17:18,560 --> 00:17:21,679 Speaker 2: feels like the ideal neural interface. Biology has given us 296 00:17:21,680 --> 00:17:24,320 Speaker 2: other examples of fiber bundles that get information in and 297 00:17:24,320 --> 00:17:26,879 Speaker 2: out of the brain for really any purpose that the 298 00:17:26,880 --> 00:17:30,639 Speaker 2: brain needs. It, is it possible to add a thirteenth 299 00:17:30,680 --> 00:17:35,000 Speaker 2: biological wire that, instead of having an eye at the 300 00:17:35,000 --> 00:17:37,359 Speaker 2: other end or having a bunch of muscles at the 301 00:17:37,359 --> 00:17:39,480 Speaker 2: other end, had a USBC port basically, And so the 302 00:17:39,560 --> 00:17:41,600 Speaker 2: high level intuition here is like, what can we add 303 00:17:41,600 --> 00:17:43,520 Speaker 2: to the brain? How does the brain do this? Like 304 00:17:43,560 --> 00:17:45,639 Speaker 2: how does nature do this on its own? And the 305 00:17:45,680 --> 00:17:49,560 Speaker 2: answers it uses neurons, And so this kind of prompts 306 00:17:49,600 --> 00:17:51,600 Speaker 2: a question what happens if we add more neurons to 307 00:17:51,640 --> 00:17:54,000 Speaker 2: the brain and the answers, they grow in and wire 308 00:17:54,119 --> 00:18:00,240 Speaker 2: up and give you these bidirectional chemical synapses. And so 309 00:18:00,320 --> 00:18:03,280 Speaker 2: this has led to an approach that we call biohybrid 310 00:18:04,240 --> 00:18:07,800 Speaker 2: like biohybrid neural interfaces, and it really feels like it 311 00:18:07,840 --> 00:18:11,600 Speaker 2: has the scalability that many conventional methods don't. Now there 312 00:18:11,600 --> 00:18:14,680 Speaker 2: are alternatives to electrodes, So tell us what a biohybrid 313 00:18:14,760 --> 00:18:18,800 Speaker 2: interface is. So a biohybrid neural interface is when we 314 00:18:18,840 --> 00:18:22,840 Speaker 2: take heavily engineered stem cell derived neurons in a dish, 315 00:18:23,160 --> 00:18:27,159 Speaker 2: we load those into the electronic device, and then what 316 00:18:27,200 --> 00:18:30,560 Speaker 2: you place into the brain is just the ingrafted cells. 317 00:18:31,040 --> 00:18:34,040 Speaker 2: So we're not placing any metal or any like, no 318 00:18:34,119 --> 00:18:36,480 Speaker 2: electronic or mechanical component goes into the brain. 319 00:18:36,640 --> 00:18:37,679 Speaker 1: Instead, you're growing. 320 00:18:38,560 --> 00:18:42,960 Speaker 2: We basically graft these these cells onto the brain through 321 00:18:43,560 --> 00:18:46,640 Speaker 2: an appropriate starting point, and then those grow out form 322 00:18:46,680 --> 00:18:50,240 Speaker 2: new connections just as kind of more more of the brain. 323 00:18:50,880 --> 00:18:53,160 Speaker 1: And this is because mother nature is really good at 324 00:18:53,240 --> 00:18:55,959 Speaker 1: growing cells into groups of other cells and so on. 325 00:18:56,000 --> 00:18:57,280 Speaker 1: So you're taking advantage of that. 326 00:18:57,400 --> 00:18:59,680 Speaker 2: Yeah, we're letting biology do as much as the heavy 327 00:18:59,720 --> 00:19:01,960 Speaker 2: lifting as we can. Now, this creates other problems, but 328 00:19:02,560 --> 00:19:04,200 Speaker 2: the and I think smart people can say, well, now 329 00:19:04,240 --> 00:19:07,200 Speaker 2: you have a really complicated selling engineering problem to solve. 330 00:19:07,520 --> 00:19:09,760 Speaker 2: But if you can solve that in the meaningful way 331 00:19:09,840 --> 00:19:12,639 Speaker 2: that you have to, yeah, you can get biologies do 332 00:19:12,640 --> 00:19:13,439 Speaker 2: a lot of work for you. 333 00:19:13,720 --> 00:19:16,679 Speaker 1: Yeah. So these cells that you're putting on there and 334 00:19:16,880 --> 00:19:20,320 Speaker 1: growing in you have heavily engineered these cells. So tell 335 00:19:20,400 --> 00:19:21,240 Speaker 1: us about that. Yeah. 336 00:19:21,280 --> 00:19:22,399 Speaker 2: So there's a couple of things you need to do. 337 00:19:22,760 --> 00:19:24,520 Speaker 2: The first is it needs to be matched to the 338 00:19:24,560 --> 00:19:27,479 Speaker 2: immune system. Now, if you don't do this, it's you 339 00:19:27,520 --> 00:19:30,639 Speaker 2: can still make a cell therapy for a patient, but 340 00:19:30,760 --> 00:19:33,920 Speaker 2: you need to do it on an individualized patient basis 341 00:19:33,960 --> 00:19:37,840 Speaker 2: per patient. This is very expensive. It can take a 342 00:19:37,920 --> 00:19:39,800 Speaker 2: very long time to make edit the other edits that 343 00:19:39,840 --> 00:19:42,560 Speaker 2: we need. And so the first set of editing that 344 00:19:42,600 --> 00:19:46,160 Speaker 2: we do is to make the neurons hypommunogenic, meaning that 345 00:19:46,240 --> 00:19:48,520 Speaker 2: they don't bother the immune system when you put them 346 00:19:48,560 --> 00:19:49,240 Speaker 2: in a patient. 347 00:19:49,359 --> 00:19:50,119 Speaker 1: So how do you do that? 348 00:19:51,520 --> 00:19:54,480 Speaker 2: This is a much longer topic. There's these things called 349 00:19:54,520 --> 00:20:01,200 Speaker 2: major histoic compatibility complexes, and we need to suppress some 350 00:20:01,200 --> 00:20:05,560 Speaker 2: protein expression and force some other protein expression to basically 351 00:20:05,600 --> 00:20:09,719 Speaker 2: tell the immune system not to eat you and and 352 00:20:09,800 --> 00:20:11,040 Speaker 2: also that you are fine. 353 00:20:11,280 --> 00:20:13,359 Speaker 1: And how far along are you on that pathway? Is 354 00:20:13,400 --> 00:20:13,960 Speaker 1: that solved? 355 00:20:14,720 --> 00:20:16,480 Speaker 2: I mean, I wouldn't say that that's a solved problem. 356 00:20:16,520 --> 00:20:20,200 Speaker 2: I would say as a as a fields, there's several 357 00:20:20,359 --> 00:20:24,240 Speaker 2: standalone companies that their ip is hypo immune agenic stem cells, 358 00:20:24,720 --> 00:20:26,920 Speaker 2: and so we are i'd say, pretty close to the 359 00:20:26,960 --> 00:20:30,560 Speaker 2: state of the art in the field, but it's not perfect. 360 00:20:30,840 --> 00:20:31,000 Speaker 3: Now. 361 00:20:31,040 --> 00:20:33,720 Speaker 2: In the brain, the immune system tends to leave you 362 00:20:33,760 --> 00:20:36,680 Speaker 2: alone more than many other areas like this is, for example, 363 00:20:36,760 --> 00:20:38,240 Speaker 2: a lot of the work that's been done in gene 364 00:20:38,240 --> 00:20:40,600 Speaker 2: therapy so far has been done in the eye because 365 00:20:40,880 --> 00:20:43,240 Speaker 2: the immune system tends not to overreact in the eye 366 00:20:43,680 --> 00:20:45,960 Speaker 2: because when it does in a patient and a subject 367 00:20:45,960 --> 00:20:49,359 Speaker 2: goes blind, this historically is a bad thing. And so 368 00:20:49,760 --> 00:20:52,200 Speaker 2: there's some areas where you tend to get more autoimmune 369 00:20:52,200 --> 00:20:55,800 Speaker 2: reactions in some are Some are anatomy where this happens less. 370 00:20:55,920 --> 00:20:58,960 Speaker 2: The brain is one of the areas where because around 371 00:20:59,000 --> 00:21:02,080 Speaker 2: the time of the surgery you're treating them with systemic 372 00:21:03,320 --> 00:21:08,040 Speaker 2: immunosuppressants anyway, and then once the bloodburned bearer has healed, 373 00:21:08,359 --> 00:21:16,720 Speaker 2: it being approximately hypominogenic is probably fine. 374 00:21:25,280 --> 00:21:28,600 Speaker 1: Okay, So you do that to these cells, you engineer 375 00:21:28,640 --> 00:21:31,679 Speaker 1: them that way, and then you stick them on so 376 00:21:31,720 --> 00:21:33,840 Speaker 1: that they grow in. But of course you're keeping the 377 00:21:33,960 --> 00:21:37,160 Speaker 1: cell bodies outside and then what are you doing with those? 378 00:21:37,520 --> 00:21:39,520 Speaker 2: Yeah, So the next edit that we make is we 379 00:21:39,520 --> 00:21:43,760 Speaker 2: add a protein called a light gated ion channel also 380 00:21:43,800 --> 00:21:46,439 Speaker 2: as an option to these cells, which allows us to 381 00:21:46,520 --> 00:21:48,360 Speaker 2: fire them using light. 382 00:21:49,000 --> 00:21:50,240 Speaker 3: And this is pretty important. 383 00:21:50,280 --> 00:21:52,880 Speaker 2: So when we have so the device that the cell 384 00:21:52,960 --> 00:21:56,840 Speaker 2: is embedded in has two components around each cell. It 385 00:21:56,880 --> 00:21:59,720 Speaker 2: has a recording electrode which allows us to detect the 386 00:22:00,080 --> 00:22:02,720 Speaker 2: date of the cell, and it has a tiny little 387 00:22:02,720 --> 00:22:04,800 Speaker 2: micro LED kind of like you'd have in like your 388 00:22:04,840 --> 00:22:08,080 Speaker 2: phone screen next to the cell. And so when we 389 00:22:08,119 --> 00:22:10,280 Speaker 2: want to fire a neuron, we turn on the LED 390 00:22:10,560 --> 00:22:12,680 Speaker 2: and that depolarizes the cell and sends a pulse into 391 00:22:12,720 --> 00:22:15,920 Speaker 2: the brain. And when that neuron receives input from the brain, 392 00:22:15,960 --> 00:22:18,400 Speaker 2: because it's grown out both inputs and outputs, we can 393 00:22:18,400 --> 00:22:20,840 Speaker 2: detect that with the electrode, and so being able to 394 00:22:20,880 --> 00:22:24,240 Speaker 2: optically stimulate using light and electrically record use it in 395 00:22:24,440 --> 00:22:27,840 Speaker 2: an electrode. A capacity of electrode allows us to minimize 396 00:22:27,920 --> 00:22:30,600 Speaker 2: crosstalk between these so that we can do them both simultaneously. 