WEBVTT 00:00:00.000 --> 00:00:02.220 I have a conviction that the brain is understandable. 00:00:02.560 --> 00:00:06.000 You look inside these AI models, and there's 00:00:06.000 --> 00:00:08.339 some clarity coming up, but it's also super confusing. 00:00:08.519 --> 00:00:12.439 So I believe that there's a huge landscape of 00:00:12.439 --> 00:00:14.699 different architectures that are able to solve 00:00:14.699 --> 00:00:17.839 these tasks, and we have two points. There's 00:00:17.839 --> 00:00:20.539 Chachi PT and Dali, and there's the brain. I 00:00:20.539 --> 00:00:22.760 think the brain is on one extreme of efficiency 00:00:22.760 --> 00:00:27.719 and understandability, and that's what I want. 00:00:27.820 --> 00:00:40.000 That's what I care about. I'm John Foxe, Director 00:00:40.000 --> 00:00:42.079 of the Del Monte Institute for Neuroscience here 00:00:42.079 --> 00:00:44.359 at the University of Rochester, and I'd like 00:00:44.359 --> 00:00:46.579 to welcome you to another episode of Neuroscience 00:00:46.579 --> 00:00:49.719 Perspectives. I'm absolutely thrilled to be joined 00:00:49.719 --> 00:00:52.740 by renowned neuroscientist Professor Doris Tsao. 00:00:53.630 --> 00:00:56.829 Her pioneering research has transformed our understanding 00:00:56.829 --> 00:00:59.109 of how the brain processes visual information. 00:00:59.390 --> 00:01:01.929 And her lab helped us discover that it takes 00:01:01.929 --> 00:01:05.090 only a few hundred neurons to encode any given 00:01:05.090 --> 00:01:07.790 face. For reference, as many of you will know, 00:01:07.969 --> 00:01:10.930 there are billions of neurons in the brain. Dr. 00:01:11.030 --> 00:01:14.549 Tsao is a professor of neurobiology. at University 00:01:14.549 --> 00:01:17.030 of California, Berkeley, and she's been the recipient 00:01:17.030 --> 00:01:19.870 of numerous awards, including most recently the 00:01:19.870 --> 00:01:22.469 Kavli Prize, one of the most prestigious international 00:01:22.469 --> 00:01:25.989 science prizes there is. She's also been elected 00:01:25.989 --> 00:01:29.030 as a very young scientist to the National Academy 00:01:29.030 --> 00:01:31.530 of Sciences and was named an investigator of 00:01:31.530 --> 00:01:33.750 the Howard Hughes Medical Institute, which is 00:01:33.750 --> 00:01:35.829 among the highest of distinctions for biomedical 00:01:35.829 --> 00:01:38.909 research. Doris, thank you so much for joining 00:01:38.909 --> 00:01:41.109 us here on Neuroscience Perspective. I'm dying 00:01:41.109 --> 00:01:43.370 to get into things with you and ask you a bunch 00:01:43.370 --> 00:01:45.930 of questions about yourself. But let's kick off 00:01:45.930 --> 00:01:48.250 with your research. You know, why faces? Why 00:01:48.250 --> 00:01:50.069 the visual system? Where did that all come from? 00:01:55.600 --> 00:02:00.879 probably high school when I realized that the 00:02:00.879 --> 00:02:03.920 visual system, the neurons are what generates 00:02:03.920 --> 00:02:07.799 our perception of everything. But I came to faces 00:02:07.799 --> 00:02:10.560 kind of by accident. You know, I was a graduate 00:02:10.560 --> 00:02:15.520 student. I was recording in V1. I was initially 00:02:15.520 --> 00:02:19.020 interested in this question of how the brain 00:02:19.020 --> 00:02:22.840 represents space. and I wasn't getting anywhere, 00:02:23.000 --> 00:02:28.819 so I got into fMRI, and I read about Nancy Kanwisher's 00:02:28.819 --> 00:02:32.460 work that reported a face area in the human brain, 00:02:32.580 --> 00:02:35.199 and it just seemed like a fun thing to do. Let's 00:02:35.199 --> 00:02:37.400 just show faces and objects to a monkey and see 00:02:37.400 --> 00:02:41.680 if monkeys also might have a face area. And then 00:02:41.680 --> 00:02:43.740 you found it. they didn't just have one face 00:02:43.740 --> 00:02:46.060 area. They had a whole system of face areas. 00:02:46.180 --> 00:02:49.560 Yes, yes. And so it took off on its own. You 00:02:49.560 --> 00:02:51.860 know, every question kind of naturally led to 00:02:51.860 --> 00:02:53.479 the next question. It's like, yeah, so there's 00:02:53.479 --> 00:02:56.080 six face areas. What are they each doing? Are 00:02:56.080 --> 00:02:57.979 they each doing something different? Yeah, yeah, 00:02:58.080 --> 00:03:01.500 yeah. Tell us about these six face areas. You 00:03:01.500 --> 00:03:05.259 know, why six in a monkey? for faces yeah we 00:03:05.259 --> 00:03:07.939 we wondered about that for a long time and um 00:03:07.939 --> 00:03:10.319 a few years ago i think we discovered the answer 00:03:10.319 --> 00:03:14.080 basically there's a whole map of object space 00:03:14.080 --> 00:03:16.900 so object space you know same