0:01
[BLANK_AUDIO].
Gotta say, I kind of like this word, Intelligence, written over my head.
Makes me feel good. Alright.
Welcome back. we're going to talk about intelligence.
Now, there, there was the desires to talk both about human intelligence and to talk
about artificial intelligence. I was trying to conceive of a way of
doing both. it's just, given my, given my track
record of 24 minute lectures when I'm shooting for 15.
I ultimately decided I just, I couldn't do both.
I may have a separate side dish on artificial intelligence.
And I may I, I do actually have a video link is at the end of this one that I
hope will help you form that connection. But in this lecture we're going to focus
on human intelligence. Alright.
So let's do that. Week 8, Lecture 6.
You know I have that feeling like, you know, when you go on vacation for, like,
two weeks, not that I remember doing that forever, but imagine you did that.
and you know the whole first week is fun but in the second week you start counting
down, oh now, only six days left. Well, there's only three lectures left
and this is one of them. This is kind of amazing and weird but
here we are. We're going to talk about intelligence.
Before I get to intelligence, I want to describe and, and really highlight a
distinction with respect to the research we're about to talk about now.
Most of the research that we've been talking about in this course has focused
on group averages. So, if you think of something, like, the
bystander interference effect. the notion is that the average person is
less likely to help if there's more people around them, et cetera.
So, we're really trying to understand the average individual.
But there's another part of psychology that is not interested in the average
human but rather is interested in what, how certain humans are different from
certain other humans. and this kind of research is called
Individual Differences research. I'll give you one other example before we
talk about intelligence. Personality is an example.
Remember way back at the beginning of the course, when we did the The Big Five
Personality Index, where we looked at your extroversion your neuroticism.
You know, all those, those, Big 5 Indices.
in personality research, they're trying to understand, now, why does one person
have a different personality than the other?
And what are the ingredients, and what predicts that?
Well, research on intelligence is very much the same in the sense that, it's
really trying to focus on, what is intelligence.
And why are some people more intelligent than others?
And is there anything we can do about that?
and because of this, and because of the importance of the question of
intelligence it's been very controversial and been, been very political.
In fact, that politicalness really stems from the fact that intelligent behavior,
to some extent, we see as really core to being a successful human being.
And, we want to do things like school our children we, the hope, of course, of
putting our children to school is that they will emerge from the education
process more intelligent than when they went in.
In fact, all of you guys are taking these MOOC courses.
Why are you taking these MOOC courses? Well, I assume you think that somehow,
through an experience like the one we've been having together, you are going to
become smarter. and that smarter may mean things like
you're just going to know more about, in this case, for example, the human mind
and human behavior. but also the very process of learning and
thinking and writing your ideas and, and memorizing, and all that.
It's like exercise for you brain and it's going to make your brain stronger.
It's going to make you a better thinker and hopefully a better communicator, and
I certainly hope that. Now, what did I just say in all those
things? All kinds of things, when you think of
the term intelligence. And if you actually ask somebody, you
know, what do you mean by that? And one of the things I like to do in, in
a class is say, think of somebody or multiple people if you can, that you
would call intelligent. Now what are the attributes of that
person that you're thinking of? You know, what is it about them that
makes them seem intelligent? Is it the fact that they seemed to know a
lot? Is it book smarts?
Or is it the fact that they can explain things clearly?
Or is it the fact that they are just very articulate?
You know, they, they have a very good command of vocabulary and can literally
speak eloquently. What about a really good musician?
5:00
Or a really good athlete? Can you call that intelligent?
You know, athletics we're probably on the no side.
Musicianship [SOUND] I don't know. Depends.
What about somebody who knows every song that's out there and can play all them
really well? Would you call that person intelligent?
Okay. So this is one of the issues with
intelligence, is it's a big concept. The other issue is that, with that
individual differences mentality, sometimes people ask questions that start
to feel a little dangerous. Like, if you could measure intelligence,
would you find out that, let's say, women are more intelligent than men?
Or that one race is more intelligent than another race?
what if you found that to be true? what are the implications for society?
Can you actually change intelligence? Can you improve intelligence And how do
you do that? so these are all questions that are not
just theoretical questions. They have the ability to cause and they
have caused a lot of controversy in society.
So we're going to kind of go down that route a little bit.
that's, that's your sort of intro. So let's now just, kind of, with that
behind us. [INAUDIBLE] I want to tell you a story
and then we'll [INAUDIBLE] we'll come back to this.
Okay? And see how it all fits in.
6:25
The story begins with history as it usually does.