397 00:22:31,280 --> 00:22:33,800 Speaker 1: And they're sandwiched in between this. So the cell body 398 00:22:33,840 --> 00:22:36,040 Speaker 1: is sandwiched in between the little light and the little 399 00:22:36,040 --> 00:22:38,640 Speaker 1: recording electrode. And so you can say, for this guy, 400 00:22:38,680 --> 00:22:40,200 Speaker 1: I want to turn him on now, and I want 401 00:22:40,240 --> 00:22:42,720 Speaker 1: to record what he's doing through time. 402 00:22:42,840 --> 00:22:45,479 Speaker 2: Yeah, it's not quite exactly one to one, but it's 403 00:22:45,520 --> 00:22:46,040 Speaker 2: pretty close. 404 00:22:46,200 --> 00:22:50,520 Speaker 1: Great, And how many neurons can you grow in there 405 00:22:50,920 --> 00:22:51,399 Speaker 1: at once? 406 00:22:51,600 --> 00:22:53,879 Speaker 2: Well, I mean this is so there's the number of 407 00:22:53,920 --> 00:22:56,399 Speaker 2: electrodes in the device, or number of channels in the device, 408 00:22:56,800 --> 00:22:58,720 Speaker 2: and then there's the number of cells, and then there's 409 00:22:58,720 --> 00:23:00,399 Speaker 2: the number of synapses that you get the brain. And 410 00:23:00,440 --> 00:23:03,880 Speaker 2: these are slightly different things. So the chips that we're 411 00:23:03,880 --> 00:23:09,160 Speaker 2: working with right now have four thousand electrodes per FIN, 412 00:23:09,840 --> 00:23:12,480 Speaker 2: and so we're thin is one of these one of 413 00:23:12,480 --> 00:23:14,760 Speaker 2: these little sandwiches. Yeah, and so it's actually it's really 414 00:23:14,760 --> 00:23:18,000 Speaker 2: eight thousand per because it's four thousand microelids and four 415 00:23:18,040 --> 00:23:20,359 Speaker 2: thousand electrodes. But we call this a four thousand channel 416 00:23:20,359 --> 00:23:24,040 Speaker 2: fin and we're working on stacks of these to scale 417 00:23:24,040 --> 00:23:27,879 Speaker 2: this up to hundreds of thousands of channels in one 418 00:23:28,240 --> 00:23:31,119 Speaker 2: in a couple millimeters by a couple millimeters. But I mean, 419 00:23:31,119 --> 00:23:34,480 Speaker 2: you could load this with a half a milli liter 420 00:23:34,600 --> 00:23:39,919 Speaker 2: of cells, which easily millions of cells, and those can 421 00:23:40,000 --> 00:23:43,280 Speaker 2: form many billions of synapses through the brain. 422 00:23:43,560 --> 00:23:46,240 Speaker 1: Do each of these cells form about let's say, ten 423 00:23:46,280 --> 00:23:47,720 Speaker 1: thousand synapses or. 424 00:23:47,960 --> 00:23:50,199 Speaker 2: I mean it's tough to count them. I mean you 425 00:23:50,240 --> 00:23:55,000 Speaker 2: can get there's the order of magnitude. People think it's 426 00:23:55,000 --> 00:23:57,159 Speaker 2: like maybe about a thousand synapses per cell, but I 427 00:23:57,160 --> 00:23:59,320 Speaker 2: mean we can't. These are tough to actually count. 428 00:23:59,440 --> 00:24:01,800 Speaker 1: Right, If you had a million neurons in there, you'd 429 00:24:01,800 --> 00:24:04,040 Speaker 1: get a billion synapses in the brain. 430 00:24:04,720 --> 00:24:05,840 Speaker 3: Yeah, back of the envelope. 431 00:24:05,880 --> 00:24:07,679 Speaker 1: Back of the envelope. And then so what you'd be 432 00:24:07,720 --> 00:24:11,720 Speaker 1: able to do is stimulate exactly as you want to, Okay, 433 00:24:11,720 --> 00:24:14,000 Speaker 1: fire number three hundred and seventy nine, now, fire number 434 00:24:14,000 --> 00:24:16,800 Speaker 1: one hundred and fifteen, and son, and then record the 435 00:24:16,880 --> 00:24:20,480 Speaker 1: activity going on there so you can read and write. 436 00:24:20,840 --> 00:24:22,359 Speaker 3: Yeah, so you can read and write. And it's a 437 00:24:22,359 --> 00:24:23,359 Speaker 3: fairly complex. 438 00:24:23,880 --> 00:24:26,800 Speaker 2: So you've got this transform between the activate the activities 439 00:24:26,840 --> 00:24:28,800 Speaker 2: and the cells and your device and what's going on 440 00:24:28,840 --> 00:24:30,800 Speaker 2: in the brain. We don't think of it in terms 441 00:24:30,840 --> 00:24:33,600 Speaker 2: of the single unit activity. In the beginning of the field, 442 00:24:33,840 --> 00:24:36,520 Speaker 2: we were really thinking in terms of single neurons, and 443 00:24:36,600 --> 00:24:40,000 Speaker 2: in the very beginning, the first experiments that were done 444 00:24:40,040 --> 00:24:43,760 Speaker 2: in animals didn't have a model of brain activity really 445 00:24:43,800 --> 00:24:45,600 Speaker 2: at all. What they did is they place electrodes in 446 00:24:45,640 --> 00:24:49,399 Speaker 2: the brain and then say, when this neuron fires, the 447 00:24:49,680 --> 00:24:51,920 Speaker 2: cursor should go up, and when this neuron fires, the 448 00:24:51,960 --> 00:24:53,960 Speaker 2: cursor should go down, and you just can learn to 449 00:24:53,960 --> 00:24:57,280 Speaker 2: separate these things. So the brain is very plastic under feedback. 450 00:24:57,800 --> 00:25:00,480 Speaker 2: Now that works for a very small number of channels, 451 00:25:00,480 --> 00:25:03,520 Speaker 2: and of course the subject isn't learning to modulate those 452 00:25:03,560 --> 00:25:06,760 Speaker 2: neurons specifically. They're actually modulating big groups of neurons around 453 00:25:06,800 --> 00:25:09,159 Speaker 2: where the electrode is, and so as you go to 454 00:25:09,240 --> 00:25:14,439 Speaker 2: higher level control, that doesn't really work anymore. But the 455 00:25:14,480 --> 00:25:22,000 Speaker 2: brain has these abstract informational representations of things like intended 456 00:25:22,000 --> 00:25:27,680 Speaker 2: motor activity or face recognition or other things that objects 457 00:25:27,720 --> 00:25:30,680 Speaker 2: that it thinks about, and so we're still at the 458 00:25:30,720 --> 00:25:33,639 Speaker 2: early stages of learning to use these devices. Really different 459 00:25:33,640 --> 00:25:38,280 Speaker 2: type of BCI, But how we think what we think 460 00:25:38,320 --> 00:25:41,040 Speaker 2: we're seeing is these cells would really join these cortical 461 00:25:41,119 --> 00:25:43,879 Speaker 2: representations and then just become part of part of the brain, 462 00:25:44,240 --> 00:25:46,480 Speaker 2: and you can do neuroscience on them like you would 463 00:25:46,480 --> 00:25:48,320 Speaker 2: any other part of the brain, except that the soel 464 00:25:48,359 --> 00:25:50,120 Speaker 2: body is right there in your device and really easy 465 00:25:50,119 --> 00:25:50,560 Speaker 2: to observe. 466 00:25:51,040 --> 00:25:53,760 Speaker 1: What are the biggest challenges that you're facing in terms 467 00:25:53,760 --> 00:25:56,800 Speaker 1: of bridging these digital systems and these biological systems. 468 00:25:57,119 --> 00:26:00,560 Speaker 2: There's many of the hard problems here are not the 469 00:26:00,600 --> 00:26:04,200 Speaker 2: really obvious sexy ones. In fact, actually I realized the 470 00:26:03,840 --> 00:26:06,200 Speaker 2: other other week that the very first piece of writing 471 00:26:06,200 --> 00:26:08,840 Speaker 2: that I put on the Internet was kind of this 472 00:26:09,000 --> 00:26:11,159 Speaker 2: like sophomore literally as a software I call it, but 473 00:26:11,600 --> 00:26:14,879 Speaker 2: the software grant about how like back in circa two 474 00:26:14,920 --> 00:26:17,560 Speaker 2: thousand and eight, everyone felt like the hard problems here 475 00:26:17,560 --> 00:26:20,760 Speaker 2: were understanding the neural code and like real science to 476 00:26:20,880 --> 00:26:24,679 Speaker 2: study these like deep neuroscience questions, and it was kind 477 00:26:24,720 --> 00:26:26,480 Speaker 2: of for the technicians to figure out how to get 478 00:26:26,480 --> 00:26:28,439 Speaker 2: the electrodes in the brain, whereas actually the problem is 479 00:26:28,440 --> 00:26:30,960 Speaker 2: how do you get these electrodes in the brain. And 480 00:26:31,480 --> 00:26:33,600 Speaker 2: certainly the neuroscience has advanced a lot, and the neuroscience 481 00:26:33,640 --> 00:26:35,840 Speaker 2: is very cool, but a lot of the problems here 482 00:26:35,840 --> 00:26:39,119 Speaker 2: are things like packaging, which is a fancy term for 483 00:26:39,760 --> 00:26:42,880 Speaker 2: when you place an electronic device in the body, it's 484 00:26:42,920 --> 00:26:45,000 Speaker 2: going to be it's going to be attacked, it's going 485 00:26:45,040 --> 00:26:48,600 Speaker 2: to be degraded, it gets encapsulated in scar tissue that 486 00:26:48,960 --> 00:26:53,119 Speaker 2: neurons are pulled away from you. There's these very harsh 487 00:26:53,520 --> 00:26:56,280 Speaker 2: chemical environments that try to attack and destroy your device. 