Video: The Neuroscience of Language and Thought, Dr. George Lakoff Professor of Linguistics (DocDrop)

00:00:09

or extremely important in the intellectual development of cognitive psychology and much of where we are now on artificial intelligence and all those sorts of things anyway and with that I will please help me join join me in

00:00:23

welcoming George Lake off [Applause] well now that you've heard the talk thank you all for coming it's really a pleasure to be here and to

00:00:44

see you can we raise the house lights a little bit because I would like to see who I'm talking to a bit I don't like speaking to a dark audience so Justin a

00:00:54

little bit of light out there this is a remarkable time and I think it's important to understand what is happening in the neurosciences and the

00:01:08

cognitive sciences and what isn't happening so let me mention for example what happens when you see one of these so-called pictures of the brain you've all seen these things they have red and blue and yellow and so on there's no red

00:01:21

and blue and yellow in the brain what happens is this if the pictures are not pictures of the brain they're pictures of gia the ratio of oxygenated to

00:01:35

deoxygenated blood in certain places and when neurons fire they need to be getting some oxygen back in there too so they can fire again and it turns out that they have different magnetic

00:01:49

properties if their oxygen if the blood is oxygenated or not and what you see actually is a bunch of pixels and each little pixel is actually 3 millimeters

00:02:03

by 3 millimeters by 3 millimeters 3 millimeters cubed and it goes over anywhere between one second and several seconds now if you ask how many neurons are in there the answer is about 125

00:02:17

million per pixel each neuron is connected to between a thousand and 10,000 other neurons so there are tens of billions of connections lots of circuitry in that one little

00:02:30

pixel and that picture doesn't show you what's going on in that circuitry very important you know we cannot see that we can say hey something is happening there we

00:02:42

don't know exactly what neural computation allows us to model what it is that's happening there and to ask a deep question and the question is how is

00:02:54

it possible for neurons which are just these things that fire and you know propel axons on to the next neuron how do you get ideas out of neurons even if

00:03:07

there are a hundred billion of them and trillions of connections how do you get ideas out of them and that is what we are have begun to figure out so we're going to talk about that idea today and

00:03:20

it turns out it has lots of consequences including consequences for politics which I've been writing about for some time and but I want to talk about the science behind it today the science

00:03:33

behind it is actually quite remarkable because it makes a radical change what it changes is our understanding of what it means to have reason what it means to

00:03:44

be a human being and what it means to be able to reason using a brain and a body now there's something shocking here I and many of you were brought up with the

00:03:57

idea of Enlightenment reason critical thinking an Enlightenment reason had a number of properties and it turns out that most of them are not true

00:04:10

Enlightenment reason is useful in many situations we'll talk about why it's useful and why it's been popular and so on but it's inadequate grossly inadequate for understanding what hew

00:04:22

means for human beings to understand the world and to think so what we're going to do is talk about real reason which is coming out of the neural and sciences and what the properties are so

00:04:36

there's a certain myth that comes out of enlightenment reason it says you know I think therefore I am says Descartes reason is conscious you know what you think it's just not true for most of

00:04:50

your thought it's unconscious mainly about 98% consciousness is a tip of the iceberg you know how do you get 98% there are two ways one if you look at

00:05:03

what you're conscious of versus what your brain is doing the roll is about fifty to one your brain is doing 50 times as much as you're conscious of and there's another way to look at it if you

00:05:15

take a sentence and you say what can the next sentence be in a paragraph and what do you have to fill in to understand all the possible next sentences the answer is that you need to fill in 50 times as

00:05:29

much as it's in that sentence roughly so it's about 98% unconscious you're not even aware that you're filling this in but you're not moreover consciousness could not in

00:05:42

principle in principle be you know you you couldn't have reason being conscious because most of your reason is done in parallel circuitry but consciousness is

00:05:55

linear so you have massively parallel circuitry but you're tracing out a linear path through it and that is means you can only be aware of a tiny portion of what you're thinking now

00:06:10

there's another myth and that is that thought is abstract that the concepts are abstract and disembodied as if they're so kind of floating in air and that's not true thought is physical it is a matter of

00:06:23

brain circuitry it's what the brain is doing and that's important because how can thought be meaningful and as you'll see it's made meaningful via connections to the body

00:06:35

in ways we'll talk about in a few minutes the next view that is a shocker is that you can reason directly about anything in the world you know you have

00:06:49

this capacity for reason and you say well anything in the world whether it's in politics and in economics in the environment in human relations we can reason directly about it

00:07:01

it's not true because you think with an embodied brain reason is indirectly connected to the world not that it's disconnected so it can be you can

00:07:14

imagine things but you can only reason about and understand what your body and brain allow you cannot understand just anything at all and this as you'll see

00:07:27

is crucial in politics if you have a certain political ideology on one side or another you may not be able to understand what other people with a different ideology are saying you may

00:07:41

not be able to hear a fact and make sense of it if you have a view of the world in which that fact does not fit you can't understand just anything and you'll get examples of that in a while

00:07:55

now ideas are main meaningful according to the old theory because they have direct connections to the external world somehow this abstract thought supposedly just goes directly to the world which

00:08:08

isn't true what really happens is that ideas are made meaningful via the connections to our body and embodied experiences now how can that work well let's look at some of the ways and can

00:08:21

work first before we get to that we need to talk a bit about emotion in the old view of enlightenment reason emotion got in the way of reason and motion was the

00:08:34

enemy of reason reason was what you know sort of like mr. Spock on Star Trek you know who is you know super reason no emotion or whatever not true suppose

00:08:49

that you had a stroke or a brain injury that wouldn't allow you to feel emotion and there are such strokes and brain injuries rep Antonio Damasio and his

00:09:01

wife Hana figured out some years ago and published in a book called des cartes error is that you can't reason without emotion emotion is necessary and it's

00:09:14

easy to see why if you cannot feel emotion then like and not like mean nothing to you and you do not know what to want think about it

00:09:27

if you couldn't feel anything if you wouldn't know what it meant to like or not like something or if somebody else or you couldn't tell if someone else would like or not like what you were doing you wouldn't know what to want you couldn't set a goal and this is what

00:09:41

happens to people with such brain injuries they act randomly they don't know how to plan they don't know how to structure their lives or set rational

00:09:53

goals because rationality requires emotion very very deep finding now the myth was that reason uses mathematical

00:10:06

logic and I spent well over a decade working to try to see if mathematical logic would work for the way that we reasoned in natural language it failed and it failed hugely and I can tell you

00:10:21

lots of reasons why it failed and where it failed and so on but will maybe go through a few of those here and there but the big result is that reason uses logics

00:10:32

of what are called image schemas image skin and I'll give you an example of this in a minute frames conceptual metaphors prototypes and narratives now what are these things

00:10:44

let's take a few minutes an image schema is a brain structure that structures imagery so for example we are in an

00:10:55

auditorium it is a container I have here a glass of water that's a container for the water there are lots of containers around the world

00:11:07

windows are containers rooms are containers your body is a container how do you understand them all as containers now what has been discovered by Terry Revere

00:11:22

who is in art worked in our lab on his PhD thesis back in the mid 90s is that there is a brain structure that can compute any kind of container but no

00:11:35

matter how big or how small wherever it's located whatever its shape is there's a simple brain structure that puts together certain vertical topographic maps of the brain that will

00:11:46

compute any container it'll tell you what the interior is the boundary the exterior now that is called an image schema it structures your mental imagery and but and there's lots of others there

00:12:01

are paths with a source path and a goal so if you talk about going into something you have a a path that begins outside the container and winds up inside no big deal but it's interesting as we'll see

00:12:14

because they're in diff parts of the brain and the question is how do they get together and we'll talk about that image schemas also structure space so there is in front of and back

00:12:26

up behind above below those are structured relative to the body the body is represented in the brain up here primary motor cortex there is an actual

00:12:39

map of the entire body neuron by neuron in the body connected to neurons up here that map allows you to locate things in space relative to your body that is what

00:12:54

allows you to talk about things being in front of or behind or next to and so on things like that those are called image schemas and then there are process schemas called aspectual schemas but

00:13:07

tell you what a process is or what is the precondition where what are you starting with what is the middle are you iterating it doing it over and over is the process long or short has it met a