The first person that really got interested in, in attempting to measure
intelligence was Sir Francis Galton. And in fact the controversy started with
him. Galton was a cousin of Darwin living in
the same period of time. He was a scientist as well.
One of the most clever things I think Galton did is he wanted to do research on
human cognitive functioning, human abilities and on humanity in general.
And so he had this clever idea, he went to local fairs and he would set up a
tent. And he would get people to pay for him to
measure them in various ways. Now, this seems kind of odd but people
in, in, at that time, 1800s, didn't necessarily know how tall they were, not
like we, most of us do. They didn't know how heavy they were.
They didn't know how long their arms were.
and here was this guy in a tent with a bunch of measuring equipment, and you
could go in, pay him a certain amount of money, and he would measure all these
attributes of you. And it wouldn't just be physical
attributes. He would ask you to, like, respond to
things as quickly as you could. when something pops up, press a button as
quickly as you can, and he was timing your reaction times.
you know, things like that. Now, the cleverness about this is, of
course, Galton wanted the data. He, he wanted to know things like, well,
what's the average circumference of a head of a human head.
does that differ between males and females?
Nobody had actually measured the human body like that.
So Galton was getting all this data, and people were paying him to give him data.
That's pretty clever, that's intelligent, we might say.
but he, he took, he, he really associated intelligence with that reaction time
measure that I told you about. he thought intelligent people somehow
could process information quicker. and maybe they were just wired better,
for example. They could respond quicker, so they were
literally quick minded, quick thinking. and he thought that was a really good way
of measuring intelligence. He also, by the way, was the one, do you
remember that eugenics movement I told you about?
That notion that maybe artificial selection should be applied to humans and
maybe we should only allow certain humans to breed and produce children, and
prevent other humans from doing that? Well, he started a lot of that.
Again, he was in Darwin's time and everybody was in love with this notion
of, of selection and, and evolution. And so, he saw this as noble.
The idea of let's let the human race evolve, this is the way to do it and one
of the [UNKNOWN] of that was he felt we should be allowing mostly the intelligent
people to reproduce. And the people who are not so
intelligent, we should be preventing them from reproducing.
10:46
Instead, Binet created a whole bunch of, of little tests.
Now, there's a whole evolution, dare I say, of IQ testing.
And there's a video at the end to kind of walk you through that.
I'm going to do all that. I'm going to take you right to the punch
line of the current version of the Binet Simon now intelligence test.
Just to give you a sense. So in the test they now use, there is
essentially four domains of, of knowledge or, or types of reasoning that are trying
to be measured. The ability of your your ability to
verbal reasoning, to deal with word kind of problems, your vocabulary, your
ability to comprehend complex written passages, things like that.
So how good are you with words? alright, that's one part.
The other one is how good are you with numbers, quantitative reasoning.
and so these were basically math tests as, as it were.
But not just math tests but just even things dealing with numbers.
Like a number series would be, I'm going to give you something like one, two,
four, eight. What's next?
Okay? Sixteen would be next.
I was doubling each number as we go. but can you kind of figure that out as
you go? And so they'd be very complex ones, not
as easy as doubling, but very complex ones.
You know, can a person figure that out. Abstract visual reasoning, so these are
things like paper folding. I give you a piece of paper.
I want you to take this big piece of paper and make it look like this funnily
shaped thing. You have to figure out how to fold it to
turn what you have now into some goal state.
So can you do that visual reasoning? You know, think of Tetris or think of if
you've ever moved and you're trying to fill the truck, trying to figure out
which box to put where. You know, that kind of thing, can you do
that abstract visual reasoning? and also your short-term memory.
Remember working memory. Working memory is very much linked to
intelligence. People with a bigger working memory at
which you can measure by things like span.
So what I mean by span; you can do things like, let's do the memory for digits.
I give you five digits to remember. Let's try that.
So, try to remember these digits. 7, 5, 9, 6, 2.
Now wait for a while. But that's only five digits, 7, 5, 9, 6,
2. Easy, right?
but I keep adding digits, and I keep adding digits.
Now, the average person Is pretty good at remembering about seven plus or minus
two. So somewhere in between five or nine
digits they can remember. They can keep them in their working
memory and then give them back. Some people are better, some are worse
and that is part of, it's considered part of your intelligence.
The more of these digits you can keep in your working memory the more intelligent
you're assumed to be. Which kind of makes sense because working
memory is what you use to problem-solve, to figure things out.
And so, we can look at your memory for sentences or your memory for digits and
get some sense of your short-term, or your working memory, capacities.