488 00:26:57,040 --> 00:26:59,479 Speaker 2: It's important to realize that there are no truly passive 489 00:26:59,520 --> 00:27:02,480 Speaker 2: surfaces anywhere in the body, like even bone is constantly 490 00:27:02,480 --> 00:27:05,960 Speaker 2: getting remodeled and turned over and regenerated, and so when 491 00:27:06,000 --> 00:27:09,000 Speaker 2: you place one of these non regenerating device in the body, 492 00:27:09,000 --> 00:27:11,480 Speaker 2: it's going to be attacked. And so now we have 493 00:27:11,600 --> 00:27:13,920 Speaker 2: we have much better materials than we did ten years ago, 494 00:27:14,440 --> 00:27:20,760 Speaker 2: specifically things like silicon carbide, which is really annoying material 495 00:27:20,800 --> 00:27:22,880 Speaker 2: to work with, but a very good encapsulant that does 496 00:27:22,920 --> 00:27:24,920 Speaker 2: not degrade in the body in the same way as 497 00:27:24,920 --> 00:27:28,199 Speaker 2: these older polymer king capsulans do. It's like, if you 498 00:27:28,200 --> 00:27:29,720 Speaker 2: look at the history of Prima part of like how 499 00:27:29,760 --> 00:27:30,840 Speaker 2: did PIXI them, the company we. 500 00:27:30,800 --> 00:27:31,719 Speaker 3: Bought this get here? 501 00:27:31,880 --> 00:27:35,320 Speaker 2: They actually had an approved device in the I want 502 00:27:35,320 --> 00:27:37,800 Speaker 2: to say twenty fourteen twenty fifteen called Iris, which was 503 00:27:37,840 --> 00:27:40,879 Speaker 2: a different retinal prosthesis and it worked very differently. It 504 00:27:40,920 --> 00:27:45,159 Speaker 2: had a conventional electronics package, It required a battery, but 505 00:27:45,400 --> 00:27:47,920 Speaker 2: it was it was got on market and then was 506 00:27:48,160 --> 00:27:52,480 Speaker 2: withdrawn and it was withdrawn because of packaging failures. Basically, 507 00:27:52,520 --> 00:27:56,040 Speaker 2: the device didn't have an acceptable lifespan in human patients 508 00:27:56,080 --> 00:27:59,480 Speaker 2: once on the market, and that was like they were 509 00:27:59,720 --> 00:28:01,720 Speaker 2: using materials that were available at the time, which was 510 00:28:01,800 --> 00:28:03,880 Speaker 2: before we figured out, as I feel, how to work 511 00:28:03,920 --> 00:28:07,479 Speaker 2: with things like silicon carbide. And that is an example 512 00:28:07,520 --> 00:28:09,760 Speaker 2: of a problem that enabled Prima to work. So Prima 513 00:28:09,880 --> 00:28:12,679 Speaker 2: is a full carbid encapsulation and it should last I 514 00:28:12,680 --> 00:28:15,280 Speaker 2: mean there's now dat out to six years and some 515 00:28:15,280 --> 00:28:16,840 Speaker 2: patients and it should outlast these patients. 516 00:28:16,840 --> 00:28:17,719 Speaker 3: It should last decades. 517 00:28:18,240 --> 00:28:18,600 Speaker 1: Amazing. 518 00:28:18,640 --> 00:28:20,400 Speaker 2: And so that's an example of like a big area 519 00:28:20,440 --> 00:28:22,119 Speaker 2: of progress in the last few years that people wouldn't 520 00:28:22,119 --> 00:28:22,520 Speaker 2: really think of. 521 00:28:23,400 --> 00:28:27,080 Speaker 1: And so, what are some surprising findings or unexpected obstacles 522 00:28:27,119 --> 00:28:30,960 Speaker 1: that you've run into while doing let's say the biohybrid electrodes. 523 00:28:31,840 --> 00:28:35,399 Speaker 2: I mean, biology is just it's when it works, it 524 00:28:35,400 --> 00:28:38,120 Speaker 2: can do a lot of things that we humanity is 525 00:28:38,200 --> 00:28:42,040 Speaker 2: just not at that level of capability yet. But also 526 00:28:42,520 --> 00:28:47,280 Speaker 2: in neural engineering or either whether that means systems neuroscience 527 00:28:47,480 --> 00:28:52,320 Speaker 2: or BCI, you'll start in mice and then maybe you'll 528 00:28:52,360 --> 00:28:54,640 Speaker 2: work in an intermediate species like pigs and then eventually 529 00:28:54,720 --> 00:28:57,000 Speaker 2: end up in monkeys, then end up in humans. And 530 00:28:57,120 --> 00:29:02,880 Speaker 2: when you have an electrode or even something like optogenetics 531 00:29:03,440 --> 00:29:05,520 Speaker 2: that works basically the same in mice as a dozen 532 00:29:05,520 --> 00:29:08,480 Speaker 2: monkeys as it does in humans. When you're engrafting neurons 533 00:29:08,520 --> 00:29:10,360 Speaker 2: into the brain, I mean, there's a big difference between 534 00:29:10,360 --> 00:29:14,520 Speaker 2: mouse neurons and human neurons and macac neurons are different 535 00:29:14,520 --> 00:29:17,240 Speaker 2: thing entirely, and so you end up having to redo 536 00:29:17,280 --> 00:29:19,280 Speaker 2: a bunch of this work in each species that you 537 00:29:19,320 --> 00:29:22,240 Speaker 2: work in, and so there's every time we switch species, 538 00:29:22,280 --> 00:29:24,800 Speaker 2: there's a lot to relearn. And fifteen years ago now 539 00:29:25,080 --> 00:29:27,520 Speaker 2: probably something like that, there is a major discovery of 540 00:29:27,920 --> 00:29:31,280 Speaker 2: the ability to turn any cell into a stem cell. Again, 541 00:29:31,320 --> 00:29:34,000 Speaker 2: this was a discovery called induced plury potency, won the 542 00:29:34,000 --> 00:29:37,640 Speaker 2: Nobel Prize a while ago, and that works really well 543 00:29:37,680 --> 00:29:41,440 Speaker 2: in rodents, it works really well in human cells. Turning 544 00:29:41,640 --> 00:29:46,920 Speaker 2: a macaque skin cell into an ips is like there's 545 00:29:46,960 --> 00:29:48,600 Speaker 2: just a bunch of little tricks that don't work as well. 546 00:29:49,040 --> 00:29:51,440 Speaker 2: And so the biology is pretty deep in all of 547 00:29:51,480 --> 00:29:51,920 Speaker 2: these areas. 548 00:29:51,920 --> 00:29:55,440 Speaker 1: It's surprising that those are different, but you know, just 549 00:29:55,440 --> 00:29:57,320 Speaker 1: given the evolutionary shared history. 550 00:29:57,400 --> 00:29:59,400 Speaker 2: But yes, yeah, I mean there's a lot that's conserved, 551 00:29:59,400 --> 00:30:00,960 Speaker 2: but there's also a lot of little things that are 552 00:30:00,960 --> 00:30:01,600 Speaker 2: slightly different. 553 00:30:01,680 --> 00:30:04,400 Speaker 1: Yeah, quite right. So big congratulations on where Prima is 554 00:30:04,480 --> 00:30:08,480 Speaker 1: right now. That's so exciting. On the biohybrid electrodes, it 555 00:30:08,520 --> 00:30:10,719 Speaker 1: said you have growing neurons into the brain and then 556 00:30:10,760 --> 00:30:12,880 Speaker 1: being able to read and write that way. When do 557 00:30:12,920 --> 00:30:14,680 Speaker 1: you think that's going to be ready in humans? What's 558 00:30:14,720 --> 00:30:15,200 Speaker 1: your prediction. 559 00:30:16,280 --> 00:30:18,840 Speaker 2: I think that the first human ingraftment will happen around 560 00:30:18,880 --> 00:30:23,360 Speaker 2: twenty thirty, okay, so I think like probably five years, okay. 561 00:30:24,040 --> 00:30:26,360 Speaker 1: And what is the first thing you're going to tackle 562 00:30:26,400 --> 00:30:27,520 Speaker 1: once it gets into humans? 563 00:30:28,040 --> 00:30:30,600 Speaker 2: Well, I mean it should be it's a communication device, 564 00:30:30,840 --> 00:30:34,320 Speaker 2: and so motory coding, speech to coding, all of that 565 00:30:34,360 --> 00:30:36,440 Speaker 2: should be possible. And so I think in the near 566 00:30:36,520 --> 00:30:39,520 Speaker 2: term it's you're looking at the figure of merit for 567 00:30:39,520 --> 00:30:43,080 Speaker 2: any brain computer interface for communication is a bandwidth measured 568 00:30:43,080 --> 00:30:47,000 Speaker 2: in bits per second. The record for keyboard and mouse 569 00:30:47,360 --> 00:30:50,120 Speaker 2: kind of low dimension motority coding is about seven bits 570 00:30:50,120 --> 00:30:53,760 Speaker 2: per second, which is I think neuralinks current participants. There's 571 00:30:53,760 --> 00:30:58,160 Speaker 2: a group at UC Davis led by saying Nick Card 572 00:30:58,280 --> 00:31:01,760 Speaker 2: and Sergey Staviski, who recently showed speech to coding from 573 00:31:01,800 --> 00:31:03,880 Speaker 2: pridal cortex that gets about twenty to twenty five e 574 00:31:03,880 --> 00:31:08,960 Speaker 2: bits per second. Human language is routinely rated at forty 575 00:31:09,040 --> 00:31:10,680 Speaker 2: bits per second, So you think that you can as 576 00:31:10,720 --> 00:31:12,520 Speaker 2: them tote towards that, and so I think in the 577 00:31:12,560 --> 00:31:14,160 Speaker 2: near term what we're looking for is a forty bit 578 00:31:14,200 --> 00:31:19,000 Speaker 2: per second communication prosthesis. Longer term, this is where neural 579 00:31:19,040 --> 00:31:22,360 Speaker 2: engineering and BCI diverge a little bit, and there's a 580 00:31:22,360 --> 00:31:24,480 Speaker 2: lot of interest internally at looking at how is this 581 00:31:24,520 --> 00:31:28,560 Speaker 2: applicable in stroke or other areas where you've lost cells 582 00:31:28,880 --> 00:31:32,480 Speaker 2: where conventional BCI techniques really won't work in the same way, 583 00:31:33,240 --> 00:31:36,680 Speaker 2: and potentially even organic nerd degenerative diseases, But those are 584 00:31:36,760 --> 00:31:39,120 Speaker 2: very hard and I don't want over promise on the 585 00:31:39,120 --> 00:31:39,880 Speaker 2: timeline there. 