AI transcript
0:00:02 (dramatic music)
0:00:05 – Hey there, it’s Stephen Dubner.
0:00:08 Today, a holiday treat, a bonus episode
0:00:10 from people I mostly admire.
0:00:12 One of the other shows we make here
0:00:13 at the Freakonomics Radio Network.
0:00:16 It is an interview show hosted by Steve Levitt,
0:00:18 my Freakonomics friend and co-author,
0:00:21 who is an economics professor emeritus now
0:00:23 at the University of Chicago.
0:00:26 On this episode, Levitt interviews David Eagleman,
0:00:28 a neuroscientist, entrepreneur,
0:00:31 and author of several books, including LiveWired,
0:00:34 the inside story of the ever-changing brain.
0:00:36 It is a fascinating conversation.
0:00:37 You are going to love it.
0:00:40 To hear more conversations like this,
0:00:44 follow people I mostly admire in your podcast app.
0:00:45 Okay, that’s it for me.
0:00:47 Here is Steve Levitt.
0:00:49 (dramatic music)
0:01:01 – I love podcast guests who change the way
0:01:04 I think about some important aspect of the world.
0:01:07 A great example is my guest today, David Eagleman.
0:01:09 He’s a Stanford neuroscientist whose work
0:01:13 on brain plasticity has completely transformed
0:01:18 my understanding of the human brain and its possibilities.
0:01:20 The human brain is about three pounds.
0:01:22 It’s locked in silence and darkness.
0:01:25 It has no idea where the information is coming from
0:01:28 because everything is just electrical spikes
0:01:30 and also chemical releases as a result of those spikes.
0:01:34 And so what you have in there is this giant symphony
0:01:36 of electrical activity going on
0:01:40 and its job is to create a model of the outside world.
0:01:47 – Welcome to people I mostly admire with Steve Levitt.
0:01:50 – According to Eagleman,
0:01:52 the brain is constantly trying
0:01:54 to predict the world around it.
0:01:56 But of course, the world is unpredictable and surprising.
0:02:00 So the brain is constantly updating its model.
0:02:02 The capacity of our brains to be ever-changing
0:02:05 is usually referred to as plasticity,
0:02:08 but Eagleman offers another term, live wired.
0:02:10 That’s where conversation begins.
0:02:18 Plasticity is the term used in the field
0:02:21 because the great neuroscientist
0:02:24 or psychologist actually, William James,
0:02:25 coined the term because he was impressed
0:02:28 with the way that plastic gets manufactured,
0:02:31 where you mold it into a shape and it holds onto that shape.
0:02:34 And he thought that’s kind of like what the brain does.
0:02:37 The great trick that mother nature figured out
0:02:39 was to drop us into the world half-baked.
0:02:42 If you look at the way an alligator drops into the world,
0:02:44 it essentially is pre-programmed.
0:02:47 It eats, mates, sleeps, does whatever it’s doing.
0:02:49 But we spent our first several years
0:02:52 absorbing the world around us based on our neighborhood
0:02:54 and our moment in time and our culture
0:02:56 and our friends and our universities.
0:03:00 We absorb all of that such that we can then springboard
0:03:03 off of that and create our own things.
0:03:05 There are many things that are essentially
0:03:09 pre-programmed in us, but we are incredibly flexible
0:03:11 and that is the key about live wiring.
0:03:13 When I ask you to think of the name
0:03:14 of your fifth grade teacher,
0:03:16 you might be able to pull it up,
0:03:18 even though it’s been years since you saw
0:03:19 that fifth grade teacher,
0:03:23 but somehow there was a change made in your brain
0:03:25 and that stayed in place.
0:03:27 You’ve got 86 billion neurons.
0:03:30 Each neuron is as complicated as a city.
0:03:33 This entire forest of neurons,
0:03:35 every moment of your life is changing.
0:03:38 It’s reconfiguring, it’s strengthening connections
0:03:41 here and there, it’s actually unplugging over here
0:03:42 and replugging over there.
0:03:45 And so that’s why I’ve started to feel
0:03:48 that the term plasticity is maybe underreporting
0:03:49 what’s going on.
0:03:51 And so that’s why I made up the term live wiring.
0:03:53 When I went to school, I feel like they taught me
0:03:56 the brain was organized around things like senses
0:03:58 and emotions, that there were these different parts
0:04:01 of the brain that were good for those things.
0:04:04 But you make the case that there’s a very different
0:04:06 organization of the brain.
0:04:08 – It is organized around the senses,
0:04:10 but the interesting thing is that the cortex,
0:04:13 this wrinkly outer bit, is actually a one-trick pony.
0:04:14 It doesn’t matter what you plug in.
0:04:17 It’ll say, okay, got it, I’ll just wrap myself
0:04:21 around that data and figure out what to do with that data.
0:04:25 It turns out that in almost everybody you have functioning
0:04:27 eyeballs that plug into the back of the head.
0:04:29 And so we end up calling the back part of the brain
0:04:30 the visual cortex.
0:04:32 We call this part the auditory cortex.
0:04:35 And this is the somatosensory cortex that takes
0:04:37 in information from the body and so on.
0:04:40 So what you learned back in high school or college
0:04:42 is correct most of the time.
0:04:45 But what it overlooks is the fact that the brain
0:04:46 is so flexible.
0:04:49 If a person goes blind or is born blind,
0:04:51 that part of the brain that we’re calling
0:04:53 the visual cortex, that gets taken over by hearing,
0:04:54 by touch, by other things.
0:04:56 And so it’s no longer visual cortex.
0:04:59 The same neurons that are there are now doing
0:05:01 a totally different job.
0:05:04 – So let me pose a question to listeners.
0:05:08 Imagine you have a newborn baby and he or she looks
0:05:10 absolutely flawless on the outside.
0:05:14 But then upon examination, the doctors discover
0:05:17 that half of his or her brain is just missing.
0:05:20 A complete hemisphere of the brain, it’s never developed.
0:05:22 It’s just empty space.
0:05:24 I would expect that would be a fatal defect
0:05:28 or best the child would be growing up profoundly
0:05:30 mentally disabled.
0:05:32 – Turns out the kid will be just fine.
0:05:34 You can be born without half the brain
0:05:36 or you can do what’s called a hemispherectomy,
0:05:39 which happens to children who have something called
0:05:42 rasmusans encephalitis, which is a form of epilepsy
0:05:45 that spreads from one hemisphere to the other.
0:05:47 The surgical intervention for that is to remove
0:05:49 half the brain.
0:05:51 You can just imagine as a parent, the horror you would feel
0:05:53 if your child had to go in for something like that.
0:05:55 But you know what, kid’s just fine.
0:05:59 I can’t take my laptop and rip out half the motherboard
0:06:01 and expect it to still function.
0:06:04 But with the brain, with a live wired system, it’ll work.
0:06:07 – So I first came to work because I was so blown away
0:06:10 by the idea of human echolocation.
0:06:13 Only to discover that echolocation is only
0:06:14 the tip of the iceberg.
0:06:17 But could you talk just a bit about echolocation,
0:06:19 how quickly with training it can start
0:06:21 to substitute for sight?
0:06:23 – So it turns out that blind people can make
0:06:27 all kinds of sounds either with their mouth like clicking
0:06:29 or the tip of their cane or snapping their fingers,
0:06:30 anything like this.
0:06:33 And they can get really good at determining
0:06:36 what is coming back as echoes and figure out,
0:06:38 oh, okay, this is an open space in front of me.
0:06:39 Here, there’s something in front of me.
0:06:41 It’s probably a parked car.
0:06:43 And oh, there’s a little gap between two parked cars here.
0:06:44 So I can go in here.
0:06:48 The key is the visual part of the brain is no longer
0:06:50 being used because for whatever reason,
0:06:52 there’s no information coming down those pipelines anymore.
0:06:55 So that part of the brain is taken over by audition,
0:06:58 by hearing and by touch and other things.
0:07:01 What happens is that the blind person becomes really good
0:07:04 at these other things because they’ve just devoted
0:07:06 more real estate to it.
0:07:09 And as a result, they can pick up on all kinds of cues
0:07:11 that would be very difficult for me and you
0:07:14 because our hearing just isn’t that good.
0:07:16 – And then in these studies,
0:07:20 you put a blindfold on a person for two or three days
0:07:22 and you try to teach them echolocation.
0:07:25 If I understand correctly, even over that timescale,
0:07:28 the echolocation starts taking over
0:07:29 the visual part of the brain.
0:07:31 Is that a fair assessment?
0:07:32 – That is exactly right.
0:07:34 This was my colleagues at Harvard.
0:07:36 They did this over the course of five days.
0:07:39 They demonstrated that people could get really good at,
0:07:41 they’re actually a number of studies like this.
0:07:42 They can get really good at reading Braille.
0:07:44 They can do things like echolocation.
0:07:48 And the speed of it was sort of the surprise.
0:07:50 But the real surprise for me came along
0:07:52 when they blindfolded people tightly
0:07:54 and put them in the brain scanner
0:07:59 and they were making sounds or touching the hand.
0:08:01 And they were starting to see activity
0:08:06 in the visual cortex after 60 minutes of being blind.
0:08:09 – So in your book, you talk about REM sleep.
0:08:11 And honestly, if I had sat down
0:08:14 and tried to come up with an explanation of REM sleep,
0:08:17 I could have listened to a thousand ideas.
0:08:20 Your pet theory would not be one of them.
0:08:22 So explain what REM sleep is
0:08:24 and then tell me why you think we do it.
0:08:26 – REM sleep is rapid eye movement sleep.
0:08:28 We have this every night, about every 90 minutes.
0:08:30 And that’s when you dream.
0:08:31 So if you wake someone up
0:08:33 when their eyes are moving rapidly
0:08:34 and you say, “Hey, what are you thinking about?”
0:08:37 they’ll say, “Whoa, I was just riding a camel across a meadow.”
0:08:39 But if you wake them up at other parts of their sleep,
0:08:42 they typically won’t have anything going on.
0:08:44 So that’s how we know we dream during REM sleep.
0:08:45 But here’s the key.
0:08:48 My student and I realized that at nighttime,
0:08:50 when the planet rotates,
0:08:52 we spend half our time in darkness.
0:08:55 And obviously we’re very used to this electricity blessed world.
0:08:57 But think about this in historical time
0:08:59 over the course of hundreds of millions of years.
0:09:00 It’s really dark.
0:09:03 I mean, half the time you are in blackness.
0:09:05 Now you can still hear and touch and taste
0:09:07 and smell in the dark.
0:09:10 But the visual system is at a disadvantage
0:09:12 whenever the planet rotates into darkness.
0:09:16 And so given the rapidity with which other systems
0:09:18 can encroach on that,
0:09:21 what we realized is it needs a way of defending itself
0:09:23 against takeover every single night.
0:09:25 And that’s what dreams are about.
0:09:28 So what happens is you have these midbrain mechanisms
0:09:31 that simply blast random activity
0:09:34 into the visual cortex every 90 minutes during the night.
0:09:37 And when you get activity in the visual cortex,
0:09:39 you say, “Oh, I’m seeing things.”
0:09:41 And because the brain is a storyteller,
0:09:43 you can’t activate all the stuff
0:09:47 without feeling like there’s a whole story going on there.
0:09:48 But the fascinating thing is when you look
0:09:51 at the circuitry carefully, it’s super specific,
0:09:53 much more specific than almost anything else in the brain.
0:09:57 It’s only hitting the primary visual cortex and nothing else.
0:10:02 And so that led us to a completely new theory about dreams.
0:10:05 We studied 25 different species of primates
0:10:07 and we looked at the amount of REM sleep
0:10:08 they have every night.
0:10:12 And we also looked at how plastic they are as a species.
0:10:14 It turns out that the amount of dream sleep
0:10:18 that a creature has exactly correlates
0:10:20 with how plastic they are, which is to say,
0:10:23 if your visual system is in danger of getting taken over
0:10:25 because your brain is very flexible,
0:10:27 then you have to have more dream sleep.
0:10:29 And by the way, when you look at human infants,
0:10:32 they have tons of dream sleep at the beginning
0:10:34 when their brains are very plastic.
0:10:37 And as they age, the amount of dream sleep goes down.
0:10:40 – Have you convinced the sleep scientists this is true
0:10:43 or is this just you believing it right now?
0:10:44 – At the moment, there are 19 papers
0:10:46 that have cited this and discussed this.
0:10:48 And I think it’s right.
0:10:49 I mean, look, everything can be wrong.
0:10:50 Everything is provisional,
0:10:54 but it’s the single theory that is quantitative.
0:10:56 It’s the single theory about dreams
0:11:00 that says not only here is a idea for why we dream,
0:11:01 but we can compare across species
0:11:04 and the predictions match exactly.
0:11:07 No one would have suspected that you’d see a relationship
0:11:11 between how long it takes you to walk or reach adolescence
0:11:12 and how much dream sleep you have.
0:11:14 But it turns out that is spun on.
0:11:23 – So we talked about echolocation,
0:11:26 which uses sound to accomplish tests
0:11:29 that are usually done by vision.
0:11:31 And you’ve started a company called NeoSensory,
0:11:35 which uses touch to accomplish tasks
0:11:37 that are usually done with hearing.
0:11:38 Can you explain the science behind that?
0:11:41 Given that all the data running around in the brain
0:11:44 is just data and the brain doesn’t know where it came from.
0:11:47 All it knows is, oh, here are electrical spikes
0:11:48 and it tries to figure out what to do with it.
0:11:51 I got really interested in this idea of sensory substitution,
0:11:53 which is can you push information into the brain
0:11:56 via an unusual channel?
0:11:58 Originally we built a vest
0:12:00 that was covered with vibratory motors
0:12:03 and we captured sound for people who are deaf.
0:12:05 So the vest captures sound,
0:12:07 breaks it up from high to low frequency
0:12:10 and you’re feeling the sound on your torso.
0:12:12 By the way, this is exactly what the inner ear does.
0:12:14 It breaks up sound from high to low frequency
0:12:16 and ships that off to the brain.
0:12:18 So we’re just transferring the inner ear
0:12:20 to the skin of the torso.
0:12:22 And it worked, people who are deaf
0:12:25 could come to hear the world that way.