00:13:19

goal have you reached the end is there a consequence of that process is it part of a bigger process those are structures that have been studied figured out and that are computed by neural circuitry of

00:13:33

your brain every single language in the world uses the same structures and the same elementary image schemas show up in languages all over the world because

00:13:47

they're part of our human neural structure so image schemas are parts of that but also they all allow certain reasoning if you know that one a

00:13:59

container a is inside container B and you put something inside container a you know it's going to be inside container B that is if you know you have a closet inside your house and you put your

00:14:13

tennis racket inside your closet your tennis racket is inside your house okay that has to do with the logic of the container schema and we know how that logic functions

00:14:25

and it functions Norley so a lot of your reasoning is coming out of those schemas very important to see that those schemas structure bigger schemas called frames

00:14:38

now frames were figured out in the mid-1970s in by a number of people but in particular Charles Fillmore a

00:14:51

linguist in my department and one of the world's great linguists figured out that every word is defined in terms of a structure like the frame and let me tell you what a frame is it's very simple

00:15:05

imagine your understanding of a hospital well there are certain roles played by people in that hospital there are doctors and patients and receptionists and nurses there there are rooms like

00:15:18

operating rooms and reset you know there are instruments like scalpels and so on and there are things you know about hospitals you know for example that surgeons operate on patients in

00:15:32

operating rooms with things like scalpels okay every frame is a structure like that and if I say the word surgery immediately everything about it will

00:15:44

come up you'll know that there's a surgeon there's a nurse there's a patient there's an operation going on etc you know something and a lot what one word is defined relative to a frame

00:15:57

what film are discovered was that every word in every language is defined relative to some frame that is a crucial discovery and what's really important

00:16:09

about that and and important for politics is that well consider the following when I teach this the first thing I do is I give my students an

00:16:21

assignment I say ok everybody ready don't think of an elephant you all can do that right you had a problem because the word elephant evokes an image and knowledge of frame in which

00:16:39

you understand what an elephant is and if you negate the frame you have to activate it anyway okay the word activates the frame even if you negate it and that's important in for political

00:16:54

reasons that we'll come back to in a while but words activate frames and fern frames are a ways in which you structure the world you cannot think without

00:17:05

frames you cannot speak without frames being there and those frames are physical there are circuitry in your brain that carries out all those inferences and imposes that structure and that

00:17:20

circuitry once you learn a frame is there mostly for life now for many of these cases they are just there all the time and we'll talk about how that can happen

00:17:32

another very important part of what is not in mathematical logic is metaphor and metaphor used to be thought of as something in poetry or rhetoric or and

00:17:44

in terms of language but what mic ready and I discovered back in 1978 was that actually it's a mode of thought that linguistic metaphors to sort of surface

00:17:57

realizations of more complicated metaphors the first example that I looked at was a very simple case in which there were linguistic expressions

00:18:09

like our relationship has hit a dead end street the marriage is on the rocks it's been a long bumpy road where love is understood as a journey now there are

00:18:22

a lot of expressions in which love is understood as a journey right rather crossroads in this relationship we're going in different directions we may have to bail out right

00:18:34

lots of linguistic expressions but there's a generalization here the generalization is that in every single case the lovers are travelers in every

00:18:46

single case the relationship is a vehicle in every single case the lovers are supposed to have common life goals that is are there understood in terms of destinations they're trying to reach and

00:18:59

in every case difficulties in their relationship are understood as impediments to motion toward those common life goals so if you say we're spinning our wheels in this relationship what happens you have an image of a car

00:19:14

the car isn't moving the wheels are turning you want to get it moving you're putting energy into getting and moving and you're frustrated that you take that and you say ok let's map that from

00:19:26

journeys to love and that says we have a love relationship but it's not going anywhere we want it to you know go somewhere to reach our common life goals not happening and we're frustrated that

00:19:40

is there's a mapping from journeys to love and that image allows that your knowledge of that image to map to love and reason about love in terms of

00:19:54

journeys ok how many of those metaphors are there in your brain tens of thousands and then we're going to point out in a little while that they have

00:20:06

elementary metaphors that that make them up there are actually smaller metaphors that fit together but we'll get to that in a while the point is none of this is in enlightenment reason none of this is

00:20:21

in logic none of this is in what was called critical theory you know when when I was going to college where they you know the idea was to say oh you have to learn to to reason logically about

00:20:35

things it turns out people think in terms of these structures that are there in your brain that are not part of enlightenment reason now there's another myth and the myth is

00:20:48

that enlightenment reason is there reason is there to serve your interests that's why we have it that's the usual assumption it's partly true reason can

00:21:00

serve your interests no question but reason can also serve empathy and social connection as well as self-interest and let me tell you how this was discovered

00:21:11

because empathy turns out to be physical back in 1996 in Parma Italy a remarkable thing happened in the neuroscience lab

00:21:24

there and I've been there Vittorio galazy was running an experiment for professor ritsu lottie's in which they had macaque monkeys and they were

00:21:36

training the monkeys too they trained them to push buttons and peel bananas and grab rings and do certain things and then they had probes in the monkeys brains neuron by neuron in what is

00:21:50

called the premotor cortex which choreographed actions it sort of puts actions together and they were looking at which neurons fired when the monkeys did which actions they had it hooked up to a computer so every time the monkey

00:22:03

moved and did this versus peel the banana etc you would see exactly which neurons were firing the monkey breast something you would see which neurons are firing and if he grasped with the right hand to be one set left hand

00:22:15

another set et cetera and this was going just fine then it was lunchtime and in Italy you take a nice lunch so you take a nice lunch for toriel comes back didn't have dessert as a pile

00:22:29

of bananas thank you take a banana he starts to peel it and he finds that the computer is registering click click click click click that the monkey's brain which we're supposed to register

00:22:42

register monkey movements is registering his movements and when they checked it out it turns out these were exactly at least 1/3 of the neurons firing 30%

00:22:54

actually will were the neurons for peeling bananas then they did further experiments and they found out that there was an interesting phenomenon that

00:23:06

there's a connection between vision and action that the same neurons that are firing when you peel a banana are firing when you see someone else peel a banana

00:23:18

well not quite all of them 30% the other 70% are doing interesting variations on that but those 30% are interesting they're called mirror neurons they mirror what you're doing and they link

00:23:31

vision in action and then they found out why there is a neural pathway Direction directly between the premotor cortex and the parietal cortex just behind it that

00:23:45

integrates a vision and they are you know tuned as you're growing up to length vision and action so that vision and action are linked together but then the other interesting thing is that

00:23:58

mirror neurons are connected to the emotional regions and emotionally something wild was discovered back in

00:24:09

the 1950s and 60s by Paul Ekman namely that we have a physiology that corresponds to our emotions Darwin first hypothesized this that around the world

00:24:23

when people are happy they smile when they're sad they frown when they're angry they bare their teeth etc and animals do a lot of the same things

00:24:35

and that basically there are other physiological adji of emotions when you're angry your skin temperature rises half a degree which is why you say my blood boils you know why you get burned

00:24:49

up and so on there's a reason for that those metaphors are for anger are based on your actual physiology and that physiology of anger is tied to the mirror neurons because the physiology

00:25:02

has to do with what your body is doing and if it's connected to vision you can see with someone else's body cuz doing and you can tell if somebody is writhing in pain or deliriously happy or really

00:25:15

sad or whatever you are emotionally connected to other people Briah your brain you have a brain and everybody else has the capacity for

00:25:29

what's called empathy it's a physical capacity linking you to others and that's how social connection works that is a physical thing and actually most

00:25:41

the reason Oh a huge amount is based on social connection and empathy it's not just self-interest and that's important for many reasons having to do with

00:25:52

morality and with politics next the it was assumed that rational thought was value free it was just you know about what was true in the world and so

00:26:06

on and it's not because you're concerned with empathy and morality when you're thinking and and with emotion rational thought is really value-based

00:26:19

now whoa let's go back one there's a further view Descartes assumed that all rationality was the same for everybody that it was universal that what made us

00:26:34

human beings was being rational and logical and that therefore everybody had the same mode of reasoning now if that's true then all you have to do let's say