way we have 3d 00:03:16.900 --> 00:03:18.680 space we can represent you know different points 00:03:18.680 --> 00:03:21.650 and locations so you This part of the brain called 00:03:21.650 --> 00:03:24.050 IT cortex represents objects also in this more 00:03:24.050 --> 00:03:26.349 abstract space that's also two -dimensional, 00:03:26.430 --> 00:03:28.729 and there's repeated copies of it, and the representation 00:03:28.729 --> 00:03:30.569 becomes more and more invariant, and so you have 00:03:30.569 --> 00:03:32.550 these multiple stages, and that's why you have 00:03:32.550 --> 00:03:34.710 these multiple copies of this map, and faces 00:03:34.710 --> 00:03:37.310 are one part of that map, and so when you just 00:03:37.310 --> 00:03:38.889 look at faces, you see these six face patches. 00:03:39.250 --> 00:03:41.849 I got you. Okay, so it's six levels of complexity? 00:03:41.990 --> 00:03:43.550 Is that the way to think about it? Yeah, I think 00:03:43.550 --> 00:03:45.810 that's... But highly... They're part of a circuit, 00:03:45.909 --> 00:03:48.189 right? Yes, yes. So they're interacting to evolve, 00:03:48.330 --> 00:03:51.840 and eventually then... an individual face I mean 00:03:51.840 --> 00:03:55.259 I think we all remember these stories about like 00:03:55.259 --> 00:03:58.719 the Jennifer Aniston neuron and that you know 00:03:58.719 --> 00:04:01.379 when you get down to the level of like dissociating 00:04:01.379 --> 00:04:05.550 two very similar faces from your life The representation 00:04:05.550 --> 00:04:07.849 there is actually very limited in terms of the 00:04:07.849 --> 00:04:09.210 number of neurons that are involved. Is that 00:04:09.210 --> 00:04:11.229 fair to say? Or is that not the right way to 00:04:11.229 --> 00:04:13.430 think about it? Yeah, so face carries so much 00:04:13.430 --> 00:04:15.250 information. There's information about the physical 00:04:15.250 --> 00:04:17.490 features. There's information about the identity, 00:04:17.790 --> 00:04:20.990 the expression. And I should say this is still, 00:04:21.110 --> 00:04:23.069 we're still figuring all this out, right? It's 00:04:23.069 --> 00:04:26.569 still early days. But indeed, you don't need... 00:04:27.240 --> 00:04:30.759 that many cells to represent the identity, which 00:04:30.759 --> 00:04:32.399 is kind of surprising. You look at a face, it 00:04:32.399 --> 00:04:35.100 just, you know, it feels so complex, right? It's 00:04:35.100 --> 00:04:37.620 like ineffable, like who knows how many neurons 00:04:37.620 --> 00:04:40.139 that would take. But it turns out that with just 00:04:40.139 --> 00:04:42.800 50 numbers, and this is something that computer 00:04:42.800 --> 00:04:44.920 vision people discovered with like 50 numbers. 00:04:45.720 --> 00:04:47.779 you can, like, recreate the face, right? So you 00:04:47.779 --> 00:04:50.040 can really compress the information down. And 00:04:50.040 --> 00:04:53.759 it turns out that the neurons use a code that 00:04:53.759 --> 00:04:56.579 is very similar to one that the computer vision 00:04:56.579 --> 00:04:58.779 people figured out for how to compress the information. 00:04:58.779 --> 00:05:00.579 Gotcha, yeah. So this sort of dimensionality 00:05:00.579 --> 00:05:03.600 reduction is instantiated neurons in much the 00:05:03.600 --> 00:05:06.879 same way that one can do it with a formula in 00:05:06.879 --> 00:05:10.160 silico in a computer. Exactly. And, you know, 00:05:10.199 --> 00:05:12.660 in recent years, like with Dali, right, where 00:05:12.660 --> 00:05:14.339 you can, like, you know, specify any prompt and 00:05:14.339 --> 00:05:16.459 it can create a picture. of anything that you 00:05:16.459 --> 00:05:19.699 can possibly imagine um that idea of being able 00:05:19.699 --> 00:05:22.699 to compress reality has really taken off right 00:05:22.699 --> 00:05:25.480 and they're kind of an existence proof that you 00:05:25.480 --> 00:05:28.519 know like like the space of all possible for 00:05:28.519 --> 00:05:31.879 example 500 by 500 you know rgb images it's like 00:05:31.879 --> 00:05:34.319 it's more than the number of atoms you know in 00:05:34.319 --> 00:05:37.139 the universe right but you know with just this 00:05:37.139 --> 00:05:40.040 like one gigabyte you know neural network like 00:05:40.040 --> 00:05:43.290 you can create any like any picture of reality 00:05:43.290 --> 00:05:47.069 and the reason is because reality is much lower 00:05:47.069 --> 00:05:49.649 dimensional than the space of all possible images 00:05:49.649 --> 00:05:52.069 right so much of that space is just like random 00:05:52.069 --> 00:05:54.949 noise right but then there's like Yeah, the space 00:05:54.949 --> 00:05:57.589 of reality, of