0:12:28 So I spun this out of my lab as a company, Neosensory,
0:12:32 and we shrunk the vest down to a wristband
0:12:34 and we’re on wrist of deaf people all over the world.
0:12:37 The other alternative for somebody who’s deaf
0:12:39 is a cochlear implant, an invasive surgery.
0:12:44 This is much cheaper and does as good a job.
0:12:45 Just to make sure I understand it.
0:12:50 Sounds happen and this wristband hears the sounds
0:12:54 and then shoots electrical impulses into your wrist
0:12:56 that correspond to the high and low frequency.
0:12:58 It’s actually just vibratory motors.
0:13:00 So it’s just like the buzzer on your cell phone
0:13:03 but we have a string of these buzzers all along your wrist
0:13:06 and we’re actually taking advantage of an illusion
0:13:09 which is if I have two motors next to each other
0:13:11 and I stimulate them both,
0:13:14 you will feel one virtual point right in between.
0:13:16 And as I change the strength of those two motors
0:13:20 relative to each other, I can move that point around.
0:13:23 So we’re actually stimulating 128 virtual points
0:13:24 along the wrist.
0:13:27 – Do people train you give them very direct feedback
0:13:29 or is it more organic?
0:13:30 – Great question.
0:13:32 It started off where we were doing a lot of training
0:13:35 on people and what we realized is it’s all the same
0:13:36 if we just let it be organic.
0:13:39 The key is we just encourage people be in the world
0:13:40 and that’s it.
0:13:42 You see the dog’s mouth moving
0:13:45 and you feel the barking on your wrist
0:13:47 or you close the door and you feel that on your wrist
0:13:49 or you say something, you know,
0:13:50 most deaf people can speak
0:13:53 and they know what their motor output is
0:13:55 and they’re feeling the input.
0:13:58 – Okay, so hearing their own voice for the first time
0:13:59 through this. – Exactly.
0:14:01 – Oh God, yeah, that’s interesting.
0:14:01 – And by the way,
0:14:03 that’s how you learned how to use your ears too.
0:14:05 You know, when you’re a baby,
0:14:06 you’re watching your mother’s mouth move
0:14:08 and you’re hearing data coming in your ears
0:14:10 and you clapped your hands together
0:14:12 and you hear something in your ears.
0:14:13 It’s the same idea.
0:14:15 You’re just training up correlations in the brain
0:14:18 about, oh, this visual thing seems to always go
0:14:20 with that auditory stimulus.
0:14:23 – So then it seems like if I’m deaf
0:14:25 and I see the dog’s mouth moving
0:14:27 and I now associate that with the sound,
0:14:31 do the people say that they hear the sound where the dog is
0:14:33 or is the sound coming from the wrist?
0:14:34 – For the first few months,
0:14:36 you’re hearing it on your wrist,
0:14:38 you can get pretty good at these correlations
0:14:40 but then after about six months,
0:14:43 if I ask somebody when the dog barks,
0:14:44 do you feel something on your wrist?
0:14:45 And you think, okay, what was that on?
0:14:46 That must have been a dog bark
0:14:47 and then you look for the dog.
0:14:51 And they say, no, I just hear the dog out there.
0:14:52 – Hmm.
0:14:53 – And that sounds so crazy.
0:14:55 But remember, that’s what your ears are doing.
0:14:58 Your ears are capturing vibrations at the eardrum
0:15:00 that moves through the middle ear to the inner ear,
0:15:01 breaks up to different frequencies,
0:15:04 goes off to your brain, goes to your auditory cord.
0:15:06 It’s this giant pathway of things.
0:15:09 And yet, even though you’re hearing my voice right now
0:15:13 inside your head, you think I’m somewhere else.
0:15:14 And that’s exactly what happens,
0:15:18 irrespective of how you feed the data in.
0:15:22 So you also have a product that helps with tinnitus.
0:15:23 Could you explain both what that is
0:15:25 and how your product helps?
0:15:27 So tinnitus is a ringing in the ears.
0:15:31 It’s like beep and about 15% of the population has this.
0:15:33 And for some people, it’s really, really bad.
0:15:38 It turns out there is a mechanism for helping with tinnitus
0:15:41 which has to do with playing tones
0:15:45 and then matching that with stimulation on the skin.
0:15:47 People wear the wristband, it’s exactly the same wristband,
0:15:51 but we have the phone play tones, boop, boop, boop, boop, boop.
0:15:53 And you’re feeling that all over your wrist.
0:15:54 And you just do that for 10 minutes a day.
0:15:56 And it drives down the tinnitus.
0:15:58 Now, why does that work?
0:16:00 There are various theories on this,
0:16:02 but I think the simplest version
0:16:05 is that your brain is figuring out,
0:16:09 okay, real sounds always cause this
0:16:11 correlating vibration on my wrist,
0:16:15 but a fake sound, beep, you know, this thing in my head,
0:16:17 that doesn’t have any verification on the wrist.
0:16:20 And so that must not be a real sound.
0:16:23 So because of issues of brain plasticity,
0:16:26 the brain just reduces the strength of the tinnitus
0:16:28 because it learns that it’s not getting any confirmation
0:16:30 that that’s a real world sound.
0:16:32 Now, how did you figure out
0:16:34 that this bracelet could be used for this?
0:16:36 This was discovered by a woman named Susan Shore,
0:16:39 who’s a researcher who discovered this about a decade ago.
0:16:42 She was using electrical shocks on the tongue.
0:16:43 And there’s actually another company
0:16:44 that’s spun out called Lanier
0:16:46 that does this with sounds in the ear
0:16:47 and shocks on the tongue.
0:16:49 They had an argument that they think
0:16:52 it had to be touched from the head and the neck.
0:16:53 And I didn’t buy that at all.
0:16:54 And that’s why I tried that with the wristband.
0:16:57 So this was not an original idea for us
0:17:01 except to try this on the wrist and it works equally as well.
0:17:06 So what we’re talking about is substituting between senses.
0:17:08 Are there other forms of this products
0:17:10 that are currently available to consumers
0:17:13 or likely to become available soon in the space?
0:17:16 For people who are blind, for example,
0:17:19 there are a few different approaches to this.
0:17:21 One is called the brain port and that’s where,
0:17:24 for a blind person, they have a little camera on their glasses
0:17:29 and that gets turned into little electrical stimulation
0:17:30 on the tongue.
0:17:33 So you’re wearing this little electro-tactile grid
0:17:36 on your tongue and it tastes like pop rocks
0:17:37 sort of in your mouth.
0:17:39 Blind people can get pretty good at this.
0:17:42 They can navigate complex obstacle courses
0:17:45 or throw a ball into a basket at a distance
0:17:48 because they can come to see the world through their tongue,
0:17:50 which if that sounds crazy,
0:17:52 it’s the same thing as seeing it through these two spheres
0:17:54 that are embedded in your skull.
0:17:57 It’s just capturing photons and information about them,
0:17:58 figuring out where the edges are
0:17:59 and then shipping that back to the brain.
0:18:01 And the brain can figure that out.
0:18:03 There’s also a colleague of mine
0:18:05 that makes an app called Voice.
0:18:07 It uses the phone’s camera
0:18:10 and it turns that into soundscape.
0:18:13 So if you’re moving the camera around,
0:18:15 you’re hearing (mimics sounds)
0:18:18 you know, it sounds like a strange cacophony,
0:18:19 but it doesn’t take long,
0:18:21 even for you as a sighted person,
0:18:23 to get used to this and say,
0:18:27 oh, okay, I’m turning the visual world into sound
0:18:28 and it’s starting to make sense.
0:18:30 When I pass over an edge
0:18:32 or when I zoom into something,
0:18:34 the pitch changes, the volume changes,
0:18:37 there’s all kinds of changes in the sound quality
0:18:39 that tells you, oh, yeah, now I’m getting close to something,
0:18:40 now I’m getting far,
0:18:43 and here’s what the world looks like in sound.
0:18:48 – Coming up after the break,
0:18:51 there’s really no shortage of theoretical ideas
0:18:52 in neuroscience,
0:18:56 but fundamentally, we don’t have enough data.
0:18:59 More of Steve Leavitt’s conversation with David Eagleman
0:19:02 in this special episode of “People I Mostly Admire.”
0:19:18 Okay, back now to this special episode of “People I Mostly Admire.”
0:19:20 This is my Freakonomics friend and co-author Steve Leavitt
0:19:24 in conversation with the neuroscientist David Eagleman.
0:19:27 (eerie music)
0:19:30 – Elon Musk’s company, Neuralink,
0:19:32 has gotten a ton of attention lately.
0:19:33 Could you explain what they’re trying to do
0:19:37 and whether you think that’s a promising avenue to explore?
0:19:38 – What they’re doing is they’re putting electrodes
0:19:43 into the brain to read from and talk to the neurons there.
0:19:45 So what we’ve been talking about so far
0:19:47 has been sending signals to the brain,
0:19:48 but what Neuralink is trying to do
0:19:51 is take signals out of the brain, is that right?
0:19:51 – That is correct.
0:19:53 Everything we’ve been talking about so far
0:19:54 with sensory substitution,
0:19:57 that’s a way of pushing information in and non-invasive.
0:20:00 And what Neuralink is, you have to drill a hole in the head
0:20:02 to get to the brain itself,
0:20:04 but then you can do reading and writing invasively.
0:20:09 That actually has been going on for 60 years.
0:20:11 The language of the brain is electrical stimulation.
0:20:14 And so with a little tiny wire, essentially,
0:20:17 you can zap a neuron and make it pop off
0:20:21 or you can listen to when it’s chattering along,
0:20:23 going pa-pa-pa-pa-pa-pa-pa-pa.
0:20:25 There’s nothing actually new about what Neuralink is doing,
0:20:28 except that they’re making a one ton robot
0:20:32 that sews the electrodes into the brain
0:20:34 so it can do it smaller and tighter and faster
0:20:36 than a neurosurgeon can.
0:20:38 And by the way, there are a lot of great companies
0:20:41 doing this sort of thing with electrodes.
0:20:44 As people get access to the brain,
0:20:46 we’re finally getting to a point, we’re not there yet,
0:20:48 but we’re getting to a point
0:20:52 where we’ll finally be able to push theory forward.
0:20:55 There’s really no shortage of theoretical ideas
0:20:58 in neuroscience, but fundamentally,
0:21:01 we don’t have enough data because, as I mentioned,
0:21:04 you’ve got these 86 billion neurons all doing their thing,
0:21:09 and we have never measured what all these things
0:21:10 are doing at the same time.
0:21:11 So we have technologies
0:21:15 like functional magnetic resonance imaging, FMRI,
0:21:19 which measures big blobby volumes of,
0:21:21 oh, there was some activity there and some activity there,
0:21:22 but that doesn’t tell us what’s happening
0:21:24 at the level of individual neurons.
0:21:26 We can currently measure some individual neurons,
0:21:28 but not many of them.
0:21:32 Be like if an alien asked one person in New York City,
0:21:33 hey, what’s going on here?
0:21:36 And then tried to extrapolate to understand
0:21:37 the entire economy of New York City
0:21:38 and how that’s all working.
0:21:43 So I think we’re finally getting closer to the point
0:21:45 where we’ll have real data about,
0:21:47 wow, this is what thousands or eventually hundreds
0:21:49 of thousands or millions of neurons
0:21:52 are actually doing in real time at the same moment,
0:21:55 and then we’ll be able to really get progress.
0:21:57 I actually think the future is not in things like Neuralink,
0:22:02 but the next level past that, which is nanorobotics.
0:22:04 This is all theoretical right now,
0:22:07 but I don’t think this is more than 20, 30 years off,
0:22:10 where you do three-dimensional printing,
0:22:14 atomically precise, you make molecular robots,
0:22:15 hundreds of millions of these,
0:22:17 and then you put them in a capsule
0:22:18 and you swallow the capsule,
0:22:19 and these little robots swim around
0:22:23 and they go into your neurons, these cells in your brain.
0:22:27 And from there, they can send out little signals saying,
0:22:29 hey, this neuron just fired.
0:22:30 And once we have that sort of thing,
0:22:32 then we can say non-invasively,
0:22:35 here’s what all these neurons are doing at the same time,
0:22:38 and then we’ll really understand the brain.
0:22:41 – I’ve worn a continuous glucose monitor a few times,
0:22:43 so you stick this thing in your arm
0:22:44 and you leave it there for 10 days,
0:22:46 and every five minutes,
0:22:49 it gives you a reading of your blood glucose level.
0:22:51 It gives you direct feedback
0:22:53 on how your body responds to the foods you eat,
0:22:55 also to stress or lack of sleep,
0:22:57 that you simply don’t get otherwise.
0:23:01 I learned more about my metabolism in 10 days
0:23:04 than I had over the entire rest of my life combined.
0:23:06 What you’re talking about with these nanorobots
0:23:07 is obviously in the future,
0:23:12 but is there anything now that I can buy
0:23:13 and I can strap on my head?
0:23:15 And I know it’s not gonna be individual neurons,
0:23:17 but that would allow me to get feedback
0:23:21 about my brainwaves and be able to learn
0:23:23 in that same way I do with the glucose monitor?
0:23:27 – What we have now is EEG, Electroencephalography,
0:23:30 and there are several really good companies
0:23:34 like MUSE and Emotive that have come out with at-home methods.
0:23:36 You just strap this thing on your head
0:23:39 and you can measure what’s going on with your brainwaves.
0:23:43 The problem is that brainwaves are still pretty distant
0:23:46 from the activity of 86 billion chattering neurons.
0:23:49 An analogy would be if you went
0:23:50 to your favorite baseball stadium
0:23:53 and you attached a few microphones
0:23:55 to the outside of the stadium
0:23:58 and you listened to a baseball game,
0:24:00 but all you could hear with these microphones
0:24:02 is occasionally the crack of the bat
0:24:03 and the roar of the crowd.
0:24:07 And then your job is to reconstruct what baseball is
0:24:10 just some of these few little signals you’re getting.
0:24:14 So I’m afraid it’s still a pretty crude technology.
0:24:17 – I could imagine that I would put one of these EEGs on
0:24:20 and I would just find some feeling I liked,
0:24:23 bliss or peace or maybe it’s a feeling
0:24:26 induced by drugs and alcohol.