00:26:47

in politics is tell something somebody to facts and everybody will reason to the same conclusion right sure not true that is if enlightenment reason were

00:27:01

right that would happen it doesn't happen it turns out that it is not the case that all concepts are you are universal and that all of them are

00:27:13

accessible to everybody and then there's another view of you coming out of the postmodern thought that says no concepts are universal they're all arbitrary that's also false many concepts are

00:27:26

universal and many are not a very and it's an empirical question which is which and they vary from language to language and that's important you need to know that it is not the case that

00:27:41

everything with all thought is universal and it's not the case that that's all language particular or particular that is you need to know empirically what part is shared with other people who

00:27:53

speak other languages and come from other cultures and what part is not and that is an important empirical study so the big myth is this but there is some objective rational structure to the

00:28:07

world out there and that human reason can fit it directly and characterize it literally without frames or metaphors and reason about it adequately with formal logic alone that's just false the

00:28:21

world is real that is whatever it is our bodies and brains provide understandings of the world which depend on frames metaphors image schemas prototypes narratives all of those things and it is

00:28:35

that that permits us to create mathematics and to create scientific theories of the world mathematics is not just out there mathematics works by the same mechanisms and if you're interested

00:28:48

there's a book called where mathematics comes from which goes through it in very very great detail now what about language as you heard a

00:29:02

little bit before Chomsky came up with the view that language is purely formal and abstract that it's just a bunch of meaning meaningless symbols that the grammatical structure of a language is independent of meaning in communication

00:29:16

and that's not true language is a physical product of the embodied human brain and dramatical structure makes crucial use of meaning and communication so let me give you

00:29:28

some examples of how meaning works in grammar so let's take an example like a simple sentence the ball went to left field now the to left field is what's

00:29:41

called a directional adverb tells you where it's going and and in there are directional adverbs usually go along with motion they have to do with direction of motion and the thing that

00:29:53

moves and in this sentence it happens that the thing that moves in the motion are represented in the sentence itself the ball moving went is has to do with going that's motion to a field and so

00:30:06

the question is is the occurrence of this directional adverb to left field just dependent upon what's in the grammar of the sentence is that true

00:30:18

well now I started looking into this in 1963 god help me long time ago and in 1963 there was a left fielder a rookie left fielder for

00:30:33

the Boston Red Sox called Carl Yastrzemski he's now in the Hall of Fame it's been a long time some of you may be old enough to remember Carl Yastrzemski great hitter and the sentence was yes

00:30:46

rum ski double to left now if you know baseball you know what that means and that the notice that the two left they will talk about why it's not left field you could say to left

00:31:00

field Yastrzemski doubled to left that directional adverb does not have a ball or a verb of motion in it double is not a verb of motion and it's yeah Yastrzemski who's going to left if you

00:31:13

say yes jump ski doubled off the wall it's not yes gem ski bouncing off the wall it's the ball the word double has within its meaning the idea of the ball

00:31:25

going somewhere and the to left field tells you where the ball went it depends upon the meaning of the sentence not just the elements in the grammar of the

00:31:37

sentence itself that is a counter example to the old theory that says it's all just arbitrary symbols it has to do with meaning and that's crucial let's

00:31:50

take another example if you're a linguist you study things like existential versus baked big-block locative constructions let me give you an example so there is a God as an

00:32:02

existentialist says it got a God exists and notice you can negate it we can say there isn't a god or you can question it is there a God that we can say there's a God isn't there all right normal that's how

00:32:16

normal sentences work but there's another kind of sentence like there's Harry on the bench over there that's a date two o'clock it's pointing to Harry and that there isn't

00:32:28

existential who doesn't say Harry exists it's pointing out where he is okay now can't negate that yet say there isn't Harry on the bench or you can say if you

00:32:42

say something here comes the bust you can say here doesn't come the bus I can't negate it you can't question it is there Harry on the bench or it does here

00:32:53

come the bus impossible you can't have a tag question like there is Harry on the bench isn't there all right impossible here comes the bus doesn't here no good

00:33:05

hey now why is this just a matter of arbitrary symbols moving around or does it have something to do with meaning well think about what addictive locative is it points out something what you're

00:33:21

doing is assuming the following you're communicating to somebody else and you share a visual field and what you're doing is is pointing to a location in that visual field you're trying to get

00:33:34

the other person to look at that location and you're saying what's in that location there is Harry on the bench here comes the bus okay that's what you're doing what would it mean to

00:33:46

negate it it couldn't work because it's existing in the visual field it's there and since it's there in front of you it's obviously in inconsistent with negation since it's there in front of

00:34:01

you and you're pointing it out it's inconsistent with questioning if you have an attack question like isn't isn't he or something like that or isn't there that is doubting something but this is

00:34:12

indubitable because it's their present for you the meaning of the construction is telling you it's grammar the meaning of the grammatical construction is

00:34:25

telling you what the grammatical possibilities are and that's profound we give you another example there's a phenomenon called negative

00:34:38

polarity items like a red cent in Harry doesn't have a red cent you don't say Harry has a red cent or Harry doesn't give a good goddamn about that not Harry gives a good goddamn about

00:34:50

that and so on now so you don't say Harry has a red cent or I wouldn't do a damn thing for him but not I would do a damn thing for him now there is an

00:35:04

interesting phenomenon where you have sentences like why paint your house purple which means you shouldn't do it unless you have a good reason versus why not paint your house purple which means

00:35:17

you should do it unless you have a good reason notice that the positive one why paint your house purple is a negative suggestion you shouldn't do it and the

00:35:29

negative one that has a negative in the sentence why not paint your house purple is a positive suggestion so the question is does do this does the negative polarity item like a red cent or a damn

00:35:42

thing and things like that does it go with the meaning or does it go with the actual element of the sentence in the grammar doesn't go with the symbols well look at the sentences why give Harry a

00:35:54

red cent but you don't say why not give Harry a red cent well I do a damn thing for him but you don't say why not do a damn thing for him that is the negative polarity items a damn thing

00:36:08

in a red cent are going with the grammatical positive but the semantic negative it's going with the meaning not with the grammatical form the meaning is

00:36:21

determining how this grammatical element works meaning is determining grammar now that's important because it goes along with all the other things we're saying

00:36:34

what is meaning meaning has to do with what your brain is doing that is grammar is going to turn out that have to do everything to do with how you understand things in the world with your brain and

00:36:47

that's very interesting so let's look at some results from neuroscience and look at some of the brain basics because at this point we want to ask how do we make sense of all of this how does it work

00:37:00

what are we doing and what is the present state of the research on it so here are some basics color is not in the world there are no colors out there in

00:37:13

the world there is no green ingress okay there is no blue in the sky there is no red in the sunset our in the Rose etc

00:37:26

color is our product of several different things there are wavelength reflectances that is each object reflect certain wavelengths but individual wavelengths aren't colors colors are

00:37:41

something new since you know and they go across many wavelengths so there are we wave length reflectances color depends upon things in you in your retina there

00:37:53

are certain color cones that respond to certain ranges of wavelengths you have to have the color cones to see color if you don't have all your color or your three color cones your works called

00:38:05

colorblind next there are neural circuitry connected to those color cones and that neural circuitry is what computes color color doesn't exist without the color

00:38:18

cones the neural circuitry and the reflectances in the world it's an interaction between the world and you and your through your body and brain but the color is only in your mind in your

00:38:32

brain in your understanding of the world it isn't out there independent of you that's wild now again go back to the idea that

00:38:46

meaning is a matter of truth in the world and that if you think about it a sentence like the chair is green in a traditional view coming out of

00:38:59

anglo-american philosophy it would say that's true just in case the entity referred by the referred to in the world by the word chair is in the set of green things in the world there is no set of

00:39:14

green things in the world that account of meaning cannot be true but it's taught in most philosophy classes in the

00:39:25

West it's taught as the theory of meaning but it's false let's continue we saw in the mirror neuron cases that action and perception are linked with the same neurons are

00:39:42

firing during actions and perceptions of the same actions that's very important because it turns out that concepts have everything to do with that and so let's

00:39:56

continue as a result that is why there are things called basic level categories so let me tell you about basic level categories it was discovered by eleanor