So these were seen as kind of the cornerstones of a person's intelligence.
14:03
And again, we're getting the notion that there is some single thing, with these
components albeit, that you can think of as intelligence.
But this is what the Binet Simon test kind of measures.
Now what do you do with that? Well you get a score on these tests.
The first thing you need to realize is that in order for that score to make any
sense You have to look at it relative to something.
So what they've done with these IQ scores is they've, first off, normed them.
Now, what we mean by norm them is they create a test that they consider
appropriate for a specific age. and then they give that test to as many
people as possible that are that age. Okay?
And they tried to get an overall sense of, how does the population do on this
test? And you see a range.
You see some people do not do very well, some people do very well, bu you almost
always see this shape, which is sometimes called a normal distribution or a
Gaussian distribution. [SOUND] this idea that, you know, most
people cluster around some middle score. So this is where most people are and, in
fact, if you add these together you know, 68% are, are on this test between 84 and
116. and then there's some that aren't doing
so well, but not that many. And then there are some that are doing
really well, but then again not that many.
Tends to be pretty symmetrical. So, once you've tested a lot of people,
and you get a sense of this distribution, you can in fact modify how you score the
test to put the average somewhere. And I hate to do this without going into
the details but I'm going to. So, they can, they can manipulate these
tests and the way they score them, so that the average is always 100 and that's
how it works on these IQ tests. Average performance is 100, standard
deviation of 16. Stats, back to stats.
Remember those stats? Digital labcoat.
So then we can, we can give the test to an individual now, and say where do you
fall on that test. So if you score, let's say 110, you're
somewhere around here, which is to say you're doing better than oh, 34 47, 49,
and then some of these. So, so you're up, you know, at 60%.
You're better than 60% of the population but maybe there's 30 or 40% that are
doing better than you. and that's the kind of idea you can get.
Where are you on this distribution? How well are you doing?
and that's how they do these tests. They norm them first.
They get a good idea of what the population does.
They transform the test, so that the mean is 100 and standard deviation is 16.
And then once they've done that, anybody can take the test and have a good sense
of where their intelligence lies relative to the population.
[SOUND] Funny little side note. When they first started doing this, they
decided to create little categories and give them names.
And these names were just names they just created.
They didn't have any valence. Any, any positive or negative
connotations. They were just names.
But here are the names. You'll recognize them.
[SOUND] Genius, if you're over 140. Very superior, if you're 120 to 139.
Superior, if you're in this 110 to 119 range.
Average if you're 90 to whatever. Dull, if you're 80 to 89.
Borderline deficiency, 70 to 79. And then it goes to moron, imbecile, and
idiot. Okay?
Now, at the time, these words were, were created, they'd did not have the
connotations they have now. They got these connotations because they
were associated with low performance on an intelligence test.
So if you're going to insult people, now this is the, this is the sort of
continuum you should use. Now if you just want a sort of insult
them, say you're borderline deficient. If you want to really insult them, you
call them a moron. No, sorry.
If you want to insult them more you call them a moron, if you want to go a step
further imbecile. Idiot is the absolute nastiest way you
can put down someone's intelligence. At least according to the statistics.
But all of these words have found their way into our common language.
And we now use these to talk about somebody who we don't think is very
intelligent. Kind of interesting.
Originally, they were just scientific categories.
Just thought I'd throw that in there. All right.
So here's where things get interesting. This is all well and good.
If you believe these tests are validly measuring somebody's intelligence.
So, you know, let's go back for a second. If somebody doesn't do very well on these
things, does that really mean they're less intelligent?
Well, here's where some more controversy came in, because there was a really kind
of interesting and I think scientifically informative notion related to all this
intelligence that happened in America in the 1960s.
and it was a time when segregation was a big political issue.
And the argument by segregatinoists said that, hey, in our school system, we need
to have separate schools for white kids versus black kids, because, the
segregationists would say, the black kids are not as smart.
And so if we mix these kids together, we're going to have to water down the
education to the level of the African American children, and that's not going
to be good for our white kids. And so, that's not good.
We should have separate schools and keep our standards high for our white kids, is
the claim. Now, there seem to be some data that
supported that. They took one of these IQ tests, Simon
Binet IQ test, and they gave it to children that were both Black and White.
same age. The Black kids were from the Harlem area
of New York. The White kids were from the, White areas
of New York. and they gave them the test and this is
what they found. On average you'll see the white kids are
outperforming [SOUND] the black kids. And so they said aha.
See? This is the evidence.