0:24:30 And I would be able to see what my brain patterns
0:24:32 looked like in those states.
0:24:35 Then I could sit around and try to work towards
0:24:37 reproducing those same patterns.
0:24:39 – No, it might not actually lead to anything good.
0:24:41 But in your professional opinion,
0:24:44 total waste of time, you trying to do that?
0:24:47 – The fact is if you felt good at some moment in your life
0:24:49 and you sat around and tried to reproduce that,
0:24:52 I think you’d do just as well thinking about that moment,
0:24:55 trying to put yourself in that state,
0:24:57 rather than try to match a squiggly line.
0:24:59 – You know, I’m a big believer in data though
0:25:03 and it seems like somebody should be building AI systems
0:25:06 that are able to look at those squiggles
0:25:07 and give me feedback.
0:25:10 The thing that I’d so hard about the brain
0:25:14 is that we don’t get direct feedback about what’s going on,
0:25:17 which is how the brain is so good at what it does.
0:25:18 If the brain didn’t get feedback from the world
0:25:20 about what it was doing,
0:25:21 it wouldn’t be any good at predicting things.
0:25:24 So I’m trying to find a way that I can get feedback,
0:25:25 but it sounds like you’re saying I gotta live
0:25:28 for 20 more years if I wanna hope to do that.
0:25:29 – I think that’s right.
0:25:31 I mean, there’s also this very deep question about
0:25:34 what kind of feedback is useful for you.
0:25:37 Most of the action in your brain is happening unconsciously.
0:25:39 It’s happening well below the surface of your awareness
0:25:41 or your ability to access it.
0:25:45 And the fact is that your brain works much better that way.
0:25:47 Do you play tennis, for example?
0:25:47 – Not well.
0:25:48 – Or golf?
0:25:49 – Golf I play.
0:25:50 – Okay, good.
0:25:51 So if I ask you, hey Stephen,
0:25:54 tell me exactly how you swing that golf club.
0:25:55 The more you start thinking about it,
0:25:56 the worse you’re gonna be at it.
0:25:59 Because consciousness, when it starts poking around
0:26:00 in areas that it doesn’t belong,
0:26:02 it’s only gonna make things worse.
0:26:04 And so it is an interesting question
0:26:06 about the kind of things
0:26:07 that we want to be more conscious of.
0:26:09 I’m trying some of these experiments now,
0:26:11 actually using my wristband,
0:26:16 wearing EEG and getting a summarized feedback on the wrist.
0:26:17 So I don’t have to stare at a screen,
0:26:19 but as I’m walking around during the day,
0:26:22 I have a sense of what’s going on with this.
0:26:23 Or with the smart watch,
0:26:26 having a sense of what’s going on with my physiology.
0:26:28 I’m not sure yet whether it’s useful
0:26:30 or whether those things are unconscious
0:26:33 because mother nature figured out a long time ago
0:26:36 that it’s just as well if it remains unconscious.
0:26:37 One thing I’m doing,
0:26:40 which is just a wacky experiment is to try it.
0:26:42 The smart watch is measuring all these things.
0:26:43 We have that data going out,
0:26:47 but the key is you have someone else wear the wristband.
0:26:48 Like your spouse wear the smart watch
0:26:51 and you’re feeling her physiology.
0:26:52 And I’m trying to figure out,
0:26:55 is this useful to be tapped into someone else’s physiology?
0:26:57 I don’t know if this is good or bad for marriages,
0:26:58 but what a nightmare.
0:27:01 But I’m just trying to really get at this question
0:27:03 of these unconscious signals that we experience.
0:27:05 Is it better if they’re exposed
0:27:07 or better to not expose them?
0:27:09 – What have you found empirically?
0:27:10 – Empirically what I found is that married couples
0:27:11 don’t want to wear it.
0:27:13 (laughs)
0:27:21 – So in my lived experience,
0:27:26 I walk around and there’s almost nonstop chatter in my head.
0:27:29 It’s like there’s a narrator who’s commenting
0:27:31 on what I’m observing in the world.
0:27:34 My particular voice does a lot of rehearsing
0:27:36 of what I’m gonna say out loud in the future
0:27:39 and a lot of rehashing of past social interactions.
0:27:40 Other people have voices in their head
0:27:44 that are constantly criticizing and belittling them.
0:27:47 But either way, there’s both a voice that’s talking
0:27:50 and there’s also some other entity in my head
0:27:52 that’s listening to that voice and reacting.
0:27:56 Does neuroscience have an explanation for this sort of thing?
0:27:59 On my book Incognito, the way I cast the whole thing
0:28:01 is that the right way to think about the brain
0:28:04 is like a team of rivals.
0:28:06 Lincoln, when he set up his presidential cabinet,
0:28:08 he set up several rivals in it
0:28:10 and they were all functioning as a team.
0:28:13 That’s really what’s going on under the hood in your head
0:28:15 is you’ve got all these drives
0:28:16 that want different things all the time.
0:28:19 So if I put a slice of chocolate cake in front of you,
0:28:21 Steven, part of your brain says,
0:28:23 “Ooh, that’s a good energy source, let’s eat it.”
0:28:25 Part of your brain says, “No, don’t eat it.
0:28:26 It’ll make me overweight.”
0:28:27 Part of your brain says, “Okay, I’ll eat it,
0:28:29 but I’ll go to the gym tonight.”
0:28:31 And the question is, who is talking with whom here?
0:28:35 It’s all you, but it’s different parts of you.
0:28:38 All these drives are constantly arguing it out.
0:28:40 It’s by the way, generating activity
0:28:43 in the same parts of the brain as listening and speaking
0:28:44 that you would normally do.
0:28:48 It’s just internal before anything comes out.
0:28:53 – Language is such an effective form of communicating
0:28:55 and of summarizing information
0:28:57 that at least my impression inside my head
0:29:02 is that a lot of this is being mediated through language.
0:29:03 But I also have this impression
0:29:06 that there are parts of my brain
0:29:07 that are not very good with language.
0:29:11 Maybe I’m crazy, but I have this working theory
0:29:13 that the language parts of my brain
0:29:16 have really co-opted power.
0:29:18 The non-speaking parts of my brain,
0:29:20 they actually feel to me like the good parts of me,
0:29:22 the interesting parts of me,
0:29:24 but I feel like they’re essentially held hostage
0:29:26 by the language parts.
0:29:27 Does that make any sense?
0:29:28 – Well, this might be a good reason
0:29:31 for you to keep pursuing possible ways
0:29:33 to tap into your brain data.
0:29:37 And by the way, it turns out that the internal voice
0:29:39 is on a big spectrum across the population,
0:29:41 which is to say some people like you
0:29:43 have a very loud internal radio.
0:29:45 I happen to be at the other end of the spectrum
0:29:48 where I have no internal radio at all.
0:29:49 I never hear anything in my head.
0:29:51 That’s called an endophagia.
0:29:55 But everyone is somewhere along this spectrum.
0:29:58 One of the points that I’ve always really concentrated
0:30:00 on neuroscience is what are the actual differences
0:30:03 between people traditionally that’s been looked at
0:30:05 in terms of disease states.
0:30:07 But the question is from person to person
0:30:09 who are in the normal part of the distribution,
0:30:10 what are the differences between us?
0:30:12 In terms of those are manifold.
0:30:15 So take something like how clearly you visualize
0:30:16 when you imagine something.
0:30:20 So if I ask you to imagine a dog running across
0:30:24 the flowery meadow towards a cat,
0:30:26 you might have something like a movie in your head.
0:30:28 Other people have no image at all.
0:30:29 They understand it conceptually,
0:30:31 but they don’t have any image in their head.
0:30:34 And it turns out when you carefully study this,
0:30:37 the whole population is smeared across the spectrum.
0:30:39 So our internal lives from person to person
0:30:40 can be quite different.
0:30:42 – So when you talk about the spectrum,
0:30:46 it makes me think of synesthesia.
0:30:49 Could you explain what that is and how that works?
0:30:52 – So I’ve spent about 25 years now studying synesthesia
0:30:55 and that has to do with some percentage of the population
0:30:57 has a mixture of the senses.
0:30:59 They might look at letters on a page
0:31:01 and that triggers a color experience for them
0:31:05 where they hear music and that causes them to see some visual
0:31:07 or they put some taste in their mouth
0:31:10 and it causes them to have a feeling on their fingertips.
0:31:13 There are dozens and dozens of forms of synesthesia,
0:31:16 but what they all have to do with is a cross blending
0:31:18 of things that are normally separate
0:31:20 in the rest of the population.
0:31:23 – And what share of the population has these patterns?
0:31:27 – So it’s about 3% of the population that has colored letters
0:31:29 or colored weekdays or months or numbers.
0:31:31 – It was big, it’s interesting.
0:31:32 I wouldn’t have thought it was so big.
0:31:35 – The crazy part is that if you have synesthesia,
0:31:37 it probably has never struck you
0:31:40 that 97% of the population does not see the world
0:31:41 the way that you see it.
0:31:44 Everyone’s got their own story going on inside
0:31:48 and it’s rare that we stop to consider the possibility
0:31:52 that other people do not have the same reality that we do.
0:31:54 – And what’s going on in the brain?
0:31:55 – In the case of synesthesia,
0:31:58 it’s just a little bit of crosstalk between two areas
0:32:01 that in the rest of the population tend to be separate
0:32:02 but neighboring.
0:32:04 So it’s like porous borders between two countries.
0:32:06 They just get a little bit of data leakage
0:32:10 and that causes them to have a joint sensation of something.
0:32:11 – People make a big deal out of it
0:32:15 when they talk about musicians having this
0:32:17 and they imply that it’s helpful,
0:32:19 that it makes them better musicians.
0:32:20 Do you think there’s truth to that
0:32:23 or is it just that if 3% of the population has this,
0:32:25 then they’re gonna be some great musicians among them?
0:32:26 – I suspect it’s the latter,
0:32:28 which is to say everyone loves
0:32:30 pointing out synesthetic musicians
0:32:34 but no one has done a study on how many deep sea divers
0:32:37 have synesthesia or how many accountants have synesthesia.
0:32:38 And so we don’t really know
0:32:41 if it’s disproportionate among musicians.
0:32:43 So you’ve created this database of people
0:32:47 who have the condition and you find a pattern
0:32:50 that is completely and totally bizarre.
0:32:53 And that’s that there’s a big bunch of people
0:32:57 who associate the letter A with red, B with orange,
0:32:59 C with yellow, it goes on and on
0:33:01 and then they start repeating it G.
0:33:03 In general though, you don’t see any patterns at all.
0:33:07 Like people can connect these colors and letters in any way.
0:33:09 Do you remember when you first found this pattern
0:33:10 and what your thought was?
0:33:13 So typically, as he said, it’s totally idiosyncratic.
0:33:17 Each synesthete has his or her own colors for letters.
0:33:20 So my A might be yellow, your A is purple and so on.
0:33:22 And then what happened is
0:33:24 with two colleagues of mine at Stanford,
0:33:27 we found in this database of tens of thousands of synesthetes
0:33:28 that I’ve collected over the years,
0:33:31 we found that starting in the late 60s,
0:33:33 there was some percentage of synesthetes
0:33:36 who happened to share exactly the same colors.
0:33:38 These synesthetes were in different locations
0:33:39 but they all had the same thing.
0:33:44 And then that percentage rose to about 15% in the mid 70s.
0:33:45 – So when you saw this,
0:33:47 you must’ve been thinking, my God, this is important, right?
0:33:48 – Exactly right.
0:33:49 The question is,
0:33:51 how could these people be sharing the same pattern?
0:33:53 What we had always suspected is that
0:33:56 maybe there was some imprinting that happens,
0:33:58 which is to say there’s a quilt in your grandmother’s house
0:34:03 that has a red A and a yellow B and a purple C and so on.
0:34:05 But everyone has different things
0:34:07 that they grew up with as little kids.
0:34:11 And so it was strange that this was going on.
0:34:13 The punchline is that we realized
0:34:16 that this is the colors of the Fisher Price Magnet set
0:34:19 on the refrigerators that were popular
0:34:22 during the 70s and 80s and then essentially died out.
0:34:23 And so it turns out that when I look across
0:34:25 all these tens of thousands of synesthetes,
0:34:28 it’s just those people who were kids
0:34:30 in the late 60s and 70s and 80s
0:34:32 that imprinted on the Fisher Price Magnet set.
0:34:34 And that’s their synesthesia.
0:34:36 And then as its popularity died out,
0:34:39 there aren’t anymore who have that particular pattern.
0:34:49 – Now I have to imagine that the way we teach
0:34:52 in traditional classrooms with a teacher or professor
0:34:54 at a Blackboard lecturing to a huge group
0:34:57 of passive students, as a neuroscientist,
0:34:59 that must make a cringe, right?
0:35:01 – It does, increasingly, yes.
0:35:02 – How should we teach?
0:35:04 I think the next generation is going to be smarter
0:35:07 than we are simply because of the broadness
0:35:09 of the diet that they can consume.
0:35:10 Whenever they’re curious about something,
0:35:13 they jump on the internet, they get the answer straight away
0:35:16 or from Alexa or from ChatGPT, they just get the answers
0:35:20 and that is massively useful for a few reasons.
0:35:23 One is that when you are curious about something,
0:35:26 you have the right cocktail of neurotransmitters present
0:35:28 to make that information stick.
0:35:31 So if you get the answer to something
0:35:32 in the context of your curiosity,
0:35:34 then it’s going to stay with you.
0:35:36 Whereas you and I grew up in an era
0:35:39 where we had lots of just-in-case information.
0:35:40 – What do you mean by that?
0:35:42 – Oh, you know, like just in case you ever need to know
0:35:45 that the Battle of Hastings happened in 1066, here you go.
0:35:46 – And you want to contrast that
0:35:47 with just-in-time information.
0:35:48 – Exactly.
0:35:51 – I need to know how to fix my car
0:35:52 and so the internet tells me
0:35:55 and then I can really remember it ’cause I need it.
0:35:56 – That’s exactly it.
0:35:58 And so, look, you know, for all of us with kids,
0:36:00 I know you’ve got kids, I’ve got kids and we feel like,
0:36:03 oh, my kid’s on YouTube and wasting time.