objects. And there's beautiful 00:05:57.589 --> 00:06:00.329 ways to discover those spaces that these machine 00:06:00.329 --> 00:06:03.170 learning people are figuring out. And obviously 00:06:03.170 --> 00:06:05.670 the brain has figured out how to do that with 00:06:05.670 --> 00:06:07.269 its limited neurons. And so that's what we're 00:06:07.269 --> 00:06:09.290 trying to figure out. You know, and all the buzz 00:06:09.290 --> 00:06:12.829 is about AI at the moment. And you made a case 00:06:12.829 --> 00:06:14.970 yesterday, and I'd be interested to hear you 00:06:14.970 --> 00:06:18.209 sort of extrapolate on that a bit, that actually 00:06:18.209 --> 00:06:20.089 it's still very important for us to understand 00:06:20.089 --> 00:06:22.569 the neurobiology of vision because we have a 00:06:22.569 --> 00:06:26.529 lot of insight. of how that's achieved to provide 00:06:26.529 --> 00:06:30.850 to AI folks and that there's still a place for 00:06:30.850 --> 00:06:34.089 neuroscience in the face of AI. I mean, it is 00:06:34.089 --> 00:06:36.009 incredible. You know, our brain runs on 30 watts 00:06:36.009 --> 00:06:38.290 while you hear about them, like, OpenAI building 00:06:38.290 --> 00:06:42.189 nuclear reactor farms to power, you know, next 00:06:42.189 --> 00:06:44.529 generation chat GPT. So I think the brain is 00:06:44.529 --> 00:06:47.370 so much more efficient. I have a conviction that 00:06:47.370 --> 00:06:50.449 the brain is understandable. And, you know, you 00:06:50.449 --> 00:06:54.199 look inside these... AI models, there's some 00:06:54.199 --> 00:06:56.300 clarity coming up, but it's also super confusing. 00:06:56.459 --> 00:07:00.379 So I believe that there's a huge landscape of 00:07:00.379 --> 00:07:02.639 different architectures that are able to solve 00:07:02.639 --> 00:07:06.629 these tasks. We have two points, you know, there's 00:07:06.629 --> 00:07:08.829 Chachi PT and Dali and there's like the brain. 00:07:08.970 --> 00:07:11.589 I think the brain is on one extreme of efficiency 00:07:11.589 --> 00:07:16.129 and understandability. Right, right. And that's 00:07:16.129 --> 00:07:18.610 what I want. Right, these AIs are built on the 00:07:18.610 --> 00:07:21.290 part that you can take brute force, super fast 00:07:21.290 --> 00:07:24.050 computers and a lot of energy and just churn 00:07:24.050 --> 00:07:26.370 through billions of calculations. Exactly. And 00:07:26.370 --> 00:07:28.509 they just impose this one arbitrary architecture. 00:07:28.689 --> 00:07:30.410 You know, what is the likelihood that that is 00:07:30.410 --> 00:07:32.490 the architecture, right? Like, I just really 00:07:32.490 --> 00:07:34.670 believe that the evolution figured out something. 00:07:34.699 --> 00:07:36.759 tricks that we can still learn from. Here's a 00:07:36.759 --> 00:07:40.240 thought I was thinking about technologies and 00:07:40.240 --> 00:07:42.279 the way in which you sort of have pulled together 00:07:42.279 --> 00:07:45.800 multiple technologies to address your question 00:07:45.800 --> 00:07:48.300 and we're sort of living in this like extraordinary 00:07:48.300 --> 00:07:50.240 time right particularly over the last couple 00:07:50.240 --> 00:07:52.139 of decades as you've sort of built out your career 00:07:52.139 --> 00:07:56.160 where you know fMRI came online high field magnets 00:07:56.160 --> 00:07:58.439 extraordinary electrophysiological techniques 00:07:58.439 --> 00:08:01.290 how much is that driving and do you see Do you 00:08:01.290 --> 00:08:03.689 see AI, for example, now interfacing with that 00:08:03.689 --> 00:08:05.529 and really driving forward science in a whole 00:08:05.529 --> 00:08:08.529 new way? Oh, yeah. I mean, absolutely. I'm sure 00:08:08.529 --> 00:08:10.110 you know this quote from Sidney Brenner, right? 00:08:10.310 --> 00:08:13.829 Progress in science is driven by new tools, new 00:08:13.829 --> 00:08:17.410 phenomena, and new ideas in that order of importance. 00:08:17.589 --> 00:08:20.529 That's good. Sadly, it's true. And I was just 00:08:20.529 --> 00:08:23.589 so lucky when I was in Boston as a grad student. 00:08:25.120 --> 00:08:27.220 landed at Mass General Hospital where there were 00:08:27.220 --> 00:08:29.680 all these experts on fMRI, you know, the Martinez 00:08:29.680 --> 00:08:32.159 Imaging Center, and that was, like, so important, 00:08:32.320 --> 00:08:34.039 right? Because these face patches, like, just 00:08:34.039 --> 00:08:36.259 sticking an electrode into the temporal lobe, 00:08:36.299 --> 00:08:38.779 you would never find them. Or you might, but 00:08:38.779 --> 00:08:41.779 you couldn't then find it again. Whereas this 00:08:41.779 --> 00:08:44.580 was, like, you know, it just gives you a systematic 00:08:44.580 --> 00:08:46.820 way, right? fMRI gives you this bird's -eye view 00:08:46.820 --> 00:08:48.799 of the brain so you can see all the, you know, 00:08:48.840 --> 00:08:50.580 loci that are important, then you can go in and 00:08:50.580 --> 00:08:52.879 study the single neurons in detail. Right, right. 