586 00:31:40,960 --> 00:31:43,280 Speaker 1: Now, if we were just going to blue sky here, 587 00:31:43,480 --> 00:31:47,080 Speaker 1: part of the mythology about BCIs is that at some 588 00:31:47,200 --> 00:31:50,400 Speaker 1: point everyone will have one of these for summer, you know, 589 00:31:50,440 --> 00:31:52,560 Speaker 1: for communicating faster with their cell phone or their computer 590 00:31:52,600 --> 00:31:54,920 Speaker 1: or whatever. To what degree do you think that's hype 591 00:31:55,120 --> 00:31:59,000 Speaker 1: versus let's imagine one hundred years from now, where do 592 00:31:59,080 --> 00:32:01,160 Speaker 1: you realistically see think it's going to be in terms 593 00:32:01,160 --> 00:32:02,720 Speaker 1: of the amount of market it has. 594 00:32:02,920 --> 00:32:05,400 Speaker 2: Yeah, I mean one hundred years from now. I have 595 00:32:05,520 --> 00:32:08,480 Speaker 2: this event horizon somewhere between twenty thirty and twenty thirty 596 00:32:08,520 --> 00:32:10,680 Speaker 2: five now that I just can't see beyond, and kind 597 00:32:10,680 --> 00:32:12,600 Speaker 2: of for my entire life, I always kind kind of 598 00:32:12,640 --> 00:32:15,840 Speaker 2: like see the future, and we are clearly in the 599 00:32:15,880 --> 00:32:19,040 Speaker 2: takeoff era now, and this is not I don't think 600 00:32:19,040 --> 00:32:21,400 Speaker 2: I'm saying anything that contrarian, at least in Silicon Valley, 601 00:32:21,440 --> 00:32:24,160 Speaker 2: but one hundred years from now is almost impossible for 602 00:32:24,200 --> 00:32:24,920 Speaker 2: me to imagine. 603 00:32:25,240 --> 00:32:25,480 Speaker 3: Now. 604 00:32:25,520 --> 00:32:28,240 Speaker 2: With that said, I don't think that healthy forty year 605 00:32:28,240 --> 00:32:30,239 Speaker 2: olds are going to be getting holes drilled in their 606 00:32:30,240 --> 00:32:33,320 Speaker 2: skull anytime soon. My view is that it'll be a 607 00:32:33,440 --> 00:32:38,120 Speaker 2: long time before these things are really augmentative, much less 608 00:32:38,120 --> 00:32:42,240 Speaker 2: elective procedures. But everybody eventually becomes a patient. There's some 609 00:32:42,360 --> 00:32:45,480 Speaker 2: point as you get older. For example, the main indication 610 00:32:45,520 --> 00:32:50,360 Speaker 2: of prima is age related macular degeneration, which is very 611 00:32:50,360 --> 00:32:53,400 Speaker 2: common and if someone lives into their late seventies or eighties, 612 00:32:54,280 --> 00:32:57,800 Speaker 2: is actually pretty prevalent, and so for many many of 613 00:32:57,840 --> 00:33:00,600 Speaker 2: these things. Eventually there will come a time when it 614 00:33:00,640 --> 00:33:03,560 Speaker 2: makes sense. I mean we consider retinal prostisis and cochlear 615 00:33:03,840 --> 00:33:09,440 Speaker 2: prosteces also BCIs when I look at, say twenty years 616 00:33:09,480 --> 00:33:14,200 Speaker 2: from now, the things that I that are very much research. 617 00:33:14,320 --> 00:33:16,400 Speaker 2: This is not a thing that's happening in the next 618 00:33:16,440 --> 00:33:19,840 Speaker 2: five years. But if you can get a neural interface 619 00:33:20,200 --> 00:33:23,920 Speaker 2: with the bandwidth of that say like the two hemispheres 620 00:33:23,920 --> 00:33:26,240 Speaker 2: are connected, which is about one hundred million fibers on 621 00:33:26,280 --> 00:33:28,520 Speaker 2: both sides that project across the midline to connect the 622 00:33:28,560 --> 00:33:32,520 Speaker 2: two hemispheres of your brain into a single thing. If 623 00:33:32,560 --> 00:33:35,280 Speaker 2: you can get something of that bandwidth, which is probably 624 00:33:35,400 --> 00:33:39,760 Speaker 2: only tens of megabits, then this takes you into really 625 00:33:39,800 --> 00:33:42,760 Speaker 2: interesting territory about really being able to redraw the borders 626 00:33:42,760 --> 00:33:45,560 Speaker 2: around brains and gets at this thing called the binding problem. 627 00:33:46,160 --> 00:33:48,880 Speaker 3: And that feels less than twenty years away for me. 628 00:33:48,960 --> 00:33:52,200 Speaker 2: This feels not like the next five years, but not 629 00:33:52,200 --> 00:33:54,720 Speaker 2: not to the distant future like within people's lifespans today. 630 00:33:55,040 --> 00:33:56,960 Speaker 1: So let's stile click on that tell us about the 631 00:33:57,000 --> 00:33:58,880 Speaker 1: binding problem and how you think this addresses that. 632 00:34:00,000 --> 00:34:01,400 Speaker 2: But I mean, I don't have a solution for the 633 00:34:01,640 --> 00:34:04,000 Speaker 2: binding problem. Is if the brain is made up of 634 00:34:04,000 --> 00:34:05,800 Speaker 2: a lot of different neurons and a lot of different 635 00:34:05,800 --> 00:34:08,680 Speaker 2: areas kind of connected together. Why do we Where does 636 00:34:08,760 --> 00:34:13,080 Speaker 2: this unified perception come from? You? You see the world, 637 00:34:13,080 --> 00:34:14,880 Speaker 2: you can think about it, you hear things. All of 638 00:34:14,880 --> 00:34:17,040 Speaker 2: this is fit together into a coherent hole for you. 639 00:34:17,120 --> 00:34:20,080 Speaker 1: When the bluebird flies past you, the blue doesn't come 640 00:34:20,120 --> 00:34:22,000 Speaker 1: off of the bird, and the chirping doesn't seem like 641 00:34:22,040 --> 00:34:24,840 Speaker 1: it's coming from somewhere else. It seems like a unified object. Yeah, exactly, 642 00:34:24,920 --> 00:34:27,399 Speaker 1: even though even though blue is processed apparently in one 643 00:34:27,440 --> 00:34:29,320 Speaker 1: part of your brain and the motion another part, and 644 00:34:29,400 --> 00:34:31,200 Speaker 1: the chirping in a different part. Yeah, okay. 645 00:34:31,360 --> 00:34:34,080 Speaker 2: And so in some there's some sense in which almost 646 00:34:34,120 --> 00:34:38,319 Speaker 2: all communication is about creating correlations between brains. There's we're 647 00:34:38,360 --> 00:34:41,360 Speaker 2: having a conversation right now. There's concept spaces in my 648 00:34:41,440 --> 00:34:45,000 Speaker 2: brain that are being active that I developed from education, 649 00:34:45,120 --> 00:34:48,680 Speaker 2: like learning English learning, math learning, science, doing these things, 650 00:34:49,200 --> 00:34:53,440 Speaker 2: and I can serialize these neural activations to vibrations over 651 00:34:53,480 --> 00:34:56,080 Speaker 2: the air, send over to you, receive through your ears 652 00:34:56,280 --> 00:34:59,120 Speaker 2: that then activate these correlations in your brain that allow 653 00:34:59,200 --> 00:35:04,120 Speaker 2: us to share these concepts. But we don't. Our brains 654 00:35:04,160 --> 00:35:09,680 Speaker 2: don't become one thing. And so there's there's some point 655 00:35:09,800 --> 00:35:12,239 Speaker 2: between the types of correlations that you get between the 656 00:35:12,239 --> 00:35:14,480 Speaker 2: hemispheres of a brain and the types of correlations that 657 00:35:14,480 --> 00:35:18,000 Speaker 2: we get between brains that are in dialogue. And where 658 00:35:18,040 --> 00:35:21,680 Speaker 2: does this Where is that crossing point? We don't know today, 659 00:35:22,560 --> 00:35:26,840 Speaker 2: but I think that biohybrid devices have the potential to 660 00:35:26,520 --> 00:35:29,440 Speaker 2: get close to there, and that takes us to really 661 00:35:29,680 --> 00:35:34,320 Speaker 2: different regimes than kind of conventional VCI technology. 662 00:35:34,440 --> 00:35:35,919 Speaker 1: Let me just make sure I understand what you said. 663 00:35:35,920 --> 00:35:39,839 Speaker 1: So the idea is, if you're reading and writing from 664 00:35:39,880 --> 00:35:42,920 Speaker 1: my brain and from your brain, we can get closer 665 00:35:42,960 --> 00:35:44,120 Speaker 1: to being a single brain. 666 00:35:45,239 --> 00:35:47,040 Speaker 2: Well, like, yeah, the question is like, where does that happen? 667 00:35:47,239 --> 00:35:50,719 Speaker 2: What makes I mean? People back in This has done 668 00:35:50,800 --> 00:35:53,640 Speaker 2: less commonly now, but it was never really done that commonly. 669 00:35:53,719 --> 00:35:56,120 Speaker 2: But people used to cut the connection between the two 670 00:35:56,160 --> 00:35:58,799 Speaker 2: hemispheres of the brain to treat epilepsy. You could prevent 671 00:35:58,800 --> 00:36:02,440 Speaker 2: a seizure from spreading from one to the other. And 672 00:36:03,320 --> 00:36:06,719 Speaker 2: those split brain patients were really interesting to study. 673 00:36:06,440 --> 00:36:07,520 Speaker 3: Because you could. 674 00:36:09,640 --> 00:36:12,879 Speaker 2: You could ask kind of the right hand a question 675 00:36:12,920 --> 00:36:15,160 Speaker 2: which would go to the left hemisphere, and then you 676 00:36:15,160 --> 00:36:18,000 Speaker 2: could ask the other hand, which was coming from the 677 00:36:18,040 --> 00:36:21,360 Speaker 2: other hemisphere, to kind of answer, and you get the 678 00:36:21,440 --> 00:36:23,759 Speaker 2: sense that there's two agents going on. 679 00:36:23,760 --> 00:36:24,279 Speaker 3: In one head. 680 00:36:24,480 --> 00:36:25,640 Speaker 1: Yeah, one in each hemisphere. 681 00:36:25,680 --> 00:36:28,680 Speaker 2: And so if you take that then the opposite direction, 682 00:36:28,880 --> 00:36:32,480 Speaker 2: what do you get? I think is really interesting. 683 00:36:32,800 --> 00:36:36,120 Speaker 1: You're saying, put four hemispheres together and yeah you get Yeah, 684 00:36:36,560 --> 00:36:38,840 Speaker 1: Now who would do this? Who would volunteer for example, 685 00:36:39,760 --> 00:36:41,080 Speaker 1: two spouses for example? 