0:36:06 There’s a lot of amazing resources
0:36:08 and things that they learn on YouTube
0:36:10 or even on TikTok, anywhere.
0:36:12 There’s lots of garbage, of course,
0:36:14 but it’s better than what we grew up with.
0:36:16 When you and I wanted to know something,
0:36:20 we would ask our mothers to drive us down to the library
0:36:22 and we would thumb through the card catalog
0:36:23 and hope there was something on it there
0:36:25 that wasn’t too outdated.
0:36:27 – You were more ambitious than me.
0:36:28 I would just ask my mother
0:36:30 and I have since learned that every single thing
0:36:33 my mother taught me was completely wrong,
0:36:34 but I still believe them.
0:36:36 Because of this part of the brain
0:36:38 that locks in things that you learn long ago,
0:36:40 I still have to fight every day
0:36:42 against the false sorts my mother taught me.
0:36:45 I wish I had told her to take me to the library.
0:36:47 – My mother was a biology teacher
0:36:48 and my father was a psychiatrist
0:36:51 and so they had all kinds of good information.
0:36:55 I’m just super optimistic about the next generation of kids.
0:36:57 Now, as far as how we teach,
0:37:00 things got complicated with the advent of Google
0:37:03 and now it’s twice as complicated with chat GPT.
0:37:06 Happily, we already learned these lessons 20 years ago.
0:37:08 What we need to do is just change the way
0:37:10 that we ask questions of students.
0:37:13 We can no longer just assume that fill in the blank
0:37:15 or even just writing a paper on something
0:37:18 is the optimal way to have them learn something,
0:37:20 but instead they need to do interactive projects
0:37:22 like run little experiments with each other
0:37:26 and the kind of thing that you and I both love to do
0:37:29 in our careers which is, okay, go out and find this data
0:37:32 and run this experiment and see what happens here.
0:37:35 That’s the kind of opportunities that kids will have now.
0:37:40 – You were listening to a special bonus episode
0:37:43 of People I Mostly Admire with Steve Levitt
0:37:45 and the neuroscientist David Eagleman.
0:37:49 After the break, what are large language models missing?
0:37:51 – It has no theory of mind.
0:37:55 It has no physical model of the world the way that we do.
0:37:56 That’s coming up after the break.
0:38:08 (gentle music)
0:38:13 – David Eagleman is a professor, a CEO,
0:38:16 leader of a nonprofit called the Center for Science and Law,
0:38:19 host of TV shows on PBS and Netflix
0:38:21 and the founder of Possibillionism.
0:38:27 Like every curious person trying to figure out
0:38:29 what we’re doing here, what’s going on,
0:38:32 it just feels like there are two stories.
0:38:35 Either there’s some religion story
0:38:38 or there’s the story of strict atheism,
0:38:39 which I tend to agree with,
0:38:41 but it tends to come with this thing of,
0:38:42 look, we’ve got it all figured out,
0:38:44 there’s nothing more to ask here.
0:38:45 There is a middle position
0:38:46 which people call agnosticism,
0:38:48 but usually that means, I don’t know,
0:38:50 I’m not committing to one thing or the other.
0:38:52 I got interested in defining this new thing
0:38:54 that I call Possibillionism,
0:38:56 which is to try to go out there
0:38:58 and do what a scientist does,
0:39:01 which is an active exploration of the possibility space.
0:39:03 What the heck is going on here?
0:39:06 We live in such a big and mysterious cosmos.
0:39:09 Everything about our existence is sort of weird.
0:39:11 Obviously the whole Judeo-Christian tradition,
0:39:14 that’s one little point in that possibility space
0:39:17 or the possibility that there’s absolutely nothing
0:39:18 or we’re just atoms when we die,
0:39:20 but there’s lots of other possibilities.
0:39:24 And so I’m not willing to commit to one team or the other
0:39:26 without having sufficient evidence.
0:39:28 So that’s why I call myself a Possibillion.
0:39:32 – And so in support of Possibillionism,
0:39:34 maybe a better name could be in order,
0:39:37 you wrote a book called SUM, that’s S-U-M.
0:39:41 So it’s SUM, 40 Tales from the Afterlives.
0:39:44 How do you describe the book to people?
0:39:45 – I call it literary fiction.
0:39:48 It’s 40 short stories that are all mutually exclusive.
0:39:51 They’re all pretty funny, I would like to think,
0:39:53 but they’re also kind of gut-wrenching.
0:39:55 And what I’m doing is shining the flashlight
0:39:57 around the possibility space.
0:39:59 None of them are meant to be taken seriously.
0:40:02 But what the exercise of having 40
0:40:07 completely different stories gives us is a sense of,
0:40:10 wow, actually there’s a lot that we don’t know here.
0:40:12 In some of the stories, God is a female.
0:40:14 In some stories, God is a married couple.
0:40:19 In some stories, God is a species of dim-witted creatures.
0:40:22 In one story, God is actually the size of a bacterium
0:40:24 and doesn’t know that we exist.
0:40:27 And in lots of stories, there’s no God at all.
0:40:29 That book is something I wrote over the course of seven years
0:40:32 and became an international bestseller.
0:40:34 It’s really had a life to it
0:40:36 that I wouldn’t have ever guessed.
0:40:38 – When I heard about the book,
0:40:40 I saw the subtitle and I thought,
0:40:44 I have zero interest in reading a book about the afterlife.
0:40:47 I totally misunderstood what the book was about.
0:40:51 And then I certainly didn’t understand that some was Latin.
0:40:54 – Some actually I chose because among other things,
0:40:56 that’s the title story.
0:40:58 In the afterlife, you relive your life,
0:41:02 but all the moments that share a quality are grouped together.
0:41:05 So you spend three months waiting in line
0:41:08 and you spend 900 hours sitting on the toilet
0:41:10 and you spend 30 years sleeping.
0:41:11 – All in a row.
0:41:13 – Exactly, and this amount of time looking for lost items
0:41:14 in this amount of time,
0:41:16 realizing you’ve forgotten someone’s name
0:41:19 and this amount of time falling and so on.
0:41:20 Part of why I used the title sum
0:41:24 is because of the sum of events in your life like that.
0:41:26 Part of it was because Kojito or Gosume.
0:41:29 So it ended up just being the perfect title for me,
0:41:31 even if it did lose a couple of readers there.
0:41:41 – People are super excited right now
0:41:44 about these generative AI models,
0:41:45 the large language models.
0:41:47 What’s your take on it?
0:41:49 – Essentially these artificial neural networks
0:41:53 took off from a very simplified version of the brain,
0:41:55 which is, hey, look, you’ve got units and they’re connected
0:41:56 and what if we can change the strength
0:41:58 between these connections?
0:42:00 And in a very short time,
0:42:02 that has now become this thing
0:42:05 that has read everything ever written on the planet
0:42:07 and can give extraordinary answers.
0:42:10 But it’s not yet the brain or anything like it.
0:42:12 It’s just taking the very first idea
0:42:14 about the brain and running with it.
0:42:16 What a large language model does not have
0:42:19 is an internal model of the world.
0:42:21 It’s just acting as a statistical parrot.
0:42:23 It’s saying, okay, given these words,
0:42:25 what is the next word most likely to be
0:42:27 given everything that I’ve ever read on the planet?
0:42:29 And so it’s really good at that,
0:42:33 but it has no model of the world, no physical model.
0:42:38 And so things that a six-year-old can answer, it is stuck on.
0:42:39 Now, this is not a criticism of it
0:42:42 in the sense that it can do all kinds of amazing stuff
0:42:43 and it’s gonna change the world,
0:42:45 but it’s not the brain yet
0:42:46 and there’s still plenty of work to be done
0:42:49 to get something that actually acts like the brain.
0:42:52 – Do you think that it is a solvable problem
0:42:55 to give these models a theory of mind, a model of the world?
0:42:58 – I suspect so because there are 8.2 billion of us
0:43:00 who have this functioning in our brains
0:43:04 and as far as we can tell, we’re just made of physical stuff.
0:43:06 We’re just very sophisticated algorithms
0:43:09 and it’s just a matter of cracking what that algorithm is.
0:43:12 – If we were to come back in 100 years,
0:43:13 what do you think would be most different?
0:43:14 I know that’s a hard prediction to make,
0:43:17 but what do you see is transforming most
0:43:19 in the areas you work in?
0:43:21 – The big textbook that we have in our field
0:43:22 is called Principles of Neuroscience
0:43:27 and it’s about 900 pages and it’s not actually principles,
0:43:30 it’s just a data dump of all this crazy stuff we know.
0:43:34 And in 100 years, I expect it’ll be like 90 pages.
0:43:37 We’ll have things where we put big theoretical frameworks
0:43:39 together and we say, ah, okay, look, all this other stuff,
0:43:42 these are just expressions of this basic principle
0:43:43 that we have now figured out.
0:43:46 – Do you pay much attention to behavioral economics?
0:43:46 – Yes, I do.
0:43:48 – And what do you think of it?
0:43:49 – Oh, it’s great and that’s probably the direction
0:43:52 that a lot of fields will go is,
0:43:54 how do humans actually behave?
0:43:56 One of the big things that I find most interesting
0:43:59 about behavioral economics comes back to this issue
0:44:01 about the team of rivals.
0:44:04 When people measure in the brain
0:44:06 how we actually make decisions about whatever,
0:44:09 there are totally separable networks going on.
0:44:12 Some networks care about the valuation of something,
0:44:13 the price point.
0:44:14 You have totally other networks
0:44:17 that care about the anticipated emotional experience
0:44:18 about something.
0:44:22 You have other networks that care about the social context.
0:44:25 Like, what do my friends think about this?
0:44:28 You have mechanisms that care about short-term gratification.
0:44:29 You have other mechanisms that are thinking about
0:44:32 the long-term, what kind of person do I want to be?
0:44:34 All these things are battling it out under the hood.
0:44:37 It’s like the three stooches sticking each other in the eye
0:44:39 and wrestling each other’s arms and stuff.
0:44:41 But what’s fascinating is when you’re standing
0:44:44 in the grocery store aisle trying to decide
0:44:47 which flavor of ice cream you’re gonna buy,
0:44:48 you don’t know about these raging battles
0:44:50 happening under the hood.
0:44:52 You just stand there for a while and then you say,
0:44:54 “Okay, I’ll grab this one over here.”
0:44:56 – There was a point in time among economists
0:44:57 that there was a lot of optimism
0:45:00 that we could really nail macroeconomics,
0:45:04 inflation and interest rates and whatnot.
0:45:06 And we could really understand how the system worked.
0:45:10 And I think there’s been a real step back from that.
0:45:12 The view now is, look, it’s enormous complex system.
0:45:16 And we’ve really, I guess, given up in the short run.
0:45:18 Are you at all worried that’s where we’re going
0:45:19 with the brain?
0:45:20 – Oh gosh, no.
0:45:23 And the reason is because we’ve got
0:45:24 all these billions of brains running around.
0:45:27 What that tells us is it has to be pretty simple
0:45:28 and principal.
0:45:30 You got 19,000 genes, that’s all you’ve got.
0:45:32 Something about it has to be as simple
0:45:37 as falling off a log for it to work out very well
0:45:39 so often, billions of times.
0:45:44 – They say as you get older,
0:45:46 it’s important to keep challenging your brain
0:45:49 by learning new things like a foreign language.
0:45:51 I can’t say I found learning German
0:45:52 to be all that much fun.
0:45:55 And I definitely have not turned out to be very good at it.
0:45:57 So I’ve been looking for a new brain challenge
0:46:02 and I have to say, I find echolocation very intriguing.
0:46:06 How cool would it be to be able to see via sound?
0:46:09 I suspect though that my aptitude for echolocation
0:46:12 will be on power with my aptitude for German.
0:46:15 So if you see me covered in bruises, you’ll know why.
0:46:18 If you wanna learn more about David Egelman’s ideas,
0:46:20 I really enjoyed a couple of his mini books
0:46:23 like LiveWired, which talks about his brain research,
0:46:26 and some four details from the afterlife,
0:46:28 his book of speculative fiction.
0:46:32 – Hey there, it’s Stephen Dubner.
0:46:34 Again, I hope you enjoyed this special episode
0:46:36 of People I Mostly Admire.
0:46:37 I loved it.
0:46:41 And I would suggest you go right now to your podcast app
0:46:43 and follow the show, People I Mostly Admire.
0:46:46 We will be back very soon with more Freakonomics Radio.
0:46:49 Until then, take care of yourself.
0:46:51 And if you can, someone else too.
0:46:55 Freakonomics Radio and People I Mostly Admire
0:46:58 are produced by Stitcher and Renbud Radio.
0:47:00 This episode was produced by Morgan Levy
0:47:02 with help from Lyric Boutich and Daniel Moritz-Rabson.
0:47:04 It was mixed by Jasmine Klinger.
0:47:07 Our staff also includes Alina Kulman, Augusta Chapman,
0:47:09 Dalvin Abouaji, Eleanor Osborne,
0:47:12 Ellen Frankman, Elsa Hernandez, Gabriel Roth, Greg Rippen,
0:47:15 Jason Gambrell, Jeremy Johnston, John Schnars,
0:47:17 Neil Coruth, Rebecca Lee Douglas,
0:47:20 Sarah Lilly, Theo Jacobs, and Zach Lipinski.
0:47:22 Our composer is Luis Guerra.
0:47:24 As always, thank you for listening.
0:47:33 – David, you got your quick time going?
0:47:35 – I do now.
0:47:42 – The Freakonomics Radio Network,
0:47:44 the hidden side of everything.
0:47:49 – Stitcher.
0:47:51 (upbeat music)
0:47:53 you
0:00:05 – Hey there, it’s Stephen Dubner.
0:00:08 Today, a holiday treat, a bonus episode
0:00:10 from people I mostly admire.
0:00:12 One of the other shows we make here
0:00:13 at the Freakonomics Radio Network.
0:00:16 It is an interview show hosted by Steve Levitt,
0:00:18 my Freakonomics friend and co-author,
0:00:21 who is an economics professor emeritus now
0:00:23 at the University of Chicago.
