Dr. Eddie Chang: The Science of Learning & Speaking Languages

00:00:00.000 | - Welcome to the Huberman Lab Podcast,

00:00:02.280 | where we discuss science and science-based tools

00:00:04.880 | for everyday life.

00:00:05.900 | I'm Andrew Huberman,

00:00:10.120 | and I'm a professor of neurobiology and ophthalmology

00:00:13.200 | at Stanford School of Medicine.

00:00:15.180 | Today, my guest is Dr. Eddie Chang.

00:00:17.600 | Dr. Eddie Chang is the chair of the neurosurgery department

00:00:20.440 | at the University of California at San Francisco.

00:00:23.720 | Dr. Chang's clinical group focuses on the treatment

00:00:26.300 | of movement disorders, including epilepsy.

00:00:29.040 | He is also a world expert in the treatment

00:00:31.080 | of speech disorders and relieving paralysis

00:00:33.680 | that prevents speech and other forms

00:00:35.440 | of movement and communication.

00:00:37.320 | Indeed, his laboratory is credited with discovering ways

00:00:40.480 | to allow people who have fully locked in syndrome,

00:00:43.060 | that is who cannot speak or move,

00:00:45.040 | to communicate through computers and AI devices

00:00:48.240 | in order to be able to speak to others in their world

00:00:51.000 | and understand what others are saying to them.

00:00:53.720 | It is a truly remarkable achievement that we discussed today

00:00:56.560 | in addition to his discoveries about critical periods,

00:00:58.760 | which are periods of time during one's life

00:01:00.860 | when one can learn things in particular languages

00:01:03.360 | with great ease, as opposed to later in life.

00:01:05.880 | And we talk about the basis of things like bilingualism

00:01:08.760 | and trilingualism.

00:01:10.140 | We talk about how the brain controls movement

00:01:12.440 | of the very muscles that allow for speech and language

00:01:15.000 | and how those can be modified over time.

00:01:17.080 | We also talk about stutter,

00:01:18.480 | and we talk about a number of aspects of speech and language

00:01:21.200 | that give insight into not just how we create

00:01:23.520 | this incredible thing called speech

00:01:25.000 | or how we understand speech and language,

00:01:27.300 | but how the brain works more generally.

00:01:29.340 | Dr. Chang is also one of the world leaders

00:01:31.160 | in bioengineering, that is the creation of devices

00:01:34.540 | that allow the brain to function

00:01:36.420 | at super physiological levels

00:01:38.040 | and that can allow people with various syndromes

00:01:40.060 | and disorders to overcome their deficits.

00:01:42.760 | So if you are somebody who is interested

00:01:44.360 | in how the brain works normally,

00:01:46.300 | how it breaks down and how it can be repaired,

00:01:48.920 | and if you are interested in speech and language,

00:01:51.700 | reading and comprehension of information of any kind,

00:01:54.600 | today's episode ought to include some information

00:01:56.840 | of deep interest to you.

00:01:58.580 | Dr. Chang is indeed the top of his field

00:02:01.320 | in terms of understanding these issues

00:02:03.020 | of how the brain encodes speech and language

00:02:04.940 | and create speech and language,

00:02:06.780 | and as I mentioned, movement disorders and epilepsy.

00:02:09.780 | We even talk about things such as the ketogenic diet,

00:02:12.160 | the future of companies like Neuralink,

00:02:14.440 | which are interested in bioengineering

00:02:16.480 | and augmenting the human brain, and much more.

00:02:19.760 | One thing that I would like to note is that

00:02:21.400 | in addition to being a world-class neuroscience researcher

00:02:24.220 | and world-class clinician neurosurgeon,

00:02:26.760 | and chair of neurosurgery, Dr. Eddie Chang

00:02:29.240 | has also been a close personal friend of mine

00:02:31.860 | since we were nine years old.

00:02:33.440 | We attended elementary school together,

00:02:35.320 | and we actually had a science club

00:02:37.460 | when we were nine years old,

00:02:38.520 | focused on a very particular topic.

00:02:40.200 | You'll have to listen in to today's episode

00:02:41.900 | to discover what that topic was

00:02:43.440 | and what membership to that club required.

00:02:46.720 | That aside, Dr. Chang is an absolute phenom

00:02:50.340 | with respect to his scientific prowess,

00:02:52.360 | that is both his research and his clinical abilities,

00:02:55.360 | and he's one of these rare individuals

00:02:57.320 | that whenever he opens his mouth, we learn.

00:03:00.440 | Before we begin, I'd like to emphasize that this podcast

00:03:03.040 | is separate from my teaching and research roles at Stanford.

00:03:05.760 | It is, however, part of my desire and effort

00:03:07.680 | to bring zero cost to consumer information

00:03:09.520 | about science and science-related tools

00:03:11.600 | to the general public.

00:03:12.880 | In keeping with that theme,

00:03:13.920 | I'd like to thank the sponsors of today's podcast.

00:03:16.680 | Our first sponsor is Levels.

00:03:18.540 | Levels is a program that helps you see

00:03:20.080 | how different foods affect your health

00:03:21.700 | by giving you real-time feedback on your diet

00:03:23.920 | using a continuous glucose monitor.

00:03:26.420 | I started using Levels about one year ago.

00:03:29.600 | The Levels Monitor allowed me to see

00:03:31.120 | how different foods change my blood sugar level

00:03:33.480 | or my blood glucose level.

00:03:35.100 | This turns out to be immensely important

00:03:36.900 | for being able to predict how, for instance,

00:03:39.040 | certain foods will affect your energy level,

00:03:41.280 | your ability to exercise,

00:03:42.560 | your ability to recover from exercise,

00:03:44.420 | and how it will affect other hormones

00:03:46.160 | like testosterone, estrogen, thyroid hormone, and so forth.

00:03:49.960 | The other thing about using a Levels Monitor

00:03:52.040 | is that it gave me insight into how food and exercise

00:03:55.060 | and other activities and even how well I was sleeping

00:03:57.160 | or how poorly I might happen to be sleeping

00:03:59.580 | impact my blood glucose levels.

00:04:01.520 | It even taught me that the sauna

00:04:03.480 | that generating a lot of heat in my body

00:04:05.960 | was changing my blood glucose levels,

00:04:07.920 | which turned out to inform

00:04:09.160 | how I should shift my eating patterns,

00:04:11.120 | foods I should eat, timing of eat, and so on and so forth.

00:04:13.840 | Really gave me great insight

00:04:14.960 | into how all the important aspects of my health

00:04:17.460 | were interlocking and affecting one another,

00:04:19.240 | not just how food was impacting my blood glucose.

00:04:22.360 | So if you're interested in learning more about levels

00:04:24.200 | and trying a continuous glucose monitor yourself,

00:04:26.840 | you can go to levels.link/huberman.

00:04:29.640 | That's levels.link/huberman.

00:04:32.080 | Today's episode is also brought to us by Eight Sleep.

00:04:34.760 | Eight Sleep makes smart mattress covers

00:04:36.400 | with cooling, heating, and sleep tracking capacity.

00:04:39.460 | I started sleeping on an Eight Sleep mattress cover

00:04:41.620 | a few months ago, and it is simply incredible.

00:04:44.420 | In fact, I don't even like traveling anymore

00:04:46.200 | because they don't have Eight Sleep mattress covers

00:04:48.700 | in hotels and Airbnbs.

00:04:50.460 | One of the reasons I love

00:04:51.340 | my Eight Sleep mattress cover so much

00:04:53.340 | is that, as you may have heard before on this podcast

00:04:55.760 | or elsewhere, in order to fall and stay deeply asleep,

00:04:58.800 | you need your body temperature to drop

00:05:00.320 | by about one to three degrees.

00:05:01.940 | And I tend to run warm at night,

00:05:03.940 | which makes it hard to sleep

00:05:05.040 | and sometimes wakes me up in the middle of the night.

00:05:07.040 | When you sleep on an Eight Sleep mattress cover,

00:05:08.680 | you can program the temperature of that mattress cover

00:05:11.300 | for specific times in the early, middle,

00:05:13.520 | and late part of your night

00:05:14.960 | so that the mattress stays cool.

00:05:16.640 | And as a consequence, you sleep very, very deeply.

00:05:19.320 | It also tracks your sleep.

00:05:21.000 | So it's paying attention to how many times you're moving,

00:05:23.320 | how deep your sleep is, it gives you a sleep score,

00:05:25.580 | all wonderful data to help you enhance your sleep.

00:05:27.760 | And of course, sleep is the foundation of mental health,

00:05:30.180 | physical health, and performance,

00:05:31.740 | which makes an Eight Sleep a terrific tool for enhancing

00:05:34.460 | not just your sleep, but all aspects of your life, really.

00:05:37.520 | If you're interested in trying the Eight Sleep mattress

00:05:39.320 | cover, you can go to eightsleep.com/huberman

00:05:42.100 | to check out the pod three cover,

00:05:43.580 | and you can save $150 at checkout.

00:05:45.980 | Eight Sleep currently ships to the USA, to Canada, the UK,

00:05:49.620 | and select countries in the EU and Australia.

00:05:51.860 | Again, that's eightsleep.com/huberman

00:05:54.300 | to save $150 at checkout.

00:05:56.560 | Today's episode is also brought to us by Inside Tracker.

00:05:59.460 | Inside Tracker is a personalized nutrition platform

00:06:01.940 | that analyzes data from your blood and DNA

00:06:04.580 | to help you better understand your body

00:06:06.220 | and help you reach your health goals.

00:06:07.940 | I've long been a believer in getting regular blood work done

00:06:10.620 | for the simple reason that many of the factors

00:06:12.940 | that impact your immediate and long-term health

00:06:15.060 | can only be analyzed with a quality blood test.

00:06:17.740 | One problem with a lot of DNA tests and blood tests,

00:06:20.300 | however, is you get data back

00:06:22.220 | about levels of metabolic factors,

00:06:24.280 | levels of hormones, et cetera,

00:06:25.900 | but you don't know what to do with that information.

00:06:27.780 | Inside Tracker makes interpreting your data

00:06:29.460 | and knowing what to do about it exceedingly easy.

00:06:32.700 | They have a personalized platform where you can go

00:06:34.820 | and you can see those levels of hormones,

00:06:36.500 | metabolic factors, lipids, et cetera,

00:06:38.380 | and they point to specific nutritional tools,

00:06:40.740 | behavioral tools, supplement-based tools, et cetera,

00:06:43.760 | that can help you bring those numbers into the ranges

00:06:45.760 | that are optimal for you.

00:06:46.980 | If you'd like to try Inside Tracker,

00:06:48.400 | you can go to insidetracker.com/huberman

00:06:51.300 | to get 20% off any of Inside Tracker's plans.

00:06:54.120 | Again, that's insidetracker.com/huberman to get 20% off.

00:06:58.360 | The Huberman Lab podcast

00:06:59.500 | is now partnered with Momentus Supplements.

00:07:01.340 | To find the supplements we discuss

00:07:02.660 | on the Huberman Lab podcast,

00:07:03.880 | you can go to Live Momentus, spelled O-U-S,

00:07:06.500 | livemomentus.com/huberman.

00:07:09.060 | And I should just mention

00:07:09.900 | that the library of those supplements is constantly expanding.

00:07:12.640 | Again, that's livemomentus.com/huberman.

00:07:15.820 | And now for my discussion with Dr. Eddie Chang.

00:07:18.760 | Eddie, welcome.

00:07:19.700 | - Hi, hi, Andrew.

00:07:21.200 | - Great to be here with you.

00:07:22.620 | This has been a long time coming.

00:07:24.220 | Just to come clean,

00:07:27.580 | we've known each other since we were nine years old.

00:07:29.660 | - Yeah.

00:07:30.480 | - But then there was a long gap

00:07:31.500 | in which we didn't talk to one another.

00:07:33.800 | I heard things about you

00:07:35.060 | and presumably you heard a thing or two about me,

00:07:38.120 | for better or for worse.

00:07:39.460 | And then we reconnected years later

00:07:41.700 | when I was a PhD student and you were a medical student.

00:07:44.660 | We literally ran into each other

00:07:46.360 | in the halls of University of California, San Francisco,

00:07:49.040 | where you're now the chair of neurosurgery.

00:07:50.780 | So it all comes full circle.

00:07:53.540 | When you were at UCSF,

00:07:54.780 | you were working with Mike Merzenich.

00:07:56.820 | And I know that name might not be familiar

00:07:58.420 | to a lot of people,

00:07:59.260 | but he's sort of synonymous with neuroplasticity,

00:08:02.280 | the ability of the brain and nervous system

00:08:03.840 | to change in response to experience.

00:08:05.940 | So for our listeners,

00:08:06.800 | I would just love for you to give a brief overview

00:08:09.800 | of what you were doing at that time,

00:08:11.160 | because I find that work so fascinating

00:08:12.980 | and it really points to some of the things

00:08:15.080 | that can promote and maybe hinder

00:08:18.460 | our brain's ability to change.

00:08:19.980 | - Oh, wow, that's fantastic.

00:08:22.600 | So we did bump into each other serendipitously back then.

00:08:26.580 | And at the time I was a medical student at UCSF

00:08:30.300 | studying with Mike Merzenich.

00:08:31.660 | In particular, I was studying how the brain organizes

00:08:36.660 | when you have patterns of sound.

00:08:40.060 | And in particular, we were studying the brain of rodents

00:08:43.740 | and trying to understand how different sound patterns

00:08:45.720 | organize the frequency representation

00:08:49.060 | from low to middle to high frequency maps

00:08:52.000 | in the brains of baby rodents.

00:08:54.200 | And one of the things that I was very interested in

00:08:57.000 | was trying to understand how the patterns

00:08:59.460 | of the natural environment,

00:09:01.460 | let's say the vocalizations of the environment

00:09:03.900 | that the rat pups were raised in,

00:09:05.920 | or just the natural sounds that they hear,

00:09:07.760 | how that shapes the structure of the brain.

00:09:11.580 | And one of the things we did was to try and experiment

00:09:14.800 | where we raised some of these rat pups in white noise,

00:09:19.020 | continuous white noise that was essentially masking

00:09:22.140 | all of those environmental sounds.

00:09:24.620 | - And what was the consequence of animals

00:09:26.620 | being raised in white noise environment?

00:09:28.760 | - Well, one of the things that we didn't expect,

00:09:30.620 | but we found, which is quite striking,

00:09:32.620 | is that there's this early period in brain development

00:09:37.660 | where we're very susceptible to the patterns

00:09:40.740 | that we hear or see.

00:09:42.380 | In neuroscience, we call this a critical period

00:09:44.660 | or a sensitive period.

00:09:46.780 | And we have this for our eyes,

00:09:49.220 | but we also have it for our ears.

00:09:51.020 | And one of the most striking examples of this

00:09:54.620 | is that any human can essentially grow up in a culture

00:09:58.240 | where they hear different speech sounds

00:10:00.200 | from one language to another.

00:10:01.740 | And it's like after a couple of years,

00:10:03.800 | you lose sensitivity to sounds that are not part

00:10:06.460 | of your native language and you have high sensitivity

00:10:09.420 | for the languages of your native culture.

00:10:13.040 | And that's pretty extraordinary

00:10:15.980 | that human brain has that flexibility,

00:10:18.900 | yet at the same time has that specialization for language.

00:10:22.260 | And so we were trying to think about how do we model this,

00:10:25.340 | for example, in rodents who obviously don't speak,

00:10:28.460 | but we're just understanding how sounds

00:10:31.660 | and environmental sounds modulate

00:10:33.660 | and organize the auditory cortex.

00:10:36.160 | And one of the things that we found that was quite striking

00:10:38.780 | was that if you basically mask environmental sounds

00:10:43.780 | from these rat pups, the critical period,

00:10:47.600 | this sensitive period where it's open to plasticity,

00:10:50.780 | it's open to change, it's open to reorganization,

00:10:54.080 | that actually, that window can stay open much, much longer.

00:10:58.100 | And in one way, it sounds like that's a good thing,

00:11:01.380 | but on the other hand, it's also a retardation.

00:11:05.200 | It's actually, it slowed the maturation

00:11:07.460 | of the auditory cortex.

00:11:09.060 | It was ready to close when these rat pups were really young,

00:11:13.000 | but by raising them in white noise,

00:11:14.700 | we found out that you could keep it open for months

00:11:18.140 | beyond the time period that it normally closes.

00:11:22.220 | And so I think one of the things it taught me

00:11:24.460 | was that it's not just about the genetic programming

00:11:28.560 | that specifies some of this sensitive period,

00:11:31.320 | but it's also a little bit about the nature

00:11:33.840 | of the sounds that we hear that help keep that window

00:11:37.940 | for the critical period open and closed.

00:11:40.420 | - It's fascinating.

00:11:41.260 | And I know it's difficult to make a direct leap

00:11:43.440 | from animal research to human research,

00:11:46.200 | but if we could speculate a little bit,

00:11:49.620 | I can imagine that some people grow up in homes

00:11:51.480 | where there's a lot of shouting and a lot of inflection.

00:11:54.580 | Maybe people are very verbose.

00:11:56.300 | Maybe others grow up in a home

00:11:58.180 | where it's quieter and more peaceful.

00:12:02.060 | Some people are going to grow up in cities.

00:12:03.860 | I just came back from New York City.

00:12:04.960 | It's like all night long, there's honking and sirens,

00:12:07.500 | and it's just nonstop.

00:12:09.260 | And then I return here where it's quite quiet at night.

00:12:12.920 | Can we imagine that the human brain

00:12:16.020 | is going to be shaped differently

00:12:17.240 | depending on whether or not one grows up

00:12:19.380 | in one environment or another,

00:12:20.480 | and would that impact their tendency to speak

00:12:23.920 | in a certain way as well as hear in a certain way?

00:12:26.760 | What do we know about that?

00:12:28.120 | Well, I think that it's, from my perspective,

00:12:32.540 | it's really clear that those sounds that we are exposed to

00:12:36.360 | from the very earliest time, even in utero in the womb,

00:12:40.700 | where the sound is hearing the mother or father or friends

00:12:45.400 | while in the womb,

00:12:46.240 | actually will influence how these things organize.

00:12:49.300 | So there's no question that the sounds that we hear

00:12:54.320 | are going to have some influence,

00:12:55.580 | and those sounds are going to structure

00:12:58.120 | the way that those neural networks actually lay down

00:13:00.820 | and will forever influence how you hear sounds.

00:13:05.140 | And speech and language is probably

00:13:06.920 | one of the most profound examples of that.

00:13:09.840 | - I get a lot of questions

00:13:10.760 | about the use of white noise during sleep.

00:13:13.800 | In particular, people want to know

00:13:15.200 | whether or not using a white noise machine

00:13:17.500 | or a machine or a program

00:13:19.560 | that makes the sound of waves, for instance,

00:13:21.580 | if it assists their infant in sleeping,

00:13:24.320 | is it going to be bad for them

00:13:25.780 | because it's flooding the auditory system

00:13:27.860 | with a bunch of essentially white noise

00:13:29.900 | or disorganized noise?

00:13:32.320 | Do we have an answer to that question?

00:13:34.760 | - Not yet.

00:13:35.600 | I think that what you're asking is a really important

00:13:37.780 | question because parents are using white noise generators

00:13:41.760 | almost universally now, and for good reasons.

00:13:44.960 | It is hard to have kids up at night.

00:13:47.400 | I've got three kids of my own

00:13:48.880 | and was very tempted to think about

00:13:51.720 | how to use some of these tools

00:13:53.280 | to just soothe them and get them to bed,

00:13:55.860 | especially when I was so tired and exhausted.

00:13:58.720 | But I think that there is a cost

00:14:03.440 | to think a little bit about.

00:14:05.600 | We're not exposed to continuous white noise naturally.

00:14:10.440 | There is a value to having really salient, structured sounds

00:14:14.880 | that are part of our natural environment

00:14:16.280 | to actually have the brain develop normally.

00:14:18.840 | So whether or not that has an impact

00:14:22.960 | while you're sleeping, it's not clear.

00:14:26.160 | I don't think that those studies have been done.

00:14:28.400 | What was really clear was that

00:14:30.480 | if you raise these baby rats in continuous white noise,

00:14:34.300 | not super loud, but just enough to mask

00:14:37.080 | the environmental sounds, that that was enough

00:14:39.200 | to keep the auditory cortex, the part of the brain

00:14:42.560 | that hears in this really delayed state,

00:14:45.640 | which could essentially slow down the development

00:14:48.440 | and maturation of the brain.

00:14:50.480 | - And one probably assumed that slowing the maturation

00:14:55.180 | of areas of the brain they're responsible for hearing

00:14:57.440 | might underscore might impact one's ability to speak, right?

00:15:01.180 | Because isn't it the case that if people can't hear,

00:15:04.740 | they actually have a harder time enunciating

00:15:07.200 | in a particular way.

00:15:08.800 | Is that right?

00:15:09.640 | If I were to not be able to hear my own voice,

00:15:12.080 | would my speech patterns change?

00:15:14.900 | - Well, I think part of it is that over time,

00:15:17.880 | we develop sensitivity to the very specific speech sounds

00:15:21.520 | in a given language, and the sensitivity improves

00:15:26.360 | as we hear more and more and more of it.

00:15:28.800 | And then on the other hand, we lose sensitivity

00:15:31.260 | to other speech sounds at the same time.

00:15:33.680 | But as part of that process, we also have a selectivity,

00:15:38.680 | again, a specialization even for those sounds,

00:15:44.240 | even relative to noise, noisy backgrounds and things like

00:15:47.640 | that, I tend to think about it like what is the signal

00:15:50.720 | to noise ratio?

00:15:52.560 | And so the brain has its own ways of trying to increase

00:15:57.160 | that signal to noise ratio in order to make it more clear.

00:16:00.680 | Part of that is how we hear and how it lays down

00:16:04.800 | a foundation for that signal to noise ratio.

00:16:07.180 | And so you can imagine a child that's raised continuously

00:16:12.480 | in white noise would be really deprived of those kinds

00:16:15.240 | of sounds that are really necessary for it

00:16:16.840 | to develop properly.

00:16:18.480 | So I think with regard to those tools for babies,

00:16:23.120 | I think we should study, we should try to understand

00:16:25.560 | this definitively.

00:16:26.400 | I think what we saw, rodents would tell us

00:16:29.120 | that there is potential, things that we should be concerned

00:16:31.880 | about, but again, it's not really clear if you're just using

00:16:34.960 | at night whether it has those effects.

00:16:37.660 | - Yes, the critical question that a number of people

00:16:39.740 | are going to be asking is did you decide to use

00:16:43.900 | a white noise machine or not to help keep your,

00:16:48.420 | any of your three children asleep?

00:16:50.940 | - Well, I think the short answer is no.

00:16:53.620 | I mean, I obviously did a lot of work thinking

00:16:55.980 | and work on this and thought about it carefully,

00:16:59.580 | but there are other kinds of noise or I wouldn't even call

00:17:04.580 | it noise, other sounds that you can use that can be equally

00:17:07.840 | soothing to a baby.