686 00:36:41,200 --> 00:36:41,880 Speaker 3: Yeah, exactly. 687 00:36:41,920 --> 00:36:46,399 Speaker 2: So I think this is in the beginning this this 688 00:36:46,440 --> 00:36:49,040 Speaker 2: is going to be something like you've got like a 689 00:36:49,120 --> 00:36:52,200 Speaker 2: long married couple one has a terminal disease. Can you 690 00:36:52,280 --> 00:36:56,320 Speaker 2: make the loss of that brain like having a stroke 691 00:36:56,360 --> 00:36:59,359 Speaker 2: you recover from, rather than the that rather than lights out? 692 00:37:00,160 --> 00:37:03,759 Speaker 1: Oh wow, and we double click on that story. What 693 00:37:03,800 --> 00:37:04,959 Speaker 1: would the narrative be there? 694 00:37:05,120 --> 00:37:09,279 Speaker 2: Well, I mean you get so the if you have 695 00:37:09,480 --> 00:37:12,200 Speaker 2: if you can build these super organisms and get kind 696 00:37:12,200 --> 00:37:15,319 Speaker 2: of equilibration of representation over some extended period of time. 697 00:37:16,080 --> 00:37:19,600 Speaker 2: I mean, people already store memories in their spouse's brains 698 00:37:19,600 --> 00:37:21,640 Speaker 2: that then they can access and recall later. Right, this 699 00:37:21,680 --> 00:37:24,040 Speaker 2: is about creating correlations between brains, and so there's some 700 00:37:25,000 --> 00:37:27,319 Speaker 2: they suspect that there's some nonlinearity in there where you 701 00:37:27,360 --> 00:37:29,960 Speaker 2: get something really different, but of course we don't know 702 00:37:30,000 --> 00:37:32,799 Speaker 2: exactly where that is yet. I mean, this is a 703 00:37:32,800 --> 00:37:36,400 Speaker 2: tricky field because there's a fine line between doing very 704 00:37:37,239 --> 00:37:39,440 Speaker 2: like we're right now in the process of preparing twelve 705 00:37:39,560 --> 00:37:43,920 Speaker 2: hundred pages of regulatory documentation that is like very nuanced 706 00:37:43,960 --> 00:37:47,080 Speaker 2: in exactly how you do these tests to verify these 707 00:37:47,120 --> 00:37:49,479 Speaker 2: like things that have passed clinical trials that are in 708 00:37:49,680 --> 00:37:53,799 Speaker 2: almost fifty patients in six countries, and then you kind 709 00:37:53,800 --> 00:37:56,360 Speaker 2: of play some of these technologies out not even that 710 00:37:56,440 --> 00:37:58,360 Speaker 2: long five ten years, and you sound like a lunatic. 711 00:37:58,840 --> 00:38:01,200 Speaker 2: But that's part of why this is such an exciting field. 712 00:38:01,239 --> 00:38:04,680 Speaker 1: I think, right, what would you see so I know, 713 00:38:04,800 --> 00:38:07,440 Speaker 1: I know the event horizon for both of us is, 714 00:38:07,480 --> 00:38:10,040 Speaker 1: you know, not more much more than a decade out. 715 00:38:10,080 --> 00:38:13,680 Speaker 1: But what would you see is the societal benefits that 716 00:38:13,680 --> 00:38:16,560 Speaker 1: could happen from this, you know, at whatever time scale, 717 00:38:17,880 --> 00:38:20,920 Speaker 1: for example, connecting brains or something. Have you thought about 718 00:38:20,960 --> 00:38:23,719 Speaker 1: what that could what that would turn into not just responuses, 719 00:38:23,760 --> 00:38:24,520 Speaker 1: but for society. 720 00:38:25,080 --> 00:38:26,800 Speaker 2: I mean at the end of that is this idea 721 00:38:26,800 --> 00:38:30,319 Speaker 2: of substrate independence, which is the thing like when I, 722 00:38:30,640 --> 00:38:32,560 Speaker 2: like I see a person, there's two parts of this. 723 00:38:32,640 --> 00:38:36,480 Speaker 2: There's the there's the robot, and there's an agent. And 724 00:38:37,080 --> 00:38:39,040 Speaker 2: I'm going to be pretty pretty disappointed if I get 725 00:38:39,120 --> 00:38:42,400 Speaker 2: murdered by my pancreas, which is like basically a support 726 00:38:42,440 --> 00:38:46,720 Speaker 2: structure for like keeping the agent going. And so there's 727 00:38:47,000 --> 00:38:48,799 Speaker 2: I think this takes us to Okay, if we're serious 728 00:38:48,800 --> 00:38:51,000 Speaker 2: about exploring the universe, I think we have to adapt 729 00:38:51,040 --> 00:38:54,000 Speaker 2: ourselves the environment rather than bringing little pressurized bottles of 730 00:38:54,000 --> 00:38:56,759 Speaker 2: Earth with us everywhere we go. Because our like once 731 00:38:56,840 --> 00:38:58,640 Speaker 2: great grandparents grew up on a planet that happened to 732 00:38:58,680 --> 00:39:01,040 Speaker 2: have those things, and so I think this is like 733 00:39:01,200 --> 00:39:02,520 Speaker 2: very profound technology. 734 00:39:02,640 --> 00:39:05,960 Speaker 1: So substrate independence, just for the audience, means getting off 735 00:39:06,000 --> 00:39:08,880 Speaker 1: of this wet biological stuff and onto something more robust, 736 00:39:08,960 --> 00:39:10,960 Speaker 1: like a silicon chip or something. In other words, getting 737 00:39:12,000 --> 00:39:17,160 Speaker 1: your mind into something that can survive space travel. 738 00:39:16,920 --> 00:39:19,839 Speaker 2: Which could be other biological brains, or it could be 739 00:39:20,160 --> 00:39:25,520 Speaker 2: an engineered system. Brains are composed of ordinary matter assembled 740 00:39:25,520 --> 00:39:26,719 Speaker 2: by the rules of chemistry. 741 00:39:26,960 --> 00:39:28,080 Speaker 3: There's no magic in there. 742 00:39:28,080 --> 00:39:31,800 Speaker 2: They're very complicated and we don't have obviously complete explanations 743 00:39:31,840 --> 00:39:35,160 Speaker 2: for how they work. But they're ultimately physical systems, and 744 00:39:35,239 --> 00:39:37,800 Speaker 2: so there's something that they're doing that's producing this experience 745 00:39:37,840 --> 00:39:39,320 Speaker 2: that ultimately must be explainable. 746 00:39:55,239 --> 00:39:58,120 Speaker 1: And so what you're doing with the electrodes, the biohybrid 747 00:39:58,120 --> 00:40:01,799 Speaker 1: electrodes in the brain. How does this lead to substrate independence? 748 00:40:02,320 --> 00:40:04,719 Speaker 2: Well, the idea is that if you can get like, 749 00:40:04,800 --> 00:40:07,640 Speaker 2: if you can really, in some profound sense, lose track 750 00:40:07,640 --> 00:40:11,160 Speaker 2: of where one brain ends in another begins, then where 751 00:40:11,200 --> 00:40:12,680 Speaker 2: does this take you. I have no idea what that 752 00:40:12,760 --> 00:40:14,759 Speaker 2: experience will feel like, but I'm pretty confident that that 753 00:40:14,800 --> 00:40:16,400 Speaker 2: device is going to get made in the next decade. 754 00:40:16,440 --> 00:40:17,720 Speaker 3: And this is research. 755 00:40:17,760 --> 00:40:20,200 Speaker 2: This is not a there's nothing to sell here yet, 756 00:40:20,280 --> 00:40:23,680 Speaker 2: but it's the type of frontier that is enabled by 757 00:40:23,760 --> 00:40:26,000 Speaker 2: the types of devices that are getting made now and 758 00:40:26,040 --> 00:40:29,000 Speaker 2: that and there's I think enough near term commercial revenue 759 00:40:29,040 --> 00:40:31,759 Speaker 2: from things like the from the visual pres thesis to 760 00:40:32,280 --> 00:40:33,000 Speaker 2: fund this. 761 00:40:33,000 --> 00:40:33,760 Speaker 3: This stuff happening. 762 00:40:34,080 --> 00:40:36,640 Speaker 1: So if you're able to read from the brain, then 763 00:40:36,680 --> 00:40:38,640 Speaker 1: you can take that data and put it into a 764 00:40:38,680 --> 00:40:40,040 Speaker 1: different substrate. 765 00:40:40,080 --> 00:40:42,600 Speaker 2: Whether that requires so to do that, that requires new 766 00:40:42,600 --> 00:40:44,600 Speaker 2: physics that we don't understand today. We'd have to really 767 00:40:44,719 --> 00:40:49,120 Speaker 2: understand what is the brain doing that is producing this 768 00:40:49,320 --> 00:40:52,839 Speaker 2: ordered experience that we have. But I strongly suspect that 769 00:40:52,960 --> 00:40:56,960 Speaker 2: intelligence and consciousness are separate or independent. Is possible to 770 00:40:56,960 --> 00:40:59,800 Speaker 2: have a pure experience in the absence of adaptive behavior 771 00:41:00,160 --> 00:41:03,680 Speaker 2: and it's possible to have very apparent adaptive behavior and 772 00:41:03,680 --> 00:41:08,080 Speaker 2: the absence of experience, So these things are separate now. 773 00:41:09,360 --> 00:41:12,080 Speaker 2: In order to have true substrate independence, like you could 774 00:41:12,120 --> 00:41:14,440 Speaker 2: build a silicon based system that is as good as 775 00:41:14,480 --> 00:41:18,279 Speaker 2: our brains. This requires a physics and neuroscience breakthrough that 776 00:41:18,320 --> 00:41:20,600 Speaker 2: will produce several Nobel prizes that we don't have yet. 777 00:41:20,960 --> 00:41:23,520 Speaker 3: But I do think that that is not one hundred 778 00:41:23,600 --> 00:41:24,000 Speaker 3: years away. 779 00:41:24,000 --> 00:41:26,440 Speaker 2: I think that there's really compelling threads of research that 780 00:41:26,440 --> 00:41:28,240 Speaker 2: are being pulled on that have the potential to produce 781 00:41:28,800 --> 00:41:31,920 Speaker 2: those equations. But even if we don't get those equations, 782 00:41:32,680 --> 00:41:35,040 Speaker 2: if you can build brain to brain connections, then you 783 00:41:35,360 --> 00:41:37,040 Speaker 2: don't need them, because you know that brains are good 784 00:41:37,160 --> 00:41:39,720 Speaker 2: enough and if you can assemble, if you can connect 785 00:41:39,760 --> 00:41:43,920 Speaker 2: them together, then that is another approach with some drawbacks and. 786 00:41:45,800 --> 00:41:48,680 Speaker 3: Some like big head starts. 