0:00:26 On this episode, Levitt interviews David Eagleman,
0:00:28 a neuroscientist, entrepreneur,
0:00:31 and author of several books, including LiveWired,
0:00:34 the inside story of the ever-changing brain.
0:00:36 It is a fascinating conversation.
0:00:37 You are going to love it.
0:00:40 To hear more conversations like this,
0:00:44 follow people I mostly admire in your podcast app.
0:00:45 Okay, that’s it for me.
0:00:47 Here is Steve Levitt.
0:00:49 (dramatic music)
0:01:01 – I love podcast guests who change the way
0:01:04 I think about some important aspect of the world.
0:01:07 A great example is my guest today, David Eagleman.
0:01:09 He’s a Stanford neuroscientist whose work
0:01:13 on brain plasticity has completely transformed
0:01:18 my understanding of the human brain and its possibilities.
0:01:20 The human brain is about three pounds.
0:01:22 It’s locked in silence and darkness.
0:01:25 It has no idea where the information is coming from
0:01:28 because everything is just electrical spikes
0:01:30 and also chemical releases as a result of those spikes.
0:01:34 And so what you have in there is this giant symphony
0:01:36 of electrical activity going on
0:01:40 and its job is to create a model of the outside world.
0:01:47 – Welcome to people I mostly admire with Steve Levitt.
0:01:50 – According to Eagleman,
0:01:52 the brain is constantly trying
0:01:54 to predict the world around it.
0:01:56 But of course, the world is unpredictable and surprising.
0:02:00 So the brain is constantly updating its model.
0:02:02 The capacity of our brains to be ever-changing
0:02:05 is usually referred to as plasticity,
0:02:08 but Eagleman offers another term, live wired.
0:02:10 That’s where conversation begins.
0:02:18 Plasticity is the term used in the field
0:02:21 because the great neuroscientist
0:02:24 or psychologist actually, William James,
0:02:25 coined the term because he was impressed
0:02:28 with the way that plastic gets manufactured,
0:02:31 where you mold it into a shape and it holds onto that shape.
0:02:34 And he thought that’s kind of like what the brain does.
0:02:37 The great trick that mother nature figured out
0:02:39 was to drop us into the world half-baked.
0:02:42 If you look at the way an alligator drops into the world,
0:02:44 it essentially is pre-programmed.
0:02:47 It eats, mates, sleeps, does whatever it’s doing.
0:02:49 But we spent our first several years
0:02:52 absorbing the world around us based on our neighborhood
0:02:54 and our moment in time and our culture
0:02:56 and our friends and our universities.
0:03:00 We absorb all of that such that we can then springboard
0:03:03 off of that and create our own things.
0:03:05 There are many things that are essentially
0:03:09 pre-programmed in us, but we are incredibly flexible
0:03:11 and that is the key about live wiring.
0:03:13 When I ask you to think of the name
0:03:14 of your fifth grade teacher,
0:03:16 you might be able to pull it up,
0:03:18 even though it’s been years since you saw
0:03:19 that fifth grade teacher,
0:03:23 but somehow there was a change made in your brain
0:03:25 and that stayed in place.
0:03:27 You’ve got 86 billion neurons.
0:03:30 Each neuron is as complicated as a city.
0:03:33 This entire forest of neurons,
0:03:35 every moment of your life is changing.
0:03:38 It’s reconfiguring, it’s strengthening connections
0:03:41 here and there, it’s actually unplugging over here
0:03:42 and replugging over there.
0:03:45 And so that’s why I’ve started to feel
0:03:48 that the term plasticity is maybe underreporting
0:03:49 what’s going on.
0:03:51 And so that’s why I made up the term live wiring.
0:03:53 When I went to school, I feel like they taught me
0:03:56 the brain was organized around things like senses
0:03:58 and emotions, that there were these different parts
0:04:01 of the brain that were good for those things.
0:04:04 But you make the case that there’s a very different
0:04:06 organization of the brain.
0:04:08 – It is organized around the senses,
0:04:10 but the interesting thing is that the cortex,
0:04:13 this wrinkly outer bit, is actually a one-trick pony.
0:04:14 It doesn’t matter what you plug in.
0:04:17 It’ll say, okay, got it, I’ll just wrap myself
0:04:21 around that data and figure out what to do with that data.
0:04:25 It turns out that in almost everybody you have functioning
0:04:27 eyeballs that plug into the back of the head.
0:04:29 And so we end up calling the back part of the brain
0:04:30 the visual cortex.
0:04:32 We call this part the auditory cortex.
0:04:35 And this is the somatosensory cortex that takes
0:04:37 in information from the body and so on.
0:04:40 So what you learned back in high school or college
0:04:42 is correct most of the time.
0:04:45 But what it overlooks is the fact that the brain
0:04:46 is so flexible.
0:04:49 If a person goes blind or is born blind,
0:04:51 that part of the brain that we’re calling
0:04:53 the visual cortex, that gets taken over by hearing,
0:04:54 by touch, by other things.
0:04:56 And so it’s no longer visual cortex.
0:04:59 The same neurons that are there are now doing
0:05:01 a totally different job.
0:05:04 – So let me pose a question to listeners.
0:05:08 Imagine you have a newborn baby and he or she looks
0:05:10 absolutely flawless on the outside.
0:05:14 But then upon examination, the doctors discover
0:05:17 that half of his or her brain is just missing.
0:05:20 A complete hemisphere of the brain, it’s never developed.
0:05:22 It’s just empty space.
0:05:24 I would expect that would be a fatal defect
0:05:28 or best the child would be growing up profoundly
0:05:30 mentally disabled.
0:05:32 – Turns out the kid will be just fine.
0:05:34 You can be born without half the brain
0:05:36 or you can do what’s called a hemispherectomy,
0:05:39 which happens to children who have something called
0:05:42 rasmusans encephalitis, which is a form of epilepsy
0:05:45 that spreads from one hemisphere to the other.
0:05:47 The surgical intervention for that is to remove
0:05:49 half the brain.
0:05:51 You can just imagine as a parent, the horror you would feel
0:05:53 if your child had to go in for something like that.
0:05:55 But you know what, kid’s just fine.
0:05:59 I can’t take my laptop and rip out half the motherboard
0:06:01 and expect it to still function.
0:06:04 But with the brain, with a live wired system, it’ll work.
0:06:07 – So I first came to work because I was so blown away
0:06:10 by the idea of human echolocation.
0:06:13 Only to discover that echolocation is only
0:06:14 the tip of the iceberg.
0:06:17 But could you talk just a bit about echolocation,
0:06:19 how quickly with training it can start
0:06:21 to substitute for sight?
0:06:23 – So it turns out that blind people can make
0:06:27 all kinds of sounds either with their mouth like clicking
0:06:29 or the tip of their cane or snapping their fingers,
0:06:30 anything like this.
0:06:33 And they can get really good at determining
0:06:36 what is coming back as echoes and figure out,
0:06:38 oh, okay, this is an open space in front of me.
0:06:39 Here, there’s something in front of me.
0:06:41 It’s probably a parked car.
0:06:43 And oh, there’s a little gap between two parked cars here.
0:06:44 So I can go in here.
0:06:48 The key is the visual part of the brain is no longer
0:06:50 being used because for whatever reason,
0:06:52 there’s no information coming down those pipelines anymore.
0:06:55 So that part of the brain is taken over by audition,
0:06:58 by hearing and by touch and other things.
0:07:01 What happens is that the blind person becomes really good
0:07:04 at these other things because they’ve just devoted
0:07:06 more real estate to it.
0:07:09 And as a result, they can pick up on all kinds of cues
0:07:11 that would be very difficult for me and you
0:07:14 because our hearing just isn’t that good.
0:07:16 – And then in these studies,
0:07:20 you put a blindfold on a person for two or three days
0:07:22 and you try to teach them echolocation.
0:07:25 If I understand correctly, even over that timescale,
0:07:28 the echolocation starts taking over
0:07:29 the visual part of the brain.
0:07:31 Is that a fair assessment?
0:07:32 – That is exactly right.
0:07:34 This was my colleagues at Harvard.
0:07:36 They did this over the course of five days.
0:07:39 They demonstrated that people could get really good at,
0:07:41 they’re actually a number of studies like this.
0:07:42 They can get really good at reading Braille.
0:07:44 They can do things like echolocation.
0:07:48 And the speed of it was sort of the surprise.
0:07:50 But the real surprise for me came along
0:07:52 when they blindfolded people tightly
0:07:54 and put them in the brain scanner
0:07:59 and they were making sounds or touching the hand.
0:08:01 And they were starting to see activity
0:08:06 in the visual cortex after 60 minutes of being blind.
0:08:09 – So in your book, you talk about REM sleep.
0:08:11 And honestly, if I had sat down
0:08:14 and tried to come up with an explanation of REM sleep,
0:08:17 I could have listened to a thousand ideas.
0:08:20 Your pet theory would not be one of them.
0:08:22 So explain what REM sleep is
0:08:24 and then tell me why you think we do it.
0:08:26 – REM sleep is rapid eye movement sleep.
0:08:28 We have this every night, about every 90 minutes.
0:08:30 And that’s when you dream.
0:08:31 So if you wake someone up
0:08:33 when their eyes are moving rapidly
0:08:34 and you say, “Hey, what are you thinking about?”
0:08:37 they’ll say, “Whoa, I was just riding a camel across a meadow.”
0:08:39 But if you wake them up at other parts of their sleep,
0:08:42 they typically won’t have anything going on.
0:08:44 So that’s how we know we dream during REM sleep.
0:08:45 But here’s the key.
0:08:48 My student and I realized that at nighttime,
0:08:50 when the planet rotates,
0:08:52 we spend half our time in darkness.
0:08:55 And obviously we’re very used to this electricity blessed world.
0:08:57 But think about this in historical time
0:08:59 over the course of hundreds of millions of years.
0:09:00 It’s really dark.
0:09:03 I mean, half the time you are in blackness.
0:09:05 Now you can still hear and touch and taste
0:09:07 and smell in the dark.
0:09:10 But the visual system is at a disadvantage
0:09:12 whenever the planet rotates into darkness.
0:09:16 And so given the rapidity with which other systems
0:09:18 can encroach on that,
0:09:21 what we realized is it needs a way of defending itself
0:09:23 against takeover every single night.
0:09:25 And that’s what dreams are about.
0:09:28 So what happens is you have these midbrain mechanisms
0:09:31 that simply blast random activity
0:09:34 into the visual cortex every 90 minutes during the night.
0:09:37 And when you get activity in the visual cortex,
0:09:39 you say, “Oh, I’m seeing things.”
0:09:41 And because the brain is a storyteller,
0:09:43 you can’t activate all the stuff
0:09:47 without feeling like there’s a whole story going on there.
0:09:48 But the fascinating thing is when you look
0:09:51 at the circuitry carefully, it’s super specific,
0:09:53 much more specific than almost anything else in the brain.
0:09:57 It’s only hitting the primary visual cortex and nothing else.
0:10:02 And so that led us to a completely new theory about dreams.
0:10:05 We studied 25 different species of primates
0:10:07 and we looked at the amount of REM sleep
0:10:08 they have every night.
0:10:12 And we also looked at how plastic they are as a species.
0:10:14 It turns out that the amount of dream sleep
0:10:18 that a creature has exactly correlates
0:10:20 with how plastic they are, which is to say,
0:10:23 if your visual system is in danger of getting taken over
0:10:25 because your brain is very flexible,
0:10:27 then you have to have more dream sleep.
0:10:29 And by the way, when you look at human infants,
0:10:32 they have tons of dream sleep at the beginning
0:10:34 when their brains are very plastic.
0:10:37 And as they age, the amount of dream sleep goes down.
0:10:40 – Have you convinced the sleep scientists this is true
0:10:43 or is this just you believing it right now?
0:10:44 – At the moment, there are 19 papers
0:10:46 that have cited this and discussed this.
0:10:48 And I think it’s right.
0:10:49 I mean, look, everything can be wrong.
0:10:50 Everything is provisional,
0:10:54 but it’s the single theory that is quantitative.
0:10:56 It’s the single theory about dreams
0:11:00 that says not only here is a idea for why we dream,
0:11:01 but we can compare across species
0:11:04 and the predictions match exactly.
0:11:07 No one would have suspected that you’d see a relationship
0:11:11 between how long it takes you to walk or reach adolescence
0:11:12 and how much dream sleep you have.
0:11:14 But it turns out that is spun on.
0:11:23 – So we talked about echolocation,
0:11:26 which uses sound to accomplish tests
0:11:29 that are usually done by vision.
0:11:31 And you’ve started a company called NeoSensory,
0:11:35 which uses touch to accomplish tasks
0:11:37 that are usually done with hearing.
0:11:38 Can you explain the science behind that?
0:11:41 Given that all the data running around in the brain
0:11:44 is just data and the brain doesn’t know where it came from.
0:11:47 All it knows is, oh, here are electrical spikes
0:11:48 and it tries to figure out what to do with it.
0:11:51 I got really interested in this idea of sensory substitution,
0:11:53 which is can you push information into the brain
0:11:56 via an unusual channel?
0:11:58 Originally we built a vest
0:12:00 that was covered with vibratory motors
0:12:03 and we captured sound for people who are deaf.
0:12:05 So the vest captures sound,
0:12:07 breaks it up from high to low frequency
0:12:10 and you’re feeling the sound on your torso.
0:12:12 By the way, this is exactly what the inner ear does.
0:12:14 It breaks up sound from high to low frequency
0:12:16 and ships that off to the brain.
0:12:18 So we’re just transferring the inner ear
0:12:20 to the skin of the torso.
0:12:22 And it worked, people who are deaf
0:12:25 could come to hear the world that way.
0:12:28 So I spun this out of my lab as a company, Neosensory,
0:12:32 and we shrunk the vest down to a wristband
0:12:34 and we’re on wrist of deaf people all over the world.
0:12:37 The other alternative for somebody who’s deaf
0:12:39 is a cochlear implant, an invasive surgery.
0:12:44 This is much cheaper and does as good a job.
0:12:45 Just to make sure I understand it.
0:12:50 Sounds happen and this wristband hears the sounds
0:12:54 and then shoots electrical impulses into your wrist
0:12:56 that correspond to the high and low frequency.
0:12:58 It’s actually just vibratory motors.