00:40:09

rosch back in the mid-1970s that category hierarchies like furniture chair and desk chair or like vehicle car

00:40:23

and sports car had different properties that at the middle level of car and chair you could get mental images all right you can all get a mental image of

00:40:36

a chair you can all get a mental image of a car can you get a mental image of a general piece of furniture that is neutral between chairs and tables and

00:40:48

beds and and lamps you can't do it okay can you get a mental image that is neutral between cars and boats and planes and motorcycles and bicycles and

00:41:04

all kinds of vehicles can't do it the highest level of which you can get a mental image is that middle level okay and not only that there are motor

00:41:17

programs you are now eggs and using your motor program for sitting in a chair if you're driving your car you have motor programs for driving a car okay those you don't have general motor

00:41:31

programs for interacting with generalized pieces of furniture you have different ones for chairs tables bed lamps and so on okay so there's a level

00:41:43

at which you have mental images and you perceive things by your stall perception at this middle level called the basic level and it's what it's the level at which children first learn concepts why

00:41:56

should you have what a motor programs have to do with your stall perception and mental images the answer is mirror neurons which link actions motor actions

00:42:08

to what you can see and what you can imagine mirror neurons explain why there should be this phenomenon right very profound result of that that shows that

00:42:22

there's something about the nature of categories themselves that has to do with the neural system now another thing has to do with verb routes here are two various languages you'll have different

00:42:35

endings like you say something like you know he ate he ate he is eating he eats things like that but the word eat is the same and they might vary a little is

00:42:49

like run and ran and things like that but in general the root of that run and runs is run the same why shouldn't now if you think about it in one case if I

00:43:03

say I'm doing it I'm running that's about my action but I say he's running it's about my perception of him or my imagination of him it's not about my action why should you have the same verb

00:43:17

forms for action and perception there are neurons the same neurons are firing there's a reason why language is structured that way

00:43:29

let's take some more we have this action and now this is really wild Martha Farah the University of Pennsylvania in 1990 made a remarkable discovery

00:43:42

when you think about it it's not so remarkable but it had to happen what she found was that when you imagined seeing something let's say she was doing mental

00:43:54

imagery experiments he would remember pictures and he asked you to memorize a picture and then imagine it in your brain and then she did fMRI studies brain scans and found that exactly the

00:44:08

same parts of the brain were active when you're actually seeing as when you were imagining seeing and then it was discovered that the same is true for moving the same part of the brain is active when you imagine kicking a ball

00:44:22

and you actually kicked well when you imagine biting or when you actually bite and this is not only true for imagining and seeing it's also true for dreaming and remembering and the use of language

00:44:36

there's the same part some of the same parts of the brain are used when you either imagine remember dream of doing something and you actually do it we're talking a little bit about why that's

00:44:48

true but it's a remarkable fact that this is the case and this allows us to get a theory of meaning what it says is that let's suppose I give you a sentence

00:45:02

like Harry drank a glass of water if you can't even imagine what it what it means to drink a glass of water then you don't understand this happens

00:45:14

what that says is that meaning is a form and understanding involves mental simulation that is the use of that same part of the brain that you would have

00:45:27

for doing it that's remarkable but it's an interesting theory of meaning and that is the basis of the neural theory of meaning in the neural theory of language and it's crucial for that and

00:45:39

it's profound in many ways which we'll see shortly so one of the ways in which you study this how do you go about studying something like this well what

00:45:54

Jerry Feldman figured out and was the following that there's a level of organization in the brain called a neuronal group or a cluster of neurons

00:46:06

for example when I was going through the data for the monkeys what the monkeys were doing when they were peeling bananas and so on in the experiments at parma vittorio believes they visited us

00:46:21

at berkeley and we went through all the basic data we could actually see that there were certain clusters of neurons that were actually used in all of these cases it's not just one neuron that's

00:46:34

used for peeling a banana it's anywhere between a few dozen and a few hundred and study these neuron by neuron and you can see how many there are for different actions but it turns out

00:46:47

that these clusters overlap so for example they studied and what the monkeys were doing when they were trained to either eat a peanut or when

00:47:00

they were trained to have a cup over here and put the peanut in the cup so in the one condition you would eat the peanut put it in the cup they notice the actions overlap and the neurons that are

00:47:14

firing overlap in the same cluster that is neurons in the same cluster could do different things the same neuron could be doing something different in a different situation but it takes

00:47:27

number of neurons between a few dozen a few hundred to carry out any action okay what that suggests is the following kind of theory that you have what are called

00:47:39

functional circuitry and that the neurons used to carry out a particular action are what are called a node that's a cluster of neurons that carry out the action and the nodes can overlap but

00:47:54

then the question is at the level of the node are there generalizations you can see in the circuitry and the answer is yes and that's where we're going that's

00:48:06

where the current theory has been so we're going to talk about nodes functional subjects and functional circuitry now to get there we need to

00:48:19

talk about some brain structure that is quite remarkable what are called topographic maps and these are no structures that preserve closeness and

00:48:33

let me try to give you any some examples of this in the primary visual cortex back here you have lots of neural connections coming from your retina going through the LGN to the opposite

00:48:47

side of the brain both sides and each neuron coming out of the retina corresponds to a neuron a single corresponding eye on here and the one right next to it is right next to it

00:48:59

here not necessarily in the same orientation but right next to it it'd be they preserve closeness this is not a miracle how does it happen

00:49:11

and none like that they can see neuroscientists have looked at cat embryos and can see the development and Cincy the axons coming from the eye going through here looking for the right

00:49:25

spot to land back here and they strike here that's right here and then they land in a certain place and when they wind up they preserve closeness how does this happen it happens because

00:49:38

coming out of the eye there is too is a chemical gradient that as to do with two perpendicular places where the chemicals are extruded in us and the closer they are to the place they're excluded that

00:49:51

thicker they are and so there's a degree of thickness at every point so every axon coming out of here going to there gets attached to a pair of chemicals of

00:50:05

with a certain degree of thickness a certain amount of chemicals and the corresponding chemicals that bind to those chemicals are extruded over here and they have to fit exactly in order

00:50:19

for them to land and bind it's not a miracle it's chemistry right that is how it's possible to have a topographic map and that it would preserve closeness

00:50:31

that is the map pops here preserves closeness in the retina now you get them also in the auditory cortex because neurons coming from the inner ear that

00:50:44

go around there the inner ear is like a big horn and there are hairs along their neurons connected to that and liquid inside that vibrates when you hear

00:50:56

something and you have different frequencies of along the inner ear that make the hair vibrate either a lot and and also the various waves going through there give you frequencies at different

00:51:09

locations there those neurons are connected to the cortex or the auditory cortex and that preserves closeness of

00:51:21

frequency and closeness of intensity so that the auditory cortex is a map of frequency and intensity in what you hear pretty well hey that's those and we have

00:51:36

maps not just in our brains for lots of these things we have maps that preserve frequency spatial location movement speed all sorts of things like that that

00:51:49

were preserving input and they're all through our brain through our cortex you have ridges where there might be seven maps one right after another of the visual field Mike silver and his lab at

00:52:04

Berkeley studies what happens in visual maps and you can see when you flash a light that moves and you say attend to this light in those maps you can actually see the activation of the

00:52:16

neurons going across the maps but they're in different locations in different maps and they're still trying to figure out what the different maps are doing but they can find the maps there more over there throughout your

00:52:28

body so what it's called phantom limb syndrome if you lose part of an arm you can feel pain in a limb that isn't there why because there's a map of the hand

00:52:41

that you were born with that comes down here and when you lose it in that map it preserves closeness of the neural connections in your fingers that's how a

00:52:53

phantom limb syndrome works okay we have maps throughout our brain and our body all over the place now that is crucial for understanding how

00:53:06

meaning works okay so now within each map there are connections in the neurons in the map and connections across maps both happen okay and there are certain

00:53:22

functional types that occur that link across maps so let me tell you about some of these circuit types because these circuit types there are four types these circuit types we think they're all

00:53:35

that you need to characterize conceptual thought in the right places and that's we don't know if this is true this is a hypothesis and the hypothesis goes like