00:08:52.940 --> 00:08:56.529 Around those same times. people like Greg McCarty 00:08:56.529 --> 00:09:01.009 and Ina Puse, Truett Allison, were using ECOG 00:09:01.009 --> 00:09:03.509 and seeing fusiform face area. Yes, they were 00:09:03.509 --> 00:09:05.529 the pioneers that really suggested there was 00:09:05.529 --> 00:09:08.149 something profound down there. It's amazing. 00:09:08.450 --> 00:09:10.269 I mean, having done some of those recordings 00:09:10.269 --> 00:09:13.570 myself, when you have an electrode sitting over 00:09:13.570 --> 00:09:16.309 the fusiform face area and you show it a face, 00:09:16.470 --> 00:09:20.149 I mean, it announces itself. It's really amazing. 00:09:20.210 --> 00:09:24.289 Okay, good. So we share that. Yeah, yeah. When 00:09:24.289 --> 00:09:26.269 did you decide you wanted to be a scientist or 00:09:26.269 --> 00:09:27.669 you think this is where I'm going to end up? 00:09:27.769 --> 00:09:29.590 You know, I wasn't interested in science very 00:09:29.590 --> 00:09:32.690 much as a kid. I was not really a curious child. 00:09:32.850 --> 00:09:36.549 I played with Barbie dolls. Really? Yeah. I love 00:09:36.549 --> 00:09:41.509 to read. And so I always found the stories, the 00:09:41.509 --> 00:09:44.730 biographies of scientists very inspiring. These 00:09:44.730 --> 00:09:47.409 were people who came from humble backgrounds, 00:09:47.570 --> 00:09:51.960 but just by working hard. they could you know 00:09:51.960 --> 00:09:54.919 achieve glory so that seemed like a very accessible 00:09:54.919 --> 00:09:59.710 path to glory to me um and then It was after 00:09:59.710 --> 00:10:02.090 I got accepted to Caltech. And, you know, I was 00:10:02.090 --> 00:10:03.570 worried I was going to fall behind. And so my 00:10:03.570 --> 00:10:05.409 senior year, I started reading the Feynman Lectures 00:10:05.409 --> 00:10:08.830 on Physics, Volume 1. And that was when I understood 00:10:08.830 --> 00:10:11.669 what science was about. It was so beautiful. 00:10:11.730 --> 00:10:14.610 It's just like this freedom to ask any question 00:10:14.610 --> 00:10:16.830 you want and use your brain and think through 00:10:16.830 --> 00:10:20.610 it logically and come to an insight. Those Feynman 00:10:20.610 --> 00:10:22.429 Lectures, you can get them on YouTube. And he 00:10:22.429 --> 00:10:25.610 has this fantastic Brooklyn accent. And they're 00:10:25.610 --> 00:10:28.779 really wonderful. He's such an entertainer. Recommend 00:10:28.779 --> 00:10:30.019 them to people. You don't need to be a physicist 00:10:30.019 --> 00:10:32.759 to enjoy Richard Feynman telling his stories. 00:10:33.600 --> 00:10:37.860 Amazing. Now, but your mom and dad were academics? 00:10:39.480 --> 00:10:42.840 So, yeah, we came to the U .S. So they were graduate 00:10:42.840 --> 00:10:44.440 students at the University of Maryland. My mom 00:10:44.440 --> 00:10:46.200 was a computer programmer. I wouldn't call her 00:10:46.200 --> 00:10:49.340 an academic. I think she was my mom. And by my 00:10:49.340 --> 00:10:53.019 dad, you know, he was like he, you know, during 00:10:53.019 --> 00:10:54.860 the Cultural Revolution, he was sent off to Manchuria 00:10:54.860 --> 00:10:57.820 to like chop trees. And he always knew that that 00:10:57.820 --> 00:10:59.519 period would end and there would be a new day. 00:10:59.659 --> 00:11:02.539 And so he, you know, brought a thick stack of 00:11:02.539 --> 00:11:06.620 math books with him, like studying. functional 00:11:06.620 --> 00:11:09.840 analysis in the forest. And so, you know, after 00:11:09.840 --> 00:11:11.700 the end of the Cultural Revolution, you know, 00:11:11.700 --> 00:11:13.940 he got into Peking University and then we came 00:11:13.940 --> 00:11:16.799 to the U .S. And so, yeah, he's a mathematician 00:11:16.799 --> 00:11:20.799 and always, like, just deeply curious. You know, 00:11:20.840 --> 00:11:23.220 he's 82 years old now. He works harder than I 00:11:23.220 --> 00:11:24.799 do. Like, whenever I go in his room, he's, like, 00:11:24.799 --> 00:11:26.860 watching a YouTube video, like, learning a new... 00:11:26.860 --> 00:11:29.820 Recently, he got excited about this area of math 00:11:29.820 --> 00:11:31.799 I've never heard of called sheaf theory. And 00:11:31.799 --> 00:11:35.340 it's, like... Apparently, it's about how you 00:11:35.340 --> 00:11:38.799 can, everything is defined by its