00:17:09.880 | It's just that white noise has no structure

00:17:12.080 | and what it's doing is essentially masking out all

00:17:15.380 | of the natural sounds and I think the goal should really

00:17:18.400 | be about how do we replace that with other more natural

00:17:21.500 | sounds that structure the brain in the way that we want

00:17:23.720 | to be more healthy.

00:17:24.680 | - Well, I know that after you finished your medical

00:17:29.040 | training, you went on to, or specialized in neurosurgery

00:17:32.240 | and last I checked, you spend your most of your days

00:17:36.320 | either running your laboratory or in the clinic

00:17:38.140 | or running the department and your clinical work

00:17:40.580 | and your laboratory work involves often removing pieces

00:17:44.100 | of the skull of humans and going in and either removing

00:17:48.260 | things or stimulating neurons, treating various ailments

00:17:53.260 | of different kinds, but your main focus these days

00:17:57.260 | of course is the neurobiology of speech and language.

00:18:01.900 | And so for those that aren't familiar, could you please

00:18:04.740 | distinguish for us speech versus language in terms

00:18:09.080 | of whether or not different brain areas control them

00:18:11.620 | and I know that there's a lot of interest in how speech

00:18:14.980 | and language and hearing all relate to one another

00:18:18.140 | and then we'll talk a bit about for instance, emotions

00:18:22.060 | and how facial expressions could play into this

00:18:24.460 | or hand gestures, et cetera, but for the uninformed person

00:18:29.460 | and for me to be quite direct, what are the brain areas

00:18:35.140 | that control speech and language?

00:18:37.240 | What are they really, and especially in humans,

00:18:41.060 | how are they different?

00:18:41.900 | I mean, we have such a sophisticated language compared

00:18:44.720 | to a number of other species.

00:18:46.400 | What does all this landscape look like in there?

00:18:48.660 | - Yeah, well, that's a fascinating question

00:18:52.620 | and I'm going to just try to connect a couple of the dots

00:18:55.580 | here, which is that in that earlier work during medical

00:18:59.700 | school, I was doing a lot of what we call neurophysiology,

00:19:02.700 | putting electrodes into the auditory cortex

00:19:04.860 | and understanding how the brain responds to sounds

00:19:08.380 | and that's how we actually mapped out these things

00:19:10.940 | about the sensitivity to sensitive periods.

00:19:14.200 | That experience with Mike Merzenich and thinking about

00:19:18.380 | how plasticity is regulating the brain in particular

00:19:21.420 | about how sound is represented by brain activity

00:19:24.900 | was something that was really formative for me

00:19:28.700 | and because I was a medical student, I was going back

00:19:31.300 | to my medical studies.

00:19:32.620 | It was that in combination with seeing some awake brain

00:19:38.840 | surgeries that our department is really well known for.

00:19:43.020 | One of my mentors, Mitch Berger, really pioneered

00:19:45.540 | these methods for taking care of patients with brain tumor

00:19:49.080 | and be able to do these surgeries safely by keeping

00:19:52.920 | patients awake and by mapping out language.

00:19:55.420 | - So they're talking and listening and you're essentially

00:19:57.860 | in conversation with these patients while there's

00:19:59.980 | a portion of their skull removed and you are stimulating

00:20:02.540 | or in some cases removing areas of their brain,

00:20:05.740 | is that right?

00:20:06.580 | - That's exactly right and the only thing off there

00:20:10.380 | is it's not essentially, it is just that.

00:20:13.080 | The only difference between the conversation that I might

00:20:15.860 | have with my patient who's undergoing awake brain surgery

00:20:19.100 | is that I can't see their face and they can't see my face.

00:20:23.060 | We actually have a sterile drape that actually separates

00:20:26.580 | the operating field and they're looking and interacting

00:20:28.760 | with our neuropsychologist, but I can talk to them

00:20:31.760 | and they can hear my voice and vice versa.

00:20:34.780 | And it's a really, really important way of how we can

00:20:38.100 | protect some of those areas that are really critical

00:20:40.080 | for language at the same time accomplish the mission

00:20:43.580 | of getting the seizures under control or getting

00:20:45.620 | a brain tumor removed.

00:20:46.900 | - And is that because occasionally you'll encounter

00:20:48.740 | a brain area, maybe you're stimulating or considering

00:20:51.140 | removing that brain area and suddenly the patient

00:20:54.900 | will start stuttering or will have a hard time

00:20:57.340 | formulating a sentence?

00:20:58.500 | Is that essentially what you're looking for?

00:21:00.640 | You're looking for regions in which it is okay

00:21:04.700 | or not okay to probe?

00:21:06.500 | - Exactly, so the first thing that we do is that we use

00:21:10.180 | a small electrical stimulator to probe different parts

00:21:12.940 | of the areas that we think might be related and important

00:21:16.080 | for language or talking or even movements of your arm

00:21:18.860 | and leg, that's what we call brain mapping.

00:21:21.420 | And we use a small electrical current that's delivered

00:21:25.180 | through a probe that we can just put at each spot.

00:21:28.380 | And the areas that we're really interested in are,

00:21:30.020 | of course, the areas that are right around the part

00:21:32.860 | that is pathological, the part that's injured

00:21:35.300 | or the part that has a brain tumor that we want to remove.

00:21:38.300 | So we can apply that probe and transiently,

00:21:41.340 | meaning temporarily, activate it.

00:21:43.960 | So if you're stimulating the part of the brain

00:21:46.420 | that controls the hand, the hand will move.

00:21:49.060 | It will jerk, sometimes a fist will be made,

00:21:51.700 | something like that.

00:21:53.380 | Other times, while someone is counting or just saying

00:21:58.360 | the days of the week, you can stimulate in a different area

00:22:01.500 | that stops their speech altogether.

00:22:04.140 | That's what we call speech arrest.

00:22:06.300 | Or if someone is looking at pictures and they're describing

00:22:10.260 | the pictures and you stimulate a particular area,

00:22:12.580 | they stop speaking or the words start coming out slurred

00:22:16.220 | or they can't remember the name of the object

00:22:19.380 | that they're seeing in the picture.

00:22:20.700 | These are all things that we're listening really carefully

00:22:23.540 | while we apply that focal stimulation.

00:22:26.100 | That's what we call brain mapping.

00:22:27.780 | - What are some of the more surprising,

00:22:29.920 | or maybe even if you want to offer

00:22:31.200 | one of the more outrageous examples of things

00:22:33.560 | that people have suddenly done or failed to be able to do

00:22:36.920 | as a consequence of this brain mapping?

00:22:39.320 | - Well, I think the thing to me that has been

00:22:41.900 | the most striking is that, you know,

00:22:45.240 | some of these areas you stimulate and altogether,

00:22:48.220 | you can shut down someone's talking.

00:22:50.020 | So person says, "I wanted to say it,

00:22:55.020 | but I couldn't get the words out."

00:22:57.080 | And even though I've seen this thousands of times now,

00:23:01.680 | it's still exciting every time that I see it

00:23:04.880 | because it's exciting because you're seeing the brain.

00:23:09.880 | It's a physical organ, it's part of the body.

00:23:14.720 | Outside of the veins, on top of it,

00:23:17.700 | doesn't look like a machine.

00:23:21.460 | But when you do something like that

00:23:23.880 | and you've vocally changed the way it works,

00:23:27.320 | and you see that because a person can't talk anymore,

00:23:29.400 | and they say, "I know what I want to say,

00:23:31.300 | but I couldn't get the words out,"

00:23:33.000 | you're confronted with this idea that that organ

00:23:39.460 | is the basis of speech and language,

00:23:42.160 | and way beyond that, obviously,

00:23:45.040 | for all the other functions that we have for thinking

00:23:47.220 | and feeling our emotions, everything.

00:23:50.580 | So that, to me, is a constant reminder

00:23:54.200 | of this really special thing that the brain does,

00:23:58.420 | was compute so many of the things that we do.

00:24:01.740 | And in particular, in the area around speech and language,

00:24:04.120 | generating words, something that is really unique

00:24:07.760 | to our species, is just extraordinary to see.

00:24:12.760 | Again, even though I've seen it thousands of times,

00:24:15.800 | it's just having that connection

00:24:18.880 | because it doesn't look like machine,

00:24:20.320 | but it is doing something that is quite complicated,

00:24:23.940 | precise, and remarkable.

00:24:25.600 | - Do you ever see emotional responses

00:24:27.200 | from stimulation in particular areas?

00:24:29.280 | And do you ever hear or see emotional responses

00:24:34.020 | that are associated with particular types of speech?

00:24:36.720 | Because for instance, curse words are known to,

00:24:41.700 | people with Tourette's often will curse, not always,

00:24:43.960 | but sometimes they left texts or other things.

00:24:46.260 | But what I learned from a colleague of ours

00:24:48.600 | is that curse words have a certain structure to them.

00:24:51.120 | There's usually a heavy or kind of a sharp consonant

00:24:54.520 | up front that allows people,

00:24:57.760 | at least as it was described to me,

00:24:59.000 | to have some sort of emotional release.

00:25:00.420 | It's not a word like murmur,

00:25:02.240 | which it has a kind of a soft entry.

00:25:03.880 | Here, I'm not using the technical language.

00:25:05.960 | And you pick your favorite curse word out there, folks.

00:25:08.160 | I'm not going to shout out any now or say any now.

00:25:11.140 | But that certain words have a structure to them

00:25:15.300 | that because of the motor patterns

00:25:16.920 | that are involved in saying that word,

00:25:19.260 | you could imagine has an emotional response unto itself.

00:25:23.640 | So when stimulating

00:25:24.760 | or when blocking these different brain areas,

00:25:26.320 | do you ever see people get angry or sad or happy

00:25:28.960 | or more relaxed?

00:25:30.860 | - Oh, well, definitely I've seen cases

00:25:35.860 | where you can invoke anxiety, stress.

00:25:40.020 | And I think that there are also areas

00:25:45.780 | that you can stimulate.

00:25:46.940 | You can also evoke the opposite of that,

00:25:50.120 | sort of like a calm state.

00:25:51.840 | - I think that brain area is slightly hyperactive in you,

00:25:55.920 | or at least more than me.

00:25:57.840 | In all the years I've known you,

00:25:59.140 | you've always been, at least externally, a very calm person.

00:26:03.220 | I mean, I always find it amusing

00:26:04.220 | that you work on speech and language

00:26:05.500 | and you have a very calming voice, right?

00:26:07.720 | And I'm being really serious.

00:26:10.060 | I think that there was a huge variation there, right,

00:26:12.680 | in terms of how people speak and how they accent words.

00:26:15.800 | - Absolutely, yeah.

00:26:16.840 | So there are areas, for example,

00:26:19.900 | the orbital frontal cortex that we showed,

00:26:23.800 | that if you stimulate there,

00:26:25.080 | the orbital frontal cortex is a part of the brain

00:26:26.960 | that's above the eyes.

00:26:27.800 | That's why they call it orbital frontal,

00:26:30.120 | meaning it's above the eye or the orbit

00:26:32.440 | and in the frontal lobe.

00:26:34.240 | And it's this area right in here.

00:26:36.640 | It has really complex functions.

00:26:38.740 | It's really important for learning and memory.

00:26:41.640 | But one of the things that we observed

00:26:43.480 | is when you stimulate there,

00:26:44.900 | people tended to have a reduction in their stress.

00:26:48.960 | And it was very much related to their state of being,

00:26:53.440 | meaning that if someone was already kind of feeling normal

00:26:57.660 | and you stimulate there, it didn't do much.

00:26:59.640 | But if someone was in a very anxious state,

00:27:01.960 | it actually relieved that.

00:27:03.840 | And then we've seen the corollary of that,

00:27:06.380 | which is true too,

00:27:07.720 | which is that there are other areas like the amygdala

00:27:10.760 | or parts of the insula that if you stimulate,

00:27:14.160 | you can cause an acute temporary anxiety,

00:27:18.280 | a nervous feeling.

00:27:21.880 | Or if you stimulate an insula,

00:27:23.480 | people can have an acute feeling of disgust.

00:27:26.240 | So, you know, the brain has different functions

00:27:30.760 | and these different nodes

00:27:32.040 | that help process the way we feel.

00:27:34.680 | Certainly, I think that to some degree,

00:27:36.600 | neuropsychiatric conditions reflect an imbalance

00:27:39.440 | of the electrical activities in these areas.

00:27:42.840 | One of the things that was something I will never forget

00:27:48.720 | was taking care of a young woman with uncontrolled seizures.

00:27:53.720 | We call that epilepsy.

00:27:55.000 | It's a medical condition where someone has

00:27:57.200 | uncontrolled electrical activity in the brain.

00:27:59.620 | Sometimes you can see that as convulsions

00:28:01.640 | where people are shaking and lose consciousness.

00:28:04.680 | There are other kinds of seizures that people can have

00:28:07.080 | where they don't lose consciousness,

00:28:08.560 | but they can have experiences that just come out of nowhere.

00:28:12.040 | And it's just a result of electrical activity

00:28:14.960 | coming from the brain.

00:28:16.740 | And about six years ago,

00:28:19.480 | I took care of a young woman who was diagnosed

00:28:24.280 | psychiatrically with anxiety disorder for several years.

00:28:27.640 | It turns out that it wasn't really an anxiety disorder.

00:28:31.040 | It was actually that she had underlying seizures

00:28:33.960 | and epilepsy activating a part of her brain

00:28:37.060 | that evokes anxious feelings.

00:28:39.780 | - How was that discovered?

00:28:41.400 | Because I know a lot of people out there have anxiety.

00:28:43.220 | I mean, in the absence of a brain scan,

00:28:45.560 | or why would one suspect that maybe they have a tumor

00:28:50.880 | or some other condition that was causing those neurons

00:28:54.900 | to become hyperactive?

00:28:55.740 | - Yeah, that's really important

00:28:56.940 | because so many people have anxiety

00:28:58.860 | and the vast, vast majority are not having that

00:29:01.960 | because they're having seizures in the brain.

00:29:04.120 | I think one of the ways that this was diagnosed

00:29:06.900 | was that the nature of when she was having

00:29:11.000 | these panic attacks was not triggered by anything.

00:29:14.860 | They would just happen spontaneously.

00:29:17.520 | And that's what can happen with seizures sometimes.

00:29:19.500 | They just come out of nowhere.

00:29:21.280 | We don't fully understand what can trigger them,

00:29:23.680 | but they weren't things that were

00:29:26.060 | typically anxiety-provoking.

00:29:28.320 | This is something that just happened all of a sudden.

00:29:31.560 | And because you brought it up,

00:29:35.400 | this is not something that you can see on an MRI.

00:29:38.640 | We could not see and look at the structure of her brain

00:29:42.140 | with an MRI that she was having seizures.

00:29:45.180 | The only way that we could actually prove this

00:29:48.000 | was actually putting electrodes into her brain

00:29:51.600 | and proving that these attacks that she was having

00:29:56.600 | were localized to a part called the amygdala.

00:30:00.400 | It's a medial part of the temporal lobe, which is here.

00:30:03.460 | And associating the electrical activity

00:30:06.700 | that we were seeing on those electrodes

00:30:08.280 | with the symptoms that she had.

00:30:10.780 | And she ultimately needed a kind of surgery

00:30:13.280 | where she was awake in order to remove this safely.

00:30:18.280 | - Speaking of epilepsy, a number of people out there

00:30:22.520 | have epilepsy or know people who do.

00:30:24.880 | Are the drugs for epilepsy satisfactory?

00:30:28.380 | I think about things like Depakote

00:30:29.720 | and adjusting the excitation and inhibition of the brain.

00:30:33.320 | I mean, are there good drugs for epilepsy?

00:30:35.840 | We know there are not great drugs

00:30:37.640 | for a lot of other conditions.

00:30:38.800 | But, and how often does one need neurosurgery

00:30:43.800 | in order to treat epilepsy

00:30:45.620 | or can it be treated most often just using pharmacology?

00:30:49.520 | - Yeah, great question.

00:30:50.640 | Well, a lot of people have seizures

00:30:55.300 | that can be completely controlled by their medications,

00:30:58.160 | a lot.

00:30:59.680 | But there's about a one third of people who have epilepsy,

00:31:03.180 | which we define as anyone who's had three or more seizures,

00:31:07.560 | that about a third of them actually don't have control

00:31:13.280 | with all of the modern medications that we have nowadays.

00:31:16.160 | And some of the data suggests that if you have two

00:31:21.240 | or three medications, it actually doesn't matter necessarily

00:31:24.160 | which of the anti-seizure medications it is,

00:31:27.500 | but there is data suggests if you've just tried two or three

00:31:32.000 | the fourth, fifth, sixth, and beyond

00:31:35.720 | is not likely to help control it.

00:31:37.980 | So we are in a situation, unfortunately,

00:31:41.320 | where a lot of the medications are great for some people,

00:31:44.560 | but for another subset, they can't control it

00:31:47.440 | and it comes from a particular part of the brain.

00:31:49.440 | Now, fortunately, in that subset,

00:31:53.680 | there's another part of that group that can benefit

00:31:56.800 | from a surgery that actually either removes

00:31:58.980 | that part of the brain,

00:32:00.340 | and nowadays we'll use stimulators now

00:32:02.620 | to sometimes put electrical stimulation

00:32:05.940 | in that part of the brain to help reduce the seizures.

00:32:08.340 | - And you said a third of people with epilepsy

00:32:10.800 | might need neurosurgery.

00:32:12.300 | - Well, what I mean by that is they continue to have seizures

00:32:16.340 | that are not controlled by all medications,

00:32:18.520 | and there's going to be another subset of those

00:32:21.380 | that may benefit from a surgery.

00:32:24.760 | It's probably not that whole third, it's a subset of that.

00:32:28.140 | It's just to say that epilepsy can be really hard

00:32:30.840 | to get fixed, and for people where the seizures come from,

00:32:34.680 | one spot or an area, then surgery can do great.

00:32:39.380 | If it comes from multiple areas

00:32:42.320 | or if it comes from the whole brain,

00:32:43.500 | then we have to think about other methods to control it.

00:32:46.160 | Fortunately, nowadays, there's actually other ways.

00:32:49.360 | Surgery now to us doesn't just mean

00:32:52.060 | we're moving part of the brain.

00:32:53.560 | Half of what we do now is use the stimulators

00:32:57.160 | that modulate the state of the brain

00:32:58.440 | that can help reduce the seizures.

00:33:01.140 | - I've heard before that the ketogenic diet

00:33:04.720 | was originally formulated in order to treat epilepsy,

00:33:09.460 | and in particular in kids.

00:33:11.400 | Is that true, and why would being in a ketogenic state

00:33:15.560 | with low blood glucose reduce seizures?

00:33:18.640 | - That's a great question, and to be honest,

00:33:21.240 | I don't know actually if it was originally designed

00:33:24.720 | to treat seizures, but I can tell you for sure

00:33:27.080 | that for some people, just like with some medications,

00:33:31.240 | it can be a life-changing thing.

00:33:32.560 | It can completely change the way that the brain works,

00:33:36.480 | and it's not something that's for everybody,

00:33:39.560 | but for some people, there's no question,

00:33:41.320 | and it has some very beneficial effects.

00:33:43.640 | I think it's to be determined still,

00:33:45.480 | like why and how that works.

00:33:48.760 | - I've heard similar things about the ketogenic diet

00:33:50.800 | for people with Alzheimer's, dementia,

00:33:53.160 | that there's nothing particularly relevant

00:33:57.580 | about ketosis to Alzheimer's per se,

00:33:59.840 | but because Alzheimer's changes the way

00:34:02.960 | that neurons metabolize energy,

00:34:06.040 | that shifting to an alternate fuel source

00:34:07.920 | can sometimes make people feel better,

00:34:09.920 | and so a number of people are now trying it,

00:34:11.340 | but it's not as if blood glucose and having carbohydrates

00:34:16.200 | is causing Alzheimer's, and people get confused often

00:34:18.900 | that just because something can help

00:34:21.600 | doesn't mean that the opposite is harming somebody.

00:34:24.720 | So I find this really interesting.

00:34:26.560 | Sometime I'll check back with you about what's happening

00:34:28.460 | in terms of ketogenic diets and epilepsy,

00:34:31.300 | but you said that in some cases it can help.

00:34:34.820 | Has that observation been made both for children

00:34:37.360 | and for adults?

00:34:38.200 | Because I thought that originally the ketogenic diet

00:34:40.720 | for epilepsy was really for pediatric epilepsy.

00:34:44.040 | - Yeah, that's right.

00:34:44.860 | So a lot of its focus has really been on kids with epilepsy,

00:34:49.440 | but certainly it's a safe thing to try,

00:34:52.200 | so a lot of adults will try it as well.

00:34:54.840 | - Interesting.

00:34:55.800 | I'd like to take a quick break

00:34:57.320 | and acknowledge one of our sponsors, Athletic Greens.

00:35:00.220 | Athletic Greens, now called AG1,

00:35:02.680 | is a vitamin mineral probiotic drink

00:35:05.040 | that covers all of your foundational nutritional needs.

00:35:07.940 | I've been taking Athletic Greens since 2012,

00:35:10.640 | so I'm delighted that they're sponsoring the podcast.

00:35:12.840 | The reason I started taking Athletic Greens

00:35:14.440 | and the reason I still take Athletic Greens

00:35:16.520 | once or usually twice a day is that it gets me

00:35:19.640 | the probiotics that I need for gut health.

00:35:22.100 | Our gut is very important.

00:35:23.200 | It's populated by gut microbiota

00:35:25.760 | that communicate with the brain, the immune system,

00:35:27.520 | and basically all the biological systems of our body

00:35:29.920 | to strongly impact our immediate and long-term health.

00:35:33.560 | And those probiotics in Athletic Greens are optimal

00:35:36.580 | and vital for microbiota health.

00:35:39.240 | In addition, Athletic Greens contains a number

00:35:41.080 | of adaptogens, vitamins, and minerals

00:35:42.720 | that make sure that all of my foundational

00:35:44.600 | nutritional needs are met, and it tastes great.

00:35:48.040 | If you'd like to try Athletic Greens,

00:35:49.520 | you can go to athleticgreens.com/huberman,

00:35:52.900 | and they'll give you five free travel packs

00:35:54.880 | that make it really easy to mix up Athletic Greens

00:35:57.200 | while you're on the road, in the car,

00:35:58.480 | on the plane, et cetera.

00:35:59.760 | And they'll give you a year's supply of vitamin D3K2.

00:36:03.180 | Again, that's athleticgreens.com/huberman

00:36:05.860 | to get the five free travel packs

00:36:07.240 | and the year's supply of vitamin D3K2.

00:36:10.320 | I'm curious about epilepsy for another reason.

00:36:14.200 | I was taught that epilepsy is an imbalance

00:36:16.440 | in the excitation and inhibition in the brain.

00:36:18.800 | So you think about these electrical storms

00:36:20.360 | that give people either grand mal, you know,

00:36:22.340 | shaking and kind of convulsions.