00:53:48

this if you think about a frame like you know for let's say the hospital frame with doctors and nurses and so on there's certain semantic roles and each of these semantic roles like doctor and

00:54:02

nurse evokes the entire frame and the hospital frame evokes each of those activates those the entire hospital frame is what we call a gestalt the

00:54:14

circuit meeting to do that if that whole thing is active let's call that a Gestalt node and the others semantic role notes if the Gestalt note is active for the hospital then all of the others

00:54:27

are active you can think of doctors nurses and so on if you talk about surgery then you're in the hospital it activates the hospital node and so on that is each of the roles activates the

00:54:39

other that creates a Gestalt of a hospital it's a structure and other things aren't in there you know so you know you talk about a hospital and a

00:54:54

pine tree or a desert is not in the hospital is that part of the hospital frame it's not there there's a gestalt for that and for every other structure that you understand you have that that

00:55:07

is what a gestalt circuit does and then we'll see that it does other things too what it does is bind things together to form whole your thoughts out of particular roles there's a binding

00:55:20

circuit now to understand a binding circuit take a simple example the example is the word into versus the

00:55:32

words in and the words word to okay in has a container schema at as an interior boundary and exterior in basically you have are having three maps

00:55:45

which were tied together one activating the interior when activating the boundary when activating the exterior and they're tied together as a single collection now that is in a different

00:55:57

part of the brain than the map that characterizes a source of path and a goal of motion so there's a map that says oh here's where you started here's all the places you've been we're going

00:56:11

to activate those neurons and here's where you wind up that's the goal okay and those are into two different parts of the brain - in two different maps there's a third map where these are

00:56:25

brought together the into map says that the goal is in the interior well how how does it goal get in there again move nothing in the brain moves from one place to another you have to have neural connections between these two maps and a

00:56:38

third map and the third map the goal maps to a certain point and the interior activates that point and if you get enough activation from both of these to make the neuron here fire then that says

00:56:53

you have an in - and this third map is going to fit your image of the world it's going to fit your visual understanding of it or your imagination so if you can imagine pouring water into a glass there's a path of the water from

00:57:07

a pitcher into the glass and the structure of the glass as a container is in one place this path of the water is in another place and there's a map that says they're coming together right here

00:57:20

that's the only way you can understand into and then it's a word into that says hey you activate that now that is how

00:57:31

it's possible for a brain a physical thing to have a concept like into which involves two different structures as simple as that is but that is how that

00:57:45

is going to work okay now one of the things these circuits do is they form these schemas frames and complex concepts and let's talk a little bit

00:57:58

about a complex concept take the concept of a restaurant what do you know about a restaurant well actually there's a frame for a restaurant you know that there are

00:58:10

menus as a maitre d there's a chef there's a dish you know there's you you eat you there's a customer etc but a restaurant is actually made up of three other frames there's a business frame

00:58:24

this is a business there is a cooking and eating a food service frame somebody cooks in search of food and there's a host guest frame so the maitre d is the host the customer is a guest and the

00:58:38

customer is also the person who is fed and there's a chef and the chef is the cook who feeds the customer etc we were binding together three frames and you've got a third frame for a restaurant and

00:58:51

these things bind together from each of these frames to the third one that's how complex frames get made up in a brain you take physical structures and you have binding circuits and you bind them

00:59:05

together in other frames okay this is physical and that's how it works now what about learning one of the things that's remarkable is

00:59:19

this when you were born you're born with about a hundred billion neurons that's a lot well each of them has about a thousand to ten thousand inputs and outputs that gives you trillions of

00:59:34

connections all over your brain when you're born now the time you are five half of them have died F of the connections have died the ones that are least used have died off if you don't

00:59:48

hear music till the time you were five you'll not become a musician what happened if you don't learn fine motor control in that point from the time you're five you'll never become a pianist you'll never become a seamstress

01:00:01

or of doing anything that requires fine motor control and so on that's why early childhood education is so crucial it's so crucial now these still trillions of connections

01:00:16

are there even though half of them die off you still have trillions left and they're around and some of them actually have to do with the concepts you learned but what does it mean to learn a concept

01:00:29

it means you've learned a neural so neural circuitry a complex circuit how do you learn a complex circuit by strengthening the synapses connecting the neurons that is in a circuit all the

01:00:42

synapses in the circuit have to be strengthened and how do they get strengthened they get strengthened in a couple of ways that we'll talk about but they have to be firing in order to get strengthened so anything that's not

01:00:54

firing it's not going to be strength and it turns out that at any point in your life you still have trillions of connections waiting the form circuitry where the synapses have not yet gotten

01:01:07

strengthened enough to form a circuit and anything you learn is a has to do with strengthening synapses linking the previous circuits in short when you

01:01:20

learn something it's on the basis of what you before it's the minimal connections needed to form new circuitry to learn that to learn that thing and learning

01:01:32

happens all the time now sometimes you get no new neurons form not too often there are very small parts of the brain where new neurons are produced but you can learn things or your life in this

01:01:47

way you can keep lying whatever your age is provided you know this hasn't started to die off now what kinds of neural learning are there there are three three

01:02:00

basic clients at least and it's going to be important to understand what they are for reasons you'll see in a minute one there's what's called heavy and learning okay what does that mean it means neurons that fire together wire together

01:02:14

because you have two neurons that are connected next to each other and they're firing at the same time they're going to learn gradually very slowly strength in the synapse is connecting them okay

01:02:26

that's called heavy and learning there's what's called long-term potentiation and long-term depression that is in very emotional situations what happens if you're really happy dopamine gets

01:02:38

flooded out there if you're sad other neurotransmitters and hormones flood parts of your brain okay what happens then is that those neurotransmitters either increase greatly right away the

01:02:53

connections the connection strengths immediately to form a new and new circuit or if there were they inhibited they they stop it from forming a new circuit they can depend oppress it or

01:03:06

activate it and long term potentiation happens in trauma in trauma you learn things you don't forget very easily think of 911 okay

01:03:20

you're not going to forget that those of you who are old enough to remember the Kennedy assassination you can probably remember exactly where you were other traumas as Colima go clinical

01:03:33

psychologists have pointed out and Freud observe affect you the rest of your life because of long-term potentiation that is you you have these things there the Third Kind is recently discovered and

01:03:46

it's really wild and very important what's called time dependent plasticity and you'll see in a minute why this is important way down when you have two neurons that are connected to each other

01:04:00

here you have two neurons the axon for one goes to here the axon from the other goes to there they're connected to each other okay and they impinge on each other now it turns out that when one of

01:04:13

them fires first strongly it increases the strength of the connection in its direction and decreases the strength in the other direction okay listen just little neurons connected to each other

01:04:28

okay so what it turns out that determines the directionality of actions and the ability to do that and it determines the structure of metaphor how

01:04:43

can something like that determine metaphor well let's look okay I connect conceptual metaphor is a mapping from frame to frame across

01:04:54

different brain regions as in love is a journey for example between journeys and love okay and they can be complex or they can be made up of smaller metaphors

01:05:07

for example intimacy is closeness but for achieving a purpose is reaching a destination these are very basic simple metaphors more is up less is down happy

01:05:20

is oriented upset as I oriented down very simple metaphors those metaphors are called primary metaphors and they are learned usually before language just

01:05:33

by living in the world how does that work how is it possible and also why is it that more is up not up is more affection is warmth why isn't warmth

01:05:46

affection why is it how do you know that when you talk about a warm person you mean someone affectionate but it's not the case if you're talking about temperature you talk about affection okay it doesn't work in the opposite

01:05:58

direction how do you learn it at all first and that's crucial you learn it when experiences come together think for a minute about learning more is up you

01:06:13

pour more water in the glass the level goes up every time you pile more stuff on your desk the level goes up there's a correlation in the world and your experience between quantity and

01:06:24

verticality we're at think of affection this warmth we held affectionately by your parents as a child to experience affection and you feel the temperature now what's interesting about this is

01:06:37

that quantity and verticality are registered in the brain in different places affection and warmth are registered in the brain in different places so how do you get a metaphor out

01:06:50

of that the answer to that is interesting what happens is when you let's take affection and warmth you feel affection and emotional regions we shall