relation to 00:11:38.799 --> 00:11:40.659 other things. And he's telling me how that's 00:11:40.659 --> 00:11:42.740 like the essence of Marxism. And so that's the 00:11:42.740 --> 00:11:47.379 kind of person he is. So mathematics to political 00:11:47.379 --> 00:11:51.799 theory. Yeah. Just a brilliant mind. Yes. And 00:11:51.799 --> 00:11:54.340 an inspiration then to you both, both the practical 00:11:54.340 --> 00:11:57.080 part of computer science that your mom. brings 00:11:57.080 --> 00:11:59.460 to the table and the inspiration yes yes you 00:11:59.460 --> 00:12:01.220 know my mom actually like programmed some stimuli 00:12:01.220 --> 00:12:04.080 for me when I was yeah setting these bases yeah 00:12:04.080 --> 00:12:07.639 so yeah so a big move though to come from China 00:12:07.639 --> 00:12:11.159 to the US in the in the 80s was it yeah 1980 00:12:11.159 --> 00:12:16.399 1980 so right yes yeah and uh how much of that 00:12:16.399 --> 00:12:17.720 do you remember you must have been extraordinarily 00:12:17.720 --> 00:12:21.100 young at that point I remember yeah I remember 00:12:21.100 --> 00:12:23.980 this night we were in Hong Kong I remember tasting 00:12:23.980 --> 00:12:25.940 chocolate for the first time on that journey 00:12:25.940 --> 00:12:29.799 somehow um yeah the U .S. you know when we first 00:12:29.799 --> 00:12:31.539 came my grandfather he had been here for many 00:12:31.539 --> 00:12:34.080 years and so he would have these like American 00:12:34.080 --> 00:12:36.220 culture evenings, apparently. He would teach, 00:12:36.360 --> 00:12:39.059 because I came with my cousins and their family, 00:12:39.100 --> 00:12:41.179 so he would teach them about all the American 00:12:41.179 --> 00:12:43.120 traditions, like this is football and here's 00:12:43.120 --> 00:12:46.879 the rules and stuff. Wow. Were you five or six 00:12:46.879 --> 00:12:49.120 or something? I was four. You were four, right. 00:12:49.559 --> 00:12:52.399 And yet, very profound memories of that, obviously, 00:12:52.399 --> 00:12:54.919 because it was such a big change. And how much 00:12:54.919 --> 00:12:58.320 of your very early childhood in China do you 00:12:58.320 --> 00:13:03.269 remember? Yeah, vague, vague glimmerings. And 00:13:03.269 --> 00:13:06.710 you studied mathematics, though, in university, 00:13:06.830 --> 00:13:08.990 right, as well? Yeah, I studied math and biology. 00:13:09.330 --> 00:13:12.149 I really love math. Yeah, I wish I could, like, 00:13:12.210 --> 00:13:14.309 if I had a secret wish, it would be, like, if 00:13:14.309 --> 00:13:16.690 I could go do a math PhD, yeah. Look at that. 00:13:16.769 --> 00:13:19.330 Well, maybe you will, like your dad at 82. Yeah. 00:13:19.870 --> 00:13:23.190 That's outstanding, yeah. So then you went to 00:13:23.190 --> 00:13:25.230 Caltech, you did your undergraduate there, and 00:13:25.230 --> 00:13:26.669 then you went to Harvard for graduate school. 00:13:27.370 --> 00:13:29.769 And then you ended up in a very important lab 00:13:29.769 --> 00:13:31.850 or very famous lab, right, with Marge Livingston. 00:13:31.850 --> 00:13:34.549 Tell us a little bit about that and the impact 00:13:34.549 --> 00:13:37.710 Marge had on you. Marge is an incredible scientist. 00:13:37.830 --> 00:13:40.370 She's such a true scientist. You know, the first 00:13:40.370 --> 00:13:43.110 time I met her, like during the graduate school 00:13:43.110 --> 00:13:46.710 interviews, she was recording and she kind of 00:13:46.710 --> 00:13:49.570 like. had to finish her experiment then she looked 00:13:49.570 --> 00:13:51.990 up at me and said hello and then she was like 00:13:51.990 --> 00:13:54.450 so um just direct you know she said what are 00:13:54.450 --> 00:13:56.490 you interested in i told her something vague 00:13:56.490 --> 00:13:58.389 about you know i want to understand neural circuits 00:13:58.389 --> 00:14:00.730 and she said well if i were interested in neural 00:14:00.730 --> 00:14:03.570 circuits i would go to larry katz's lab maybe 00:14:03.570 --> 00:14:06.230 this is not going so well but you know she just 00:14:06.230 --> 00:14:08.309 like intuitively understood me you know it was 00:14:08.309 --> 00:14:10.509 yeah i am very shy i was very shy i remember 00:14:10.509 --> 00:14:13.809 i was like Marge, can I talk to you about something? 