787 00:41:48,880 --> 00:41:51,160 Speaker 1: Do you think people would volunteer to connect their brain 788 00:41:51,200 --> 00:41:53,440 Speaker 1: to someone else's I'm not sure. I'm not sure I 789 00:41:53,440 --> 00:41:54,840 Speaker 1: would enjoy connect with everything. 790 00:41:54,880 --> 00:41:55,160 Speaker 3: I don't know. 791 00:41:55,200 --> 00:41:57,680 Speaker 2: I mean, I don't think that this is for everybody. Also, 792 00:41:57,719 --> 00:41:59,440 Speaker 2: this is on a thing that exists today. I think 793 00:41:59,440 --> 00:42:03,759 Speaker 2: that this is a really interesting thing on the horizon 794 00:42:03,960 --> 00:42:06,040 Speaker 2: that is like enough to notice. So like, oh that, like, 795 00:42:06,120 --> 00:42:10,879 Speaker 2: if that's possible, what does that mean? But I think 796 00:42:10,920 --> 00:42:15,040 Speaker 2: it's tough to to I think really anticipate it too much. 797 00:42:15,080 --> 00:42:15,759 Speaker 3: Right now, you. 798 00:42:15,800 --> 00:42:17,759 Speaker 1: Onz wrote that one of the main goals in neuroscience 799 00:42:17,840 --> 00:42:20,480 Speaker 1: is to understand the physics of consciousness so that we 800 00:42:20,520 --> 00:42:23,560 Speaker 1: can engineer experience. So tell us what you mean by that. 801 00:42:23,800 --> 00:42:25,279 Speaker 2: Yeah, So to be clear, I don't think that's like 802 00:42:25,320 --> 00:42:26,960 Speaker 2: the only goal of neuroscience. I think there's lots people 803 00:42:27,000 --> 00:42:28,719 Speaker 2: working in neuroscience that are thinking about other stuff and 804 00:42:28,760 --> 00:42:33,000 Speaker 2: have never asked themselves those questions. But I think that, 805 00:42:33,120 --> 00:42:35,840 Speaker 2: I mean, arguably one of the kind of end goals 806 00:42:35,840 --> 00:42:41,279 Speaker 2: of technology is is recursion in the sense of we 807 00:42:43,239 --> 00:42:47,160 Speaker 2: gain the gain the ability to observe and manipulate kind 808 00:42:47,200 --> 00:42:53,840 Speaker 2: of our own existence, and we I think, like Earth 809 00:42:53,920 --> 00:42:56,560 Speaker 2: is small and intensely contested, and space is large, and 810 00:42:56,560 --> 00:43:00,560 Speaker 2: the speed of light is low, and there's like you 811 00:43:00,640 --> 00:43:03,520 Speaker 2: never run out of real estate, and like in the matrix, 812 00:43:04,520 --> 00:43:06,520 Speaker 2: and so getting to a point where we can we 813 00:43:06,600 --> 00:43:10,440 Speaker 2: really have we have control over our like the nature 814 00:43:10,480 --> 00:43:14,520 Speaker 2: of our experience feels like kind of a logical endpoint 815 00:43:14,520 --> 00:43:16,080 Speaker 2: of a lot of what we've seen over the last 816 00:43:16,160 --> 00:43:18,440 Speaker 2: like since the beginning of the technological revolution. 817 00:43:19,040 --> 00:43:22,359 Speaker 1: So, how is what you're doing with the biohybrid electrodes. 818 00:43:22,880 --> 00:43:25,279 Speaker 1: How will this get us closer to understanding something about 819 00:43:25,320 --> 00:43:27,160 Speaker 1: the physics of consciousness? Oh? 820 00:43:27,200 --> 00:43:29,440 Speaker 2: Well, I mean one thing, one thing that I think 821 00:43:29,520 --> 00:43:32,600 Speaker 2: is true about about consciousness is that there's a good 822 00:43:32,680 --> 00:43:35,279 Speaker 2: chance that to really know one will have to see 823 00:43:35,280 --> 00:43:37,360 Speaker 2: it for yourself. I think that the problem, one of 824 00:43:37,360 --> 00:43:39,200 Speaker 2: the problems that has made it so hard to study 825 00:43:39,800 --> 00:43:42,239 Speaker 2: is not it's not that it's magic or that there's 826 00:43:42,239 --> 00:43:46,120 Speaker 2: like some metaphysical thing that makes it inherently impossible, but 827 00:43:46,239 --> 00:43:48,799 Speaker 2: that there's no measurements that we can take that will 828 00:43:48,800 --> 00:43:50,960 Speaker 2: tell us things, because you can always if you believe 829 00:43:50,960 --> 00:43:54,520 Speaker 2: that intelligence and adaptive behavior is separate from phenomenal experience, 830 00:43:55,040 --> 00:43:57,360 Speaker 2: then if you run a behavioral experiment in an animal, 831 00:43:57,960 --> 00:44:00,520 Speaker 2: you can always see some explanation for what's happening without 832 00:44:00,520 --> 00:44:02,360 Speaker 2: resorting to saying anything about conciousness. And when we do 833 00:44:02,400 --> 00:44:05,520 Speaker 2: experiments in animals, we don't talk about what they see 834 00:44:05,640 --> 00:44:07,480 Speaker 2: or perceive. We say they can use the information or 835 00:44:07,480 --> 00:44:11,160 Speaker 2: they can learn the information. And so when you think 836 00:44:11,160 --> 00:44:13,520 Speaker 2: about what experiments can you really run that would allow 837 00:44:13,560 --> 00:44:18,120 Speaker 2: you to know if you've learned something this This often 838 00:44:18,120 --> 00:44:21,439 Speaker 2: looks like, can we add a new sensory mode? Can 839 00:44:21,480 --> 00:44:24,839 Speaker 2: we It's also pretty tough to imagine a sense that 840 00:44:25,280 --> 00:44:27,400 Speaker 2: you don't have, because again evolution is very good at 841 00:44:27,440 --> 00:44:29,680 Speaker 2: its job and it's really fit filled this available time 842 00:44:29,680 --> 00:44:31,719 Speaker 2: and space. But for example, it's sense that you don't have. 843 00:44:32,280 --> 00:44:35,200 Speaker 2: Is a true vector sense, So the ability to see 844 00:44:35,520 --> 00:44:37,920 Speaker 2: a field, like a three D field out in the environment, 845 00:44:38,000 --> 00:44:39,440 Speaker 2: and we don't have this because you don't have the 846 00:44:39,480 --> 00:44:40,600 Speaker 2: sense organs to do this. 847 00:44:40,640 --> 00:44:42,680 Speaker 3: We don't make measurements out of a distance. We only get 848 00:44:42,719 --> 00:44:43,799 Speaker 3: measurements that arrive to you. 849 00:44:44,239 --> 00:44:46,319 Speaker 2: If we had some way to get this signal, say 850 00:44:46,320 --> 00:44:48,600 Speaker 2: from remote sensors or other things, then you could get 851 00:44:48,640 --> 00:44:51,320 Speaker 2: the information. So what would a true vector sense feel 852 00:44:51,360 --> 00:44:53,880 Speaker 2: like to experience? And so at the point where we 853 00:44:53,960 --> 00:44:58,360 Speaker 2: can implement that and make and then make that available 854 00:44:58,480 --> 00:44:59,920 Speaker 2: to you, and then way that you see it, you're 855 00:45:00,000 --> 00:45:03,200 Speaker 2: I guess this was this was a new information, and 856 00:45:03,239 --> 00:45:05,880 Speaker 2: I'm experiencing it directly and I can use it intuitively, 857 00:45:05,960 --> 00:45:06,880 Speaker 2: and there's no other way. 858 00:45:06,719 --> 00:45:07,719 Speaker 3: I could have experienced this. 859 00:45:08,120 --> 00:45:09,520 Speaker 2: I think that is like the type of proof of 860 00:45:09,560 --> 00:45:13,600 Speaker 2: concept for knowing that you've gotten some of that model. 861 00:45:14,000 --> 00:45:15,719 Speaker 2: And I think that this isn't I don't think that 862 00:45:15,760 --> 00:45:17,400 Speaker 2: you can do this with conventional electrodes. I think that 863 00:45:17,440 --> 00:45:19,520 Speaker 2: you need something like a biohybrid neural interface to get 864 00:45:19,560 --> 00:45:23,359 Speaker 2: to that level. Why when you electrically stimulate vision into 865 00:45:23,360 --> 00:45:25,040 Speaker 2: the brain. So let's say that you put an electrode 866 00:45:25,080 --> 00:45:28,759 Speaker 2: in primary visual cortex. If you inject charge through this, 867 00:45:28,960 --> 00:45:30,920 Speaker 2: you can absolutely get a flash of light somewhere in 868 00:45:30,920 --> 00:45:32,760 Speaker 2: the visual field. And if you do this in an animal, 869 00:45:32,800 --> 00:45:34,040 Speaker 2: you can get them to look to the way you 870 00:45:34,040 --> 00:45:36,080 Speaker 2: put the flash of light, and so you can say, okay, 871 00:45:36,120 --> 00:45:38,720 Speaker 2: I got some visual signal into the brain. The problem 872 00:45:38,760 --> 00:45:41,359 Speaker 2: is that these flashes of light, these are known as phosphenes. 873 00:45:41,960 --> 00:45:44,240 Speaker 2: And what a phosphine really is is when you stimulate 874 00:45:44,280 --> 00:45:48,200 Speaker 2: lots of neurons simultaneously, you average them together. And so 875 00:45:48,280 --> 00:45:50,840 Speaker 2: if you have a neuron that represents like red and 876 00:45:50,880 --> 00:45:53,000 Speaker 2: some part of the visual field, next to something that 877 00:45:53,080 --> 00:45:57,360 Speaker 2: represents like a spatial frequency, next to something that represents 878 00:45:57,440 --> 00:46:01,000 Speaker 2: like emotion, like an orientation emotion, and you drive all 879 00:46:01,000 --> 00:46:04,919 Speaker 2: of these simultaneously, you kind of average them. Basically nothing 880 00:46:04,960 --> 00:46:07,560 Speaker 2: that the only information that's remaining is is i thing 881 00:46:07,600 --> 00:46:09,400 Speaker 2: called writing a topic, which is where in the visual 882 00:46:09,400 --> 00:46:11,920 Speaker 2: field was it? And if you do that, then you're limited. 