01:07:03

warmth and temperature we have firing in both places at once and every time your health affectionately you fire again or let's take mores up you experience verticality in one part of your brain

01:07:16

quantity and other every time you see here you know water poured into a glass the child may not even notice it consciously but the brain notices it it will notice the level going up and it

01:07:28

will notice there's more quantity and two parts of the brain will fire simultaneously but what happens when that happens over and over again to places fire you get strengthening but

01:07:40

because they're connected each neuron to a thousand of 10,000 others you get spreading activation to a long existing pathways so it spreads and that gets stronger and it spreads further and that

01:07:53

gets stronger and sooner or later the shortest pathway between these two parts of the brain is found when that shortest pathway is found you get a circuit formed and it's form via heavy and

01:08:06

learning but when it reaches when you reach the other sides what happens what's interesting is that the one that fires first is going to get strengthened

01:08:19

in one direction and become the source domain and the one that doesn't fire first will be weakened and become the target domain so why is it more is up

01:08:29

not up is more the reason is that your brain is always computing verticality even when you're asleep parts your brain are telling you or computing what's up where the gravitate area or where the

01:08:43

gravitational field is okay but not always computing quantity your brain is always computing temperature but not always computing affection so that it's

01:08:55

affection is warmth not warmth is affection this works from the hundreds of cases of primary metaphor we have studied hundreds of cases work like this and

01:09:08

they work by time dependent plasticity at the level of individual neural pairings something that low can determine the entire metaphorical

01:09:19

structure that structures how you understand the world metaphorically that's amazing I never knew when I started doing this stuff I've been now working on it for I won't tell you how

01:09:32

many years 50 after what I never understood I never imagined that that could be true and yet it is it's a remarkable remarkable fact now that's

01:09:48

how primary metaphors work just as I described it and we have very interesting experimental evidence that's been done enough in a new field the field is called embodied cognition and

01:10:00

what an embodied cognition does is do experiments that show that the metaphorical structure in your brain that's fixed they're linking to parts of the brain allowing you to reason about

01:10:14

one in terms of the other determines behavior determines what you do okay so for example at the University of Amsterdam there was a study which showed

01:10:29

that you subjects who are thinking about the future they said okay okay guys think about something in the future and then they took cameras very carefully and notice that they were thinking about the future they leaned forward and if

01:10:42

they were thinking about the past they leaned backwards a little bit but they leaned you can measure it the future is ahead the past is behind let's look ahead to the future let's put it well

01:10:54

behind us that metaphor structure is a physical structure in the brain that will determine when you're thinking about the future that you lean forward a little bit and have to pass as you lean

01:11:07

backward a little bit think about how wild that is so this metaphorical structure is not just about thought and reasoning it's not just about language it determines how you

01:11:19

live in the world how you function at Yale researchers brought in subjects and they said and they gave half of them warm cup coffee

01:11:32

it was winter in Yale the in half of them warm coffee and half of them cold coffee and then they asked them okay now the experiment starts that was supposed to be before the experiment now the experiment starts the experiment is

01:11:44

imagine you've just met somebody and described the person you met on the whole statistically the ones who got the warm coffee describes someone who is nice to them a friendly person and the

01:11:58

ones who got the cold coffee described an unfriendly person affection is one right behavior is being determined there let's take some more at Toronto subjects

01:12:11

were asked to remember a time when they're either socially accepted or socially snubbed those with the warm memories of acceptance judged the room to bare asked how warm is the room they judged through them to be 5 degrees

01:12:24

warmer than the people who remember being snubbed right affection is once right the affecting how how you determine and how you judge the

01:12:37

temperature in a room their subjects who are asked to think about a moral transgression like adultery or cheating on a test and then at the end as they

01:12:50

left they said oh we have a gift for you you can choose one of two things you can get either an antiseptic cloth or a pencil the ones who thought about in-world things got the antiseptic cloth

01:13:04

another example of this a similar one was subjects were brought in and half of them were told to whisper in the ear they're in a group whisper in other people's ears and say something nasty about somebody else the other guys would

01:13:18

say we're told to say something nice about somebody else okay on the way out they had a chance to either wash their hands or not just to wash their hands the people who said

01:13:31

immoral things right immorality is purity right we find this over and over there are now hundreds of these experiments I'll give you another example

01:13:45

Daniel casa santo has a very nice experimental apparatus it's three pieces of wood with a groove in each and he has marbles in the central one and he asked subjects to move the marbles either up

01:13:58

for some subjects or to move it down for other subjects and then he has these guys moving these marbles up and he says as you move the marbles up or down whichever it is tell me a story attached

01:14:11

who moved the marbles up tell happy stories the guys who move the marbles down tell sad stories happy is upset is down right because of your physiology this is going on all the time the

01:14:23

experiments keep coming out day after day almost that's called embodied cognition and it works because the metaphors are physically there in your brain

01:14:35

structuring your behavior not just your understanding not just your language but what you do let me give you another example the wonder there's a group in

01:14:45

Germany where the they some people okay pick a number don't tell me tell me later but just pick a number okay now pick another

01:14:59

number now what they do is they measure eye movements and what they found was that when the eyes go up into the right they're picking up number higher than

01:15:13

the first number and when they go down into the left they're picking a number lower why more is up less is down and why left and right think of speedometers or the number line

01:15:28

you write you do it left to right that is the lower numbers on the left the higher numbers are on the right okay those metaphors are structuring how you think about these things how you think

01:15:41

about numbers and the experiments go on and on but the point is that behavior makes sense and it makes sense because the message

01:15:53

metaphors are physically there in your brain structuring what you do so we now looked at these there are various kinds of circuits I won't go through all of

01:16:03

this stuff one of the things that my colleagues rena narayan and does he's the guy who figured out how this primary metaphors work in how they were learned originally he does neural computation

01:16:18

and what he does is he takes these neural cell clusters they correspond to nodes over here and the nodes you know are understood in certain ways

01:16:30

and the dendrites are input links and so on and what he does is he takes a computational structure let's go back to that he takes this kind of physical

01:16:42

structure which is the actual physical neuron by neuron thing this looks at them in terms of nodes which is functional clusters and then there's a form of computation that he uses that

01:16:54

has input links and output links in certain places and so on they're called Petri Nets and he models how the neural system works in terms of these computational networks and the

01:17:07

modeling is very accurate that is he can tweak it in such a way that it does what the brain is doing it can do no modeling that that that is based on these

01:17:19

correlations that's how you do no remodeling and with a again the basic functional circuits are Gestalt circuits binding circuits and then what I've called activated activation than

01:17:31

inhibitions but the idea here is that just as we've seen these basic simple circuits allow you to connect things so let me talk a little bit about gated

01:17:45

activations what does that mean right think about a metaphor when you're learning a structure you have a frame which is a hierarchical structure and you're activating another structure

01:17:59

someplace else you'll be mapping from node to node which means you have to have activating connections going from here to here but they have to fit together as one thing all right and that one thing has a gestalt circuit that

01:18:11

says these all fit together they're all one metaphor and they gate that is when the gestalt circuit is on when you know that this is a metaphor this allows the activation to go through if it's not a metaphor if this is literal it's all if

01:18:24

it doesn't allow the activation to go through very very simple structure that's the kind of ways that these things fit together and I could go through this for all the kinds of things that are involved in cognition but that's the sort of

01:18:37

modeling that gets done if you have I won't go through all this if you have a frame the schema name for the whole thing is a control node in the gestalt circuit etc I'm not going to go through

01:18:49

all that but technically what you can do is take the structure of a frame or a schema and then show that every part of the linguistic description of it can map

01:19:01

one to one onto a circuit of a certain kind and that can map on to a neuro computational mechanism that tells you what the circuit does that's what this research is about that's what we that's

01:19:12

what we sit around doing at the neural theory of language project at Berkeley and we you know this is pretty cool it's fun now the big deal has to do with the

01:19:26

difference between rows called modularity and what is called Cascades and this matters there's an old view of

01:19:37

fMRIs and the brain these brain scan pictures that you see what says the following let's suppose that there's some task that you're studying and you have somebody doing this task