00:14:14.029 --> 00:14:16.450 Because I wanted to join her lab. And I didn't 00:14:16.450 --> 00:14:19.210 have to say anything more. She said, I ordered 00:14:19.210 --> 00:14:24.690 an amplifier for you already. Yes. So it was 00:14:24.690 --> 00:14:27.769 wonderful. She just does her own experiments 00:14:27.769 --> 00:14:31.710 every day. And she's taught us by example. Yeah, 00:14:31.710 --> 00:14:35.210 true. one of the great leading vision scientists 00:14:35.210 --> 00:14:38.710 in the history. She's incredible. She started 00:14:38.710 --> 00:14:41.509 as a Drosophila neuroscientist, discovered these 00:14:41.509 --> 00:14:45.750 learning mutants, and then she went on to, she 00:14:45.750 --> 00:14:48.070 became so interested in vision, she felt like 00:14:48.070 --> 00:14:52.269 she found her calling. And she's also done incredible 00:14:52.269 --> 00:14:54.490 work understanding the relationship between art 00:14:54.490 --> 00:14:58.649 and vision, how artists... And so her mentorship 00:14:58.649 --> 00:15:02.850 style, she's sleeves rolled up, in the lab making 00:15:02.850 --> 00:15:04.629 recordings. Oh, yeah, yeah. And have you taken 00:15:04.629 --> 00:15:06.629 that into your own lab? How do you approach it? 00:15:06.629 --> 00:15:09.309 Not as much as Marge. Yeah, I don't know how. 00:15:09.389 --> 00:15:14.190 I think I'd like to. Yeah, I do try, but just, 00:15:14.250 --> 00:15:17.389 yeah, time management. That's an amazing trajectory. 00:15:17.529 --> 00:15:20.389 And then you did something. A little strange, 00:15:20.509 --> 00:15:22.590 too, because, you know, most people, you know, 00:15:22.590 --> 00:15:26.350 the American science engine. But you took off 00:15:26.350 --> 00:15:29.049 for Germany, right? And I went to, was it Bremen? 00:15:29.250 --> 00:15:31.750 Bremen, yeah. And what was the, like, why? What 00:15:31.750 --> 00:15:33.429 was the decision there and how did that work? 00:15:33.590 --> 00:15:35.789 I got a fellowship from the Humboldt Foundation 00:15:35.789 --> 00:15:39.830 to set up my own little lab. This was, you know, 00:15:39.850 --> 00:15:42.230 I got my PhD and I spent two more years in Marge's 00:15:42.230 --> 00:15:44.309 lab. But then, yeah, it was like an amazing opportunity. 00:15:44.730 --> 00:15:46.149 So you didn't do an official postdoc. Yeah, yeah. 00:15:47.039 --> 00:15:49.460 So, you know, we found these space patches and 00:15:49.460 --> 00:15:51.980 it just like there's so many questions. And so 00:15:51.980 --> 00:15:54.580 that was like an amazing opportunity to keep 00:15:54.580 --> 00:15:57.799 studying this on my own. Yeah. So off to Germany. 00:15:58.440 --> 00:16:00.480 And I'd always been interested in German culture. 00:16:00.639 --> 00:16:02.779 You were. Okay. Where does that come from? I 00:16:02.779 --> 00:16:04.600 think it started with music. Like my father, 00:16:04.799 --> 00:16:07.200 he was always listening to Beethoven. I see. 00:16:07.259 --> 00:16:10.759 Yeah, yeah, yeah. And was that culture shock? 00:16:10.899 --> 00:16:14.080 Was it different? Did you settle easily? I loved 00:16:14.080 --> 00:16:16.440 Germany. It was like a much simpler place. You 00:16:16.440 --> 00:16:18.879 know, Bremen's a small town. There's like one 00:16:18.879 --> 00:16:21.860 department store in Karstadt. So there wasn't 00:16:21.860 --> 00:16:23.759 much to do. So I found a violin teacher. I fell 00:16:23.759 --> 00:16:26.259 in love with the violin again. I loved it. Yeah, 00:16:26.299 --> 00:16:30.500 yeah, yeah. It's a brave move. I mean, to get 00:16:30.500 --> 00:16:34.620 out of your PhD and your own milieu and the place 00:16:34.620 --> 00:16:37.220 that you've sort of trained and to take off for 00:16:37.220 --> 00:16:40.740 Europe to a foreign country. launch a science 00:16:40.740 --> 00:16:42.960 career I mean and are there lessons in that for 00:16:42.960 --> 00:16:45.139 example for for your trainees is there something 00:16:45.139 --> 00:16:47.700 about that streak in you that's willing to take 00:16:47.700 --> 00:16:50.200 a chance that's part of how you approach science 00:16:50.200 --> 00:16:53.100 I mean every path is unique but I think um yeah 00:16:53.100 --> 00:16:55.840 I've never worried too much about I've just never 00:16:55.840 --> 00:16:59.519 worried too much you know about the future I've 00:16:59.519 --> 00:17:01.980 always had this like irrational exuberance and 00:17:01.980 --> 00:17:05.380 so I try to like tell my trainees yeah yeah just 00:17:05.380 --> 00:17:07.660 like follow follow your heart what you want to 00:17:07.660 --> 00:17:09.700 do everything else will take care of itself yeah 00:17:09.700 --> 00:17:12.240 yeah you gave a wonderful talk yesterday and 00:17:12.940 --> 00:17:15.519 I really appreciated the part where you made 00:17:15.519 --> 00:17:17.799 sure to point out exactly who the folks were 00:17:17.799 --> 00:17:19.640 that did the work, the graduate students and 00:17:19.640 --> 00:17:21.799 the postdocs, and give credit where credit's 00:17:21.799 --> 00:17:23.420 due. And it's obvious that you've put some very, 00:17:23.480 --> 00:17:25.339 very talented people through your lab. Oh, yeah. 00:17:25.619 --> 00:17:29.480 Everything is due to them. It is such a privilege 00:17:29.480 --> 00:17:30.880 to be able to work with these brilliant young 00:17:30.880 --> 00:17:32.700 people. Right. It begins and ends with that sort 00:17:32.700 --> 00:17:36.579 of vibrancy in the ranks and training people. 