00:36:25.680 | But years ago, I was reading a book,

00:36:27.760 | a wonderful book, actually,

00:36:29.160 | called "Einstein in Love" by Dennis Overby.

00:36:31.520 | It was about Einstein and his more,

00:36:34.400 | I guess, his personal life.

00:36:35.920 | Then people who knew him claimed

00:36:38.380 | that he would sometimes walk along

00:36:40.600 | and then every once in a while would just stop

00:36:42.840 | and kind of stare off into space for anywhere

00:36:44.600 | from a minute to three or five minutes.

00:36:47.360 | And it was speculated that he had absence seizures.

00:36:51.720 | What is an absence seizure?

00:36:53.320 | And the reason I ask is I occasionally will be walking along

00:36:56.680 | and I'll be thinking about something and I'll stop.

00:36:58.640 | But in my mind, I think I'm thinking during that time.

00:37:03.240 | But I realized that if I were to see myself

00:37:04.720 | from the outside, it might appear

00:37:06.000 | that I was just kind of absent.

00:37:07.920 | What is an absence seizure?

00:37:09.980 | Because it's so strikingly different

00:37:11.840 | in its description from, say, a grand malconvulsive seizure.

00:37:15.600 | - Sure.

00:37:16.440 | Well, like I mentioned before,

00:37:19.760 | depending on how the seizure activity spreads in the brain

00:37:23.040 | or how it actually propagates,

00:37:26.220 | if it stays in one particular spot

00:37:27.960 | and doesn't spread to the entire brain,

00:37:29.640 | it can have really different manifestation.

00:37:32.620 | It can represent really differently.

00:37:34.520 | So, absence seizure is just one category

00:37:37.480 | of different kind of seizures

00:37:38.640 | where you can lose consciousness

00:37:41.760 | basically and what I mean by that

00:37:43.240 | is that you're not fully aware

00:37:45.920 | of what's going on in your environment.

00:37:47.880 | Okay, so you're sort of taken offline temporarily

00:37:51.000 | from consciousness.

00:37:52.160 | But you could still be, for example, standing.

00:37:55.420 | And to people who are not paying attention,

00:37:58.000 | they may not even be aware that that's happening.

00:38:00.600 | - What are some other types of seizures?

00:38:03.440 | - Well, you know, I think some of the other kinds are,

00:38:07.760 | the classic ones are temporal lobe seizures.

00:38:09.920 | So these are ones that come from the medial structures

00:38:13.360 | like the amygdala and hippocampus.

00:38:15.680 | Oftentimes people, when they have seizures coming from that,

00:38:18.760 | they may taste something very unusual like a metallic taste

00:38:23.760 | or smell something like the smell of burning toast,

00:38:27.400 | something like that.

00:38:28.440 | There are some people with temporal lobe seizures

00:38:32.160 | will have deja vu.

00:38:33.320 | They will have that experience

00:38:34.600 | that you've been somewhere before.

00:38:36.680 | But that's just a precursor to the seizure.

00:38:39.840 | And it just highlights that when people have seizures

00:38:43.120 | coming from these areas, they sometimes hijack

00:38:46.840 | what that part of the brain is really for.

00:38:49.480 | So the amygdala and hippocampus, for example,

00:38:52.680 | are really important for learning and memory.

00:38:55.120 | It's not surprising that when people have seizures there

00:38:57.600 | that it can invoke a feeling of deja vu

00:39:00.720 | or that it can invoke a feeling of anxiety.

00:39:03.800 | And in the areas that are right next to it, for example,

00:39:10.060 | these areas are really important for processing smell.

00:39:14.740 | So these areas are right next to each other.

00:39:17.380 | So you can have these kind of complex set of symptoms,

00:39:20.100 | the weird taste, the smell of toast,

00:39:23.920 | and then a feeling of deja vu.

00:39:25.260 | That's classic for temporal lobe seizure.

00:39:28.100 | And it's because those parts of the brain that process

00:39:30.300 | those functions are right next to each other.

00:39:32.660 | - I'm told that I've had nocturnal seizures

00:39:37.020 | and I've woken up sometimes from sleep

00:39:38.900 | having felt as if I was having a convulsion,

00:39:41.180 | the sort of sense of buzzing in the back of the head.

00:39:44.260 | This happened to me two or three times in college.

00:39:46.340 | My girlfriend, well, I woke up and my girlfriend

00:39:49.300 | was very distraught, like you were having a seizure.

00:39:51.700 | I was having full convulsion in my sleep.

00:39:54.060 | What are, is that correct?

00:39:56.620 | Is there such a thing as nocturnal seizures?

00:39:58.980 | What do they reflect?

00:40:00.400 | They eventually stopped happening

00:40:02.460 | and I couldn't tether them to any kind of life event.

00:40:04.940 | I wasn't doing any kind of combat sport

00:40:07.460 | or anything at the time.

00:40:08.440 | I wasn't drinking alcohol much.

00:40:11.220 | It's never really been my thing.

00:40:12.820 | What are nocturnal seizures about?

00:40:16.000 | - Oh, well.

00:40:17.660 | - And do I need brain surgery?

00:40:18.940 | [laughing]

00:40:20.540 | - Nocturnal seizures are just another form.

00:40:23.620 | Like again, epilepsy and seizures

00:40:25.680 | can have so many different forms

00:40:27.200 | and not just like where in the brain,

00:40:30.140 | but also when they happen.

00:40:31.700 | And there are some people who, for whatever reason,

00:40:35.620 | it's very timed to the circadian rhythm.

00:40:38.480 | There's actually not just happening at night,

00:40:40.140 | but a certain period at night

00:40:41.480 | when people are in a certain stage of sleep

00:40:44.400 | that the brain is in a state that it's vulnerable

00:40:48.280 | to having a seizure.

00:40:49.820 | And so that's basically just one form of that.

00:40:53.020 | Again, it's not just about where it's coming from,

00:40:55.060 | but also when it's happening and how that's timed

00:40:57.600 | with other things that are happening with the body.

00:41:00.380 | - Interesting.

00:41:01.220 | Well, it eventually stopped happening,

00:41:02.440 | so I stopped worrying about it,

00:41:03.960 | but I haven't had seizures since.

00:41:07.160 | Returning to speech and language.

00:41:10.360 | When I was getting weaned in neuroscience,

00:41:13.720 | I learned that we have an area of the brain

00:41:15.320 | for producing speech,

00:41:16.520 | and we have an area of the brain for comprehending speech.

00:41:19.320 | What's the story there?

00:41:22.020 | Is it still true that we have a Broca's

00:41:24.340 | and a Wernicke's area?

00:41:26.160 | Those are names of neurologists, presumably,

00:41:28.320 | or neurosurgeons that discover these different brain areas.

00:41:31.180 | Maybe you could familiarize us

00:41:32.460 | with some of the sort of textbook version

00:41:35.940 | of how speech and language are organized in the brain.

00:41:39.020 | Maybe share with us a little bit of the lesion studies

00:41:40.940 | that led to that understanding.

00:41:42.460 | And then I would love to hear a bit

00:41:44.460 | about what your laboratory is discovering

00:41:46.180 | about how things are actually organized,

00:41:48.620 | because from some discussions you and I have had

00:41:51.740 | over the last year or so, it seems like,

00:41:54.120 | well, let's just be blunt.

00:41:55.020 | It seems that much of what we know

00:41:56.740 | from the textbooks could be wrong.

00:41:59.620 | - Well, I love that question,

00:42:02.860 | because for me, it's very central to the research we do,

00:42:07.020 | and it's where the intersection

00:42:08.140 | between what we do in the laboratory

00:42:10.980 | and our research interfaces with what I see in patients.

00:42:14.980 | And one of the things that fascinated me early on

00:42:19.320 | in my medical training was,

00:42:21.780 | in doing some of these brain mapping,

00:42:23.340 | or watching them with my mentor,

00:42:24.940 | or taking care of patients that had brain tumors

00:42:28.820 | in certain part of the brain,

00:42:30.740 | was that a lot of times what I was seeing in a patient

00:42:33.260 | did not correlate with what I was taught in medical school.

00:42:36.280 | And some people will think,

00:42:40.220 | well, this might be an exception.

00:42:43.040 | But after you see it for a couple of times,

00:42:45.100 | and if you're kind of interested in this problem,

00:42:47.780 | it poses a serious challenge to what you've learned

00:42:52.780 | and how you think about how these things operate.

00:42:57.020 | And that actually got me really interested

00:42:59.780 | in trying to figure this out,

00:43:01.640 | because earlier we talked about

00:43:04.060 | just this extraordinary thing

00:43:06.140 | that the brain is doing to create words and sentences,

00:43:09.780 | and that's the process by which I'm getting ideas out

00:43:13.240 | from my mind into yours.

00:43:15.900 | It's an incredible thing, right?

00:43:16.940 | It's the basis of communication,

00:43:19.080 | high information communication between two individuals

00:43:25.020 | that's really unique to humans.

00:43:27.180 | So in historical times,

00:43:32.180 | how this works has been very controversial

00:43:36.100 | from day one of neuroscience.

00:43:38.680 | Long time ago, people thought the bumps on your head

00:43:45.060 | corresponded to the different faculties of the mind.

00:43:49.400 | So for example, if you had a bump here,

00:43:51.380 | it might be corresponding to intelligence

00:43:54.080 | or another one over here to vision and these kind of things.

00:43:58.820 | That's what we nowadays call phrenology.

00:44:01.080 | And that was kind of the starting point.

00:44:05.080 | A lot of that has been, of course, debunked,

00:44:07.220 | but when you see those little statues

00:44:08.860 | of different brain partitions on someone's head,

00:44:12.600 | that's essentially how people were thinking

00:44:15.940 | about how the brain worked back then a couple hundred years ago.

00:44:21.520 | Modern neuroscience began when,

00:44:25.300 | actually was very much related to the discovery of language.

00:44:28.780 | So modern neuroscience, meaning moving beyond this idea

00:44:32.340 | that the bumps on the scalp corresponded

00:44:34.540 | to the faculties of the mind,

00:44:36.140 | but there were things that actually

00:44:37.420 | were in the brain themselves,

00:44:38.860 | and they weren't corresponding to things

00:44:40.580 | that you could see superficially,

00:44:42.620 | like on the scalp or externally,

00:44:46.240 | that it was something about the brain itself.

00:44:48.340 | I mean, it seems so obvious now,

00:44:50.020 | but back then, this was the big academic debate.

00:44:55.020 | And the first observation that I think really

00:44:59.080 | was really impactful in the area of language

00:45:02.020 | was an observation by a neurosurgeon,

00:45:04.940 | a French neurosurgeon named Pierre Broca.

00:45:07.000 | And what he observed was that in a patient,

00:45:11.000 | not that he did surgery,

00:45:12.120 | but that he had seen and taken care of,

00:45:15.380 | that the person couldn't talk.

00:45:17.780 | And in particular, they called this individual tan,

00:45:21.180 | because the only words that he could produce was tan, tan.

00:45:26.180 | For the most part, he could generally understand

00:45:28.740 | the kind of things that people were asking him about,

00:45:31.980 | but the only thing that he could utter from his mouth

00:45:35.380 | were these words, tan, tan.

00:45:37.840 | And what eventually had happened

00:45:41.380 | was this individual passed away.

00:45:43.420 | And the way that neuroscience was done back then

00:45:47.100 | was basically to wait until that happened,

00:45:49.860 | and then to remove the brain,

00:45:51.140 | and to see what part of the brain was affected

00:45:54.260 | in this patient that they called tan.

00:45:56.580 | And what Broca found was that there was a part

00:46:00.440 | in the left frontal lobe.

00:46:02.240 | So the frontal lobe is this area like I described earlier,

00:46:05.780 | which is behind our forehead up here.

00:46:09.940 | And in the back of that frontal lobe,

00:46:15.780 | he claimed that this was the seat of articulation

00:46:20.780 | in the brain.

00:46:21.860 | He literally used something like that in France,

00:46:23.620 | the seat of articulation,

00:46:25.380 | meaning that this is the part of the brain

00:46:27.100 | that is responsible for us to generate words.

00:46:30.620 | About 50 years later, the story becomes more complicated

00:46:36.780 | with a German neurologist named Karl Wernicke.

00:46:41.040 | And what Wernicke described was a different set of symptoms

00:46:46.040 | in patients that he observed, a different phenomenon,

00:46:50.840 | where people could produce words, but a lot of the words,

00:46:55.840 | and they were fluent in the sense that they had like,

00:47:00.460 | they sound like they could be real words,

00:47:04.520 | but from a different language, for example.

00:47:07.400 | And some of us call that like word salad or jargon.

00:47:11.000 | It's essentially, they were essentially making up words,

00:47:14.100 | but it was not intentional.

00:47:15.180 | It was just the way that the words came out.

00:47:17.640 | But in addition to that, he observed that these people

00:47:20.960 | also could not understand what was being said to them.

00:47:24.900 | So we could be having a conversation

00:47:27.800 | and I'd be asking you, am I a woman?

00:47:30.800 | And you might nod your head, you know,

00:47:32.480 | just because you're not processing the question, you know.

00:47:36.560 | And so here are two observations.

00:47:41.560 | One is that the frontal lobe is important

00:47:45.640 | for articulating speech, creating the words

00:47:49.120 | and expressing them fluently.

00:47:51.160 | And then a different part of the brain

00:47:52.720 | called the left temporal lobe,

00:47:55.120 | which is this area right above my ear.

00:47:56.960 | That is an area that I think was claimed

00:48:03.280 | to be really important for understanding.

00:48:05.980 | So the two major functions in language,

00:48:09.000 | to speak and to understand, were kind of pinned down to that.

00:48:13.300 | And we've had that basic idea in the textbooks

00:48:18.300 | for, you know, over 200 years.

00:48:20.560 | - Certainly what I was taught.

00:48:22.000 | - Is that right?

00:48:22.840 | - Oh, every, yeah.

00:48:23.780 | And certainly what we still teach undergraduates,

00:48:26.880 | graduate students and medical students that.

00:48:29.240 | - Well, that's what I learned too in medical school.

00:48:31.520 | And what I saw in reality,

00:48:33.820 | when I started taking care of patients,

00:48:35.240 | was that it's not so simple.

00:48:37.200 | In fact, part of it is fundamentally wrong.

00:48:40.800 | So just in a nutshell, nowadays,

00:48:43.940 | after looking at this very carefully

00:48:46.200 | over hundreds of patients,

00:48:47.460 | we've shown that surgeries, for example,

00:48:50.480 | in the posterior part of the frontal lobe,

00:48:52.840 | a lot of times people have no problem

00:48:55.020 | talking at all whatsoever after those kinds of surgeries.

00:48:59.880 | And that it's a different part of the brain,

00:49:02.440 | what we call the precentral gyrus.

00:49:05.280 | The precentral gyrus is a part of the brain

00:49:07.900 | that is intimately associated with the motor cortex.

00:49:12.040 | The motor cortex is the part of the brain

00:49:13.560 | that has a map of your entire body.

00:49:16.720 | So it has a part that corresponds to your feet,

00:49:19.340 | it has a part that corresponds to your hands.

00:49:21.520 | But then there's another part that comes out more laterally

00:49:24.040 | on the side of the brain that corresponds to your lips,

00:49:27.080 | your jaw, your larynx.

00:49:29.520 | And we have seen that when patients have surgeries

00:49:33.600 | or injuries to that part of the brain,

00:49:35.560 | it actually can really interrupt language.

00:49:37.440 | So it's not as simple as just moving the muscles

00:49:40.400 | of the vocal tract, but it's also important

00:49:42.280 | for formulating and expressing words.

00:49:45.400 | So that's Broca's area that I think the field now recognizes,

00:49:50.400 | not just because of our work,

00:49:53.240 | but many other people that have studied this

00:49:55.220 | in stroke and beyond,

00:49:56.880 | is that the idea that that is the basis of speaking

00:50:00.680 | in Broca's area is fundamentally wrong right now.

00:50:05.680 | And we have to figure out how to correct the textbooks

00:50:08.080 | that we kind of understand that

00:50:09.760 | so that we can continue to make progress.

00:50:13.240 | Now, in terms of the other major area

00:50:15.720 | that we call Wernicke's area, the posterior temporal lobe,

00:50:18.560 | that has held,

00:50:24.080 | held I think quite legitimately for some time.

00:50:27.720 | So that is an area that you have to be super careful

00:50:32.240 | when you do surgery there.

00:50:34.180 | That's an area where if you have a mistake there

00:50:38.240 | and you cause a stroke or you remove too much

00:50:41.160 | of the tumor there, you go too far beyond it,

00:50:44.380 | then the person can be really, really hurt.

00:50:48.640 | Like they'll have a condition that we call aphasia

00:50:51.560 | where they may not be able to understand words,

00:50:53.960 | they may not be able to remember the word

00:50:56.920 | that they're trying to say.

00:50:59.120 | They know what they're trying to say,

00:51:00.640 | but they can't remember the precise word

00:51:03.080 | that goes with the object that they're trying to think of.

00:51:05.560 | They may even produce words that I described before,

00:51:09.440 | like word salad or very jargony.

00:51:12.020 | So they might say something like tamirani.

00:51:15.600 | That's not a real word, but it sounds like it could be.

00:51:21.400 | And that's just because that part of the brain

00:51:23.200 | has some role, not just in understanding what we hear,

00:51:27.440 | but also actually has a really important role

00:51:30.100 | in sending the commands to different parts of the brain

00:51:32.340 | to control what we say.

00:51:34.720 | - Not long ago, you and me

00:51:38.560 | and my good friend, Rick Rubin,

00:51:40.960 | were having a conversation about medicine and science.

00:51:44.960 | And Rick asked the question,

00:51:46.740 | what percentage of what you learned in graduate

00:51:52.120 | and/or medical school do you think is correct?

00:51:56.960 | And you had a very interesting answer.

00:51:58.840 | Would you share it with us?

00:52:01.240 | - I don't know, I don't remember the exact,

00:52:04.440 | but I would say that with regard to the brain in particular,

00:52:09.440 | I would say about 50% gets it right and accurate

00:52:16.400 | and is helpful.

00:52:17.960 | But another 50% is just the approximation

00:52:21.160 | and oversimplification of what's going on.

00:52:24.280 | The example that we talked about, language,

00:52:25.760 | just an example of that.

00:52:26.960 | It's just there are things that make it easier to learn

00:52:31.680 | and easier to teach and easier to even think about.

00:52:34.540 | And that's probably why we continue teaching

00:52:37.760 | in the way that we do.

00:52:39.260 | But I think as time goes on,

00:52:41.360 | the complexity of reality of how the brain works is,

00:52:45.480 | well, first of all, we're still trying to figure it out.

00:52:49.540 | And second of all, it is complex

00:52:51.400 | and it's still incomplete story.

00:52:54.200 | - It's early days.

00:52:55.040 | And we get into some of the technical advances

00:52:57.560 | that are allowing some correction of the errors

00:53:01.120 | that the field has made.

00:53:02.800 | And look, no disrespect to the brain explorers

00:53:05.880 | that came before us and the ones that come after us

00:53:09.160 | will correct us, right?

00:53:10.560 | That's the way the game is played.

00:53:12.560 | But what I'm hearing is that there are certain truths

00:53:15.140 | that people accept and then there's about half

00:53:17.680 | of the information that is still open for debate

00:53:19.920 | and maybe even for complete revision.

00:53:21.820 | One thing that I learned about language

00:53:25.380 | and the neural circuits underlying language

00:53:27.660 | is that it's heavily lateralized,

00:53:29.580 | that these structures Broca's and Wernicke's

00:53:31.700 | and other structures in the brain responsible for speech

00:53:34.160 | and comprehension of speech,

00:53:35.880 | sit mainly on one side of the brain,

00:53:38.640 | but they do not have a mirror representation

00:53:42.020 | or another equivalent area

00:53:43.980 | on the opposite side of the brain.

00:53:45.080 | And for those that haven't poked around in a lot of brains,

00:53:50.080 | certainly you, Eddie, have done far more of that

00:53:52.760 | than I have, but I've done my fair share

00:53:55.000 | in non-human species and a little bit in humans,

00:53:58.080 | almost every structure, almost every structure

00:54:00.560 | has a matching structure on the other side of the brain.

00:54:03.920 | So when we say the hippocampus,

00:54:05.280 | we really mean two hippocampi,

00:54:07.160 | one on each side of the brain.

00:54:08.740 | But language I was taught is heavily lateralized.

00:54:11.600 | That is that there's only one.

00:54:13.440 | So that raises two questions.

00:54:14.580 | One, is that true?

00:54:16.120 | And if it is true,

00:54:18.080 | then what is the equivalent real estate

00:54:20.500 | on the opposite side of the brain doing

00:54:22.580 | if it's not doing the same function

00:54:23.980 | that the one on the, say, the left side is performing?

00:54:26.400 | - Well, that's one of those things that is,

00:54:28.280 | again, like mostly true, not 100%.

00:54:31.300 | And what I mean by that is that it's complicated.

00:54:35.140 | So for people who are right-handed,

00:54:38.480 | 99% of the time, the language part of the brain

00:54:41.120 | is on the left side.

00:54:43.720 | - And what is the equivalent brain area

00:54:45.840 | on the right side doing if it's not doing language?

00:54:48.900 | - Well, you know, the thing that's incredible

00:54:50.380 | is if you look at the right side

00:54:52.540 | and you look at it very carefully,

00:54:53.620 | either under an MRI or you actually look at the brain

00:54:57.240 | under slides at a microscope,

00:54:59.260 | it looks very, very similar.

00:55:00.860 | It's not identical, but it looks very, very similar.

00:55:03.880 | All the gyri, which are the bumps on the brain

00:55:06.880 | that have the different contours

00:55:08.880 | and the valleys that we call sulci,

00:55:10.720 | those all look basically the same.

00:55:12.500 | Like there is a mirror anatomy on the left and right side.

00:55:17.220 | And so it's not been so clear what's so special actually

00:55:20.900 | about the left side to house language.

00:55:25.140 | But what we do know,

00:55:26.620 | and this is what we use all the time in assessing

00:55:29.460 | and figuring out this before surgery,

00:55:32.420 | is if you're right-handed, 99% of the time,

00:55:35.380 | the language is going to be on the left side of the brain.

00:55:38.660 | - Is handedness genetic in any way?

00:55:40.660 | I mean, when I grew up by a pen or pencil

00:55:43.100 | or crayon was placed into my hand, presumably,

00:55:45.020 | or I started using it, my father was left-handed

00:55:48.180 | and then where he grew up in South America,

00:55:50.220 | they forced him to force himself to become right-handed

00:55:54.700 | if she used to restrict the movement of his left hand

00:55:57.180 | so he was forced to write.

00:55:58.440 | So, and then you have hook lefties and hook righties.