01:19:50

and whenever they do this task a certain part of the brain lights up a lot it activates a lot and you see a lot of activation you don't know what's going on but there's a whole lot of oxygen being put into the neurons

01:20:05

and then suppose you have a patient who has a lesion in that spot and they can't do the tasks so the old idea was saying ah the task is done their language works

01:20:17

in this part of the brain you know you know this emotion is in that part of the brain judgment is in this part of the brain and you see this discuss a lot in the popular literature and sometimes

01:20:30

even in the scientific journals that's the old view what has been found out is that that's not true rather what's involved as many parts of the brain and the lis and what's in that particular

01:20:43

place is where the activation is going through and if there's a lesion there it doesn't go through to the rest of the brain but it has to go through to lots of parts of the brain to get meaning and

01:20:54

that's what's cool so the old view says is what's called modularity so for example in language you'll hear people saying there is a language organ or a language modules module and all language

01:21:09

is done their language is done all over the brain because language is very complicated and does all kinds of different things there is no such thing as a module where there is language in what in the whole place lots of

01:21:21

different things are done all over linking these things together in what are called Cascades so let me tell you and give you some feel for what a cascade is and this

01:21:34

is important Cascades link their circuits that link across maps and they either form assaults or bindings across Maps or activations across maps now

01:21:49

we've talked about into that's one example of that where you have a map for motion and a map for containment and then they map together to form a cascade

01:22:02

linking motion and containment and then if I say he put the ring into the drawer then your you have put has to do with a force that applies to the ring that

01:22:15

moves it into the drawer and you have to put together your understanding of application of force to your understanding of motion to an interior thing and they have to bind together to

01:22:26

one complex meaning in a cascade okay now so binding circuits link entities in two different maps to a single entity in a third map now Cascades are interesting

01:22:41

and so let's take an example from Stanislas Dehaene the great neuroscientist in paris who has a great book I recommend it highly and it's very

01:22:54

readable called reading in the brain it's a book about reading and if you want to find out how neuroscientists think this is a great book to read here's how it works

01:23:05

in the visual cortex you get inputs from the retina here get neurons coming in there the length immediately now in there those are connected to another layer so they come here they've

01:23:19

connected to another layer and these ones in the second of the ones compute what is called a center-surround structure what does that mean means neurons in the second layer that are in

01:23:31

the center in the second layer that what that says that the second layer of neurons light up best when they see let's say a black dot with a white surround or a white dot with a black surround or a red dot with a green

01:23:44

surround etcetera the center and is surround how does that work it means they get active from the black dot coming in an inhibition from the whites around out

01:23:57

there and that says if you get lots of black bear with you know a weaker inhibition around it you're going to get activation of that black dot and it will fire best when you see a dot with the

01:24:09

whites around okay so far so good but now those things move to the next layer and the next layer they form lines how do you form a line you have overlapping centers if you have

01:24:23

overlapping centers in a certain direction you're going to form a little line how do you form that line well in the third layer you're going to have a single let's say a single neuron which

01:24:34

will get input from all of those centers in the center surrounded thing and that will light up if you have an orient if you have that oriented in a certain way and then there it turns out that there

01:24:47

are these columns in the brain where you have sixteen types of or sixteen types of orientations and depending on how these are linked up so when you see little line segments that's how you see

01:25:00

them if they're longer lines then there's another layer where the lines are connected one after another now when you have these things together how do you see the letter e how is it possible

01:25:12

to see the letter E well it turns out that there's a part of the brain over here called what is called the letterbox what that means is in the part that's closest to the back you can you you know

01:25:25

whenever use you read a letter something will be activated there and they can see those activations and and so on well how does that work well if you have these lines there's another layer in between

01:25:39

where the lines come together to form combinations of lines like in the letter E where there's one line here hitting this and another line coming out those are called G they are parts of the structures of

01:25:53

perceived objects so if you see an object that's how you fit it together by fitting these lines together and they are fitted together by looking at the parts that intersect are connected to a

01:26:06

single neuron in the next layer okay then when you put the jeans together you get the letter e okay to get the letter e you have to have the entire cascade

01:26:19

activated but at each point it's simple simple connections all the way down because the letter e requires that you see every dot every line every connection of lines and the way the

01:26:31

lines fit together that's just the letter E okay but that's what a cascade is it's like what's happening within to or with other things and let me give you

01:26:43

a more complicated example suppose you have a commercial event any you know like buying or selling or something every commercial event has certain roles as a buyer seller goods and money and

01:26:56

action of buying an action of selling and so on each of those have that but every commercial event is an exchange all right so you need to know that if

01:27:09

there's a commercial event you have another layer in which there's an exchange or what's in exchange in exchange is a mutual transfer well mutual transfer says there's a transfer going from here to here and a transfer

01:27:20

going from here to here and the transfer or here is the transfer iver and there's a binding in another layer the exchange letter layer where these things fit together in a certain way so there's a

01:27:33

binding to the exchange layer from the to transfer layers we want to transfer well transfer in the exchange has to do with the transfer of possessions so you have to know things transferred are

01:27:46

fitting into a possession later somewhere what's a possession well physically a possession has two parts has something you have access to physically and something you control so

01:27:59

there has to be a layer for access and a layer for control and they have to fit together okay so then you have a transfer what is a transfer physically a transfer is take physically moving

01:28:10

something applying force to it so that there's a motion from one place to another so you have to have a part where you apply the force and a part where there's movement and they have to be bound together in the transfer layer

01:28:23

okay that's all of that is known when you say I bought a book for 10 bucks everything all of it is lit up everything is activated entire cascade

01:28:37

and it comes down to physical things but then metaphors can apply to that if I try to tell you this story and you say I don't buy that you're understanding it

01:28:49

metaphorically or you can you know sell people property you can sell people ideas you can have you know you can you can patent an idea and then sell it or

01:29:03

copyright something and sell it abstractly that is their metaphorical concepts that take this and take it further to understand those you have to understand the entire cascade down to

01:29:14

the connections to the body and that's what it means from meaning to be embody now okay so we have lots of those

01:29:28

commercial event bla bla bla now every word there that language every word is defined in terms of a frame what does that mean it means if you have a word

01:29:42

which is sounds a sequence of sounds and you have a meaning let's say the meaning is buying to buy a verb like buy which has to do with this entire cascade over

01:29:55

here there has to be a link but what kind of link is it well it's gotta go in both directions that is the concept of buying if you want to say it is you're going to have to have the idea and you're going to say

01:30:06

the word buy you hear the word buy it's got to go the other way what you need is a gestalt circuit where either one activates the Gestalt node which activates the rest

01:30:18

so the his thought circuit which structures frames in the right place linking the part of the language that has to do with phonology over here with the rest of the brain over there that

01:30:30

has to be in a stalled circuit which links words to their meanings very very important how does grammar work what is

01:30:43

it now well now it's a name of a person place or thing either a literal one or a metaphorical one that is just a name of an entity but what's a name and they missed something in the phonology which

01:30:57

is linked to some entity and enters an entity frame an entity schema that has every entity has a certain kind of structure which I won't go into here but you know a lot about that structure

01:31:09

every entity has a certain structure and on the what you're doing is linking a single phonological entity over here to a conceptual entity over there via a

01:31:23

Gestalt node that Gestalt note is a noun the category noun is what links these two so grammatical categories are in the Gestalt notes that do the linking between the phonology and the meaning

01:31:38

that's part of the neural theory of language that we've been developing now that's the idea in general and what you

01:31:50

have in grammar is hierarchical structure that comes from the semantics and conceptual Vistal notes and linear structure that comes from speaking and when you're putting the hierarchical

01:32:03

structure in the linear structure together via these Gestalt notes that link it you get grammar that is our theory of how this works and what we're trying to do now is work out the

01:32:15

complete details and that's a long job and we have some of it done and lots of it not done but that's where this is going and that's what what this is all about so

01:32:28

then the question is why bother okay one it's cool it's beautiful all right you can answer the question how is it possible to get meaning and language out of neurons you can begin to see that

01:32:45

there's really an answer to this question and it has to do with all this stuff that's amazing that this could possibly happen with all of these people contributing all of these pieces to this really incredible