00:17:36.680 --> 00:17:38.259 There's nothing better, right? There's nothing 00:17:38.259 --> 00:17:41.380 better than to be slacking furiously about a 00:17:41.380 --> 00:17:45.059 new idea. with a student and just fantasizing, 00:17:45.059 --> 00:17:47.000 it's, yeah, it doesn't get better than that. 00:17:47.180 --> 00:17:49.500 I love it, love it. Here, I read a quote, just 00:17:49.500 --> 00:17:51.640 going back to mentorship, and we were talking 00:17:51.640 --> 00:17:54.299 about some of these leading elders in the field, 00:17:54.359 --> 00:17:58.000 you know, and this is a quote, I love this. It 00:17:58.000 --> 00:18:00.359 said, all of us desire to find a mentor who sees 00:18:00.359 --> 00:18:04.819 us as we wish to be rather than as we are. That's 00:18:04.819 --> 00:18:07.660 quite profound. Would you unpack that for us? 00:18:07.700 --> 00:18:11.500 What's that? What drove that? That was about 00:18:11.500 --> 00:18:16.140 Marge, my PhD advisor. I felt that's how she... 00:18:16.720 --> 00:18:19.900 treated me like i was like a child not her like 00:18:19.900 --> 00:18:23.000 you know oars women for her grant yeah right 00:18:23.000 --> 00:18:25.839 right um not just yeah the engine room right 00:18:25.839 --> 00:18:28.819 yeah she like even when i was a young faculty 00:18:28.819 --> 00:18:30.839 member like you know caltech was starting out 00:18:30.839 --> 00:18:33.640 and she she said i said something about i was 00:18:33.640 --> 00:18:35.900 like still studying math and she she sent me 00:18:35.900 --> 00:18:38.119 an email and she's like you know i i you should 00:18:38.119 --> 00:18:41.250 keep doing that i believe you can do something 00:18:41.250 --> 00:18:43.470 great you know in that realm and it's like she's 00:18:43.470 --> 00:18:46.470 always encouraged me like this side of me that's 00:18:46.470 --> 00:18:49.630 not like maybe conventional I don't know maybe 00:18:49.630 --> 00:18:54.410 is it fair to say then you know there's there's 00:18:54.410 --> 00:18:57.900 a pragmatics part of mentorship in a PhD you 00:18:57.900 --> 00:18:59.599 need to learn this material you need to know 00:18:59.599 --> 00:19:01.119 how to do this technique you need to be able 00:19:01.119 --> 00:19:05.160 to and then there's the the humanocentric piece 00:19:05.160 --> 00:19:07.160 of it there's the development of the individual 00:19:07.160 --> 00:19:10.880 through these you know early 20s into a fully 00:19:10.880 --> 00:19:14.059 formed human and and that real mentorship sees 00:19:14.059 --> 00:19:17.039 those two things equally yes like I think that 00:19:17.039 --> 00:19:20.710 like you know everyone in our field comes in 00:19:20.710 --> 00:19:23.009 with this like ambition right this raw ambition 00:19:23.009 --> 00:19:25.670 and like you have to nurture that like yeah and 00:19:25.670 --> 00:19:28.049 i think like somehow you know like on the outside 00:19:28.049 --> 00:19:31.210 i think it's like this like soup crazy shy you 00:19:31.210 --> 00:19:33.630 know girl like just shoveled hair and all that 00:19:33.630 --> 00:19:35.529 and it was like but when march like saw this 00:19:35.529 --> 00:19:38.529 like you know this is like inner flame and she 00:19:38.529 --> 00:19:41.430 like nurtured that and i i i hope i i do that 00:19:41.430 --> 00:19:43.809 for my trainees because they yeah they have their 00:19:43.809 --> 00:19:46.069 wild dreams and i think it's so important right 00:19:46.069 --> 00:19:49.240 yeah yeah developing the human as well The shyness. 00:19:49.259 --> 00:19:52.079 Tell me about the shyness. Is it gone or is it 00:19:52.079 --> 00:19:56.900 still there? It's totally still there. But you've 00:19:56.900 --> 00:19:58.380 learned to conquer it. I've learned to cope with 00:19:58.380 --> 00:20:02.240 it. We're talking right now somehow. I've seen 00:20:02.240 --> 00:20:04.119 you give in talks so many times and you're a 00:20:04.119 --> 00:20:06.839 very confident presenter. And I think if somebody 00:20:06.839 --> 00:20:10.160 just saw the talk... they wouldn't know the shy 00:20:10.160 --> 00:20:12.740 Doris. Oh yeah. No, when I go to a conference, 00:20:12.799 --> 00:20:14.640 like there's always like this moment when hotel 00:20:14.640 --> 00:20:16.319 lobby, I see all these colleagues and I just 00:20:16.319 --> 00:20:20.440 like feel this dread. Oh my God. Even the most 00:20:20.440 --> 00:20:22.640 extroverted people. Well, I think even the most 00:20:22.640 --> 00:20:24.799 extroverted people get that. Oh really? I do. 00:20:24.920 --> 00:20:27.720 Yeah. And then, yeah, I do. That's true. Oh, 00:20:27.779 --> 00:20:30.220 I certainly do. Yeah. I was very, very shy youngster. 