00:56:02.620 | I know this is a deep dive

00:56:04.720 | and we probably don't want to go

00:56:05.640 | into every derivation of this,

00:56:07.480 | but so for somebody who's left-handed,

00:56:10.260 | naturally just starts writing with the left hand,

00:56:12.540 | there's some genetic predisposition to being left-handed?

00:56:16.380 | - Absolutely, no question about it.

00:56:18.580 | Handedness is not entirely, but strongly genetic.

00:56:23.580 | So there is something about that ties all of this.

00:56:27.420 | And what does handedness, for example,

00:56:28.860 | have to do with where the part of your brain

00:56:31.620 | that controls language?

00:56:32.720 | Well, it turns out that the parts that control the hand

00:56:35.740 | are very close to the areas that really are responsible

00:56:38.480 | for the vocal tract, again, part of the motor cortex

00:56:41.820 | and part of this brain area called the precentral gyrus.

00:56:44.960 | And there are some theories that because of their proximity,

00:56:48.520 | that these parts of the brain might develop together early

00:56:52.720 | in utero and they might have a head start

00:56:54.580 | compared to the right side.

00:56:56.060 | And because they have a head start,

00:56:57.820 | that things solidify there.

00:56:59.880 | This is one theory of why this happens.

00:57:02.920 | And people who are left-handed,

00:57:05.180 | it still turns out that the vast majority of people

00:57:07.920 | have language on the left side, but it's not 99%.

00:57:10.940 | It's more like 70%.

00:57:13.340 | So if you're left-handed, it's still more likely

00:57:16.540 | that the language part of your brain

00:57:18.700 | is gonna be on the left side,

00:57:20.860 | but there's gonna be a greater proportion, maybe 20, 30%,

00:57:24.540 | where it's either in both hemispheres or on the right side.

00:57:29.480 | And just to make this a little bit more interesting

00:57:34.060 | is that when people have strokes on the left side,

00:57:38.940 | and if they're lucky enough to recover from those strokes,

00:57:42.360 | sometimes that involves reorganization,

00:57:44.460 | this term that we call plasticity earlier,

00:57:46.980 | where the areas around where the stroke

00:57:49.800 | take on that new function in a way

00:57:51.560 | that they didn't have before.

00:57:53.000 | That can certainly happen in the left hemisphere,

00:57:56.480 | but there are also instances where the right hemisphere

00:58:00.920 | can also start to take on the function of language,

00:58:03.720 | when it was once on the left

00:58:06.580 | and then transfers to the right.

00:58:08.760 | So the thing that I think about a lot

00:58:12.180 | is that the machinery probably exists on both sides,

00:58:17.180 | but we don't use them together all the time.

00:58:22.460 | In fact, we may strongly bias one side or the other,

00:58:26.300 | just like we use our two hands in very, very different ways.

00:58:29.860 | It's a little bit the same with the brain.

00:58:31.740 | Well, it's because of what we do with the brain

00:58:33.600 | that actually is why we use the hands in different ways.

00:58:36.480 | And the same thing goes for language,

00:58:38.180 | which is, again, the substrates, the organ,

00:58:41.220 | the language organ, the part of the brain.

00:58:43.320 | The process, it probably has very similar machinery

00:58:47.720 | on the left side as the right,

00:58:50.520 | and the right may have the capability to do it,

00:58:53.200 | but in real everyday use,

00:58:55.680 | the brain specializes one of the sides

00:58:59.140 | in order for us to use it functionally.

00:59:03.000 | That's a theory.

00:59:04.740 | - You're bilingual, correct?

00:59:06.080 | - Yeah.

00:59:06.920 | - You speak English and Chinese.

00:59:08.200 | - Yeah.

00:59:09.700 | - For people that are bilingual

00:59:11.100 | and that learn two or more,

00:59:13.240 | well, bilingual is two obviously,

00:59:14.360 | but learn both languages,

00:59:16.500 | or let's say more languages from an early time in life,

00:59:20.880 | do they use the same brain area to generate that language,

00:59:23.800 | or perhaps they use the left side to speak English

00:59:27.720 | and the right side to speak Chinese?

00:59:29.320 | Do we know anything about bilingualism in the brain?

00:59:32.920 | - I think we know a lot about bilingualism in the brain.

00:59:35.680 | The answers are still out there,

00:59:37.520 | the final answer is on it,

00:59:38.840 | and part of the answer is yes, absolutely,

00:59:42.160 | we use some parts of the brain.

00:59:43.960 | Very similarly, we actually have a study in the lab right now

00:59:48.580 | where we're looking at this,

00:59:49.600 | where people who speak one language or another,

00:59:53.300 | or bilingual, and we're looking at

00:59:56.160 | how the brain activity patterns occur

00:59:59.320 | when they're hearing one language versus the other,

01:00:01.080 | and what's striking to see actually

01:00:03.600 | is how overlapping they really can be,

01:00:06.580 | even though the person may have no idea

01:00:09.720 | of the language that they're hearing,

01:00:12.240 | the English part of the brain is still processing that

01:00:15.200 | and maybe trying to interpret it

01:00:16.940 | through an English lens, for example.

01:00:20.040 | So the short answer is that with bilingualism,

01:00:23.380 | there are shared circuitry,

01:00:25.760 | there's this shared machinery in the brain

01:00:28.720 | that allows us to process both,

01:00:31.560 | but it's not identical.

01:00:33.840 | It's the same part of the brain,

01:00:35.920 | but what it's doing with the signals

01:00:38.240 | can be very, very different,

01:00:40.360 | and what I mean by that precisely

01:00:41.960 | is not the instantaneous detecting of one sound to the next,

01:00:46.460 | but the memory of the sequences of those particular sounds

01:00:51.460 | that give rise to things like words and meaning.

01:00:54.380 | That can be highly variable from one individual to the next,

01:00:58.420 | and those neurons are very, very sensitive

01:01:01.340 | to the sequences of the sounds,

01:01:03.380 | even though the sounds themselves

01:01:05.540 | might have some overlap between languages.

01:01:07.880 | - Fascinating.

01:01:09.120 | Okay, so we've talked about brain areas

01:01:11.140 | and a little bit about lateralization.

01:01:13.780 | I want to get back to the hands

01:01:15.060 | and some things related to emotion in a little bit,

01:01:17.000 | but maybe now we could go into those brain areas

01:01:20.760 | and start to ask the question,

01:01:24.060 | what exactly is represented or mapped there?

01:01:27.340 | And for people who perhaps aren't familiar

01:01:29.880 | with brain mapping and representation in receptive fields,

01:01:33.080 | perhaps the simplest analogy might be the visual system

01:01:38.020 | where I look at your face, I know you, I recognize you,

01:01:40.520 | and certainly there are brain areas

01:01:41.760 | that are responsible for face recognition,

01:01:45.360 | but the fact that I know that that's your face,

01:01:48.940 | and for those listening, I'm looking into Eddie's face,

01:01:51.560 | the fact that I know that that's your face at all

01:01:53.200 | is because we are well aware

01:01:55.600 | that there are cells that represent edges

01:01:57.740 | and that represent dark and light,

01:01:59.520 | and those all combine in what we call a hierarchical structure

01:02:02.340 | they sort of build up from basic elements,

01:02:05.220 | as simple as little dots,

01:02:06.460 | but then lines and things that move, et cetera,

01:02:08.360 | to give a coherent representation of the face.

01:02:11.260 | When I think about language,

01:02:13.640 | I think about words and just talking.

01:02:15.460 | If I sit down to do a long podcast

01:02:17.160 | or I think about asking you a question,

01:02:18.460 | I don't even think about the words I want to say very much.

01:02:21.880 | I mean, I have to think about them a little bit,

01:02:23.240 | one would hope,

01:02:24.360 | but I don't think about individual syllables

01:02:26.560 | unless I'm trying to accent something,

01:02:30.120 | or it's a word that I have a particular difficulty saying,

01:02:34.240 | where I want to change the cadence, et cetera.

01:02:36.840 | So what's represented in the neurons,

01:02:39.160 | the nerve cells in these areas,

01:02:40.360 | are they representing vowels, consonants,

01:02:43.000 | and how do things like inflection,

01:02:44.480 | like I occasionally will poke fun at up-speak,

01:02:47.520 | but there's a, I think, a healthy, normal version

01:02:52.520 | of up-speak where somebody is asking a question,

01:02:54.200 | like for instance, what is that?

01:02:56.480 | That's an appropriate use of up-speak,

01:02:58.640 | as opposed to saying something that is not a question

01:03:01.840 | and putting a lilt at the end of the sentence,

01:03:03.720 | then we call that up-speak,

01:03:05.140 | which it doesn't fit with what the person is saying.

01:03:08.880 | So what in the world is contained in these brain areas?

01:03:11.900 | What is represented to me is perhaps

01:03:15.840 | one of the most interesting questions,

01:03:17.080 | and I know this lands square in your wheelhouse.

01:03:19.400 | - Sure, let's get into this, Andrew,

01:03:22.400 | because this is one of the most exciting stuff

01:03:25.280 | that's happening right now,

01:03:26.240 | is understanding how the brain processes

01:03:28.680 | these exact questions.

01:03:30.720 | And you asked me earlier,

01:03:32.600 | what is the difference between speech and language?

01:03:34.920 | Speech corresponds to the communication signal.

01:03:39.760 | It corresponds to me moving my mouth and my vocal tract

01:03:43.080 | to generate words,

01:03:44.800 | and your hearing these is an auditory signal.

01:03:48.400 | Language is something much broader.

01:03:52.240 | So it refers to what you're extracting

01:03:55.320 | from the words that I'm saying.

01:03:56.680 | We call that pragmatics,

01:03:57.840 | and sort of are you getting the gist of what I'm saying?

01:04:00.460 | There's another aspect of it that we call semantics.

01:04:02.800 | Do you understand the meaning

01:04:04.800 | of these words and the sentences?

01:04:08.240 | There's another part that we call syntax,

01:04:09.860 | which refers to how the words are assembled

01:04:12.440 | in a grammatical form.

01:04:14.040 | So those are all really critical parts of language,

01:04:17.100 | and speech is just one form of language.

01:04:20.280 | There's many other forms like sign language, reading.

01:04:24.080 | Those are all important modalities for reading.

01:04:27.060 | Our research really focuses on this area

01:04:31.580 | that we're calling speech.

01:04:32.580 | Again, the production of this audio signal,

01:04:36.960 | which you can't see, but your microphones are picking up.

01:04:41.160 | There are these vibrations in the air

01:04:43.840 | that are created by my vocal tract,

01:04:45.720 | that are picked up by the microphone

01:04:48.000 | in the case of this recording,

01:04:49.760 | but also picked up by the sensors in your ear.

01:04:53.020 | The very tiny vibrations in your ear are picking that up

01:04:56.880 | and translating that into electrical activity.

01:05:00.880 | And what the ear does at the periphery

01:05:04.160 | is translates all sounds into different frequencies.

01:05:09.160 | So its main thing to do is to take a speech signal

01:05:14.360 | or any other kind of sound and decompose it,

01:05:17.600 | meaning separate that sound into different kind of signals.

01:05:22.600 | In the case of hearing, what it's doing

01:05:25.440 | is separating it out into low, middle, high frequencies

01:05:30.440 | at a very, very high resolution.

01:05:33.580 | It's doing it very quickly,

01:05:35.040 | and it's doing it in a really fine way

01:05:36.320 | to separate all of those different sounds.

01:05:38.720 | So if you look at the periphery near the nerve

01:05:41.820 | that goes to your ear, those nerve fibers,

01:05:44.640 | some of them are tuned to low frequencies.

01:05:47.440 | Some of them are tuned to high frequency.

01:05:49.400 | Some of them are tuned to the middle frequencies.

01:05:51.960 | And that is what your ear is doing.

01:05:53.700 | It's taking these words and splitting them up

01:05:55.720 | into different frequencies.

01:05:58.000 | - And for those of you out there that aren't familiar

01:05:59.780 | with thinking about things in the so-called frequency space,

01:06:03.360 | bass tones would be lower frequencies

01:06:05.300 | and high pitch tones would be higher frequencies

01:06:07.560 | just to make sure everyone's on the same page.

01:06:09.780 | So the sound of my voice, the sound of your voice

01:06:12.440 | or any sound in the environment

01:06:13.440 | is being broken down into these frequencies.

01:06:15.600 | Are they being broken down into very narrow channels

01:06:19.320 | of frequency or they, I want to avoid nomenclature here,

01:06:23.480 | or are they being binned as fairly broad frequencies?

01:06:27.220 | 'Cause we know low, medium, and high,

01:06:29.280 | but for instance, I can detect whether or not

01:06:32.200 | something's approaching me or moving away from me

01:06:35.040 | depending on whether or not it sweeps louder

01:06:38.040 | or towards or away.

01:06:42.080 | It's subtle, and of course it's combined with what I see

01:06:44.760 | in my own movement, but how finely sliced

01:06:49.760 | is our perception of the auditory world?

01:06:53.140 | - Oh, extraordinarily precise.

01:06:55.760 | I mean, we take these millisecond cues,

01:07:00.040 | the millisecond differences between the sound

01:07:02.640 | coming to one ear, let's say your right ear

01:07:04.520 | versus your left, to understand what direction

01:07:08.880 | that sound came from.

01:07:10.080 | Those are only millisecond differences.

01:07:12.520 | And that's how precise this works.

01:07:15.040 | But on the other hand, it does a lot of computation on this.

01:07:18.880 | It does a lot of analysis as you go up.

01:07:21.520 | And a lot of our work is focused on the part of the brain

01:07:25.760 | that we call the cortex.

01:07:26.860 | The cortex is the outermost part of the brain

01:07:30.320 | where we believe that sounds are actually converted

01:07:34.520 | into words and language.

01:07:36.580 | So there's this transformation where at the ear,

01:07:40.360 | words are decomposed and turned

01:07:43.400 | into these elemental frequency channels.

01:07:46.760 | And then as it goes up through the auditory system,

01:07:49.580 | hits the cortex, there are some things that happen obviously

01:07:53.720 | before it gets to the cortex.

01:07:55.480 | But when it gets to cortex,

01:07:56.920 | there's something special going on,

01:07:59.240 | which is that that part of the brain

01:08:00.860 | is looking for specific sounds.

01:08:04.260 | And specifically, what I mean by that is the sounds

01:08:07.800 | of human language.

01:08:09.280 | So the ones that are the different consonants and vowels

01:08:12.520 | in a different language.

01:08:14.000 | One of the ways that we have studied this

01:08:16.600 | is looking in patients who have epilepsy.

01:08:21.000 | And in a lot of these cases

01:08:22.400 | where the MRI looks completely normal,

01:08:25.460 | we have to put electrodes surgically on a part of the brain.

01:08:29.660 | The temporal lobe is a very, very common place.

01:08:32.480 | So we've done a lot of our work looking at how

01:08:34.760 | the temporal lobe processes speech sounds.

01:08:37.800 | Because we're looking for where the seizures start,

01:08:42.240 | but then we're also doing brain mapping

01:08:44.400 | for language and speech so we can protect those areas.

01:08:47.040 | We want to identify the areas that we want to remove

01:08:49.040 | to cure someone's seizures,

01:08:50.460 | but we also want to figure out the areas

01:08:52.120 | that are important for speech and language

01:08:53.860 | to protect those so that we can do a surgery

01:08:56.060 | that's effective and safe.

01:08:58.360 | And so in our research,

01:09:01.380 | and why it's become a really important addition

01:09:04.560 | to our knowledge is that we have electrodes directly

01:09:08.440 | recording from the human brain surface.

01:09:12.080 | A lot of technology we work with right now

01:09:14.400 | is recording on the order of millimeters.

01:09:17.680 | And they can record millisecond time resolution

01:09:22.120 | of neural activity.

01:09:24.000 | And what we see is extraordinary patterns of activity

01:09:29.000 | when people hear words and sentences.

01:09:32.800 | If you look at that part of the brain that we call

01:09:34.840 | Wernicke's area, in this part of the temporal lobe,

01:09:37.660 | this whole area lights up when you hear words or speech.

01:09:42.660 | And it's not in a way that is like a general light bulb

01:09:47.040 | warming up and it's generally lit up,

01:09:50.740 | but what you actually see is something much,

01:09:52.980 | much more complicated, which is a pattern of activity.

01:09:56.560 | And what we've done in the last 10 years

01:09:58.920 | is to try to understand what does that pattern come from?

01:10:02.960 | And if we were to look at each individual site

01:10:05.360 | from that part of the brain, what would we see?

01:10:08.840 | What parts of words are being coded by electrical activity

01:10:14.480 | in those parts of the brain?

01:10:15.320 | Remember the cortex is using electrical activity

01:10:18.400 | to transmit information and do analysis.

01:10:22.420 | And what we're doing is we're eavesdropping

01:10:24.480 | on this part of the brain as it's processing speech

01:10:27.960 | to try to understand what each individual site is doing.

01:10:31.000 | - And what are those sites doing?

01:10:32.840 | Or could you give us some examples

01:10:34.220 | of what those sites are doing?

01:10:35.160 | So for instance, are they sites that are specific for,

01:10:39.960 | or we could say even listening for consonants or for vowels

01:10:43.560 | or for inflection or for emotionality?

01:10:47.660 | What's in there?

01:10:49.800 | - Okay, well-

01:10:50.640 | - What makes these cells fire?

01:10:53.680 | - Yeah, what gets them excited?

01:10:55.440 | - Yeah.

01:10:56.280 | - Where I'm going is hearing speech in particular,

01:10:59.760 | there are some of these really focal sites,

01:11:02.760 | again, just on the order of millimeter

01:11:04.580 | or at some level, single neurons

01:11:07.620 | that are tuned to consonants, some are tuned to vowels,

01:11:12.240 | some are tuned to particular features of consonants.

01:11:16.240 | What I mean by that are different categories of consonants.

01:11:21.240 | There's a class of consonants that we call plosive consonants.

01:11:25.040 | There's a little bit of linguistic jargon,

01:11:26.960 | but I'm going to make a point here with that

01:11:28.320 | is that certain classes of sounds, when you make them,

01:11:31.300 | it requires you to actually close your mouth temporarily.

01:11:34.760 | - Now I'm going to be thinking about this.

01:11:37.520 | So plosive, like plosives,

01:11:39.160 | like saying the word plosive does requires that.

01:11:41.680 | - Exactly, so what's cool about that

01:11:44.000 | is that we actually have no idea what's going on

01:11:47.480 | in our mouth when we speak.

01:11:49.240 | We really have no idea.

01:11:50.520 | - Some people definitely have no idea.

01:11:52.160 | [laughing]

01:11:53.400 | - Well, not just like in terms of what you're saying

01:11:55.360 | sometimes, but actually like how you're actually moving

01:11:58.780 | the different parts of vocal track.

01:12:01.040 | And I have a feeling if we actually required understanding,

01:12:05.880 | we would never be able to speak because it's so complex.

01:12:08.500 | It's such a complex feat.

01:12:09.700 | Some people would say it's the most complex motor thing

01:12:12.780 | that we do as a species is just speaking,

01:12:15.380 | not the extreme feats of acrobatics or athleticism,

01:12:19.880 | but speaking.

01:12:21.240 | - Well, and especially when one observes opera

01:12:24.800 | or people who, freestyle rappers,

01:12:28.520 | and of course it's not just the lips, it's the tongue.

01:12:31.800 | And you've mentioned two other structures,

01:12:35.160 | pharynx and larynx are the main ones that,

01:12:37.960 | can you tell us, just educate us at a superficial level,

01:12:41.460 | what does pharynx and larynx do differentially?

01:12:44.600 | 'Cause I think most people aren't going to be familiar.

01:12:46.440 | - Okay, sure.

01:12:47.600 | So I'll talk primarily about the larynx here for a second,

01:12:52.600 | which is that if you think about when we're speaking,

01:12:56.360 | really what we're doing is we're shaping the breath.

01:13:00.760 | So even before you get to the larynx,

01:13:02.380 | you got to start with the expiration.

01:13:04.820 | So we fill up our lungs and then we push the air out.

01:13:11.080 | That's a normal part of breathing.

01:13:14.040 | And what is really amazing about speech and language

01:13:17.000 | is that we evolved to take advantage

01:13:19.560 | of that normal physiologic thing at a larynx.

01:13:23.440 | And what the larynx does is that when you're exhaling,

01:13:26.440 | it brings the vocal folds together.

01:13:28.800 | Some people call them vocal cords.

01:13:30.840 | They're not really cords, they're really vocal folds.

01:13:33.400 | They're two pieces of tissue that come together

01:13:35.840 | and a muscle brings them together.

01:13:37.640 | And then what happens is when the air comes

01:13:39.520 | through the vocal folds, when they're together, they vibrate

01:13:44.040 | at really high frequencies, like 100 to 200 hertz.

01:13:47.360 | Yours is probably about 100 hertz.

01:13:49.940 | - Whereas yours is 200.

01:13:51.160 | [laughs]

01:13:52.160 | - No, no.

01:13:54.040 | Most male voices are around 100, okay?

01:13:56.440 | And then the average female voice is around 200 hertz.

01:13:58.680 | - And as you know, I've always had the same voice.

01:14:00.680 | This was a point of shame when I was a kid.

01:14:03.200 | Folks, my voice never changed.

01:14:04.600 | I always had the same voice.

01:14:06.100 | This is a discussion for another time.

01:14:07.680 | - Yeah, well, it's a great voice, a great baritone voice.

01:14:11.660 | But I know in your voice, it's a low frequency voice.

01:14:15.860 | And the reason why men and women generally

01:14:18.580 | have different voice qualities is it has to do

01:14:22.740 | with the size of the larynx and the shape of it.

01:14:26.100 | Okay, so in general, men have a larger voice box or larynx.

01:14:30.660 | And the vibrating frequency, the resonance frequency

01:14:33.340 | of the vocal folds when the air comes through them

01:14:36.740 | is about 100 hertz for men and about 200 for women.

01:14:40.220 | So what happens is, okay, so you're taking,

01:14:44.840 | you take a breath in, and then as the air is coming out,

01:14:49.740 | the vocal folds come together and the air goes through.

01:14:52.420 | That creates the sound of the voice that we call voicing.

01:14:56.540 | And that's the energy of your voice.

01:14:59.820 | It's not just your voice characteristic.

01:15:02.620 | It's the energy of your voice.

01:15:05.020 | It's coming from the larynx there.

01:15:07.880 | It's a noise.

01:15:09.860 | And then it's the source of the voice.

01:15:12.980 | And then what happens is that energy,

01:15:15.540 | that sound goes up through the parts of the vocal tract,

01:15:19.620 | like the pharynx, into the oral cavity,

01:15:23.780 | which is your mouth and your tongue and your lips.

01:15:26.820 | And what those things are doing is that they're shaping

01:15:30.180 | the air in particular ways that create consonants and vowels.

01:15:36.900 | So that's what I mean by shaping the breath.

01:15:40.100 | It just starts with this exhalation.