01:32:57

thing is is happening but it turns out that it matters in our lives it matters in our politics as I've been writing as a book out so ID you can pick

01:33:08

up called the political mind it's the fifth book in a series of how these things apply to politics it matters for a number of reasons conservatives and

01:33:22

progressives have different ways of understanding the world via different metaphorical systems and this is a crucial thing they both understand think

01:33:37

for a minute when you are first governed we're first governed in your family by your parents so you learned that a governing institution is a family and it could be a religion or a classroom or a

01:33:50

team or a nation but the governing or or the market a governing institution is a family what kind of family is it well there are different moral systems governing morality and that govern types

01:34:05

of families and that has to do with well-being as a child as a child you learn certain conditions of well-being that lead to metaphors for morality via the mechanisms we've just seen so how do

01:34:19

you get metaphors for morality and morality turns out to all be based on a set of about two dozen metaphors and if you want to see what they are they're in a book outside called philosophy in the flesh

01:34:32

chapter 14 now and they're also there summarized in this shorter book now the thing about it is this right I'll just

01:34:49

tell you briefly and then we'll go to Q&A the rate let's go to Q&A you can ask okay because I'd better be good thanks so the three microphones out there and

01:35:13

[Applause] thank you yesterday morning I was in Berkeley and drove home twelve hours in the car gave me a chance to experience I think what you were talking about my

01:35:29

wife and I are always turned north what left right you know that kind of thing does that mean she never looked at a map when she before she was five how does that perception come across and also I

01:35:42

was thinking some of what you were talking about is was somewhat judgmental in terms of the coffee for instance in some societies coffees considered a drug that has a rather negative connotation

01:35:54

versus others in Seattle it's a you know entirely different so would that affect the results of your study when someone was given coffee regardless whether it was cold or warm if whether they felt it

01:36:07

was a positive thing or a negative thing well luckily this was at Yale where it was all negative positive there were students involved and every student does

01:36:17

the positive version of it but you know I it could very well be that if you looked at a certain culture where you know if you did this study in Utah for example it might come out very

01:36:30

differently yes I was wondering more about the ride and you say our brain you know input like feel like AE looks at the lines and how it all connects what about handwriting because I mean that's very different than type like how does it

01:36:46

does it process it in the same way well what's going on there is you have a connection between what you how you're moving your hand and a mental image of that connected via the mirror neuron

01:37:00

structures so what you're doing is trying to reproduce a certain structure and they you know cursive writing is not the same as other things they have curves but there are jion's for curves

01:37:13

as well and the question then is how do you link those up so how do you link up a cursive a with a printed a or a little a with a capital a what is what makes

01:37:25

them all a and the answer to that is a Gestalt circuit that is if you have each of those as this different role if they're for survey of printed a capital

01:37:37

A etc and the gestalt circuit says these are all one thing but it's but it's interesting about it is that Gestalt circuit does that work there whereas in another part of the brain it creates a

01:37:49

frame and another part of the brain it links forming meaning the same circuitry can do different jobs in different parts of the brain thank you you are here yes hello professor perhaps this will give

01:38:03

you the door to walk through into that last piece you were just not quite able to finish but going back to very early in what you said in terms of the ability

01:38:14

to imagine creating the ability to act ok does the lack of action toward peace in the Middle East simply mean the lack of ability on the part of the

01:38:28

protagonists to imagine peace in the Middle East probably yes sadly about the learning I mean how does that change okay

01:38:42

it changes via the ability to form empathy to imagine what it is to be in someone else's shoes and that that's a

01:38:56

deep thing that is murder on stuff but it is a fundamental ability that we all have and the question is how do you generate empathy with other people one

01:39:07

is by actually meeting other people you know when you have Palestinians and Israelis who actually live in the same village and our friends they like they like each other just as any other two people would like each other they're

01:39:20

human beings they may help each other and care about each other and you know be in each other's lives and this is the sort of thing that you need to do across

01:39:32

national lines and across religious lines and so on you know one of the great things that Barack Obama did in his campaign unfortunately he didn't do it in administering but in

01:39:46

his campaign he said something really deep things the first thing he kept saying over and over is that the most important thing is mother taught us our endless empathy to think about being in someone

01:40:00

else's shoes to understand it to feel it right that that was the most one of the most important things in life and then he pointed out that democracy is based on empathy the reason that we have

01:40:13

principles like fairness and equality for everybody not just for the powerful is that we have we care about other people that's what it means to have a democracy the functions that democracy

01:40:27

is inherently caring about other people and like taking the responsibility to act on that care and to make yourself as good as possible to work hard at doing it at the level of the family at the

01:40:40

level of the community and the level of the nation and something that he repeated over and over if you want the most beautiful discussion of it it's in his father's day speech in 2008 but he repeated it again on April 13th of this

01:40:54

year with a speech with very few people heard because it was like 10:30 in the morning on the west coast but if you go back go to whitehouse.gov and look up his 8 April 2:13 speech he repeats it

01:41:07

again and he explains how his approach to the budget and the deficit etc comes out of an understanding of democracy in those terms it's a beautiful beautiful

01:41:20

account of what it means to think about how you spend your money that budgets are moral documents that have to do with the country and you don't give up the ideals of the country when you're making up a budget

01:41:35

yes yes are there patterns in cascade orientation to signify difference

01:41:48

between concept and percept the you can tell the difference the answer is that it's not there it's actually in a connection to the prefrontal cortex

01:42:03

there's an experiment done at UCLA a set of experiments done by a professor who does brain operations on split brain patients and they've located neurons

01:42:18

that fire that increased their firing when you perform an action and decrease their firing when the same action is performed by somebody else that is the mirror neurons when you look at the actual mirror neuron data that came out

01:42:31

of Parma which I did with with galazy what you find is that the degree of activation is more when you're doing it than when you're seeing it and these this this difference is exacerbated by

01:42:44

these neurons in the prefrontal cortex and that's the part of the cortex that that has to do with judgment and so it's important to judge whether you know something is happening to you or somebody else and of course you have the

01:42:57

ability to do that yes regarding the frames you were talking about how can one learn the frame for example the hospital was really using if that person would grow

01:43:13

up on the deserted island with no hospitals no health care nothing can we deconstruct it frame to the some basic building blocks axioms like that and how we learn those axioms are we learning

01:43:25

them from the beginning of the at what point of our life are there genetically-based or they're starting to learn ones did the first neuron style to fire okay a very very important question

01:43:40

let's reduce it first to a simple case imagine a flying pig right now pigs don't fly but you can imagine a flying pig okay

01:43:53

it's called Pegasus and all right where does Pegasus what is biggest it's like Pegasus is a pig in pig has wings where the wings are attached or attached to the side where is the snout going it's

01:44:06

going forward okay but there are two kinds of flying pigs there's Pegasus and super swine super swine is like Mighty Mouse he's Superman it goes like this and as a cape but it's a pig now what

01:44:18

you can do is put together things from your imagination parts that are a man like Superman and a pig and do an oral binding from two different maps onto a

01:44:29

third map that allows for imagination for putting ideas together which is how you put together extends furs and exchanges to get a notion of buying for example now somebody born in a deserted

01:44:43

island may not be able to understand what a defibrillator is you know they you know they may not ever be able to to get enough stuff to do it but they may

01:44:55

for the following reason that is if you break down the concepts into parts you have an illness there's something that makes you feel bad that decreases your sense of well-being which has to do

01:45:08

with the emotional reasons where you have a decrease in senses of well-being that or you have pain there is there are certain kinds of things that take pain away or allow you to increase your

01:45:21

well-being there are people who may know about how to do that often called shaman's or in that thing okay they're they're like primitive doctors then you

01:45:33

can generalize that to other kinds of things and extend it in that way but it depends upon the ability to understand well-being and pain verses and things that take that pain away that that is

01:45:46

our causal you know so you understand pain is something you have and causal is something that removes it by a metaphor and then there's a person who exert some force or does some action that does that

01:46:00

so causal action is understood as exerting forces so as metaphors are universal and they're used to patch these things together into basic ideas of what a shame and us then that can go

01:46:13

further as you learn more more stuff in Western medicine and so on but that's basically how these things get built up [Applause] great thank you all for coming

01:46:43

[Music]