00:20:30.279 --> 00:20:32.480 I sometimes say that to people and they're like, 00:20:32.519 --> 00:20:35.119 no, that's ridiculous. But, but it was, you know, 00:20:35.140 --> 00:20:37.680 but, and I, and I, I must say that's a, that's, 00:20:37.740 --> 00:20:39.559 I often have that, like, you know, you walk into 00:20:39.559 --> 00:20:41.720 the reception at one of those meetings and I 00:20:41.720 --> 00:20:44.200 go, oh, gosh, what if I'm just standing there 00:20:44.200 --> 00:20:46.500 alone as well? Just like, you know, you wade 00:20:46.500 --> 00:20:48.259 in, I suppose. Yeah. So you've had to conquer 00:20:48.259 --> 00:20:51.480 that as part of it. Because obviously the other 00:20:51.480 --> 00:20:54.019 pieces, you know, there's the doing the science, 00:20:54.119 --> 00:20:56.799 but the communicating part is equally important, 00:20:57.000 --> 00:21:01.440 right? Yeah, yeah. I have to. Yeah, again, thank 00:21:01.440 --> 00:21:03.140 my mentors for that, like teaching me how to 00:21:03.140 --> 00:21:06.039 give talks and so on. You know, we're living 00:21:06.039 --> 00:21:09.960 in strange political times for science. And they 00:21:09.960 --> 00:21:13.019 sort of, you know, the public perception of science 00:21:13.019 --> 00:21:15.480 as, you know, there was a time I think I read 00:21:15.480 --> 00:21:18.740 recently, you know, if you go back 10, 15 years, 00:21:18.940 --> 00:21:22.339 public trust in scientists was in the high 90%. 00:21:22.339 --> 00:21:25.240 98 % of people had complete trust in the system. 00:21:25.279 --> 00:21:27.220 I think those numbers would not be there today. 00:21:28.539 --> 00:21:32.019 What are we doing wrong and how do we get at 00:21:32.019 --> 00:21:33.500 that? How do we get at science communication 00:21:33.500 --> 00:21:36.980 that brings our public along with us? Do you 00:21:36.980 --> 00:21:39.319 have thoughts around that? Yes, I think we need 00:21:39.319 --> 00:21:43.619 to tell people about our science, make many more 00:21:43.619 --> 00:21:45.700 efforts to do that. And I think this podcast 00:21:45.700 --> 00:21:51.940 is a great way to do that. I myself am like writing 00:21:51.940 --> 00:21:55.859 a book now. I'm trying to link neuroscience, 00:21:56.140 --> 00:21:59.839 philosophy, and self -help. Because I think that 00:21:59.839 --> 00:22:02.740 there's this whole genre of neuroscience -based 00:22:02.740 --> 00:22:04.519 self -help that's around how you can sleep better, 00:22:04.579 --> 00:22:06.779 how you can eat better, and so on. But I'm interested. 00:22:06.980 --> 00:22:09.480 I believe that the mind -body problem, insights 00:22:09.480 --> 00:22:12.819 into consciousness, into free will, those insights. 00:22:13.420 --> 00:22:15.240 don't live in some pure realm, right? They're 00:22:15.240 --> 00:22:17.039 about the brain and the brain is what we used 00:22:17.039 --> 00:22:19.160 to behave with. And they have like profound implications 00:22:19.160 --> 00:22:21.539 for how we should live our life. And I want to 00:22:21.539 --> 00:22:24.119 work that out. And I, so yeah. So you're waiting. 00:22:24.779 --> 00:22:26.960 solidly into this space to bring neuroscience. 00:22:26.960 --> 00:22:30.880 Yes, I think yes. And I hope then to go out into 00:22:30.880 --> 00:22:33.220 the Oakland Unified School District where there 00:22:33.220 --> 00:22:35.319 are many underrepresented students who may not 00:22:35.319 --> 00:22:38.039 be thinking about careers in neuroscience, but 00:22:38.039 --> 00:22:41.319 they do care about how should I live my day tomorrow? 00:22:41.660 --> 00:22:43.680 Do I have free will? I think they would be totally 00:22:43.680 --> 00:22:46.000 captivated by that question if they could discover 00:22:46.000 --> 00:22:47.859 that there's a link between that and electrical 00:22:47.859 --> 00:22:50.119 activity in the brain. Maybe they will change 00:22:50.119 --> 00:22:54.259 their career trajectory. Outstanding. Thank you 00:22:54.259 --> 00:22:56.519 so much. Thanks so much for spending some time 00:22:56.519 --> 00:22:58.859 with us and allowing us to get to know you a 00:22:58.859 --> 00:23:01.380 little bit better, Doris. You're a major star 00:23:01.380 --> 00:23:03.319 in the field. Look forward to watching you in 00:23:03.319 --> 00:23:05.940 the decades ahead and see how the work evolves. 00:23:06.160 --> 00:23:07.279 Thank you so much. This was fun.