01:15:43.060 | You generate the voice in the larynx,

01:15:45.580 | and then everything above the larynx is moving around,

01:15:49.020 | just like the way my mouth is doing right now,

01:15:51.280 | to shape that air into particular patterns

01:15:54.820 | that you can hear as words.

01:15:57.980 | - Fascinating.

01:15:59.980 | And immediately makes me wonder about more primitive

01:16:06.700 | or non-learned vocalizations like crying or laughter.

01:16:11.060 | Babies will cry, babies will show laughter.

01:16:14.700 | Are those sorts of vocalizations produced

01:16:19.620 | by the language areas like Wernicke's,

01:16:22.300 | or do they have their own unique neural structures?

01:16:25.500 | - Yeah, interesting question.

01:16:26.940 | So we call those vocalizations.

01:16:29.940 | A vocalization is basically where someone can create a sound

01:16:35.020 | like a cry or a moan, that kind of sound.

01:16:40.020 | And it also involves the exhalation of air.

01:16:44.820 | It also involves some phonation at the level of larynx

01:16:48.180 | where the vocal folds come together

01:16:50.180 | to create that audible sound.

01:16:51.960 | But it turns out that those are actually different areas.

01:16:56.300 | So people who have injuries in the speech and language areas

01:17:00.300 | oftentimes can still moan, they can still vocalize.

01:17:03.920 | And it is a different part of the brain.

01:17:05.820 | I would say an area that even non-human primates have

01:17:09.740 | that can be specialized for vocalization.

01:17:12.860 | It's a different form of communication than words,

01:17:16.200 | for example.

01:17:17.660 | - The intricacy of these circuits in the brain

01:17:20.020 | and their connections to the pharynx and larynx is just,

01:17:25.020 | it's almost overwhelming in terms of thinking

01:17:27.500 | about just how complicated it must be.

01:17:29.360 | And yet some general features and principles

01:17:31.880 | are starting to emerge from your work

01:17:33.460 | and from the work of others.

01:17:34.920 | If we think about that work and we think about,

01:17:39.280 | for instance, Wernicke's area,

01:17:40.760 | if I were to record from neurons in Wernicke's area

01:17:46.240 | at different locations,

01:17:48.240 | would I find that there's any kind of systematic layout?

01:17:52.640 | For instance, in terms of,

01:17:53.920 | we've talked about sound frequency,

01:17:55.600 | we know that low frequencies are represented

01:17:57.320 | at one end of a structure and high frequencies at the other.

01:17:59.520 | This is true, actually, at least from my earlier training

01:18:02.160 | within the year itself, within the cochlea,

01:18:04.700 | the early work of von Bekasy and from cadavers, right?

01:18:07.520 | They actually figured this out from dead people,

01:18:09.380 | which is incredible.

01:18:11.060 | A fascinating literature people should look up.

01:18:13.400 | And in the visual system, we know that, for instance,

01:18:17.840 | you know, visual position where things are

01:18:21.640 | is mapped systematically.

01:18:23.320 | In other words, neurons that sit next to each other

01:18:25.540 | in the brain represent portions of visual space

01:18:28.600 | that are next to each other in the real world.

01:18:31.020 | What is the organization of language

01:18:33.900 | in areas like Wernicke's and Broca's?

01:18:36.160 | For instance, I think of the vowels, A, E, I, O, U,

01:18:39.920 | as kind of a coherent unit,

01:18:43.000 | but do I find the A neurons are next to the E neurons

01:18:46.580 | are next to the, or the A, E, I, O, U?

01:18:50.660 | Is that vowel representation also laid out in order,

01:18:55.000 | or is it kind of salt and pepper, is it random?

01:18:57.940 | - That's been one of the most important questions

01:19:00.620 | we've been trying to answer for the past decade.

01:19:03.400 | So there is a part of the brain

01:19:06.000 | that we call the primary auditory cortex,

01:19:08.240 | and the primary auditory cortex is deep in the temporal lobe.

01:19:11.400 | And if you looked at that part of the brain,

01:19:15.200 | there is a map of different sound frequencies.

01:19:18.360 | So if you look at the front of that primary auditory cortex,

01:19:21.680 | you'll find low frequency sounds.

01:19:23.440 | And then as you march backwards in that cortex,

01:19:28.320 | it goes from low to medium to high frequencies.

01:19:30.980 | It's organized in this really nice and orderly way.

01:19:34.800 | And it turns out there's just not just one,

01:19:36.580 | there's like mirrors of that tone frequency map

01:19:40.900 | in the primary auditory cortex.

01:19:43.640 | The areas that are really important for speech

01:19:46.800 | are on the side of that.

01:19:48.540 | And we now think that speech

01:19:51.860 | can go straight to the speech cortex

01:19:53.900 | without having to go through the primary auditory cortex.

01:19:57.320 | That it has its own pathway to get to the part of the brain

01:20:02.320 | that processes speech.

01:20:04.660 | And when we've looked at that question about,

01:20:07.900 | is there a map?

01:20:09.320 | The short answer is yes.

01:20:11.400 | There is a map, and it is,

01:20:14.020 | but it is not structured universally across all people

01:20:18.000 | in a way that we can clearly see right now.

01:20:21.120 | It is like a salt and pepper map

01:20:23.680 | of the different features in speech.

01:20:25.920 | So before we talked about these sounds

01:20:27.640 | that are called plosives.

01:20:29.400 | You make a plosive when the mouth

01:20:32.080 | or something in the oral cavity closes temporarily.

01:20:35.040 | And when it opens,

01:20:37.440 | that creates that fast plosive sound.

01:20:41.160 | So when you say dad,

01:20:43.760 | or the ball,

01:20:48.680 | like the B in ball, that kind of thing,

01:20:51.340 | you will notice that your lips actually close.

01:20:54.040 | And then it's the release of that

01:20:55.300 | that creates that particular sounds.

01:20:57.400 | Okay, so those are the sounds that we call plosives.

01:20:59.720 | Those are like badaga, pataka.

01:21:02.640 | Those are a certain class of consonants

01:21:05.960 | that we call plosive sounds.

01:21:08.160 | There's another class of sounds

01:21:09.840 | that we call fricatives in linguistics.

01:21:12.960 | Fricatives are created by turbulence in the air stream

01:21:17.960 | as it comes out through the mouth.

01:21:20.560 | And the way that we make that turbulence

01:21:24.560 | is getting the mouth and the lips to close

01:21:27.480 | almost until they're completely shut,

01:21:29.340 | or putting the tongue near the teeth

01:21:33.740 | to almost get it completely shut,

01:21:36.260 | but just have a narrow aperture.

01:21:38.300 | That creates a turbulence in the airflow

01:21:40.540 | that we perceive as a high frequency sound.

01:21:42.160 | So those are the sounds like sha and thuh,

01:21:46.160 | those kind of things.

01:21:47.800 | If you look at the frequencies, they're higher frequencies,

01:21:49.920 | and those are created by specific movements

01:21:52.920 | that you constrict the airflow to create turbulence.

01:21:55.180 | And we hear it as sha, sa, thuh.

01:22:00.180 | - So if I say that.

01:22:02.240 | - Exactly.

01:22:03.080 | - And as opposed to a plosive where I'd say explosive.

01:22:07.200 | - Right.

01:22:08.040 | - I mean, now, of course, I'm emphasizing here.

01:22:09.120 | Well, this explains something and solves a mystery,

01:22:11.860 | which is recently I've been fascinated by the work

01:22:14.040 | of a physician scientist back East, Dr. Shaina Swan,

01:22:19.040 | who's done a lot of work on things that are contained

01:22:22.160 | in pesticides and foods that are changing hormone levels.

01:22:25.320 | And she refers to phthalates, which is spelled pea.

01:22:28.860 | So it's both a plosive and a thuh.

01:22:31.100 | So it's combining the two.

01:22:32.080 | And it's one of the most difficult words

01:22:33.880 | in the English language to pronounce.

01:22:35.640 | Second only perhaps to the correct pronunciation

01:22:38.080 | of ophthalmology.

01:22:39.520 | So it's a combination of a plosive

01:22:43.920 | and one of these thuh sounds.

01:22:45.920 | And that's probably why it's difficult.

01:22:47.540 | - That's exactly right.

01:22:48.480 | In fact, we have a term for that.

01:22:50.720 | That's called a consonant cluster.

01:22:52.240 | So sometimes syllables will just have one consonant.

01:22:55.380 | But when we start stacking certain syllables in a sequence,

01:22:58.620 | and there's rules that actually govern which consonants

01:23:02.300 | can be in a particular sequence for a given language,

01:23:05.900 | that makes it more complicated.

01:23:07.800 | And certain languages have a lot more consonant clusters

01:23:11.440 | than others.

01:23:12.280 | - For instance.

01:23:13.120 | - So for instance, Russian, for example,

01:23:14.840 | has a lot of consonant clusters.

01:23:16.700 | English has a lot of them.

01:23:18.440 | There are other languages that have very, very few.

01:23:22.220 | For example, Hawaiian.

01:23:24.240 | Hawaiian has an inventory of about 12 to 14

01:23:26.720 | different phonemes, 14 different consonants and vowels.

01:23:30.600 | English, on contrast, has about 40

01:23:33.240 | different consonants and vowels.

01:23:35.220 | So languages have different inventories.

01:23:37.800 | They can overlap for sure.

01:23:39.840 | But different languages use different sound elements,

01:23:43.800 | combine and recombine those elements

01:23:45.560 | to give rise to different words and meanings.

01:23:48.240 | - Can we say that there's a most complicated language

01:23:51.480 | out there or among the most complicated,

01:23:53.080 | would it be Russian?

01:23:54.160 | - It's definitely high up there.

01:23:55.500 | English is up there too, actually.

01:23:57.480 | Yeah, German as well.

01:23:59.120 | - And in terms of learning multiple languages

01:24:00.880 | during development, my understanding is that

01:24:02.620 | if one wants to become bilingual or trilingual,

01:24:05.880 | best to learn those languages simultaneously

01:24:08.480 | during development, ideally before age 12,

01:24:11.320 | if one hopes to not have an accent in speaking them later.

01:24:14.520 | Is that correct or do you want to revise that?

01:24:16.400 | - Well, basically, the earlier and the earlier is better,

01:24:21.400 | the more intense it is and the more immersive it is.

01:24:25.780 | The longer that you can be exposed to that

01:24:29.860 | is really important.

01:24:30.700 | A lot of people can get exposed to it early

01:24:32.400 | and basically lose it.

01:24:33.480 | Even though it's "during that sensitive period,"

01:24:36.720 | unless it's maintained, it can be very easily lost.

01:24:40.160 | Then I think another aspect of it that's very interesting

01:24:42.840 | is some of the social requirements for it too.

01:24:47.280 | It's pretty clear that you can only go so far

01:24:51.000 | just listening to these sounds from a tape recording

01:24:54.640 | or something like that.

01:24:56.220 | There's something extra about real human interactions

01:25:00.180 | that activates the brain's sensitivity

01:25:02.320 | to different speech bounds that allows us

01:25:04.480 | to become specialized for them for a given language.

01:25:07.800 | - So returning to what's mapped,

01:25:10.200 | what the representations are in the brain,

01:25:13.000 | and I'm starting to get a picture now

01:25:15.160 | based on these plosives and these sounds.

01:25:17.740 | And what I find so interesting and logical about that

01:25:22.620 | is it maps to the motor structures

01:25:24.980 | and the actual pronunciation of the sounds,

01:25:26.840 | not necessarily to the meaning of the individual words.

01:25:31.240 | Now, of course, it's related to the meaning

01:25:32.620 | of the individual words,

01:25:34.240 | but it makes good sense to me

01:25:36.780 | why something as complex as language both to understand

01:25:41.460 | and to generate would map to something

01:25:43.820 | that is essentially motor in design,

01:25:46.320 | because as you point out, I have to generate these sounds

01:25:50.300 | and I have to hear them generated from others.

01:25:53.220 | However, there's reading and there's writing,

01:25:56.860 | and writing is certainly motor.

01:25:59.760 | Reading involves some motor commands of the eyes

01:26:02.180 | and et cetera.

01:26:04.380 | Where do reading and writing come into this picture?

01:26:07.140 | Are they in parallel with,

01:26:09.520 | as we would say in neuroscience,

01:26:10.940 | or are they embedded within the same structures?

01:26:13.900 | Are they part of the same series of computations?

01:26:18.500 | - Yeah, so to address the first part

01:26:23.500 | is that we've got this map of these different parts

01:26:27.500 | of consonants and vowels,

01:26:28.920 | and when we look at how they lay out

01:26:33.420 | in this part of the brain that we call word accusary,

01:26:36.180 | we've spent a lot of time really just dissecting

01:26:38.340 | this millimeter by millimeter.

01:26:40.900 | The term that you use is very apropos.

01:26:43.780 | It's salt and pepper.

01:26:45.220 | It's not random.

01:26:46.100 | There is this kind of selectivity

01:26:48.180 | to these individual speech sounds,

01:26:49.940 | and one point I want to make about it is this,

01:26:52.080 | is that in English, for example,

01:26:54.920 | there are about 40 different phonemes.

01:26:56.620 | Phonemes are just consonants or vowels

01:26:58.500 | or individual speech segments,

01:27:01.260 | but these articulatory features that you refer to,

01:27:05.660 | for example, the characteristic sounds

01:27:09.000 | that are generated by specific movements in the mouth,

01:27:12.120 | you can more or less reduce that

01:27:15.060 | to about 12 different features.

01:27:18.600 | Okay, these are specific movements of the tongue,

01:27:21.540 | the jaw, the lips, the larynx.

01:27:23.560 | There are about 12 of these movements,

01:27:27.420 | and just like you said, Andrew,

01:27:29.900 | by themselves, they have no meaning.

01:27:32.020 | They're just movements, but what's incredible about it

01:27:36.480 | is that you take these 12 movements

01:27:38.460 | and you put them in combinations,

01:27:40.980 | and you start putting them in sequence.

01:27:42.940 | We as humans use those 12 set of features

01:27:47.840 | to generate all words, and because we can generate

01:27:51.780 | nearly an infinite number of words with that code

01:27:55.900 | of just 12 features, we have something that generates,

01:27:59.620 | essentially, all possible meaning,

01:28:01.620 | because that's what we do as humans.

01:28:03.300 | We generate meanings.

01:28:04.820 | I'm trying to communicate one idea to another,

01:28:07.040 | which, to me, is extraordinary.

01:28:09.380 | A parallel would be, for example, DNA.

01:28:11.840 | There's four base pairs in DNA,

01:28:14.500 | but with those four base pairs in a specific sequence,

01:28:18.220 | can generate an entire code for life.

01:28:21.220 | In speech, it's the same way.

01:28:22.320 | It's like you've got these fundamental elements

01:28:25.020 | that by themselves have no meaning,

01:28:26.540 | but when you put them together,

01:28:27.780 | give rise to every possible meaning.

01:28:29.800 | So with regard to your second point about reading and writing

01:28:36.120 | it's a fascinating question.

01:28:38.180 | Speech and language is part of who we are as humans.

01:28:43.460 | That's part of how we evolved,

01:28:47.220 | and it's hardwired and molded by experience.

01:28:54.460 | Reading and writing are a human invention.

01:28:57.460 | It's something that was added on

01:29:01.660 | to the architecture of the brain.

01:29:03.700 | And because reading and writing are fairly recent

01:29:07.140 | in human evolution, it's essentially too quick

01:29:10.780 | for anything to, like, have a dramatic change

01:29:13.140 | in, let's say, a new brain area

01:29:15.580 | or some kind of specialization.

01:29:16.980 | Instead, what happens is that whenever any kind of behavior

01:29:21.240 | becomes ultra specialized in any of us or any organism,

01:29:26.240 | we can sort of take some areas

01:29:29.520 | that are normally involved with vision, for example,

01:29:32.820 | and specialize it for the purpose of reading.

01:29:35.800 | So all of us have a part of our brain

01:29:38.520 | in the back of the temporal lobe

01:29:39.680 | that interface with the cipital visual cortex

01:29:43.640 | that we call a visual word form area.

01:29:45.980 | There's actually a part of the brain

01:29:48.480 | that is very sensitive to seeing words,

01:29:52.700 | like either typed or handwritten.

01:29:55.180 | There's a part of the brain

01:29:56.420 | that also is sensitive to seeing things like faces.

01:30:00.740 | So these are things that are all conditioned

01:30:02.860 | on what's important to survive.

01:30:06.620 | So reading and writing are an invention,

01:30:10.500 | and there are things that have mapped

01:30:12.980 | to functions that the brain already has.

01:30:17.020 | And one of the really important things

01:30:19.780 | about reading and writing

01:30:21.000 | is that when we learn to read and write,

01:30:25.500 | especially with the reading part,

01:30:27.280 | it maps to the part of the brain

01:30:29.140 | that we've been talking about,

01:30:30.380 | which is the part that's processing speech sounds.

01:30:33.100 | So some of us kind of think about it.

01:30:36.040 | These are two different things.

01:30:37.020 | One is hearing sounds through your ears.

01:30:40.160 | The other is reading,

01:30:41.140 | where you're actually seeing things through your eyes

01:30:44.260 | and then getting into the language system.

01:30:46.120 | Well, it turns out that the auditory speech cortex

01:30:50.760 | is the primal and primitive fundamental area

01:30:54.680 | that's really important for speech.

01:30:56.000 | And what happens with the reading

01:30:57.920 | is once it gets through that visual cortex,

01:31:01.640 | it's going to try to map those reading signals

01:31:05.180 | to the part of the brain

01:31:06.100 | that's trying to make sense of sounds,

01:31:08.780 | the sounds of words, what we call phonology.

01:31:12.540 | Now, why is this important?

01:31:14.220 | It has a lot of relevance to how we learn to write.

01:31:18.700 | And in some kids with dyslexia,

01:31:23.720 | dyslexia is a neurological condition

01:31:27.000 | where a child, in some cases an adult,

01:31:30.820 | has trouble reading, for example.

01:31:34.620 | And in many of those cases,

01:31:37.260 | it's because that mapping between how we see the words

01:31:41.900 | to the way that the brain processes the sounds

01:31:45.820 | is something different.

01:31:47.140 | It's a little bit different

01:31:48.340 | than people who can read really well.

01:31:51.780 | So when you're reading,

01:31:53.240 | a lot of times you're actually activating

01:31:55.060 | the part of the brain that is processing

01:31:57.380 | the words that you hear.

01:31:58.860 | - What is the current treatment for dyslexia?

01:32:04.220 | I've heard that it's a deficit

01:32:07.900 | in some of the motion processing systems

01:32:10.220 | of the visual system.

01:32:12.020 | People are, their eyes are jumping

01:32:13.360 | as opposed to more linear reading across,

01:32:15.540 | or I suppose if it were Chinese, it would be...

01:32:18.060 | I'm going to presume people are always reading English,

01:32:20.160 | or I suppose if it's Hebrew,

01:32:21.260 | they're going from the opposite side of the page.

01:32:24.420 | What can be done for dyslexia?

01:32:27.720 | And do any of the modern treatments for dyslexia

01:32:31.060 | involve changing things from the speech side

01:32:34.760 | as opposed to just the quote-unquote reading side,

01:32:36.960 | given that speech and reading are interconnected?

01:32:39.300 | - Yeah, absolutely.

01:32:40.300 | So again, I think in the beginning,

01:32:42.380 | people might have thought this was purely

01:32:44.140 | a visual abstraction or something

01:32:47.180 | really just about the visual system,

01:32:48.700 | but there's been more recognition

01:32:50.940 | that it could be both, or it could be either,

01:32:53.860 | depending on the particular instance.

01:32:56.160 | It's very clear that there are many kids with dyslexia

01:33:00.900 | where the problem is a problem of phonological awareness.

01:33:05.260 | So it can be very hard to detect

01:33:09.680 | because they may understand the words that you were saying,

01:33:12.140 | but because the brain is so good at pattern recognition,

01:33:16.060 | sometimes even if the individual speech sounds

01:33:18.220 | are not crystal clear, it can compensate that

01:33:20.760 | so that you can have an individual who can hear the words,

01:33:24.420 | but not be able to essentially hear them

01:33:28.580 | when they're reading those same words.

01:33:30.940 | And so what can happen with that

01:33:33.260 | is that you can have this disconnection

01:33:35.740 | between what they're seeing and what they need

01:33:38.920 | in order to hear it as words and process it as language.

01:33:42.660 | And so skilled readers usually need that route first.

01:33:47.660 | They've gotta map the vision to the sound

01:33:51.420 | in order to get that sort of like foundation.

01:33:53.460 | But then over time, the reading has its direct connection

01:33:57.260 | to the language parts of the brain,

01:33:59.140 | and we don't necessarily always need to map to sounds.

01:34:03.180 | You can basically develop a parallel route.

01:34:06.060 | And we as readers actually use both all the time.

01:34:10.180 | So for example, if it's a new word

01:34:11.940 | that you've never seen before,

01:34:13.620 | sometimes you try to like pronounce it in your mind,

01:34:16.580 | you know, and try to hear what that word is.

01:34:19.580 | Even though you're not actually saying it,

01:34:21.180 | you're trying to just generate

01:34:22.280 | what those sounds might be like.

01:34:25.180 | And that's the part where we're kind of relying

01:34:28.420 | on how we learned to read in the first place,

01:34:31.040 | which is mapping those word images

01:34:33.540 | to the sounds that go along with them.

01:34:36.340 | But in other times, if you're a really proficient reader,

01:34:39.400 | you're just seeing the words

01:34:40.520 | and you can map them directly to meaning

01:34:42.420 | without having to go through that process.

01:34:46.980 | - Yeah, I'm a big fan of listening to audio books.

01:34:49.400 | And of course I also listen to podcasts quite a lot.

01:34:53.360 | But I also am a strong believer

01:34:55.400 | based on the research that I've seen

01:34:57.500 | that reading books, physical books,

01:34:59.840 | could be on Kindle, I suppose,

01:35:00.980 | but reading a physical book is useful

01:35:04.300 | for being able to articulate well

01:35:07.740 | and structure sentences

01:35:09.480 | and build what are essentially paragraphs,

01:35:11.340 | which is what I'm required to do

01:35:12.340 | when I do solo episodes of the podcast.

01:35:14.960 | I've noticed over the years

01:35:17.420 | as text messaging has become more popular

01:35:21.260 | and there's essentially an erosion of punctuation

01:35:26.180 | or the need to have complete sentences.

01:35:27.920 | And now that's sort of transferred to email as well.

01:35:31.420 | It's become acceptable to just say,

01:35:34.360 | fragmented sentences in email.

01:35:36.140 | It seems likely that it's starting to impact

01:35:40.820 | the way that people speak as well.

01:35:42.900 | And I don't think this has anything to do

01:35:44.260 | with intelligence or education level,

01:35:46.980 | but are you aware of any evidence

01:35:50.240 | that how we read and what we read

01:35:53.780 | and whether or not we consume information

01:35:56.660 | purely through reading or mainly through auditory sources,

01:36:00.260 | does it change the way that we speak?