0:13:00 So it’s just like the buzzer on your cell phone
0:13:03 but we have a string of these buzzers all along your wrist
0:13:06 and we’re actually taking advantage of an illusion
0:13:09 which is if I have two motors next to each other
0:13:11 and I stimulate them both,
0:13:14 you will feel one virtual point right in between.
0:13:16 And as I change the strength of those two motors
0:13:20 relative to each other, I can move that point around.
0:13:23 So we’re actually stimulating 128 virtual points
0:13:24 along the wrist.
0:13:27 – Do people train you give them very direct feedback
0:13:29 or is it more organic?
0:13:30 – Great question.
0:13:32 It started off where we were doing a lot of training
0:13:35 on people and what we realized is it’s all the same
0:13:36 if we just let it be organic.
0:13:39 The key is we just encourage people be in the world
0:13:40 and that’s it.
0:13:42 You see the dog’s mouth moving
0:13:45 and you feel the barking on your wrist
0:13:47 or you close the door and you feel that on your wrist
0:13:49 or you say something, you know,
0:13:50 most deaf people can speak
0:13:53 and they know what their motor output is
0:13:55 and they’re feeling the input.
0:13:58 – Okay, so hearing their own voice for the first time
0:13:59 through this. – Exactly.
0:14:01 – Oh God, yeah, that’s interesting.
0:14:01 – And by the way,
0:14:03 that’s how you learned how to use your ears too.
0:14:05 You know, when you’re a baby,
0:14:06 you’re watching your mother’s mouth move
0:14:08 and you’re hearing data coming in your ears
0:14:10 and you clapped your hands together
0:14:12 and you hear something in your ears.
0:14:13 It’s the same idea.
0:14:15 You’re just training up correlations in the brain
0:14:18 about, oh, this visual thing seems to always go
0:14:20 with that auditory stimulus.
0:14:23 – So then it seems like if I’m deaf
0:14:25 and I see the dog’s mouth moving
0:14:27 and I now associate that with the sound,
0:14:31 do the people say that they hear the sound where the dog is
0:14:33 or is the sound coming from the wrist?
0:14:34 – For the first few months,
0:14:36 you’re hearing it on your wrist,
0:14:38 you can get pretty good at these correlations
0:14:40 but then after about six months,
0:14:43 if I ask somebody when the dog barks,
0:14:44 do you feel something on your wrist?
0:14:45 And you think, okay, what was that on?
0:14:46 That must have been a dog bark
0:14:47 and then you look for the dog.
0:14:51 And they say, no, I just hear the dog out there.
0:14:52 – Hmm.
0:14:53 – And that sounds so crazy.
0:14:55 But remember, that’s what your ears are doing.
0:14:58 Your ears are capturing vibrations at the eardrum
0:15:00 that moves through the middle ear to the inner ear,
0:15:01 breaks up to different frequencies,
0:15:04 goes off to your brain, goes to your auditory cord.
0:15:06 It’s this giant pathway of things.
0:15:09 And yet, even though you’re hearing my voice right now
0:15:13 inside your head, you think I’m somewhere else.
0:15:14 And that’s exactly what happens,
0:15:18 irrespective of how you feed the data in.
0:15:22 So you also have a product that helps with tinnitus.
0:15:23 Could you explain both what that is
0:15:25 and how your product helps?
0:15:27 So tinnitus is a ringing in the ears.
0:15:31 It’s like beep and about 15% of the population has this.
0:15:33 And for some people, it’s really, really bad.
0:15:38 It turns out there is a mechanism for helping with tinnitus
0:15:41 which has to do with playing tones
0:15:45 and then matching that with stimulation on the skin.
0:15:47 People wear the wristband, it’s exactly the same wristband,
0:15:51 but we have the phone play tones, boop, boop, boop, boop, boop.
0:15:53 And you’re feeling that all over your wrist.
0:15:54 And you just do that for 10 minutes a day.
0:15:56 And it drives down the tinnitus.
0:15:58 Now, why does that work?
0:16:00 There are various theories on this,
0:16:02 but I think the simplest version
0:16:05 is that your brain is figuring out,
0:16:09 okay, real sounds always cause this
0:16:11 correlating vibration on my wrist,
0:16:15 but a fake sound, beep, you know, this thing in my head,
0:16:17 that doesn’t have any verification on the wrist.
0:16:20 And so that must not be a real sound.
0:16:23 So because of issues of brain plasticity,
0:16:26 the brain just reduces the strength of the tinnitus
0:16:28 because it learns that it’s not getting any confirmation
0:16:30 that that’s a real world sound.
0:16:32 Now, how did you figure out
0:16:34 that this bracelet could be used for this?
0:16:36 This was discovered by a woman named Susan Shore,
0:16:39 who’s a researcher who discovered this about a decade ago.
0:16:42 She was using electrical shocks on the tongue.
0:16:43 And there’s actually another company
0:16:44 that’s spun out called Lanier
0:16:46 that does this with sounds in the ear
0:16:47 and shocks on the tongue.
0:16:49 They had an argument that they think
0:16:52 it had to be touched from the head and the neck.
0:16:53 And I didn’t buy that at all.
0:16:54 And that’s why I tried that with the wristband.
0:16:57 So this was not an original idea for us
0:17:01 except to try this on the wrist and it works equally as well.
0:17:06 So what we’re talking about is substituting between senses.
0:17:08 Are there other forms of this products
0:17:10 that are currently available to consumers
0:17:13 or likely to become available soon in the space?
0:17:16 For people who are blind, for example,
0:17:19 there are a few different approaches to this.
0:17:21 One is called the brain port and that’s where,
0:17:24 for a blind person, they have a little camera on their glasses
0:17:29 and that gets turned into little electrical stimulation
0:17:30 on the tongue.
0:17:33 So you’re wearing this little electro-tactile grid
0:17:36 on your tongue and it tastes like pop rocks
0:17:37 sort of in your mouth.
0:17:39 Blind people can get pretty good at this.
0:17:42 They can navigate complex obstacle courses
0:17:45 or throw a ball into a basket at a distance
0:17:48 because they can come to see the world through their tongue,
0:17:50 which if that sounds crazy,
0:17:52 it’s the same thing as seeing it through these two spheres
0:17:54 that are embedded in your skull.
0:17:57 It’s just capturing photons and information about them,
0:17:58 figuring out where the edges are
0:17:59 and then shipping that back to the brain.
0:18:01 And the brain can figure that out.
0:18:03 There’s also a colleague of mine
0:18:05 that makes an app called Voice.
0:18:07 It uses the phone’s camera
0:18:10 and it turns that into soundscape.
0:18:13 So if you’re moving the camera around,
0:18:15 you’re hearing (mimics sounds)
0:18:18 you know, it sounds like a strange cacophony,
0:18:19 but it doesn’t take long,
0:18:21 even for you as a sighted person,
0:18:23 to get used to this and say,
0:18:27 oh, okay, I’m turning the visual world into sound
0:18:28 and it’s starting to make sense.
0:18:30 When I pass over an edge
0:18:32 or when I zoom into something,
0:18:34 the pitch changes, the volume changes,
0:18:37 there’s all kinds of changes in the sound quality
0:18:39 that tells you, oh, yeah, now I’m getting close to something,
0:18:40 now I’m getting far,
0:18:43 and here’s what the world looks like in sound.
0:18:48 – Coming up after the break,
0:18:51 there’s really no shortage of theoretical ideas
0:18:52 in neuroscience,
0:18:56 but fundamentally, we don’t have enough data.
0:18:59 More of Steve Leavitt’s conversation with David Eagleman
0:19:02 in this special episode of “People I Mostly Admire.”
0:19:18 Okay, back now to this special episode of “People I Mostly Admire.”
0:19:20 This is my Freakonomics friend and co-author Steve Leavitt
0:19:24 in conversation with the neuroscientist David Eagleman.
0:19:27 (eerie music)
0:19:30 – Elon Musk’s company, Neuralink,
0:19:32 has gotten a ton of attention lately.
0:19:33 Could you explain what they’re trying to do
0:19:37 and whether you think that’s a promising avenue to explore?
0:19:38 – What they’re doing is they’re putting electrodes
0:19:43 into the brain to read from and talk to the neurons there.
0:19:45 So what we’ve been talking about so far
0:19:47 has been sending signals to the brain,
0:19:48 but what Neuralink is trying to do
0:19:51 is take signals out of the brain, is that right?
0:19:51 – That is correct.
0:19:53 Everything we’ve been talking about so far
0:19:54 with sensory substitution,
0:19:57 that’s a way of pushing information in and non-invasive.
0:20:00 And what Neuralink is, you have to drill a hole in the head
0:20:02 to get to the brain itself,
0:20:04 but then you can do reading and writing invasively.
0:20:09 That actually has been going on for 60 years.
0:20:11 The language of the brain is electrical stimulation.
0:20:14 And so with a little tiny wire, essentially,
0:20:17 you can zap a neuron and make it pop off
0:20:21 or you can listen to when it’s chattering along,
0:20:23 going pa-pa-pa-pa-pa-pa-pa-pa.
0:20:25 There’s nothing actually new about what Neuralink is doing,
0:20:28 except that they’re making a one ton robot
0:20:32 that sews the electrodes into the brain
0:20:34 so it can do it smaller and tighter and faster
0:20:36 than a neurosurgeon can.
0:20:38 And by the way, there are a lot of great companies
0:20:41 doing this sort of thing with electrodes.
0:20:44 As people get access to the brain,
0:20:46 we’re finally getting to a point, we’re not there yet,
0:20:48 but we’re getting to a point
0:20:52 where we’ll finally be able to push theory forward.
0:20:55 There’s really no shortage of theoretical ideas
0:20:58 in neuroscience, but fundamentally,
0:21:01 we don’t have enough data because, as I mentioned,
0:21:04 you’ve got these 86 billion neurons all doing their thing,
0:21:09 and we have never measured what all these things
0:21:10 are doing at the same time.
0:21:11 So we have technologies
0:21:15 like functional magnetic resonance imaging, FMRI,
0:21:19 which measures big blobby volumes of,
0:21:21 oh, there was some activity there and some activity there,
0:21:22 but that doesn’t tell us what’s happening
0:21:24 at the level of individual neurons.
0:21:26 We can currently measure some individual neurons,
0:21:28 but not many of them.
0:21:32 Be like if an alien asked one person in New York City,
0:21:33 hey, what’s going on here?
0:21:36 And then tried to extrapolate to understand
0:21:37 the entire economy of New York City
0:21:38 and how that’s all working.
0:21:43 So I think we’re finally getting closer to the point
0:21:45 where we’ll have real data about,
0:21:47 wow, this is what thousands or eventually hundreds
0:21:49 of thousands or millions of neurons
0:21:52 are actually doing in real time at the same moment,
0:21:55 and then we’ll be able to really get progress.
0:21:57 I actually think the future is not in things like Neuralink,
0:22:02 but the next level past that, which is nanorobotics.
0:22:04 This is all theoretical right now,
0:22:07 but I don’t think this is more than 20, 30 years off,
0:22:10 where you do three-dimensional printing,
0:22:14 atomically precise, you make molecular robots,
0:22:15 hundreds of millions of these,
0:22:17 and then you put them in a capsule
0:22:18 and you swallow the capsule,
0:22:19 and these little robots swim around
0:22:23 and they go into your neurons, these cells in your brain.
0:22:27 And from there, they can send out little signals saying,
0:22:29 hey, this neuron just fired.
0:22:30 And once we have that sort of thing,
0:22:32 then we can say non-invasively,
0:22:35 here’s what all these neurons are doing at the same time,
0:22:38 and then we’ll really understand the brain.
0:22:41 – I’ve worn a continuous glucose monitor a few times,
0:22:43 so you stick this thing in your arm
0:22:44 and you leave it there for 10 days,
0:22:46 and every five minutes,
0:22:49 it gives you a reading of your blood glucose level.
0:22:51 It gives you direct feedback
0:22:53 on how your body responds to the foods you eat,
0:22:55 also to stress or lack of sleep,
0:22:57 that you simply don’t get otherwise.
0:23:01 I learned more about my metabolism in 10 days
0:23:04 than I had over the entire rest of my life combined.
0:23:06 What you’re talking about with these nanorobots
0:23:07 is obviously in the future,
0:23:12 but is there anything now that I can buy
0:23:13 and I can strap on my head?
0:23:15 And I know it’s not gonna be individual neurons,
0:23:17 but that would allow me to get feedback
0:23:21 about my brainwaves and be able to learn
0:23:23 in that same way I do with the glucose monitor?
0:23:27 – What we have now is EEG, Electroencephalography,
0:23:30 and there are several really good companies
0:23:34 like MUSE and Emotive that have come out with at-home methods.
0:23:36 You just strap this thing on your head
0:23:39 and you can measure what’s going on with your brainwaves.
0:23:43 The problem is that brainwaves are still pretty distant
0:23:46 from the activity of 86 billion chattering neurons.
0:23:49 An analogy would be if you went
0:23:50 to your favorite baseball stadium
0:23:53 and you attached a few microphones
0:23:55 to the outside of the stadium
0:23:58 and you listened to a baseball game,
0:24:00 but all you could hear with these microphones
0:24:02 is occasionally the crack of the bat
0:24:03 and the roar of the crowd.
0:24:07 And then your job is to reconstruct what baseball is
0:24:10 just some of these few little signals you’re getting.
0:24:14 So I’m afraid it’s still a pretty crude technology.
0:24:17 – I could imagine that I would put one of these EEGs on
0:24:20 and I would just find some feeling I liked,
0:24:23 bliss or peace or maybe it’s a feeling
0:24:26 induced by drugs and alcohol.
0:24:30 And I would be able to see what my brain patterns
0:24:32 looked like in those states.
0:24:35 Then I could sit around and try to work towards
0:24:37 reproducing those same patterns.
0:24:39 – No, it might not actually lead to anything good.
0:24:41 But in your professional opinion,
0:24:44 total waste of time, you trying to do that?
0:24:47 – The fact is if you felt good at some moment in your life
0:24:49 and you sat around and tried to reproduce that,
0:24:52 I think you’d do just as well thinking about that moment,
0:24:55 trying to put yourself in that state,
0:24:57 rather than try to match a squiggly line.