883 00:46:12,160 --> 00:46:15,239 Speaker 2: You throw away almost all of the information that you 884 00:46:15,280 --> 00:46:18,719 Speaker 2: could have conveyed. And when you do this, also, like 885 00:46:18,760 --> 00:46:21,800 Speaker 2: this very continuous stimulation tends to produce the most intense 886 00:46:21,960 --> 00:46:25,320 Speaker 2: immune responses to electrodes that you get. And so these 887 00:46:25,920 --> 00:46:29,719 Speaker 2: these writing electrodes tend to be very encapsulated. And so 888 00:46:29,800 --> 00:46:32,000 Speaker 2: you want something that gives you access to hundreds of 889 00:46:32,000 --> 00:46:37,120 Speaker 2: thousands or millions of neurons and the at single cell 890 00:46:37,200 --> 00:46:42,560 Speaker 2: informational resolution in ways that will the brain will really 891 00:46:43,120 --> 00:46:48,160 Speaker 2: adapt to informationally, and electrodes don't get that type of 892 00:46:48,239 --> 00:46:51,040 Speaker 2: specific stimulation, certainly not at the hundreds and nobody's ever 893 00:46:51,120 --> 00:46:55,680 Speaker 2: done something like one hundred thousand electrodes for stimulation. And 894 00:46:56,560 --> 00:46:59,520 Speaker 2: there's the other technique, optogenetics, where you do this with 895 00:46:59,520 --> 00:47:00,920 Speaker 2: an optic stimulator. 896 00:47:00,920 --> 00:47:02,919 Speaker 3: This requires genetically modifying the host brain. 897 00:47:02,960 --> 00:47:06,000 Speaker 2: You have to use a gene therapy to deliver this 898 00:47:06,040 --> 00:47:09,279 Speaker 2: new protein to the cells of the brain. This is 899 00:47:09,320 --> 00:47:11,680 Speaker 2: not a thing that is really done in humans and cortex, 900 00:47:11,719 --> 00:47:13,440 Speaker 2: and there's reasons that that is that it's going to 901 00:47:13,480 --> 00:47:16,080 Speaker 2: be really difficult, and so there isn't I don't as 902 00:47:16,120 --> 00:47:18,560 Speaker 2: from where I said, I don't see another technology that 903 00:47:18,600 --> 00:47:22,320 Speaker 2: is really capable of getting hundreds of thousands or millions 904 00:47:22,360 --> 00:47:24,560 Speaker 2: of neurons at single cell resolution in a way that 905 00:47:24,640 --> 00:47:26,799 Speaker 2: is long term stable, in a way that allows those 906 00:47:26,800 --> 00:47:30,400 Speaker 2: neurons to learn the signal that you're trying to give them. 907 00:47:30,440 --> 00:47:32,759 Speaker 1: What philosophical questions keep you up at night? 908 00:47:34,400 --> 00:47:38,279 Speaker 2: So there's a question that whenever I go to like 909 00:47:38,320 --> 00:47:40,560 Speaker 2: things where I see my friends, there's a question that 910 00:47:40,600 --> 00:47:43,680 Speaker 2: splits the table evenly every time, which is is a 911 00:47:43,760 --> 00:47:48,439 Speaker 2: destructively scanned upload you? So these expanded you with think 912 00:47:49,280 --> 00:47:50,759 Speaker 2: side of things that my friends and I call the 913 00:47:50,760 --> 00:47:54,759 Speaker 2: transporter problems. And in some sense they're very simple, which 914 00:47:54,800 --> 00:47:58,000 Speaker 2: is like if you have if you take like a 915 00:47:58,040 --> 00:47:59,879 Speaker 2: scan of a brain, but at the end the brain 916 00:47:59,920 --> 00:48:02,759 Speaker 2: is no more, and then you can use this to 917 00:48:02,760 --> 00:48:07,160 Speaker 2: build a perfectly biophysically accurate like atomic simulation of that person. 918 00:48:07,760 --> 00:48:09,800 Speaker 2: Does this make you feel better about dying of cancer? 919 00:48:10,520 --> 00:48:11,960 Speaker 2: And for me, the answer to that is no. And 920 00:48:12,000 --> 00:48:14,920 Speaker 2: I think many people and faced actually with that situation, 921 00:48:15,000 --> 00:48:16,319 Speaker 2: would conclude no. 922 00:48:17,120 --> 00:48:20,160 Speaker 1: This is no as in you feel you will have 923 00:48:20,239 --> 00:48:22,920 Speaker 1: died if you got destroyed, yet there was a replica 924 00:48:22,960 --> 00:48:24,560 Speaker 1: of you that got booted up a second later. 925 00:48:25,280 --> 00:48:25,960 Speaker 3: Yeah, exactly. 926 00:48:26,000 --> 00:48:28,440 Speaker 2: This is like I'll be survived by my friends, which 927 00:48:28,880 --> 00:48:31,040 Speaker 2: is great, but doesn't necessarily make me feel a lot 928 00:48:31,120 --> 00:48:32,640 Speaker 2: better about my specific situation. 929 00:48:32,920 --> 00:48:35,719 Speaker 1: Right. In other words, the replica that gets booted up 930 00:48:35,760 --> 00:48:38,279 Speaker 1: a second later thinks, wow, I'm max. It was just 931 00:48:38,320 --> 00:48:40,280 Speaker 1: over there and now I'm over here. But the question 932 00:48:40,320 --> 00:48:41,759 Speaker 1: is do you get any benefit from that? 933 00:48:41,920 --> 00:48:42,240 Speaker 3: Exactly? 934 00:48:42,320 --> 00:48:45,560 Speaker 2: And so from its perspective, it's probably right. And I 935 00:48:45,560 --> 00:48:48,000 Speaker 2: think that people respond to this while saying like, well, 936 00:48:48,000 --> 00:48:49,840 Speaker 2: every night you lose consciousness, you wake up in the 937 00:48:49,880 --> 00:48:52,239 Speaker 2: next morning, you've broken some continuity there, which I think 938 00:48:52,320 --> 00:48:55,200 Speaker 2: is like also totally fair. That's like also not that's true, 939 00:48:55,400 --> 00:48:57,759 Speaker 2: but still doesn't really make me feel better. And so 940 00:48:57,800 --> 00:49:03,560 Speaker 2: the two camps here are my agency living on in 941 00:49:03,600 --> 00:49:06,920 Speaker 2: the world, which can be done through some other, some model, 942 00:49:07,000 --> 00:49:08,920 Speaker 2: some replication of me that makes me feel like my 943 00:49:08,920 --> 00:49:12,520 Speaker 2: influence will persist, versus I will accept drift in the 944 00:49:12,560 --> 00:49:14,759 Speaker 2: personality in the agency as long as I get continuity. 945 00:49:15,640 --> 00:49:20,000 Speaker 2: And so that's like the brain to brain connection there 946 00:49:20,120 --> 00:49:22,440 Speaker 2: is like you'll get significant personality drift because you're kind 947 00:49:22,440 --> 00:49:26,600 Speaker 2: of averaging together to people to some degree, but you 948 00:49:26,640 --> 00:49:34,160 Speaker 2: get continuity or is this living on an agency without continuity? 949 00:49:34,680 --> 00:49:35,239 Speaker 3: Is that good? 950 00:49:35,520 --> 00:49:39,440 Speaker 2: And what's interesting is people's brains seem to make a 951 00:49:39,520 --> 00:49:41,560 Speaker 2: choice on this early in their life and they are 952 00:49:41,600 --> 00:49:44,319 Speaker 2: unable to see the other one. They're very convinced that 953 00:49:44,360 --> 00:49:46,640 Speaker 2: this is like one of these two things is nonsensical. 954 00:49:47,320 --> 00:49:49,279 Speaker 2: And so my read on this is that this is 955 00:49:49,320 --> 00:49:52,239 Speaker 2: a there's a choice of metaphysics that's being made here 956 00:49:52,840 --> 00:49:55,000 Speaker 2: and from which you reason. So this is a kind 957 00:49:55,000 --> 00:49:57,040 Speaker 2: of a choice that your brain has made that allows 958 00:49:57,080 --> 00:49:59,919 Speaker 2: you to see something and then from there you start reason. 959 00:50:00,239 --> 00:50:02,120 Speaker 2: And so you can't like really talk your way through this, 960 00:50:02,200 --> 00:50:03,720 Speaker 2: But I think these are kind of the two tribes 961 00:50:03,760 --> 00:50:08,759 Speaker 2: that like metaphysical tribes here, And my guess is that 962 00:50:08,840 --> 00:50:11,200 Speaker 2: kind of people get converted to continuity when faced when 963 00:50:11,200 --> 00:50:13,680 Speaker 2: it becomes like a real thing. But that's the that's 964 00:50:13,680 --> 00:50:16,000 Speaker 2: a philosophical question for which I don't know there's a 965 00:50:16,080 --> 00:50:18,080 Speaker 2: right answer that keeps debate going. 966 00:50:18,680 --> 00:50:21,320 Speaker 1: And do you feel any differently about the problem if 967 00:50:21,800 --> 00:50:25,840 Speaker 1: you were degraded into your atoms and then those atoms 968 00:50:25,880 --> 00:50:29,880 Speaker 1: were beamed over somewhere and then reconstructed, But it's still you. 969 00:50:29,880 --> 00:50:31,880 Speaker 1: You're degraded and you're rebuilt up. Does that make a 970 00:50:31,880 --> 00:50:32,400 Speaker 1: difference for you? 971 00:50:32,600 --> 00:50:32,799 Speaker 3: Yeah? 972 00:50:32,800 --> 00:50:35,280 Speaker 2: I mean, so this is this is the second transporter problem. 973 00:50:35,400 --> 00:50:37,239 Speaker 2: Is if you send the atoms, does this make it better? 974 00:50:37,520 --> 00:50:39,719 Speaker 2: And I think really the thing that I don't know? 975 00:50:39,800 --> 00:50:42,239 Speaker 2: I mean, I think so. There's a show that I 976 00:50:42,280 --> 00:50:44,880 Speaker 2: love that recently came to Netflix, was really hard to 977 00:50:44,880 --> 00:50:46,040 Speaker 2: watch for a while called Pantheon. 