White kids are smarter. We need segregation.
But here's the twist and here's the really kind of clever thing.
Somebody looked at that test and they, and they looked at the way the questions
were worded. This is hard for me to really explain
properly because I don't know the vernacular of, of Harlem.
but the claim was the following, that the kids who [NOISE] lived in Harlem did not
speak, did not communicate the way these questions were worded.
21:09
The example that I'm told about is one where, for example, a question might say
something like, you go to a store and you buy some things and the total is 59
cents, and you give the cashier a dollar. how many, if they give you the least
possible coins back, how many quarters, nickels, dimes, and pennies will they
give you? Okay?
That's some level a good intelligence test.
But here's the problem. Apparently, kids in Harlem did not call
them quarters, nickels, dimes, and pennies.
They had different words that people used to talk about those coins.
And so when the black kid was looking at quarters, nickels, dimes, pennies, that
language was unfamiliar to them. And that made the test more difficult, or
at least that was a theory. But the theory was put to the test.
The researcher who was worried about this reworded the whole test.
He asked the exact same questions, but he asked them in Harlem kid language.
So he translated the test to one that used the words that the Harlem kids were
used to and, of course, that the white kids were not used to.
And he gave a game to test the two sets of kids.
And what he found is in this case, the white kids were worse than the black
kids. The black kids outperformed the white
kids. So his point was, you gotta be careful
with these tests. That they are not necessarily
culture-free. They may be measuring intelligence, but
they may also have other things that are affecting performance, including
familiarity with the language. In fact, there's this little thing that,
I know many of you speak multiple languages.
There's this saying in psychology that when you're not speaking your native
language, it feels like your IQ drops 10 or 20 points.
that it's just harder to be intelligent in your non-native language.
I don't know if any of you feel that way. but that's sort of the same idea that
language matters. And so it's hard to perform well on a
test when you're having trouble understanding the test.
So that was, that, that was and is an issue.
Although a lot of intelligence tests, [SOUND], have tried to become what they
call culture free. And I'm going to end by giving you a
taste of one of those, and a challenge. This is one of those, it's something
called Raven's Progressive Matrices. It's kind of like that paper folding
thing but different, here's the task. You're supposed to look at the pattern
here and there's one missing and it's one of these ones and can you figure out
which one belongs here? they start out easy.
This is an easy one. And so if you just kind of look, you'll
see that there's a certain pattern that happens.
So for example, this, the one's colored on the left, this one's colored on the
right, this one's colored on both. Okay.
Colored on left, colored on the right, colored on both.
24:14
Okay now this one's, the whole le-, left side's colored, the whole right side's
colored. Hmm.
Kind of looks like eight to me. Right?
Eight seems like it would fit, this plus this would be that.
So that's one way of thinking about it, and it's like yeah ok, looks like eight
would fit there. That's the easy one.
Now, try it on this one. This is a much harder one.
These get harder and harder as they go on.
So what fits in there? And I'm not going to tell you.
we'll have this as a discussion forum thread.
What's the right answer, and why? Not just what's the right answer, but
why. What's the pattern?
So there's some pattern that is consistently changing as you move across
these items. And similarly as you move across these
items, there's some pattern of change. And if you know that pattern of change,
then that will tell you which of these fits there.
Oh. [SOUND] Figure it out.
I dare you. See if you can.
The feeling, the idea here is you don't need language.
You don't have to ask these questions in any form of language.
Once the people understand the basic instructions of the task, then they can
just look at these picture grams and try to figure stuff out.
So this is supposed to give you a more culture free Measure of intelligence.
25:38
We've only scratched the surface of intelligence research but hopefully I've
whet your appetite a little bit and, here's now some follow up.
So first of all remember when I started and said you know when you think of an
intelligent person you think of multiple things.
Well in fact there's, there's a gentleman named Gardner.
Who's very much on the forefront of that, of saying, you know, it's not enough to
say how well can you solve a Raven's progressive matrix.
That there are all these components of intelligence.
Then link on to neuropsych, parts of the brain and what they do, and that's how we
have to think about intelligence. So I put a link to that so you can see a
much more complex view of it. this is a history of measuring
intelligence, just to kind of go back to the Galton Binet stuff.
this talks a little bit about the genetic basis of intelligence, you know, how much
of your intelligence is bound by your parents' intelligence.
Fascinating things there. [SOUND].
And here's my little, nod to the, to the artificial intelligence interests out
there. a link.
And, and, you know? You can try to draw the parallels between
machine and artificial intelligence in that video.