0:24:59 – You know, I’m a big believer in data though
0:25:03 and it seems like somebody should be building AI systems
0:25:06 that are able to look at those squiggles
0:25:07 and give me feedback.
0:25:10 The thing that I’d so hard about the brain
0:25:14 is that we don’t get direct feedback about what’s going on,
0:25:17 which is how the brain is so good at what it does.
0:25:18 If the brain didn’t get feedback from the world
0:25:20 about what it was doing,
0:25:21 it wouldn’t be any good at predicting things.
0:25:24 So I’m trying to find a way that I can get feedback,
0:25:25 but it sounds like you’re saying I gotta live
0:25:28 for 20 more years if I wanna hope to do that.
0:25:29 – I think that’s right.
0:25:31 I mean, there’s also this very deep question about
0:25:34 what kind of feedback is useful for you.
0:25:37 Most of the action in your brain is happening unconsciously.
0:25:39 It’s happening well below the surface of your awareness
0:25:41 or your ability to access it.
0:25:45 And the fact is that your brain works much better that way.
0:25:47 Do you play tennis, for example?
0:25:47 – Not well.
0:25:48 – Or golf?
0:25:49 – Golf I play.
0:25:50 – Okay, good.
0:25:51 So if I ask you, hey Stephen,
0:25:54 tell me exactly how you swing that golf club.
0:25:55 The more you start thinking about it,
0:25:56 the worse you’re gonna be at it.
0:25:59 Because consciousness, when it starts poking around
0:26:00 in areas that it doesn’t belong,
0:26:02 it’s only gonna make things worse.
0:26:04 And so it is an interesting question
0:26:06 about the kind of things
0:26:07 that we want to be more conscious of.
0:26:09 I’m trying some of these experiments now,
0:26:11 actually using my wristband,
0:26:16 wearing EEG and getting a summarized feedback on the wrist.
0:26:17 So I don’t have to stare at a screen,
0:26:19 but as I’m walking around during the day,
0:26:22 I have a sense of what’s going on with this.
0:26:23 Or with the smart watch,
0:26:26 having a sense of what’s going on with my physiology.
0:26:28 I’m not sure yet whether it’s useful
0:26:30 or whether those things are unconscious
0:26:33 because mother nature figured out a long time ago
0:26:36 that it’s just as well if it remains unconscious.
0:26:37 One thing I’m doing,
0:26:40 which is just a wacky experiment is to try it.
0:26:42 The smart watch is measuring all these things.
0:26:43 We have that data going out,
0:26:47 but the key is you have someone else wear the wristband.
0:26:48 Like your spouse wear the smart watch
0:26:51 and you’re feeling her physiology.
0:26:52 And I’m trying to figure out,
0:26:55 is this useful to be tapped into someone else’s physiology?
0:26:57 I don’t know if this is good or bad for marriages,
0:26:58 but what a nightmare.
0:27:01 But I’m just trying to really get at this question
0:27:03 of these unconscious signals that we experience.
0:27:05 Is it better if they’re exposed
0:27:07 or better to not expose them?
0:27:09 – What have you found empirically?
0:27:10 – Empirically what I found is that married couples
0:27:11 don’t want to wear it.
0:27:13 (laughs)
0:27:21 – So in my lived experience,
0:27:26 I walk around and there’s almost nonstop chatter in my head.
0:27:29 It’s like there’s a narrator who’s commenting
0:27:31 on what I’m observing in the world.
0:27:34 My particular voice does a lot of rehearsing
0:27:36 of what I’m gonna say out loud in the future
0:27:39 and a lot of rehashing of past social interactions.
0:27:40 Other people have voices in their head
0:27:44 that are constantly criticizing and belittling them.
0:27:47 But either way, there’s both a voice that’s talking
0:27:50 and there’s also some other entity in my head
0:27:52 that’s listening to that voice and reacting.
0:27:56 Does neuroscience have an explanation for this sort of thing?
0:27:59 On my book Incognito, the way I cast the whole thing
0:28:01 is that the right way to think about the brain
0:28:04 is like a team of rivals.
0:28:06 Lincoln, when he set up his presidential cabinet,
0:28:08 he set up several rivals in it
0:28:10 and they were all functioning as a team.
0:28:13 That’s really what’s going on under the hood in your head
0:28:15 is you’ve got all these drives
0:28:16 that want different things all the time.
0:28:19 So if I put a slice of chocolate cake in front of you,
0:28:21 Steven, part of your brain says,
0:28:23 “Ooh, that’s a good energy source, let’s eat it.”
0:28:25 Part of your brain says, “No, don’t eat it.
0:28:26 It’ll make me overweight.”
0:28:27 Part of your brain says, “Okay, I’ll eat it,
0:28:29 but I’ll go to the gym tonight.”
0:28:31 And the question is, who is talking with whom here?
0:28:35 It’s all you, but it’s different parts of you.
0:28:38 All these drives are constantly arguing it out.
0:28:40 It’s by the way, generating activity
0:28:43 in the same parts of the brain as listening and speaking
0:28:44 that you would normally do.
0:28:48 It’s just internal before anything comes out.
0:28:53 – Language is such an effective form of communicating
0:28:55 and of summarizing information
0:28:57 that at least my impression inside my head
0:29:02 is that a lot of this is being mediated through language.
0:29:03 But I also have this impression
0:29:06 that there are parts of my brain
0:29:07 that are not very good with language.
0:29:11 Maybe I’m crazy, but I have this working theory
0:29:13 that the language parts of my brain
0:29:16 have really co-opted power.
0:29:18 The non-speaking parts of my brain,
0:29:20 they actually feel to me like the good parts of me,
0:29:22 the interesting parts of me,
0:29:24 but I feel like they’re essentially held hostage
0:29:26 by the language parts.
0:29:27 Does that make any sense?
0:29:28 – Well, this might be a good reason
0:29:31 for you to keep pursuing possible ways
0:29:33 to tap into your brain data.
0:29:37 And by the way, it turns out that the internal voice
0:29:39 is on a big spectrum across the population,
0:29:41 which is to say some people like you
0:29:43 have a very loud internal radio.
0:29:45 I happen to be at the other end of the spectrum
0:29:48 where I have no internal radio at all.
0:29:49 I never hear anything in my head.
0:29:51 That’s called an endophagia.
0:29:55 But everyone is somewhere along this spectrum.
0:29:58 One of the points that I’ve always really concentrated
0:30:00 on neuroscience is what are the actual differences
0:30:03 between people traditionally that’s been looked at
0:30:05 in terms of disease states.
0:30:07 But the question is from person to person
0:30:09 who are in the normal part of the distribution,
0:30:10 what are the differences between us?
0:30:12 In terms of those are manifold.
0:30:15 So take something like how clearly you visualize
0:30:16 when you imagine something.
0:30:20 So if I ask you to imagine a dog running across
0:30:24 the flowery meadow towards a cat,
0:30:26 you might have something like a movie in your head.
0:30:28 Other people have no image at all.
0:30:29 They understand it conceptually,
0:30:31 but they don’t have any image in their head.
0:30:34 And it turns out when you carefully study this,
0:30:37 the whole population is smeared across the spectrum.
0:30:39 So our internal lives from person to person
0:30:40 can be quite different.
0:30:42 – So when you talk about the spectrum,
0:30:46 it makes me think of synesthesia.
0:30:49 Could you explain what that is and how that works?
0:30:52 – So I’ve spent about 25 years now studying synesthesia
0:30:55 and that has to do with some percentage of the population
0:30:57 has a mixture of the senses.
0:30:59 They might look at letters on a page
0:31:01 and that triggers a color experience for them
0:31:05 where they hear music and that causes them to see some visual
0:31:07 or they put some taste in their mouth
0:31:10 and it causes them to have a feeling on their fingertips.
0:31:13 There are dozens and dozens of forms of synesthesia,
0:31:16 but what they all have to do with is a cross blending
0:31:18 of things that are normally separate
0:31:20 in the rest of the population.
0:31:23 – And what share of the population has these patterns?
0:31:27 – So it’s about 3% of the population that has colored letters
0:31:29 or colored weekdays or months or numbers.
0:31:31 – It was big, it’s interesting.
0:31:32 I wouldn’t have thought it was so big.
0:31:35 – The crazy part is that if you have synesthesia,
0:31:37 it probably has never struck you
0:31:40 that 97% of the population does not see the world
0:31:41 the way that you see it.
0:31:44 Everyone’s got their own story going on inside
0:31:48 and it’s rare that we stop to consider the possibility
0:31:52 that other people do not have the same reality that we do.
0:31:54 – And what’s going on in the brain?
0:31:55 – In the case of synesthesia,
0:31:58 it’s just a little bit of crosstalk between two areas
0:32:01 that in the rest of the population tend to be separate
0:32:02 but neighboring.
0:32:04 So it’s like porous borders between two countries.
0:32:06 They just get a little bit of data leakage
0:32:10 and that causes them to have a joint sensation of something.
0:32:11 – People make a big deal out of it
0:32:15 when they talk about musicians having this
0:32:17 and they imply that it’s helpful,
0:32:19 that it makes them better musicians.
0:32:20 Do you think there’s truth to that
0:32:23 or is it just that if 3% of the population has this,
0:32:25 then they’re gonna be some great musicians among them?
0:32:26 – I suspect it’s the latter,
0:32:28 which is to say everyone loves
0:32:30 pointing out synesthetic musicians
0:32:34 but no one has done a study on how many deep sea divers
0:32:37 have synesthesia or how many accountants have synesthesia.
0:32:38 And so we don’t really know
0:32:41 if it’s disproportionate among musicians.
0:32:43 So you’ve created this database of people
0:32:47 who have the condition and you find a pattern
0:32:50 that is completely and totally bizarre.
0:32:53 And that’s that there’s a big bunch of people
0:32:57 who associate the letter A with red, B with orange,
0:32:59 C with yellow, it goes on and on
0:33:01 and then they start repeating it G.
0:33:03 In general though, you don’t see any patterns at all.
0:33:07 Like people can connect these colors and letters in any way.
0:33:09 Do you remember when you first found this pattern
0:33:10 and what your thought was?
0:33:13 So typically, as he said, it’s totally idiosyncratic.
0:33:17 Each synesthete has his or her own colors for letters.
0:33:20 So my A might be yellow, your A is purple and so on.
0:33:22 And then what happened is
0:33:24 with two colleagues of mine at Stanford,
0:33:27 we found in this database of tens of thousands of synesthetes
0:33:28 that I’ve collected over the years,
0:33:31 we found that starting in the late 60s,
0:33:33 there was some percentage of synesthetes
0:33:36 who happened to share exactly the same colors.
0:33:38 These synesthetes were in different locations
0:33:39 but they all had the same thing.
0:33:44 And then that percentage rose to about 15% in the mid 70s.
0:33:45 – So when you saw this,
0:33:47 you must’ve been thinking, my God, this is important, right?
0:33:48 – Exactly right.
0:33:49 The question is,
0:33:51 how could these people be sharing the same pattern?
0:33:53 What we had always suspected is that
0:33:56 maybe there was some imprinting that happens,
0:33:58 which is to say there’s a quilt in your grandmother’s house
0:34:03 that has a red A and a yellow B and a purple C and so on.
0:34:05 But everyone has different things
0:34:07 that they grew up with as little kids.
0:34:11 And so it was strange that this was going on.
0:34:13 The punchline is that we realized
0:34:16 that this is the colors of the Fisher Price Magnet set
0:34:19 on the refrigerators that were popular
0:34:22 during the 70s and 80s and then essentially died out.
0:34:23 And so it turns out that when I look across
0:34:25 all these tens of thousands of synesthetes,
0:34:28 it’s just those people who were kids
0:34:30 in the late 60s and 70s and 80s
0:34:32 that imprinted on the Fisher Price Magnet set.
0:34:34 And that’s their synesthesia.
0:34:36 And then as its popularity died out,
0:34:39 there aren’t anymore who have that particular pattern.
0:34:49 – Now I have to imagine that the way we teach
0:34:52 in traditional classrooms with a teacher or professor
0:34:54 at a Blackboard lecturing to a huge group
0:34:57 of passive students, as a neuroscientist,
0:34:59 that must make a cringe, right?
0:35:01 – It does, increasingly, yes.
0:35:02 – How should we teach?
0:35:04 I think the next generation is going to be smarter
0:35:07 than we are simply because of the broadness
0:35:09 of the diet that they can consume.
0:35:10 Whenever they’re curious about something,
0:35:13 they jump on the internet, they get the answer straight away
0:35:16 or from Alexa or from ChatGPT, they just get the answers
0:35:20 and that is massively useful for a few reasons.
0:35:23 One is that when you are curious about something,
0:35:26 you have the right cocktail of neurotransmitters present
0:35:28 to make that information stick.
0:35:31 So if you get the answer to something
0:35:32 in the context of your curiosity,
0:35:34 then it’s going to stay with you.
0:35:36 Whereas you and I grew up in an era
0:35:39 where we had lots of just-in-case information.
0:35:40 – What do you mean by that?
0:35:42 – Oh, you know, like just in case you ever need to know
0:35:45 that the Battle of Hastings happened in 1066, here you go.
0:35:46 – And you want to contrast that
0:35:47 with just-in-time information.
0:35:48 – Exactly.
0:35:51 – I need to know how to fix my car
0:35:52 and so the internet tells me
0:35:55 and then I can really remember it ’cause I need it.
0:35:56 – That’s exactly it.
0:35:58 And so, look, you know, for all of us with kids,
0:36:00 I know you’ve got kids, I’ve got kids and we feel like,
0:36:03 oh, my kid’s on YouTube and wasting time.
0:36:06 There’s a lot of amazing resources
0:36:08 and things that they learn on YouTube
0:36:10 or even on TikTok, anywhere.
0:36:12 There’s lots of garbage, of course,
0:36:14 but it’s better than what we grew up with.
0:36:16 When you and I wanted to know something,
0:36:20 we would ask our mothers to drive us down to the library
0:36:22 and we would thumb through the card catalog
0:36:23 and hope there was something on it there
0:36:25 that wasn’t too outdated.
0:36:27 – You were more ambitious than me.