01:36:01.680 | Because after all, Wernicke's and Broca's area

01:36:04.400 | and the other auditory and speech production areas

01:36:07.220 | are heavily intermeshed.

01:36:10.080 | And so it would make perfect sense to me

01:36:11.620 | that what we hear and the patterns of sound

01:36:15.280 | that are being communicated to us

01:36:16.440 | would also change the way that we speak.

01:36:18.940 | - Yeah, that's a really fascinating point.

01:36:23.540 | There's this idea that there's like this proper way to speak,

01:36:27.340 | like that there's the right way, for example.

01:36:30.020 | What are the appropriate, you know,

01:36:32.660 | like for example, in school, you're oftentimes told like,

01:36:37.540 | you should say it like this, not say it like that, you know.

01:36:40.140 | And every language kind of has that.

01:36:42.040 | It turns out that that's really unnatural.

01:36:45.200 | Languages and speech in particular change over time.

01:36:49.380 | It evolves, and it can happen very quickly.

01:36:54.380 | The things that we call dialects, for example,

01:36:56.780 | are just different ways of speaking,

01:36:58.040 | and someone can just be in one environment

01:36:59.860 | and change from one dialect to another,

01:37:02.280 | or in some people, it kind of is really fixed.

01:37:06.540 | And there's this idea that, you know, like in school,

01:37:11.000 | that we're like told that there's this right way,

01:37:13.280 | but in reality, that's not true.

01:37:16.960 | Like language change and speech change is completely normal

01:37:21.160 | and happens all the time, and it can be really dramatic.

01:37:25.080 | Like certain cultures and communities,

01:37:27.280 | if they are isolated, they can develop a whole new language,

01:37:30.540 | a whole new set of words, for example,

01:37:33.300 | and new ways and dialects that are independent from people

01:37:37.040 | to the point where it's unintelligible even to others.

01:37:40.980 | And so the basic idea is that sound change

01:37:46.100 | is part of the way it works,

01:37:48.800 | and the brain is very sensitive to those kinds of changes.

01:37:52.680 | - Speaking of learning new languages,

01:37:55.400 | I'm assuming it's possible to learn new languages

01:37:58.880 | throughout the lifespan, correct?

01:38:00.980 | - Yeah.

01:38:01.960 | - I've also heard these kind of fantastical stories

01:38:05.300 | of somebody has a stroke and then suddenly, spontaneously

01:38:10.300 | can speak French fluently,

01:38:12.140 | whereas prior to the stroke, they could not.

01:38:14.880 | Is there any merit to those stories whatsoever?

01:38:17.780 | I find it very hard to believe

01:38:20.280 | that there was a complete map representation of a language

01:38:23.360 | in somebody's brain that they were completely unaware of,

01:38:26.720 | and then because of damage to a brain area,

01:38:30.600 | that capacity to speak that language was somehow unveiled.

01:38:34.260 | It just seems too wild and I don't want to say good to be

01:38:38.200 | true because nobody wants a stroke,

01:38:40.000 | but it just seems outrageously implausible.

01:38:44.000 | - Well, there are aspects of that

01:38:45.960 | that certainly are implausible.

01:38:47.360 | So I don't know of any true case that I've ever seen

01:38:51.440 | or experienced myself or even read about

01:38:54.280 | where, for example, there was an injury to the brain

01:38:56.860 | that resulted in loss of,

01:39:00.040 | well, essentially a gain of function,

01:39:03.160 | meaning like this art of just all of a sudden

01:39:05.920 | starts speaking another language.

01:39:08.360 | So for example, if you had a stroke

01:39:09.620 | and you never spoke French and then you had it,

01:39:12.180 | and then all of a sudden you're speaking,

01:39:14.280 | that I've never heard of, never seen.

01:39:16.840 | However, there is a condition that is well acknowledged,

01:39:21.720 | and I have seen one case of this

01:39:23.020 | called a foreign accent syndrome,

01:39:25.680 | which is peculiar because there are people

01:39:29.540 | who have an injury to the part of the brain

01:39:31.800 | where it sounds like they're starting

01:39:34.560 | to speak this other language,

01:39:36.080 | but they're not actually speaking the language.

01:39:38.040 | It just sounds like it.

01:39:40.580 | And this goes back to what we were talking about earlier

01:39:43.600 | about these areas that are really important

01:39:46.400 | for speech control of the vocal tract,

01:39:49.760 | this area in the precentral gyrus.

01:39:51.580 | People have documented where patients

01:39:55.640 | have had strokes there, and after that,

01:39:59.080 | it sounds like they're speaking Spanish

01:40:01.720 | as opposed to English, or it sounds like

01:40:04.040 | they have the intonational properties of French or Russian

01:40:08.300 | as compared to their original native language.

01:40:11.560 | They're not learning all the rest of it,

01:40:13.080 | like the meaning and the grammar, et cetera,

01:40:16.600 | but they're adopting some of the phonology,

01:40:20.360 | and part of that is just because

01:40:22.800 | it's not working the way it normally does.

01:40:25.560 | So there is something actually called

01:40:26.920 | a foreign accent syndrome

01:40:28.320 | that people can have after a stroke.

01:40:30.120 | - Interesting.

01:40:32.760 | I'm curious about auditory memory.

01:40:34.720 | When I was a kid, I used to get into bed at night

01:40:37.960 | and I'd close my eyes and I would replay conversations

01:40:42.020 | that I'd heard during the day or people's voices.

01:40:44.720 | I actually can remember calling your house

01:40:47.000 | when we were young kids,

01:40:48.360 | and because I don't speak any Chinese,

01:40:50.060 | but I'd have to ask for you, I'd say,

01:40:52.400 | I think it was [speaks in foreign language]

01:40:54.640 | - Yeah. - Yeah.

01:40:55.700 | And then someone, whoever answered the phone would say,

01:40:59.240 | would go get you in there and say, [speaks in foreign language]

01:41:00.680 | which I believe means thank you, right?

01:41:02.560 | That's the total of the Chinese that I speak, by the way.

01:41:06.020 | But I will never forget that.

01:41:09.040 | I'll just never forget it, I hope.

01:41:10.660 | I suppose if I have a stroke or something of that sort,

01:41:12.780 | at some point I'll forget it

01:41:13.700 | and I won't know that I've forgotten it.

01:41:15.060 | But in all seriousness, I remember that to this day.

01:41:18.300 | I couldn't spell that out.

01:41:19.880 | I wouldn't know how, certainly not in Chinese,

01:41:21.780 | but even a transliteration I couldn't do

01:41:25.120 | using English letters.

01:41:28.880 | Where are memories of sounds stored?

01:41:34.860 | Because within our days and across our lives,

01:41:38.000 | we have an infinite number of auditory experiences.

01:41:42.720 | Just like we have an infinite number of visual experiences.

01:41:46.160 | Where are they stored

01:41:46.980 | and what is the structure of their storage?

01:41:48.760 | What am I calling upon?

01:41:51.180 | Besides, of course, the motor commands

01:41:53.340 | that are required to say what I just said in Chinese,

01:41:55.980 | which I won't repeat again,

01:41:57.020 | 'cause somehow I managed to get it right the first time,

01:41:59.580 | or at least not terribly wrong,

01:42:01.260 | then I don't want to botch it the second time.

01:42:03.560 | Where is that stored and how does that work?

01:42:06.240 | And more importantly, as I speak my native language, English,

01:42:10.200 | am I pulling from a memory bank?

01:42:12.380 | Because it doesn't feel like it.

01:42:14.000 | I'm just telling you what I want to say.

01:42:17.080 | I'm doing my best to communicate clearly and succinctly.

01:42:20.820 | Usually not so good at the succinct part.

01:42:22.780 | But where is the bank of information?

01:42:26.980 | On my keyboard, on my computer, I have the letters

01:42:30.980 | and I have certain elements of punctuation in the space.

01:42:33.740 | What am I pulling from?

01:42:34.940 | Am I pulling from those plosives?

01:42:37.620 | But if so, how can I do it so quickly?

01:42:42.360 | Even for people that speak slowly,

01:42:44.420 | it appears more or less fluid.

01:42:47.400 | This to me is overwhelmingly impressive

01:42:51.480 | that the brain can do that.

01:42:52.980 | How does it do that?

01:42:54.940 | - Well, first of all, I am impressed

01:42:59.520 | that 35 years later.

01:43:01.560 | - Well, I had to get ahold of you.

01:43:02.680 | - Yeah, so I am impressed 35 years later

01:43:06.040 | that you can still remember that.

01:43:09.880 | - But only that.

01:43:11.440 | - That's fine.

01:43:12.280 | But I'm still very impressed in,

01:43:16.080 | but it clearly was something important to you.

01:43:17.920 | And so the short answer is that memory is very distributed.

01:43:22.920 | So it's almost like the question that you asked me

01:43:27.880 | is ill-posed.

01:43:30.940 | 'Cause you asked me where.

01:43:31.880 | Well, it's not one specific area.

01:43:34.040 | It's actually really distributed.

01:43:36.840 | It's not just one particular area.

01:43:38.040 | In fact, I'm fairly certain that if we were to injure

01:43:41.000 | that part of the brain called the wernicke's area,

01:43:43.100 | you may still even have memories of that.

01:43:46.420 | People can have injuries of Broca's area

01:43:48.880 | or certainly the precentral gyrus

01:43:50.840 | and be able to sing "Happy Birthday," for example,

01:43:54.960 | when it's embedded in melody or highly rehearsed things

01:43:58.360 | like counting despite not being able to speak,

01:44:01.960 | which is incredible, right?

01:44:03.480 | It's like you can see a patient, for example,

01:44:07.560 | who can't really put together a sentence.

01:44:10.880 | You ask them, "How are you feeling today?"

01:44:12.140 | They can't even utter a word,

01:44:14.440 | but then you ask them to count sometimes

01:44:16.160 | and they'll get up to any number, really.

01:44:19.320 | And so there are some things that are really built

01:44:22.800 | into our motor memory and it's distributed.

01:44:25.440 | It's not one particular part of the brain.

01:44:27.520 | It's actually multiple areas

01:44:29.440 | where that memory is distributed.

01:44:30.840 | And thank God that's the way it is

01:44:32.600 | because it's very rare in the kind of surgeries that I do

01:44:37.600 | where you go and you remove a part of a piece of the brain

01:44:43.380 | that someone forgets these kind of long-term memories

01:44:47.040 | or these long-term motor skills that they have.

01:44:51.760 | That's very, very rare.

01:44:53.760 | It's the number one question a patient will ask me,

01:44:56.660 | like, "Am I gonna be the same,

01:44:57.800 | and am I gonna remember my wife?"

01:45:01.940 | Or, "Am I gonna remember these thoughts of my birthday

01:45:06.140 | when I was 10 years old?"

01:45:07.480 | And I've never really seen that kind of severe amnesia

01:45:13.800 | unless it's a very, very severe injury

01:45:15.820 | that involves almost the entire brain.

01:45:17.720 | And thank God.

01:45:19.460 | So a lot of that information is really distributed

01:45:24.020 | across the entire brain.

01:45:26.140 | - Speaking of storage of and ability to speak,

01:45:31.140 | you are doing some amazing work

01:45:33.780 | and have achieved some pretty incredible,

01:45:37.220 | well-deserved recognition for your work

01:45:39.520 | in bringing language out of paralyzed people,

01:45:43.700 | essentially allowing people who are locked in

01:45:46.340 | to a paralyzed state or otherwise unable

01:45:50.060 | to articulate speech using brain-machine interface,

01:45:54.260 | essentially translating the neural activity

01:45:57.320 | of areas of the brain that would produce speech

01:46:00.780 | into hardware, wires, and things of that sort,

01:46:05.780 | artificial non-biological tools

01:46:09.500 | in order to allow paralyzed people to communicate.

01:46:12.100 | We will provide a link

01:46:14.180 | to some of the popular press coverage of that work

01:46:16.860 | in the original papers.

01:46:17.840 | But if you would be so kind as to tell us

01:46:21.020 | what those experiments look like,

01:46:23.780 | who these people are who are locked in

01:46:26.020 | and that you allow to communicate.

01:46:27.780 | And then especially interesting to me

01:46:30.660 | are some of the directions that you're taking this now,

01:46:33.500 | which is beyond just people being able to think

01:46:37.200 | about what they want to say

01:46:38.420 | and words coming out on a screen or through a microphone,

01:46:40.900 | but actually making the interactions

01:46:43.060 | between these people in the real world

01:46:45.340 | more elaborate and more real.

01:46:49.300 | If that seems mysterious to people,

01:46:50.760 | I'm going to let Eddie tell you what they're doing with this

01:46:53.220 | rather than put any more detail on it.

01:46:56.220 | - Oh, okay.

01:46:57.060 | Well, thanks for asking about this.

01:46:58.880 | This has really been some of the exciting,

01:47:01.700 | recent work from the lab.

01:47:04.080 | So for the last decade,

01:47:06.860 | we've really been focusing on the basic science,

01:47:09.580 | meaning trying to understand how the brain extracts

01:47:13.020 | and produces speech sounds and words.

01:47:16.900 | We've done a lot of work trying to figure out

01:47:20.260 | how these parts of the brain

01:47:21.700 | control these individual elements

01:47:24.340 | that give rise to all words and meanings.

01:47:27.060 | And so it was about six years ago

01:47:30.900 | where we realized we actually have a pretty good idea

01:47:35.820 | of how this code works.

01:47:37.680 | We had identified all of these different elements

01:47:42.060 | that we could decode in epilepsy patients.

01:47:44.660 | For example, when they had electrodes on the brain

01:47:47.420 | as part of their surgeries,

01:47:49.220 | we could decode all of the different consonants

01:47:51.540 | and vowels of English.

01:47:52.660 | That was about six years ago.

01:47:54.640 | So a natural question was this,

01:47:57.300 | which is if we understand that electrical code,

01:48:00.980 | can we use that to help someone who is paralyzed

01:48:05.500 | and can't get those signals out of the brain

01:48:08.180 | to speak normally?

01:48:10.600 | And that's in the setting of people who are paralyzed.

01:48:13.500 | So there are a series of conditions.

01:48:16.820 | They include things like brainstem stroke.

01:48:18.980 | The brainstem is the part of the brain

01:48:20.920 | that connects the cerebrum, which is the top part,

01:48:23.860 | does our thinking and a lot of the motor control,

01:48:26.300 | speech, language, everything.

01:48:28.300 | And the brainstem is what connects that to the spinal cord

01:48:30.860 | and the nerves that go out to the face and vocal tract.

01:48:32.920 | So if you have a stroke there,

01:48:34.900 | basically you could be thinking all the wild,

01:48:37.940 | creative, intelligent thoughts.

01:48:40.380 | You have in the mind and the cerebrum,

01:48:41.880 | but you can't get them out into words.

01:48:44.640 | Or you can't get them out to your hand to write them down.

01:48:48.420 | So that's a very severe form of paralysis

01:48:50.400 | called brainstem stroke.

01:48:52.420 | There's another kind of conditions

01:48:54.440 | that we call neurodegenerative,

01:48:56.020 | where the nerve cells die, basically,

01:48:59.620 | or atrophy in a condition called ALS.

01:49:03.640 | And that's a very severe form of paralysis.

01:49:08.400 | And it's extreme form, people essentially lose

01:49:10.580 | all voluntary movement.

01:49:12.180 | >> So Stephen Hawking would be a good example

01:49:14.040 | of someone with ALS, Lou Gehrig's disease.

01:49:16.060 | >> He's an example of someone who had ALS,

01:49:18.600 | but not a great example of what typical course of ALS.

01:49:22.060 | So for reasons not clear,

01:49:25.500 | the progression of his disease largely stabilized

01:49:28.420 | to the point where he could twitch, you know,

01:49:31.020 | a cheek muscle or move his eyes, let's say.

01:49:34.820 | In most people, it's very rapid.

01:49:38.060 | And many people, they die from it actually,

01:49:40.540 | you know, within a couple of years of diagnosis.

01:49:43.800 | >> Yeah, he lived a long time in that.

01:49:45.360 | >> He lived a long time.

01:49:46.300 | >> That's like slanted overstate in his wheelchair.

01:49:48.620 | >> Exactly, but he wasn't breathing, you know,

01:49:53.300 | through a tube in his throat, for example,

01:49:55.180 | because people with severe ALS,

01:49:56.920 | the muscles to their diaphragm and their lungs

01:50:01.300 | essentially give out as well.

01:50:02.800 | They get weakness there and then they can't breathe anymore.

01:50:05.260 | So that's another form of paralysis.

01:50:08.860 | And so, in our field, these are kind of like

01:50:12.460 | the most devastating things that can happen.

01:50:14.760 | I'm not gonna really try to compare like what's worse,

01:50:18.460 | you know, having a brain tumor or stroke, it's all bad.

01:50:21.900 | But this condition of what we call being locked in

01:50:26.380 | refers to this idea that you can have

01:50:28.920 | completely intact cognition and awareness,

01:50:35.020 | but have no way to express that.

01:50:37.620 | No voluntary movement, no ability to speak.

01:50:40.980 | And that is devastating because psychologically

01:50:44.420 | and socially, you know, you're completely isolated.

01:50:47.140 | That's what we call locked in syndrome.

01:50:49.120 | And it's devastating.

01:50:50.300 | I've seen that throughout my career.

01:50:52.180 | And it's really heartbreaking because

01:50:55.140 | you know that the person is there,

01:50:59.940 | but you can't see, they can't communicate.

01:51:02.880 | So, we've been studying this patterning

01:51:07.060 | of electrical activity for consonants and vowels.

01:51:09.720 | And essentially once we figured out a lot of these codes

01:51:13.200 | for the individual phonetic elements,

01:51:16.760 | we took a little bit of a detour,

01:51:20.500 | or at least part of the lab started to focus

01:51:22.320 | on this very specific question.

01:51:24.660 | For people who have these kind of paralysis,

01:51:27.180 | could we intercept those signals from the brain,

01:51:32.240 | the cerebral cortex, as someone is trying

01:51:35.020 | to say those words?

01:51:36.960 | And then can we intercept them and then have them

01:51:39.800 | taken out of the brain through wires to a computer

01:51:43.860 | that are gonna interpret those signals

01:51:46.160 | and translate them into words?

01:51:48.900 | So, about three years ago, we started a clinical trial.

01:51:53.900 | It's called the BRAVO trial, it's still underway.

01:51:57.460 | And the first participant in the BRAVO trial

01:52:00.680 | was a man who had been paralyzed for 15 years.

01:52:04.140 | When he was about 20 years old, he came to the United States,

01:52:10.060 | was actually working in Sonoma area,

01:52:13.720 | and he was in a car accident.

01:52:15.700 | And he actually walked out of the hospital

01:52:18.900 | the day after that car accident.

01:52:21.220 | But the next day had a complication related to it

01:52:25.300 | where he had a very large stroke in the brain stem.

01:52:29.540 | And that turned out to be devastating.

01:52:31.960 | He didn't wake up from that stroke for about a week.

01:52:34.900 | He was in a coma for about a week.

01:52:37.140 | And when he woke up from that coma,

01:52:38.840 | he realized that he couldn't speak

01:52:40.480 | or move his arms or legs.

01:52:42.060 | And as he told me or communicated to us,

01:52:47.760 | that was absolutely devastating.

01:52:49.240 | He wanted really to die at that time.

01:52:52.360 | - Could he blink his eyes or move his mouth in any way?

01:52:55.180 | - He could blink his eyes.

01:52:56.920 | He had some limited mouth movements,

01:52:58.340 | but couldn't produce any intelligible speech.

01:53:01.240 | He was, like, completely slurred and incomprehensible.

01:53:04.480 | And he survived this injury.

01:53:08.380 | A lot of people who have that kind of stroke

01:53:09.780 | just don't survive.

01:53:11.500 | But he survived, and I also realized

01:53:15.120 | that he's just an incredible person,

01:53:17.060 | like a force of nature in terms of his optimism,

01:53:20.660 | in terms of his ability to make friends

01:53:22.440 | despite his condition.

01:53:23.380 | The way he actually communicates,

01:53:25.680 | because he has a little bit of residual neck movements,

01:53:28.140 | is that he improvised and had his friends

01:53:32.340 | basically put a stick attached to his baseball cap.

01:53:37.020 | Because he could move his neck,

01:53:38.740 | he would essentially type out letters

01:53:41.380 | on a keyboard screen to get out words.

01:53:43.900 | In fact, this is how he communicated,

01:53:46.280 | was through a device that he would essentially

01:53:48.260 | peck out letters one by one by moving his neck

01:53:51.940 | to control the stick attached to his baseball cap.

01:53:54.860 | - How many years did he use that method of communication?

01:53:58.380 | - For about 15 years.

01:53:59.700 | He hadn't really spoken for about 15 years.

01:54:02.180 | - Oh, goodness. - Yeah.

01:54:03.700 | So it was a devastating injury,

01:54:06.780 | but there's something to be said about the human spirit.

01:54:11.780 | And if there's anyone who embodies it,

01:54:14.700 | it is Pancho, that's his nickname,

01:54:17.180 | the first participant in our trial.

01:54:19.000 | He has that human spirit.

01:54:21.620 | He persevered and in fact could thrive in his community,

01:54:25.380 | basically, and friends, being able to communicate

01:54:27.700 | in this very slow and inefficient way.

01:54:29.700 | Maybe part of that spirit is why he volunteered

01:54:33.740 | to be the first person in this trial.

01:54:36.060 | It was a clinical trial, an experiment.

01:54:39.140 | It was a study.

01:54:39.980 | This is not an approved therapy by any means.

01:54:42.660 | This was really something that had not been done before.

01:54:45.940 | And we had a lot of ideas about it, but we didn't know.

01:54:50.780 | We had proven a lot of this could be true

01:54:52.540 | in some people who were normally speaking.

01:54:55.020 | But to actually put it into someone who's paralyzed,

01:54:57.100 | number one, where we don't know the code is the same.

01:55:00.440 | Number two is someone who's not been speaking for 15 years,

01:55:04.860 | whether those signals are actually still there or not.

01:55:08.080 | So it was part of a clinical trial.

01:55:11.540 | It was, you know, something that our hospital

01:55:15.460 | and also the FDA, you know, had to approve

01:55:17.920 | and looked at very carefully.

01:55:19.060 | But given a lot of the work that we had done,

01:55:21.180 | there were some basis for why this might work.

01:55:24.900 | And so about two and a half years ago,

01:55:28.260 | we did a surgery where we implanted electrodes

01:55:32.300 | onto the parts of the brain that we've been talking about,

01:55:35.000 | these areas that control the vocal tract,

01:55:37.580 | the areas that control the larynx,

01:55:38.980 | the areas that control the lips and tongue and jaw movements

01:55:43.180 | when we normally speak.

01:55:44.220 | These are areas that presumably may be active.

01:55:47.060 | That was our hope in his brain,

01:55:49.060 | but he just couldn't get those out

01:55:50.940 | to control his mouth in a normal way.