978 00:50:46,520 --> 00:50:47,560 Speaker 3: Highly highly recommended. 979 00:50:47,560 --> 00:50:49,640 Speaker 2: I think Pantheon is probably the best depiction of how 980 00:50:49,680 --> 00:50:51,640 Speaker 2: I think the next like fifteen years might go that 981 00:50:51,680 --> 00:50:52,520 Speaker 2: I've ever seen in fiction. 982 00:50:54,400 --> 00:50:55,320 Speaker 3: It's adult animation. 983 00:50:55,880 --> 00:50:57,919 Speaker 2: It's by that it's based on a series of short 984 00:50:57,960 --> 00:51:00,560 Speaker 2: stories by Ken Lew who is probably best known as 985 00:51:00,600 --> 00:51:03,600 Speaker 2: the English language translator for the Three Body Problem series. 986 00:51:04,440 --> 00:51:08,040 Speaker 2: And that show is amazing but also terrible metaphysics. It's 987 00:51:08,080 --> 00:51:11,560 Speaker 2: a destructive upload, it's like, but the characters also realize this. 988 00:51:11,600 --> 00:51:13,719 Speaker 2: There's graffiti on a building at one point that says 989 00:51:13,760 --> 00:51:16,600 Speaker 2: like Dina live forever, which I don't find that compelling 990 00:51:16,600 --> 00:51:19,560 Speaker 2: of value proposition, but it's an interesting depiction of a 991 00:51:19,600 --> 00:51:21,560 Speaker 2: world where you kind of get to the other side 992 00:51:21,560 --> 00:51:24,439 Speaker 2: of that of that choice of metaphysics to the degree 993 00:51:24,440 --> 00:51:26,640 Speaker 2: that people aren't worrying about it anymore, and from the 994 00:51:26,800 --> 00:51:30,919 Speaker 2: backwards looking perspective it works out fine. And so that's 995 00:51:30,920 --> 00:51:34,680 Speaker 2: certainly one potential view there. The other is that what 996 00:51:34,760 --> 00:51:36,640 Speaker 2: if you really believe what matters is continuity, then what 997 00:51:36,640 --> 00:51:37,759 Speaker 2: you have to do is you kind of have to 998 00:51:37,800 --> 00:51:41,440 Speaker 2: get a seed brain on both sides of the transporter, 999 00:51:41,800 --> 00:51:45,160 Speaker 2: briefly establish brain to brain link to get the continuity 1000 00:51:45,200 --> 00:51:47,919 Speaker 2: through it, and then that's enough. As long as there's 1001 00:51:47,920 --> 00:51:52,359 Speaker 2: a brief moment of continuity, then that kind of gets 1002 00:51:52,360 --> 00:51:53,920 Speaker 2: you through that philosophically. 1003 00:51:54,239 --> 00:51:56,719 Speaker 1: Oh interesting, So this is where you might do your 1004 00:51:56,880 --> 00:51:58,360 Speaker 1: four hemisphere trick. 1005 00:51:58,400 --> 00:52:00,839 Speaker 2: Exactly, Well, yeah, I mean typically mean and that in 1006 00:52:00,880 --> 00:52:03,560 Speaker 2: the case where it's really like an adom for adam reconstruction, 1007 00:52:03,800 --> 00:52:07,200 Speaker 2: and the representations are already shared, then you wouldn't need 1008 00:52:07,239 --> 00:52:10,399 Speaker 2: any time. If you did this with two people do 1009 00:52:12,280 --> 00:52:14,280 Speaker 2: for that to really make sense, there'd be some time 1010 00:52:14,320 --> 00:52:19,439 Speaker 2: to get representational like drift between them. It's funny because 1011 00:52:19,440 --> 00:52:21,719 Speaker 2: if we talk about these things are interesting and are 1012 00:52:21,800 --> 00:52:25,719 Speaker 2: genuinely becoming from the realm of science fiction where they 1013 00:52:25,920 --> 00:52:27,319 Speaker 2: some of them still are today, and to the realm 1014 00:52:27,320 --> 00:52:30,239 Speaker 2: of engineering, which not all of this is today, but 1015 00:52:30,360 --> 00:52:33,600 Speaker 2: also only clear Like we don't at work, we don't 1016 00:52:33,880 --> 00:52:35,440 Speaker 2: really spend a lot of time thinking about like the 1017 00:52:35,440 --> 00:52:37,680 Speaker 2: future of humanity. It is mostly, as I often say, 1018 00:52:37,719 --> 00:52:41,160 Speaker 2: debugging Linux drivers, yeah, and writing regulatory documentation. 1019 00:52:42,320 --> 00:52:45,800 Speaker 1: So what drives you in your work? 1020 00:52:46,400 --> 00:52:48,480 Speaker 2: I mean, look, if you really believe that these things 1021 00:52:48,480 --> 00:52:51,960 Speaker 2: are possible within our lifetimes, I just like AI is 1022 00:52:52,000 --> 00:52:54,800 Speaker 2: also very exciting. There are other exciting things happening. 1023 00:52:54,440 --> 00:52:54,960 Speaker 3: In the world. 1024 00:52:55,000 --> 00:52:57,480 Speaker 2: But when you really believe that these things could actually 1025 00:52:57,520 --> 00:52:59,960 Speaker 2: be possible, I think it is tough to think about 1026 00:53:00,040 --> 00:53:00,520 Speaker 2: a lot else. 1027 00:53:05,200 --> 00:53:08,840 Speaker 1: That was Max Odak, founder and CEO of Science Corporation. 1028 00:53:09,440 --> 00:53:11,680 Speaker 1: He's working on the challenge of how to read and 1029 00:53:11,680 --> 00:53:14,719 Speaker 1: write from the brain, and really there are only a 1030 00:53:14,760 --> 00:53:18,279 Speaker 1: handful of people who are doing that. With the smarts 1031 00:53:18,360 --> 00:53:22,880 Speaker 1: and entrepreneurial bravery of Max, he and his team are 1032 00:53:22,920 --> 00:53:25,640 Speaker 1: at the cutting edge of integrating with the brain, whether 1033 00:53:25,760 --> 00:53:29,360 Speaker 1: that's by turning pixels into lasers and stimulating a tiny 1034 00:53:29,400 --> 00:53:33,719 Speaker 1: implant in the back of the eye, or growing neurons 1035 00:53:33,760 --> 00:53:38,200 Speaker 1: into the brain that ingratiate themselves into the network in 1036 00:53:38,239 --> 00:53:41,200 Speaker 1: a way that you can spy on the activity there. 1037 00:53:41,520 --> 00:53:43,400 Speaker 1: You can check out more about his company in the 1038 00:53:43,400 --> 00:53:46,400 Speaker 1: show notes at Eagleman dot com, slash podcast, and Max's 1039 00:53:46,440 --> 00:53:51,359 Speaker 1: website is science dot xyz. So let's wrap up at 1040 00:53:51,400 --> 00:53:55,800 Speaker 1: its core, the idea of growing cells into the brain 1041 00:53:56,080 --> 00:54:00,680 Speaker 1: as a brain computer interface. This challenges the comment intuition 1042 00:54:00,880 --> 00:54:06,400 Speaker 1: of a division between biology and machinery, and more generally, however, 1043 00:54:06,480 --> 00:54:10,760 Speaker 1: we make interfaces to the brain, these open the possibility 1044 00:54:11,040 --> 00:54:14,000 Speaker 1: that we'll be able to someday not only interpret what 1045 00:54:14,120 --> 00:54:17,080 Speaker 1: it is to be a human, but also enhance that 1046 00:54:17,920 --> 00:54:21,000 Speaker 1: and that in the future, even things like our thoughts, 1047 00:54:21,040 --> 00:54:26,480 Speaker 1: which seem unassailably private and ineffable. Things like thoughts might 1048 00:54:26,560 --> 00:54:30,480 Speaker 1: soon traverse digital pathways the way any data flows through 1049 00:54:30,520 --> 00:54:34,480 Speaker 1: a network. What does it mean when a thought leaves 1050 00:54:34,520 --> 00:54:38,000 Speaker 1: the confines of the skull? The story of BCIs is 1051 00:54:38,200 --> 00:54:41,320 Speaker 1: just beginning, and it's not just a story about the technology. 1052 00:54:41,320 --> 00:54:44,920 Speaker 1: It's the story of a whole new channel of communication. 1053 00:54:45,040 --> 00:54:49,040 Speaker 1: It's about translating the language of neurons into the language 1054 00:54:49,080 --> 00:54:53,120 Speaker 1: of computers, or perhaps eventually into the brains of other people. 1055 00:54:53,640 --> 00:54:56,880 Speaker 1: It's about giving voice to the mute, it's about giving 1056 00:54:56,960 --> 00:55:01,200 Speaker 1: movement to the paralyzed, and it's about giving wings to 1057 00:55:01,239 --> 00:55:04,880 Speaker 1: our imagination. So The work by Max and others in 1058 00:55:04,920 --> 00:55:09,719 Speaker 1: the BCI space invites us to consider whether our brains 1059 00:55:10,080 --> 00:55:15,040 Speaker 1: have to always remain isolated entities, or whether they can 1060 00:55:15,080 --> 00:55:19,640 Speaker 1: interface with a broader universe. This work reminds us that 1061 00:55:19,680 --> 00:55:23,640 Speaker 1: the brain doesn't always have to be merely an imprisoned 1062 00:55:23,760 --> 00:55:28,600 Speaker 1: container for thought, but instead a living, dynamic interface with 1063 00:55:28,680 --> 00:55:31,960 Speaker 1: the world, one that's going to soon enough, maybe in 1064 00:55:32,000 --> 00:55:37,880 Speaker 1: our lifetimes, reach far beyond the biological limits to which 1065 00:55:37,920 --> 00:55:45,960 Speaker 1: we have become accustomed. Go to Eagleman dot com slash 1066 00:55:46,000 --> 00:55:49,879 Speaker 1: podcast for more information and to find further reading. Send 1067 00:55:49,920 --> 00:55:52,799 Speaker 1: me an email at podcasts at eagleman dot com with 1068 00:55:52,920 --> 00:55:56,239 Speaker 1: questions or discussion, and check out and subscribe to Inner 1069 00:55:56,280 --> 00:55:59,400 Speaker 1: Cosmos on YouTube for videos of each episode and to 1070 00:55:59,480 --> 00:56:03,399 Speaker 1: leave comma until next time. I'm David Eagleman, and this 1071 00:56:03,560 --> 00:56:04,759 Speaker 1: is Inner Cosmos.