0:36:28 I would just ask my mother
0:36:30 and I have since learned that every single thing
0:36:33 my mother taught me was completely wrong,
0:36:34 but I still believe them.
0:36:36 Because of this part of the brain
0:36:38 that locks in things that you learn long ago,
0:36:40 I still have to fight every day
0:36:42 against the false sorts my mother taught me.
0:36:45 I wish I had told her to take me to the library.
0:36:47 – My mother was a biology teacher
0:36:48 and my father was a psychiatrist
0:36:51 and so they had all kinds of good information.
0:36:55 I’m just super optimistic about the next generation of kids.
0:36:57 Now, as far as how we teach,
0:37:00 things got complicated with the advent of Google
0:37:03 and now it’s twice as complicated with chat GPT.
0:37:06 Happily, we already learned these lessons 20 years ago.
0:37:08 What we need to do is just change the way
0:37:10 that we ask questions of students.
0:37:13 We can no longer just assume that fill in the blank
0:37:15 or even just writing a paper on something
0:37:18 is the optimal way to have them learn something,
0:37:20 but instead they need to do interactive projects
0:37:22 like run little experiments with each other
0:37:26 and the kind of thing that you and I both love to do
0:37:29 in our careers which is, okay, go out and find this data
0:37:32 and run this experiment and see what happens here.
0:37:35 That’s the kind of opportunities that kids will have now.
0:37:40 – You were listening to a special bonus episode
0:37:43 of People I Mostly Admire with Steve Levitt
0:37:45 and the neuroscientist David Eagleman.
0:37:49 After the break, what are large language models missing?
0:37:51 – It has no theory of mind.
0:37:55 It has no physical model of the world the way that we do.
0:37:56 That’s coming up after the break.
0:38:08 (gentle music)
0:38:13 – David Eagleman is a professor, a CEO,
0:38:16 leader of a nonprofit called the Center for Science and Law,
0:38:19 host of TV shows on PBS and Netflix
0:38:21 and the founder of Possibillionism.
0:38:27 Like every curious person trying to figure out
0:38:29 what we’re doing here, what’s going on,
0:38:32 it just feels like there are two stories.
0:38:35 Either there’s some religion story
0:38:38 or there’s the story of strict atheism,
0:38:39 which I tend to agree with,
0:38:41 but it tends to come with this thing of,
0:38:42 look, we’ve got it all figured out,
0:38:44 there’s nothing more to ask here.
0:38:45 There is a middle position
0:38:46 which people call agnosticism,
0:38:48 but usually that means, I don’t know,
0:38:50 I’m not committing to one thing or the other.
0:38:52 I got interested in defining this new thing
0:38:54 that I call Possibillionism,
0:38:56 which is to try to go out there
0:38:58 and do what a scientist does,
0:39:01 which is an active exploration of the possibility space.
0:39:03 What the heck is going on here?
0:39:06 We live in such a big and mysterious cosmos.
0:39:09 Everything about our existence is sort of weird.
0:39:11 Obviously the whole Judeo-Christian tradition,
0:39:14 that’s one little point in that possibility space
0:39:17 or the possibility that there’s absolutely nothing
0:39:18 or we’re just atoms when we die,
0:39:20 but there’s lots of other possibilities.
0:39:24 And so I’m not willing to commit to one team or the other
0:39:26 without having sufficient evidence.
0:39:28 So that’s why I call myself a Possibillion.
0:39:32 – And so in support of Possibillionism,
0:39:34 maybe a better name could be in order,
0:39:37 you wrote a book called SUM, that’s S-U-M.
0:39:41 So it’s SUM, 40 Tales from the Afterlives.
0:39:44 How do you describe the book to people?
0:39:45 – I call it literary fiction.
0:39:48 It’s 40 short stories that are all mutually exclusive.
0:39:51 They’re all pretty funny, I would like to think,
0:39:53 but they’re also kind of gut-wrenching.
0:39:55 And what I’m doing is shining the flashlight
0:39:57 around the possibility space.
0:39:59 None of them are meant to be taken seriously.
0:40:02 But what the exercise of having 40
0:40:07 completely different stories gives us is a sense of,
0:40:10 wow, actually there’s a lot that we don’t know here.
0:40:12 In some of the stories, God is a female.
0:40:14 In some stories, God is a married couple.
0:40:19 In some stories, God is a species of dim-witted creatures.
0:40:22 In one story, God is actually the size of a bacterium
0:40:24 and doesn’t know that we exist.
0:40:27 And in lots of stories, there’s no God at all.
0:40:29 That book is something I wrote over the course of seven years
0:40:32 and became an international bestseller.
0:40:34 It’s really had a life to it
0:40:36 that I wouldn’t have ever guessed.
0:40:38 – When I heard about the book,
0:40:40 I saw the subtitle and I thought,
0:40:44 I have zero interest in reading a book about the afterlife.
0:40:47 I totally misunderstood what the book was about.
0:40:51 And then I certainly didn’t understand that some was Latin.
0:40:54 – Some actually I chose because among other things,
0:40:56 that’s the title story.
0:40:58 In the afterlife, you relive your life,
0:41:02 but all the moments that share a quality are grouped together.
0:41:05 So you spend three months waiting in line
0:41:08 and you spend 900 hours sitting on the toilet
0:41:10 and you spend 30 years sleeping.
0:41:11 – All in a row.
0:41:13 – Exactly, and this amount of time looking for lost items
0:41:14 in this amount of time,
0:41:16 realizing you’ve forgotten someone’s name
0:41:19 and this amount of time falling and so on.
0:41:20 Part of why I used the title sum
0:41:24 is because of the sum of events in your life like that.
0:41:26 Part of it was because Kojito or Gosume.
0:41:29 So it ended up just being the perfect title for me,
0:41:31 even if it did lose a couple of readers there.
0:41:41 – People are super excited right now
0:41:44 about these generative AI models,
0:41:45 the large language models.
0:41:47 What’s your take on it?
0:41:49 – Essentially these artificial neural networks
0:41:53 took off from a very simplified version of the brain,
0:41:55 which is, hey, look, you’ve got units and they’re connected
0:41:56 and what if we can change the strength
0:41:58 between these connections?
0:42:00 And in a very short time,
0:42:02 that has now become this thing
0:42:05 that has read everything ever written on the planet
0:42:07 and can give extraordinary answers.
0:42:10 But it’s not yet the brain or anything like it.
0:42:12 It’s just taking the very first idea
0:42:14 about the brain and running with it.
0:42:16 What a large language model does not have
0:42:19 is an internal model of the world.
0:42:21 It’s just acting as a statistical parrot.
0:42:23 It’s saying, okay, given these words,
0:42:25 what is the next word most likely to be
0:42:27 given everything that I’ve ever read on the planet?
0:42:29 And so it’s really good at that,
0:42:33 but it has no model of the world, no physical model.
0:42:38 And so things that a six-year-old can answer, it is stuck on.
0:42:39 Now, this is not a criticism of it
0:42:42 in the sense that it can do all kinds of amazing stuff
0:42:43 and it’s gonna change the world,
0:42:45 but it’s not the brain yet
0:42:46 and there’s still plenty of work to be done
0:42:49 to get something that actually acts like the brain.
0:42:52 – Do you think that it is a solvable problem
0:42:55 to give these models a theory of mind, a model of the world?
0:42:58 – I suspect so because there are 8.2 billion of us
0:43:00 who have this functioning in our brains
0:43:04 and as far as we can tell, we’re just made of physical stuff.
0:43:06 We’re just very sophisticated algorithms
0:43:09 and it’s just a matter of cracking what that algorithm is.
0:43:12 – If we were to come back in 100 years,
0:43:13 what do you think would be most different?
0:43:14 I know that’s a hard prediction to make,
0:43:17 but what do you see is transforming most
0:43:19 in the areas you work in?
0:43:21 – The big textbook that we have in our field
0:43:22 is called Principles of Neuroscience
0:43:27 and it’s about 900 pages and it’s not actually principles,
0:43:30 it’s just a data dump of all this crazy stuff we know.
0:43:34 And in 100 years, I expect it’ll be like 90 pages.
0:43:37 We’ll have things where we put big theoretical frameworks
0:43:39 together and we say, ah, okay, look, all this other stuff,
0:43:42 these are just expressions of this basic principle
0:43:43 that we have now figured out.
0:43:46 – Do you pay much attention to behavioral economics?
0:43:46 – Yes, I do.
0:43:48 – And what do you think of it?
0:43:49 – Oh, it’s great and that’s probably the direction
0:43:52 that a lot of fields will go is,
0:43:54 how do humans actually behave?
0:43:56 One of the big things that I find most interesting
0:43:59 about behavioral economics comes back to this issue
0:44:01 about the team of rivals.
0:44:04 When people measure in the brain
0:44:06 how we actually make decisions about whatever,
0:44:09 there are totally separable networks going on.
0:44:12 Some networks care about the valuation of something,
0:44:13 the price point.
0:44:14 You have totally other networks
0:44:17 that care about the anticipated emotional experience
0:44:18 about something.
0:44:22 You have other networks that care about the social context.
0:44:25 Like, what do my friends think about this?
0:44:28 You have mechanisms that care about short-term gratification.
0:44:29 You have other mechanisms that are thinking about
0:44:32 the long-term, what kind of person do I want to be?
0:44:34 All these things are battling it out under the hood.
0:44:37 It’s like the three stooches sticking each other in the eye
0:44:39 and wrestling each other’s arms and stuff.
0:44:41 But what’s fascinating is when you’re standing
0:44:44 in the grocery store aisle trying to decide
0:44:47 which flavor of ice cream you’re gonna buy,
0:44:48 you don’t know about these raging battles
0:44:50 happening under the hood.
0:44:52 You just stand there for a while and then you say,
0:44:54 “Okay, I’ll grab this one over here.”
0:44:56 – There was a point in time among economists
0:44:57 that there was a lot of optimism
0:45:00 that we could really nail macroeconomics,
0:45:04 inflation and interest rates and whatnot.
0:45:06 And we could really understand how the system worked.
0:45:10 And I think there’s been a real step back from that.
0:45:12 The view now is, look, it’s enormous complex system.
0:45:16 And we’ve really, I guess, given up in the short run.
0:45:18 Are you at all worried that’s where we’re going
0:45:19 with the brain?
0:45:20 – Oh gosh, no.
0:45:23 And the reason is because we’ve got
0:45:24 all these billions of brains running around.
0:45:27 What that tells us is it has to be pretty simple
0:45:28 and principal.
0:45:30 You got 19,000 genes, that’s all you’ve got.
0:45:32 Something about it has to be as simple
0:45:37 as falling off a log for it to work out very well
0:45:39 so often, billions of times.
0:45:44 – They say as you get older,
0:45:46 it’s important to keep challenging your brain
0:45:49 by learning new things like a foreign language.
0:45:51 I can’t say I found learning German
0:45:52 to be all that much fun.
0:45:55 And I definitely have not turned out to be very good at it.
0:45:57 So I’ve been looking for a new brain challenge
0:46:02 and I have to say, I find echolocation very intriguing.
0:46:06 How cool would it be to be able to see via sound?
0:46:09 I suspect though that my aptitude for echolocation
0:46:12 will be on power with my aptitude for German.
0:46:15 So if you see me covered in bruises, you’ll know why.
0:46:18 If you wanna learn more about David Egelman’s ideas,
0:46:20 I really enjoyed a couple of his mini books
0:46:23 like LiveWired, which talks about his brain research,
0:46:26 and some four details from the afterlife,
0:46:28 his book of speculative fiction.
0:46:32 – Hey there, it’s Stephen Dubner.
0:46:34 Again, I hope you enjoyed this special episode
0:46:36 of People I Mostly Admire.
0:46:37 I loved it.
0:46:41 And I would suggest you go right now to your podcast app
0:46:43 and follow the show, People I Mostly Admire.
0:46:46 We will be back very soon with more Freakonomics Radio.
0:46:49 Until then, take care of yourself.
0:46:51 And if you can, someone else too.
0:46:55 Freakonomics Radio and People I Mostly Admire
0:46:58 are produced by Stitcher and Renbud Radio.
0:47:00 This episode was produced by Morgan Levy
0:47:02 with help from Lyric Boutich and Daniel Moritz-Rabson.
0:47:04 It was mixed by Jasmine Klinger.
0:47:07 Our staff also includes Alina Kulman, Augusta Chapman,
0:47:09 Dalvin Abouaji, Eleanor Osborne,
0:47:12 Ellen Frankman, Elsa Hernandez, Gabriel Roth, Greg Rippen,
0:47:15 Jason Gambrell, Jeremy Johnston, John Schnars,
0:47:17 Neil Coruth, Rebecca Lee Douglas,
0:47:20 Sarah Lilly, Theo Jacobs, and Zach Lipinski.
0:47:22 Our composer is Luis Guerra.
0:47:24 As always, thank you for listening.
0:47:33 – David, you got your quick time going?
0:47:35 – I do now.
0:47:42 – The Freakonomics Radio Network,
0:47:44 the hidden side of everything.
0:47:49 – Stitcher.
0:47:51 (upbeat music)
0:47:53 you
David Eagleman upends myths and describes the vast possibilities of a brainscape that even neuroscientists are only beginning to understand. Steve Levitt interviews him in this special episode of People I (Mostly) Admire.
- SOURCES:
- David Eagleman, professor of cognitive neuroscience at Stanford University and C.E.O. of Neosensory.
- RESOURCES:
- Livewired: The Inside Story of the Ever-Changing Brain, by David Eagleman (2020).
- “Why Do We Dream? A New Theory on How It Protects Our Brains,” by David Eagleman and Don Vaughn (TIME, 2020).
- “Prevalence of Learned Grapheme-Color Pairings in a Large Online Sample of Synesthetes,” by Nathan Witthoft, Jonathan Winawer, and David Eagleman (PLoS One, 2015).
- Sum: Forty Tales from the Afterlives, by David Eagleman (2009).
- The vOICe app.
- Neosensory.
- EXTRAS:
- “Feeling Sound and Hearing Color,” by People I (Mostly) Admire (2024).
- “What’s Impacting American Workers?” by People I (Mostly) Admire (2024).
- “This Is Your Brain on Podcasts,” by Freakonomics Radio (2016).