01:55:54.260 | And he underwent a surgery, a brain surgery,

01:55:56.900 | where we put an electrode array

01:55:59.340 | and we connected it to a port that was screwed to his skull.

01:56:04.180 | And the port actually goes through his scalp.

01:56:08.240 | And he's lived with this now for the last three years.

01:56:11.740 | It is a risk of infection.

01:56:13.300 | These ports eventually have to become wireless

01:56:15.600 | in the future, but we figured out a way

01:56:17.980 | to keep that port there where we can essentially

01:56:21.280 | connect him to a computer through that port.

01:56:24.580 | So he has an electrode array that's implanted

01:56:29.200 | over the part of his brain that's important for speech.

01:56:32.740 | It's connected to a port.

01:56:34.300 | And then we connect a wire to that port

01:56:38.240 | that translates those, what we call analog brainwaves,

01:56:43.120 | and converts them into digital signals.

01:56:46.120 | And then a computer takes those digital signals

01:56:49.900 | from those individual sites from the speech cortex

01:56:53.400 | and translates those into words.

01:56:56.020 | - Can you describe for us the first time

01:56:58.300 | that Poncho spoke through this engineered device?

01:57:03.300 | What was that experience like for you?

01:57:05.620 | And at least from what he conveyed to you,

01:57:07.900 | what was that experience like for him?

01:57:10.360 | Is this somebody who was essentially locked in

01:57:12.300 | except for this rather crude pecking device?

01:57:15.320 | Although I'm thoroughly impressed by how adaptive

01:57:19.560 | or adaptable Poncho was in his friend's engineering

01:57:22.820 | that device for him is really nothing short of clever

01:57:26.600 | because otherwise he would be truly locked in, right?

01:57:30.920 | But what was that moment like?

01:57:32.840 | I can only imagine.

01:57:34.440 | - That moment was incredible.

01:57:35.940 | It was truly incredible to be able to see him

01:57:40.060 | try to get out a word that was,

01:57:43.400 | for all practical purposes, unintelligible,

01:57:46.280 | but to be able to take the brain activity

01:57:48.220 | and to translate it into text on a screen.

01:57:51.400 | And that's what we did.

01:57:52.240 | We took those brainwaves, we put them through

01:57:55.440 | machine learning or artificial intelligence algorithm

01:57:58.140 | that can pick up these very, very subtle patterns.

01:58:01.480 | You can't actually see them with your eye

01:58:03.380 | in the brain activity and translate those into words.

01:58:08.000 | And I remember seeing this happening for the first time.

01:58:12.920 | It doesn't happen like immediately.

01:58:14.580 | This is something that took weeks to train the algorithm

01:58:18.960 | to interpret it correctly.

01:58:20.420 | But what was incredible about it was to see how he reacted.

01:58:26.740 | And he would be prompted to say a given word,

01:58:31.580 | like outside, for example.

01:58:35.260 | And then he would think about it, try to say it,

01:58:37.760 | and finally those words would appear on the screen.

01:58:41.020 | And what was really amazing about it was

01:58:44.480 | you could really tell that he got a kick out of that

01:58:48.400 | because he would start to giggle.

01:58:51.520 | His body would shake in a way

01:58:54.960 | and his head would shake in a way

01:58:56.180 | that he would start to giggle.

01:58:58.180 | And that was cool to see.

01:59:00.100 | But then I also realized that when he was giggling,

01:59:03.380 | it kind of screwed up the next word's decoding.

01:59:05.880 | (laughing)

01:59:07.420 | - Is that a bug you've since fixed?

01:59:09.700 | - No, we haven't fixed that.

01:59:10.980 | - Interesting. - We haven't fixed that.

01:59:12.300 | So it's easier just to tell him to stop giggling.

01:59:15.100 | (laughing)

01:59:16.580 | - So what was the first word that he said?

01:59:18.800 | - Well, I think one of the first sentences

01:59:20.660 | that he put together was, you know,

01:59:22.480 | can you get my family outside?

01:59:24.380 | - You mean get them out of the room?

01:59:26.820 | - No, no.

01:59:27.660 | - All these years, he wanted to get away from his family?

01:59:29.500 | - No, I think what he meant was can you get them--

01:59:31.580 | - Bring them in. - Bring them in.

01:59:32.740 | And so the way this worked was we trained

01:59:37.660 | this computer to recognize 50 words.

01:59:39.700 | We started with a very small vocabulary

01:59:42.460 | that's expanding as we speak.

01:59:44.260 | I think that this is just a matter of time

01:59:46.300 | before these vocabularies become much, much larger.

01:59:50.300 | But we started with a 50 set of words.

01:59:52.620 | We created essentially all the possible sentences

01:59:56.260 | that you could generate from those 50 words.

01:59:59.260 | Why that was important was you can use those,

02:00:02.420 | all those possible sentences to create

02:00:04.420 | a computational model, computer model

02:00:06.820 | of all the different word combinations

02:00:09.200 | to give different sentences, given those 50 words.

02:00:12.940 | And then you can essentially do what we call autocorrect.

02:00:16.540 | It's the same kind of thing that we do

02:00:18.660 | when you're texting, for example,

02:00:20.620 | you get the right, the wrong letter in there,

02:00:23.220 | but your phone actually knows, you know,

02:00:26.220 | because it's context what corrected.

02:00:29.060 | So because the decoding's not 100% correct all the time,

02:00:32.840 | in fact, it's far from that, it's really helpful

02:00:35.980 | to have these other features like autocorrect,

02:00:38.100 | the stuff that we use routinely now with texting

02:00:40.900 | that makes it correct and then updates it.

02:00:43.700 | So it's a combination of a lot of things.

02:00:45.860 | It's the AI that is translating

02:00:47.860 | those brain activity patterns.

02:00:49.980 | But it's also things that we've learned from speech

02:00:53.220 | and speech technologies that, you know,

02:00:55.940 | you put all together and then all of a sudden

02:00:57.780 | it starts to work.

02:00:59.400 | And so we were really excited

02:01:01.740 | because that was the first time that someone was paralyzed

02:01:04.400 | and could create words and sentences

02:01:07.920 | that was just decoded from the brain activity.

02:01:10.480 | - Incredible.

02:01:11.360 | And I know you're very humble,

02:01:13.460 | but I'm going to embarrass you by saying,

02:01:14.700 | I always knew you were destined for great things

02:01:17.460 | since the early age of nine when we first became friends.

02:01:21.440 | But when I read that, the news coverage of your work

02:01:25.760 | with Poncho and the release of this language

02:01:29.040 | from this locked in patient, it literally, you know,

02:01:32.460 | it brought tears to my eyes because it's, you know,

02:01:34.460 | it's an interesting thing as fellow neuroscientists, right?

02:01:37.820 | We explore the brain and we try and find mechanisms

02:01:41.640 | and we try and compare those to what other people find

02:01:44.100 | and find truths and principles and build up from those.

02:01:47.020 | But pretty rarely is there a case where that route

02:01:52.020 | of exploration leads to something of clinical significance

02:01:56.820 | within one's own lifetime.

02:01:58.500 | I mean, that's the reality of science.

02:02:00.140 | And oftentimes it's a very distributed process and,

02:02:03.260 | but in this case, it's been a magnificent thing

02:02:06.240 | to see you move along this trajectory,

02:02:09.400 | parsing these language and speech areas,

02:02:11.200 | and then to also do the clinical work in parallel.

02:02:14.200 | Speaking of which, these days,

02:02:17.380 | we hear a lot about Neuralink, Elon Musk's company,

02:02:21.560 | a neurosurgeon that came up briefly through my lab,

02:02:24.460 | but I can't take any credit for what he knows or does,

02:02:26.800 | which is Matt McDougall is the neurosurgeon at Neuralink.

02:02:29.700 | There's some other excellent neuroscientists there

02:02:31.400 | and engineers there.

02:02:32.600 | We hear a lot about Neuralink because

02:02:34.400 | while brain machine interface of the sort that you do

02:02:38.720 | and that other laboratories do

02:02:40.860 | has been going on for a long time,

02:02:42.560 | there's been some press around Neuralink

02:02:47.060 | about the promise of what brain machine interface could do.

02:02:50.400 | For instance, early in our discussion,

02:02:52.320 | you talked about how, you know,

02:02:53.500 | language is constrained by these sound waves.

02:02:55.640 | And typically it's a few people communicating

02:02:58.120 | or one person with many people

02:03:00.360 | through a podcast, for instance, or a speech.

02:03:02.600 | But the idea has been thrown out there

02:03:05.660 | that through the use of stimulating chips

02:03:08.520 | or through other brain machine devising,

02:03:11.960 | that perhaps one could internalize

02:03:15.120 | 50 conversations in parallel, right?

02:03:17.920 | 50x communication.

02:03:19.960 | Or that the memory systems could be augmented

02:03:24.160 | to remember 10 times as much information

02:03:26.800 | or even twice as much information in a given period of time.

02:03:29.920 | My understanding of what they're doing at Neuralink,

02:03:32.360 | which is admittedly crude and from the outside,

02:03:35.720 | few discussions with people there,

02:03:37.240 | is that they too are going to pursue clinical goals first.

02:03:41.520 | Things like trying to generate smooth movement

02:03:43.880 | in a Parkinsonian patient,

02:03:45.460 | trying to adjust movement patterns

02:03:49.240 | in someone with Huntington's disease, for instance,

02:03:51.560 | things of that sort.

02:03:53.800 | Before they embark on the more sci-fi-like

02:03:58.800 | explorations of 50x-ing communication

02:04:01.840 | or doubling memory capacity and these kinds of things.

02:04:04.360 | Although I don't know, they may be doing

02:04:05.440 | all of those things in parallel.

02:04:07.040 | What are your thoughts about super capabilities of the brain

02:04:13.220 | or I don't even know what word to use.

02:04:16.440 | Supercharging the brain, giving the brain functions

02:04:19.400 | for which we've never observed before in human history.

02:04:23.100 | We have our Einsteins and our Feynmans and our Merzenics

02:04:26.760 | and it's unclear who to put in along that line side by side,

02:04:30.860 | but they're Michael Jordans and et cetera.

02:04:34.200 | But we've never heard of or seen somebody

02:04:37.540 | who can jump 20 feet in the air.

02:04:40.040 | Or we've heard of people who have photographic memories,

02:04:44.320 | but I don't know that we are aware of any human being

02:04:47.020 | in history who could memorize the entire library of Congress

02:04:51.620 | or all the works within the Vatican within an hour.

02:04:55.260 | Anyway, you get the idea.

02:04:57.320 | What are your thoughts about manipulating neural circuitry

02:05:01.700 | to achieve superhuman or superhuman

02:05:05.080 | or super physiological functions?

02:05:07.080 | Are we there or should we even be thinking about that?

02:05:10.420 | Is it possible given that neurons simply communicate

02:05:13.440 | through electrical activity and electrical activity

02:05:15.420 | can be engineered outside of the brain?

02:05:17.880 | How do you think about it?

02:05:19.420 | And here we don't even have to think about

02:05:20.640 | Neural Link in particular, it's just but one example

02:05:24.580 | of companies and people in laboratories

02:05:26.420 | that are quite understandably considering all this.

02:05:30.460 | - Well, it's a really interesting time right now.

02:05:34.580 | The science has been going on for decades.

02:05:37.580 | The work that we've done in this field

02:05:39.420 | that you call brain-machine interface

02:05:41.220 | has been going on for a while.

02:05:44.020 | And a lot of the early work was just trying to restore

02:05:47.260 | things like arm movement or having people

02:05:50.100 | or monkeys control a computer cursor, for example,

02:05:53.000 | on the screen, that's been going on for decades.

02:05:56.260 | What's been really new is that industry is now involved

02:06:01.180 | and some of this is now becoming commercialized

02:06:03.640 | and we're starting to see us now cross over to this field

02:06:08.180 | where it's no longer just research

02:06:11.020 | that we're talking about medical products

02:06:13.580 | that are designed to be surgically implanted.

02:06:16.740 | In some cases, there's people doing this kind of work

02:06:19.360 | non-invasively as well that don't require surgery.

02:06:22.780 | The specific question that you're asking about

02:06:26.800 | is an area that we call augmentation.

02:06:29.940 | So can you build a device that essentially enhances

02:06:34.940 | someone's ability beyond super normal, super memory,

02:06:41.340 | super communication speeds beyond speech, for example?

02:06:46.980 | I guess superior precision athletic abilities.

02:06:51.980 | I think that these are very serious kind of questions

02:06:58.440 | to be asking now because, as you mentioned,

02:07:01.900 | the pathway so far is really to focus

02:07:05.000 | on these medical applications.

02:07:07.860 | I personally don't think that we've thought enough,

02:07:10.200 | actually, about what these kind of scenarios

02:07:12.400 | are gonna look like.

02:07:13.840 | And I don't think we've thought through

02:07:14.920 | all the ethical implications of what this means

02:07:17.440 | for augmentation in particular.

02:07:20.960 | There's part of this that is not new at all.

02:07:26.780 | Humans throughout history have been doing things

02:07:30.860 | to augment our function.

02:07:32.500 | Coffee, nicotine, all kinds of things,

02:07:36.940 | all kinds of medications that cross over

02:07:39.720 | from medical to consumer, that is everywhere.

02:07:44.240 | So the pursuit of augmentation or performance

02:07:47.880 | or enhancement is really not a new thing.

02:07:50.700 | The questions really, as they relate to neurotechnologies,

02:07:55.680 | for example, have to do with the invasive nature.

02:07:59.840 | For example, if these technologies require surgery,

02:08:03.200 | for example, to do something that is not

02:08:05.700 | for a medical application.

02:08:07.880 | Again, there, that is not exactly new territory either.

02:08:12.140 | People do that routinely for cosmetic kind of procedures,

02:08:15.760 | for physical appearance, not necessarily cognitive.

02:08:18.840 | So I do think that provided the technology continues

02:08:23.840 | to emerge the way that it does,

02:08:26.680 | that it's gonna be around the corner.

02:08:29.480 | And it probably is not gonna be in ways

02:08:30.880 | that are super obvious.

02:08:32.480 | I don't think it's gonna be like,

02:08:34.320 | can we easily memorize every fact in the world?

02:08:37.200 | But in forms that are gonna be much more incremental

02:08:40.240 | and maybe more subtle, in many ways,

02:08:43.680 | we already have that now.

02:08:44.860 | Like, for example, you don't have to have

02:08:46.540 | a neural interface embedded in your brain

02:08:48.600 | to get information, essentially access

02:08:51.380 | to all information in the world.

02:08:53.140 | You just have to have your iPhone.

02:08:55.780 | Whether you could do it faster through a brain interface,

02:09:00.520 | I definitely wouldn't rule that out.

02:09:04.440 | But think about this, that the systems that we have already

02:09:08.220 | to speak and to communicate have evolved over,

02:09:11.940 | you know, thousands and millions of years.

02:09:14.560 | And they're supported by neural structures

02:09:17.240 | that have bandwidth of millions of neurons.

02:09:20.560 | There's no technology that exists right now

02:09:24.960 | that people are thinking about

02:09:26.240 | that are in commercial form, certainly.

02:09:28.400 | Not even in research labs that come anywhere close

02:09:32.060 | to what has been evolved for those natural purposes.

02:09:35.880 | So I'm essentially saying two sides of this,

02:09:39.520 | which is we're already getting into this now.

02:09:43.200 | This is not new territory, this topic of augmentation,

02:09:46.460 | both physical and cognitive.

02:09:48.900 | We've already surpassed that.

02:09:49.860 | That's part of what humans do in general.

02:09:53.840 | But we are entering this area of enhanced cognition,

02:09:58.000 | these areas that I think the technology

02:10:01.600 | is gonna be the rate-limiting step in how far we can go.

02:10:04.760 | We have not had the full conversations about,

02:10:07.580 | number one, is this what we actually want?

02:10:11.260 | Is this gonna be good for society?

02:10:13.000 | Who gets access to this technology?

02:10:15.600 | These are all things that are gonna

02:10:16.880 | become real-world problems.

02:10:19.700 | - There's certainly a lot to consider.

02:10:21.380 | In thinking about augmentation,

02:10:23.500 | and another theme that I've yet to ask you about,

02:10:26.880 | but I'm extremely curious about,

02:10:28.280 | which is facial expressions.

02:10:30.640 | Before we talk about the relationship

02:10:32.200 | between the musculature of the face and language

02:10:34.880 | and the communication of emotion,

02:10:36.620 | I'd love for you to, if you would,

02:10:40.280 | touch on a little bit of what you're doing

02:10:41.640 | with patients like Pancho,

02:10:43.000 | to move beyond somebody who's locked in,

02:10:46.260 | being able to type out words on a screen

02:10:48.520 | with their thoughts.

02:10:49.580 | There's a rich array of information

02:10:53.120 | contained within the face and facial expression.

02:10:55.800 | And while somebody like Pancho,

02:10:57.600 | going from having to be completely locked in

02:11:00.980 | to being able to peck out letters on a keyboard,

02:11:03.700 | to being able to just think of those letters

02:11:05.420 | and having them spelled out,

02:11:06.360 | that's a tremendous set of leaps forward towards normalcy.

02:11:11.320 | It's still far and away different

02:11:13.760 | than Pancho speaking with his mouth,

02:11:16.260 | which I think knowing some people who are restricted,

02:11:19.720 | who are quadriplegic,

02:11:22.080 | a lot of what they struggle with in the real world

02:11:24.040 | is actually a height difference sometimes

02:11:25.800 | because they're seated while other people are standing.

02:11:27.840 | This actually, we don't often think about this,

02:11:29.820 | but always have to look up to communicate with people.

02:11:31.920 | It's a very different interface in the world.

02:11:34.600 | They manage quite well, of course,

02:11:35.840 | but could you tell us what you're doing

02:11:38.300 | in terms of merging the brain machine interface

02:11:41.360 | with extraction of speech signals

02:11:43.200 | from people who are locked in like Pancho

02:11:45.600 | with facial expressions?

02:11:47.020 | - Sure, yeah.

02:11:48.100 | Well, like we described before, progress is being made.

02:11:53.100 | The proof of principle is out there

02:11:56.240 | that you can decode speech.

02:11:57.780 | That will continue to optimize,

02:11:59.220 | and I'm very confident that that's gonna improve

02:12:02.400 | very, very quickly in the coming years

02:12:04.880 | to the point where it's like not just a small vocabulary

02:12:08.360 | but a large vocabulary and at reasonable rates,

02:12:12.180 | at a level that's gonna be really helpful.

02:12:14.040 | I'm very optimistic about that.

02:12:15.740 | I think it's the right time

02:12:17.800 | to start really thinking about a broader vision

02:12:20.640 | of what communication really is.

02:12:22.600 | So for example, I'm here with you in person.

02:12:26.560 | We could have done this virtually, probably.

02:12:29.080 | It's pretty easy to do that.

02:12:30.320 | We could have recorded this really separate,

02:12:32.560 | but there is something about being able

02:12:34.220 | to actually see your expressions

02:12:36.860 | and to understand other forms of communication.

02:12:41.440 | So another really important one is nonverbal,

02:12:45.280 | the expressions that you're making.

02:12:49.020 | For example, if you have a quizzical look on your face,

02:12:51.260 | if I'm saying something not clear,

02:12:53.240 | that's a sign to me that I need to rephrase it

02:12:55.760 | or to say it in a different way or to slow down.

02:12:58.740 | For example, or if there's something

02:13:01.260 | that really excites you,

02:13:02.320 | I wanna continue to say more about it

02:13:04.080 | and talk more in detail, essentially, about a given thing.

02:13:08.880 | So facial expressions actually are a really important part

02:13:13.300 | of the way we speak and there's two things.

02:13:15.400 | It's not just the expressions of how you're feeling

02:13:19.280 | and perceiving what I'm saying,

02:13:21.520 | but it's also seeing my mouth move.

02:13:25.120 | In your eyes, I actually see my mouth move

02:13:27.200 | and my jaw move in a particular way

02:13:29.080 | that actually allows you to hear those sounds better.

02:13:32.760 | So having both the visual information,

02:13:36.880 | but also the sounds go into your brain

02:13:39.720 | is going to improve intelligibility,

02:13:41.680 | also make it more natural.

02:13:43.680 | - And memory for what is spoken?

02:13:46.720 | - Perhaps.

02:13:47.560 | - So here's a call for people,

02:13:48.920 | not just listening to podcasts,

02:13:50.200 | but watching them and listening to them on YouTube,

02:13:52.720 | I suppose, if we were to translate this to the real world.

02:13:56.720 | - Exactly, and the reason why we're also very interested

02:13:59.760 | in this idea of not just having text on a screen,

02:14:04.680 | but essentially a fully computer-animated face,

02:14:09.640 | like an avatar of the person's speech movements

02:14:13.840 | and their facial expressions,

02:14:16.100 | is gonna be a more complete form of expression.

02:14:19.340 | Now, you can imagine right now,

02:14:23.060 | that might just be someone looking at a computer screen

02:14:25.560 | interpreting these signals,

02:14:26.840 | but I think the way things are going

02:14:29.720 | in the next couple of years,

02:14:31.120 | a lot more of our social interactions,

02:14:33.320 | more than even now,

02:14:34.920 | are gonna move into this digital virtual space.

02:14:37.900 | And of course, most people are thinking about

02:14:40.480 | what that means for most consumers,

02:14:43.340 | but it also has really important implications

02:14:45.320 | for people who are disabled, right?

02:14:47.080 | And how are they gonna participate in that?

02:14:50.400 | And so we're thinking really about,

02:14:52.120 | for people like Pancho and other people who are paralyzed,

02:14:55.880 | what other forms of BCI can we do

02:15:00.600 | in order to help improve their ability to communicate?

02:15:02.960 | So one is essentially building out more holistic avatars,

02:15:07.080 | things that can essentially decode,

02:15:11.840 | essentially, their expressions

02:15:14.100 | or the movements associated with their mouth and jaw

02:15:18.080 | when they actually speak to improve that communication.

02:15:21.500 | - So do you envision a time not too long from now

02:15:24.260 | where instead of tweeting out something in text,

02:15:26.900 | my avatar will, I'll type it out,

02:15:29.340 | but my avatar will just say it.

02:15:31.120 | It'll be an image of my avatar saying,

02:15:33.680 | whatever it is I happen to be tweeting at that moment.

02:15:35.360 | - That's what we're working on, yeah.

02:15:36.620 | So I don't think that that is gonna happen

02:15:39.800 | and it's gonna happen soon.

02:15:40.920 | And there's a lot of progress in that.

02:15:43.340 | And again, we're just trying to enrich

02:15:46.100 | the field of communication expression

02:15:51.500 | to make it more normal.

02:15:52.760 | And we actually think that having that kind of avatar

02:15:56.680 | is a way of getting feedback to people

02:15:58.760 | learning how to speak through a speech neuroprosthetic,

02:16:02.120 | that's the device that we call it,

02:16:03.320 | it's a speech neuroprosthetic,

02:16:05.320 | that that is gonna be the way that can help people

02:16:08.240 | learn how to do it the quickest,

02:16:09.800 | not necessarily like trying to say words

02:16:11.800 | and having it come on a screen,

02:16:13.600 | but actually have people embody,

02:16:16.400 | feel like it's part of themselves

02:16:18.320 | or that they are directly controlling that,

02:16:21.040 | that illustration or animation.

02:16:22.960 | - This idea of an avatar speaking out

02:16:28.480 | what we would otherwise write

02:16:30.480 | is fascinating to me on Instagram.

02:16:32.700 | I post videos, I don't filter them,

02:16:35.520 | but I know there's a lot of discussion nowadays

02:16:38.000 | about people using filters

02:16:39.920 | to make their skin look different

02:16:41.120 | or the lighting look different,

02:16:42.260 | a lot of filtering and also the use of captions.

02:16:44.920 | So that essentially what you end up with

02:16:46.300 | is somewhere between an actual raw video

02:16:49.960 | of what was spoken and an avatar version of it.

02:16:53.040 | I mean, if the mismatch between what's spoken

02:16:55.080 | and what's in the caption is too dramatic,

02:16:57.660 | then it doesn't quite work.

02:16:58.720 | But I watch these carefully when people use captions

02:17:01.980 | and oftentimes there's a smoothing

02:17:04.520 | of what was said into the caption

02:17:05.760 | so it seems much more succinct and accurate.

02:17:08.160 | Oftentimes the reverse is also true

02:17:10.200 | where the caption is inaccurate

02:17:11.580 | and then it creates this kind of jarring mismatch.

02:17:14.560 | In any case, I think this aspect in the clinical realm

02:17:17.440 | of using an avatar to allow people like Poncho

02:17:19.760 | to essentially be a face that communicates

02:17:22.060 | through spoken language from an avatar that looks like them

02:17:25.440 | is fascinating and indeed important.

02:17:28.360 | And I think how avatars emerge in social spaces

02:17:32.300 | is going to be really fascinating.

02:17:34.660 | I get a lot of questions about stutter.

02:17:37.120 | I think that for people who have a stutter,

02:17:39.580 | it is itself anxiety provoking.

02:17:43.220 | - Is stutter related to anxiety?

02:17:45.860 | If one has a stutter, what can they do?

02:17:49.280 | Does stutter reflect some underlying neurologic phenomenon

02:17:55.520 | that might distinguish between one kind of stutter

02:17:58.500 | and another?

02:17:59.760 | What can people with stutter do

02:18:01.180 | if they'd like to relieve their stutter?

02:18:02.660 | - Yeah, great question.

02:18:04.120 | Stutter is a condition where the words

02:18:10.440 | can't come out fluently.

02:18:11.580 | So you have all the ideas.

02:18:13.060 | You've got the language intact.

02:18:14.600 | You know, remember we talked about this distinction

02:18:16.140 | between language and speech.

02:18:18.080 | Stuttering is a problem of speech, right?

02:18:20.240 | So the ideas, the meanings, the grammar,

02:18:23.240 | it's all there and people stutter,

02:18:25.400 | but they can't get the words out fluently.

02:18:28.300 | So that's a speech condition.

02:18:30.940 | And in particular, it's a condition

02:18:32.820 | that affects articulation,

02:18:35.880 | specifically about controlling the production of words

02:18:40.140 | in this really coordinated kind of movements

02:18:42.380 | that have to happen in the vocal tract

02:18:44.420 | to produce fluent speech.

02:18:47.480 | And stuttering is a condition

02:18:50.860 | where people have a predisposition to it.

02:18:55.680 | So there's an aspect of stuttering.

02:18:58.660 | You are a stutterer or you're not a stutterer, right?

02:19:02.200 | But people who stutter don't stutter all the time either.

02:19:06.080 | So you could be a stutterer who stutters at some times,

02:19:09.380 | but not others.

02:19:11.180 | And really the main link between stuttering and anxiety

02:19:15.480 | is that anxiety can provoke it and make it worse.

02:19:20.480 | That's certainly true,

02:19:23.100 | but it's not necessarily caused by anxiety.

02:19:26.860 | It can essentially trigger it or make it worse,

02:19:30.680 | but it's not the cause of it per se.

02:19:33.360 | So the cause of it is still really not clear,

02:19:36.460 | but it does have to do with these kind of brain functions

02:19:40.180 | that we've been talking about earlier,

02:19:42.300 | which is that in order to produce normal fluent speech,

02:19:47.300 | we're not even conscious of what is going on in our mouths,

02:19:51.880 | in our larynx.

02:19:53.060 | We're not conscious, and if we were,

02:19:54.380 | we'd not be able to speak because it's too complex.

02:19:57.240 | It's too precise.

02:19:58.800 | It's something that we have really developed

02:20:02.780 | the abilities to do, and we do it naturally, right?

02:20:04.900 | It's part of our programming

02:20:06.060 | and part of what we learn inherently.

02:20:08.300 | And it's just through exposure.

02:20:11.380 | So stuttering is essentially a breakdown at certain times

02:20:16.380 | in that machinery being able to work

02:20:21.220 | in a really coordinated way.

02:20:22.800 | You can think about the operations of these areas

02:20:26.580 | that are controlling the vocal tract.

02:20:28.460 | Let's say speech is like a symphony.

02:20:30.060 | In order for it to come out normally,

02:20:31.740 | you've got to have not just one part, the larynx,

02:20:35.740 | but the lips, the jaw.

02:20:37.000 | They can't be doing their own thing.

02:20:38.980 | They have to be very, very precisely activated

02:20:43.940 | and very, very precisely controlled in a way

02:20:45.860 | to actually create words.

02:20:48.020 | And so in stuttering,

02:20:49.140 | there's a breakdown of that coordination.

02:20:51.620 | - If somebody has a stutter,

02:20:53.100 | is it better to address that early in life

02:20:55.880 | when there's still neuroplasticity that is very robust?

02:20:59.720 | And if so, what's the typical route for treatment?

02:21:02.340 | I have to imagine it's not brain surgery typically.

02:21:05.780 | I'm guessing there are speech therapists

02:21:07.280 | that people can talk to and they can help them work out

02:21:12.280 | where they're getting stuck

02:21:14.100 | in the relationship to anxiety.

02:21:15.840 | - Yeah, exactly.

02:21:16.680 | I mean, part of it is about that anxiety,

02:21:19.040 | but a lot of it really has to do with therapy

02:21:23.880 | to sort of like work through and think of tricks basically

02:21:28.140 | sometimes to create conditions

02:21:30.500 | where you can actually get the words to come out.

02:21:32.280 | A lot of some forms of stuttering

02:21:34.220 | are really initiation problems.

02:21:36.620 | Just getting started itself is very hard.

02:21:39.340 | You want to start with initial vowel or consonant,

02:21:43.260 | but it won't emit.

02:21:44.760 | And so a lot of that therapy is really just focusing on

02:21:49.580 | like how do you create the conditions for that to happen?

02:21:54.580 | There's another aspect to it that I find very interesting

02:21:57.540 | is that the feedback, essentially,

02:22:00.360 | what we hear ourselves say, for example,

02:22:04.700 | and every time that I say a word,

02:22:08.060 | I'm also hearing what I'm saying,

02:22:09.460 | so that's what we call auditory feedback.

02:22:11.980 | That turns out to be very important,

02:22:13.700 | and sometimes when you change that,

02:22:15.860 | it can actually change the amount someone stutters

02:22:18.220 | for better or for worse.

02:22:19.900 | And it's giving us a clue that the brain

02:22:23.820 | is not just focused on sending the commands out,

02:22:27.300 | but it's also possibly interacting with the part

02:22:30.180 | that is hearing the sounds,

02:22:32.080 | and there's something might be going on in that connection

02:22:34.620 | that breaks down when stuttering occurs.

02:22:37.660 | So there are individuals that are stutterers,

02:22:41.140 | but they don't stutter all the time.

02:22:43.200 | In those instances, there's something happening

02:22:45.320 | in those particular moments where this very,

02:22:49.120 | very precise coordination needs to happen in the brain

02:22:53.140 | in order to get the words out fluently.

02:22:54.980 | - We talked a little bit about caffeine

02:22:56.540 | and why you avoid it.

02:22:57.680 | Because your work requires such precision and calm

02:23:01.760 | and frankly, to me, it seems like you're running

02:23:04.640 | a lot of operations, and no pun intended,

02:23:06.880 | in parallel when you're doing surgery,

02:23:08.980 | not just thinking about where to direct the instruments,

02:23:11.100 | but also thinking like a chess player

02:23:13.460 | several steps down the line, what could happen,

02:23:15.660 | what if, if then type thinking.

02:23:18.060 | What are some of the other practices and tools

02:23:20.520 | that you use to put yourself into state

02:23:23.360 | for optimal neurosurgery or for, you know,

02:23:27.440 | thinking about scientific problems for that matter?

02:23:29.620 | We keep threatening to go running together,

02:23:31.340 | but I know you run, correct?

02:23:33.840 | - Yeah.

02:23:35.040 | - Do you find running to be an essential part

02:23:37.460 | of your state regulation?

02:23:40.320 | - Absolutely.

02:23:41.640 | Yeah, so for me, most exercise that I do,

02:23:45.760 | I really don't do for physical reasons.

02:23:48.120 | I do it for mental reasons.

02:23:50.780 | I can tell, for example, if I don't go on a run or a swim

02:23:55.780 | just after a day or two,

02:23:58.140 | and it can have translation, for example,

02:24:00.320 | in the way I feel in the operating room

02:24:02.360 | or even the way I interact with other people.

02:24:04.760 | So there's no question that those, you know,

02:24:07.900 | the mind and body are deeply connected.

02:24:09.820 | And for me personally,

02:24:11.720 | being able to have opportunity to disconnect for a while,

02:24:15.760 | it turns out to be really, really important.

02:24:18.360 | Now, the operating room for me is another space,

02:24:22.960 | kind of like running or swimming,

02:24:26.400 | where I'm disconnected from the rest of the world.

02:24:30.160 | I don't bring my cellphone into the operating room.

02:24:33.120 | I'm disconnected from the external world

02:24:37.080 | for that time that I'm in the surgery.

02:24:39.280 | And all I am doing is just focusing.

02:24:41.860 | Now, that doesn't mean that I'm having complex thoughts

02:24:44.920 | or doing something very complicated.

02:24:47.280 | Sometimes it is like that, but it's not always like that.

02:24:50.780 | There are things that we do in surgery

02:24:53.480 | that are like routine and rote and are from muscle memory.

02:24:59.000 | So for example, suturing skin

02:25:01.120 | or doing certain kinds of dissection

02:25:04.180 | or drilling part of the bone, for example,

02:25:06.060 | these are all things that become very rote after a time.

02:25:09.240 | So for me, even being in the operating room

02:25:12.200 | actually can sometimes fulfill that purpose.

02:25:14.240 | So I really look forward to being in the operating room

02:25:18.560 | because that intense focus allows me to sort of disconnect

02:25:23.560 | from all the other things that I'm worrying about,

02:25:26.200 | you know, that are happening on the outside world.

02:25:30.100 | You know, we all have those kinds of things that happen

02:25:32.280 | and I'm certainly no exception to that.

02:25:35.360 | But strangely, the operating room for me is a sanctuary.

02:25:40.000 | I love being there because we have some control

02:25:44.880 | over the environment.

02:25:45.940 | I know what is there.

02:25:51.120 | I know the anatomy of the brain.

02:25:54.160 | My motions are going through routines.

02:25:57.560 | And so for me, that's not actually very different

02:26:00.520 | than going on a run and letting my, you know,

02:26:03.200 | legs move in specific ways.

02:26:05.840 | It's just the same thing for my hands.

02:26:08.480 | - Do you listen to music or audio books when you run

02:26:11.260 | or are you divorced from technology when you run?

02:26:14.560 | - Well, music helps me like just stay motivated

02:26:17.920 | and distracted from being out of breath and other things.

02:26:21.560 | And for me, it's a way just to catch up with like the world.

02:26:25.400 | So sometimes I do, but I do notice

02:26:28.080 | that like I don't run as well, for example.

02:26:30.640 | In the operating room, it's a little different.

02:26:32.660 | You know, different surgeons have preferences.

02:26:35.500 | I'm more of the camp where I don't like

02:26:39.520 | any distraction whatsoever.

02:26:41.220 | I'd like people to be able to hear the words

02:26:43.600 | that I'm saying without having background noise.

02:26:46.440 | I don't really think about relying on music

02:26:51.320 | or other things to try to put me in a state of mind.

02:26:54.320 | You know, I think just being there alone

02:26:57.160 | and just, you know, trying to treat it the way it is.

02:27:00.280 | It's a sacred moment where someone's life

02:27:03.280 | is really directly under your hands.

02:27:06.200 | That enough kind of focuses me very quickly.

02:27:09.240 | And I like that.

02:27:11.200 | It really detaches me from a lot of things

02:27:13.480 | that are preoccupying me.

02:27:15.320 | And for those couple of hours that we have a surgery,

02:27:18.880 | we're just focused on one thing only.

02:27:21.220 | - That's fantastic.

02:27:22.360 | Again, I think of in the range of brain explorers

02:27:24.940 | that the neurosurgeons, those of your profession,

02:27:28.880 | are to me like the astronauts of neuroscience

02:27:31.900 | because they're really going

02:27:33.080 | to the farthest reaches possible.

02:27:35.300 | And they're testing and probing

02:27:36.720 | and really at the front edge of discovering

02:27:39.760 | from the species that we arguably care about the most,

02:27:42.940 | which is humans.

02:27:45.180 | Eddie, I have to say from the first time we became friends,

02:27:50.960 | 38 years ago.

02:27:52.560 | - Something like that. - Something like that.

02:27:55.320 | I'm almost reluctant to say,

02:27:56.660 | but so I only reveal it in part

02:27:59.180 | that Eddie and I became friends

02:28:00.960 | because both he and I shared a love of birds

02:28:05.000 | and we had a club at our school

02:28:07.920 | of which there were only two members, Eddie and I.

02:28:11.060 | - Small club. - Small club.

02:28:12.760 | There was one honorary member

02:28:14.240 | and there were certain requirements for being in this club

02:28:16.360 | that we won't reveal.

02:28:17.320 | We took a pact of secrecy

02:28:19.040 | and we're going to obey that pact of secrecy.

02:28:21.920 | But to be sitting here with you today

02:28:23.720 | for me is a absolute thrill.

02:28:27.540 | Not just because we've been friends for that long

02:28:30.160 | or that we got reacquainted

02:28:31.500 | through literally the halls of medicine and science,

02:28:34.020 | but because I really do see what you're doing

02:28:37.280 | as really representing that front absolute cutting edge

02:28:42.280 | of exploration and application.

02:28:45.000 | I mean, the story of Poncho is but one of your many patients

02:28:48.300 | that has derived tremendous benefit from your work.

02:28:51.680 | And now as a chair of a department,

02:28:53.660 | you of course work alongside individuals

02:28:58.360 | who are also doing incredible work

02:28:59.560 | in the spinal cord, et cetera.

02:29:01.720 | So now on behalf of myself and everyone listening,

02:29:04.360 | I just really want to thank you

02:29:05.480 | for joining us today to share this information.

02:29:08.600 | We will certainly have you back

02:29:09.920 | because there's an entire list of other questions

02:29:12.520 | we didn't have time to get to,

02:29:14.080 | but also just for the work you do.

02:29:15.680 | It's truly spectacular.

02:29:18.000 | - Andrew, thanks so much.

02:29:19.780 | You know, I'm very humbled basically by what you just said.

02:29:23.820 | And I feel that it's really an extraordinary honor,

02:29:28.000 | actually I'm privileged to be here with you

02:29:30.260 | to reconnect and talk about all these ideas.

02:29:33.740 | It's probably not random, you know,

02:29:35.760 | that we ended up in similar spots and interests.

02:29:39.480 | I think when we were kids, you know,

02:29:41.000 | it starts with some deep interests

02:29:42.660 | and kind of nerding out on topics.

02:29:47.340 | And it's probably not a coincidence, you know,

02:29:50.760 | that we have such deep interest in this work now.

02:29:54.800 | I just feel really lucky to be able to do what I do.

02:29:57.320 | It's fun every day, almost every day,

02:30:00.920 | be able to go to work and take care of folks

02:30:03.420 | and learn at the same time and then just close the loop.

02:30:06.500 | You know, how do we apply the knowledge

02:30:08.200 | that we learned one day to someone who comes in next week?

02:30:11.020 | It's really fun.

02:30:12.560 | And we don't know everything, we're not even close to it,

02:30:17.140 | but the journey to figure this out is really extraordinary.

02:30:22.560 | I mean, it's like you said, it's exploring new lands,

02:30:29.680 | literally in the operating room when

02:30:31.640 | I'm looking at the exposed cortex trying to understand,

02:30:35.600 | is it safe to walk down this part of the cortical landscape

02:30:40.600 | or this other trail?

02:30:42.520 | You know, which one is going to be

02:30:43.920 | the one that is going to be safe versus the other that

02:30:46.800 | results in paralysis and inability to talk?

02:30:49.360 | Well, maybe I shouldn't call it fun,

02:30:52.160 | but it's very important too,

02:30:56.000 | in addition to being really intellectually important

02:30:58.380 | for how we understand how the brain works.

02:31:00.680 | And so, yeah, I feel just really lucky

02:31:02.400 | to be in that opportunity.

02:31:03.760 | - And we're lucky to have you being

02:31:05.660 | one of the people doing it, so thank you ever so much.

02:31:08.380 | - Thanks.

02:31:09.380 | - Thank you for joining me today

02:31:10.500 | for my discussion with Dr. Eddie Chang.

02:31:12.840 | If you'd like to learn more about his research

02:31:14.860 | into the neuroscience of speech and language

02:31:17.120 | and bioengineering, his treatment of epilepsy

02:31:20.320 | and other aspects and diseases and disorders of the brain,

02:31:23.660 | please check out the links in our show note captions.

02:31:25.960 | We have links to his laboratory website,

02:31:28.080 | his clinical website and other resources

02:31:30.740 | related to his critical research as well.

02:31:33.200 | If you're learning from and are enjoying

02:31:34.600 | the Huberman Lab Podcast,

02:31:36.000 | please subscribe to our YouTube channel.

02:31:37.720 | That's a terrific zero cost way to support us.

02:31:40.280 | In addition, please subscribe to the Huberman Lab Podcast

02:31:43.120 | on Spotify and Apple.

02:31:44.760 | And on both Spotify and Apple,

02:31:46.220 | you also have the opportunity

02:31:47.700 | to leave us up to a five-star review.

02:31:49.760 | If you have questions for us or comments

02:31:51.280 | about the information we've covered

02:31:52.860 | or suggestions about future guests,

02:31:54.900 | please put those in the comment section on YouTube.

02:31:57.300 | We do read all the comments.

02:31:59.120 | Please also check out the sponsors mentioned

02:32:00.700 | at the beginning of today's episode.

02:32:02.340 | That's the best way to support the Huberman Lab Podcast.

02:32:05.200 | Not so much today, but in many previous episodes

02:32:07.360 | of the Huberman Lab Podcast, we talk about supplements.

02:32:09.760 | While supplements aren't necessary for everybody,

02:32:11.960 | many people derive tremendous benefit from them

02:32:13.980 | for things like enhancing sleep and focus

02:32:15.960 | and hormone optimization.

02:32:17.640 | The Huberman Lab Podcast

02:32:18.660 | has partnered with Momentous Supplements.

02:32:20.280 | If you'd like to see the supplements

02:32:21.120 | that the Huberman Lab Podcast

02:32:22.440 | has partnered with Momentous on,

02:32:23.900 | you can go to Live Momentous, spelled O-U-S,

02:32:26.280 | so livemomentous.com/huberman.

02:32:28.800 | And there you'll see a number of the supplements

02:32:30.720 | that we talk about regularly on the podcast.

02:32:32.680 | I should just mention that that catalog of supplements

02:32:34.640 | is constantly being updated.

02:32:36.240 | As mentioned at the beginning of today's episode,

02:32:38.040 | the Huberman Lab Podcast has now launched a premium channel.

02:32:41.100 | That premium channel will feature monthly AMAs

02:32:43.640 | or ask me anythings where I answer your questions in depth,

02:32:46.520 | as well as other premium resources.

02:32:48.480 | If you'd like to subscribe to the premium channel,

02:32:50.320 | you can simply go to HubermanLab.com/premium.

02:32:53.920 | I should mention that the proceeds from the premium channel

02:32:56.520 | go to support the standard Huberman Lab Podcast,

02:32:58.720 | which will continue to be released every Monday per usual,

02:33:02.040 | as well as supporting various research projects

02:33:04.920 | done on humans to create the sorts of tools

02:33:07.300 | for mental health, physical health, and performance

02:33:09.120 | that you hear about on the Huberman Lab Podcast.

02:33:11.420 | Again, it's HubermanLab.com/premium to subscribe.

02:33:14.520 | It's $10 a month or $100 per year.

02:33:17.600 | If you haven't already subscribed

02:33:18.820 | to our zero cost newsletter,

02:33:20.500 | we have what is called the Neural Network Newsletter.

02:33:23.100 | You can subscribe by going to HubermanLab.com,

02:33:25.540 | go to the menu and click on newsletter.

02:33:27.380 | Those newsletters include summaries of podcast episodes,

02:33:30.140 | lists of tools from the Huberman Lab Podcast.

02:33:32.680 | And if you'd like to see previous newsletters

02:33:34.920 | we've released, you can also just go to HubermanLab.com,

02:33:37.160 | click on newsletter in the menu,

02:33:38.520 | and you'll see various downloadable PDFs.

02:33:40.760 | If you want to sign up for the newsletter,

02:33:42.440 | we just ask for your email.

02:33:43.560 | We do not share your email with anybody.

02:33:45.480 | And again, it's completely zero cost.

02:33:47.740 | If you're not already following me on social media,

02:33:49.720 | it's Huberman Lab on Twitter, on Facebook, and on Instagram.

02:33:53.200 | And at all three of those places,

02:33:54.960 | I cover topics and subject matter

02:33:57.400 | that are sometimes overlapping with the information

02:33:59.800 | covered on the Huberman Lab Podcast,

02:34:01.240 | but that's often distinct from information

02:34:03.280 | on the Huberman Lab Podcast.

02:34:04.720 | Again, it's Huberman Lab on all social media channels.

02:34:07.520 | Thank you once again for joining me today

02:34:09.560 | for the discussion about the neuroscience of speech,

02:34:12.080 | language, epilepsy, and much more with Dr. Eddie Chang.

02:34:16.040 | And as always, thank you for your interest in science.

02:34:19.100 | [upbeat music]

02:34:21.680 | you

Dr. Eddie Chang: The Science of Learning & Speaking Languages | Huberman Lab Podcast #95

Chapters