There's so much misinformation out there

about intelligence. So today we're going

to be debunking claims on YouTube about

IQ and intelligence. We're going to look

at what's really true. This video is

from Sihow Psych, which is a Hank Green

channel with 860,000 subscribers. They

do a lot of content about psychology.

People love to bring up how Einstein had

a genius level IQ of 160. And to join

Mensah, you need to have an IQ of at

least 130.

>> Yeah, it's funny. So many people think

that Einstein had an IQ of 160. This

idea is actually a little bit silly

because as far as anyone knows, Einstein

never took an IQ test and we can't just

say because he was a great genius who

developed all these really fascinating

and important ideas in physics that he

necessarily had an IQ of 160. There are

many different things that go into

creative success. Einstein had these

incredibly profound and important

theories of physics and the fact that he

was able to produce that does imply that

he was probably incredibly intelligent.

But intelligence is not the same as IQ

and we actually don't know his IQ score.

IQ is a measure of intelligence. And we

can't just say because he graded these

great theories, he necessarily had an

extremely high IQ. But is IQ even a good

way to measure intelligence? Well, that

depends on how you define intelligence.

IQ scores may be a useful shortorthhand

to talk about education strategies for

big groups of people like when

discussing public policy. But IQ can be

affected by a lot of factors, even

things as subjective as your motivation

while taking the test. Yeah, it's

absolutely true that your motivation

taking a test can affect your IQ scores.

This is a source of variability where a

test will always depend on how hard you

try. If you don't try at all, you're

probably not going to do very well. So,

this doesn't make the idea of a Q

invalid. It just means it's a source of

noise in IQ testing. IQ tests will be

limited in their ability to measure

people's actual IQ scores based on

whether people try during the test. But

psychologists seem to agree that one

thing that seems to help people with

learning and academic achievement is

thinking about intelligence as a thing

that can change. What he's talking about

here is the idea of growth mindset,

which is essentially the view that you

can always improve your abilities at

anything. So suppose you do badly on a

test. If you have a growth mindset,

you'll think, well, next time I can work

harder, I can learn the subject better,

and therefore do better at the test. On

the other hand, if you have a fixed

mindset, you'll just think, well, this

test shows how good or bad I am at the

subject. since I did bad, it means I'm

bad at the subject and therefore I'm

going to be doomed to do badly in the

future as well. And there's a lot of

research that suggests it is a bit

helpful to have a growth mindset. When

you randomize people and teach them

about growth mindset, they do tend to do

a little bit better in certain ways. So,

it's a useful mindset to have. That

being said, it's important to

distinguish between things we can

improve at and things we can't. No

matter how much growth mindset you have,

it's not going to make you taller.

However, you can improve your muscle

growth. So, growth mindset about how

strong you are, that makes a lot of

sense. And so it's important to

distinguish there's some things that we

have control over and some things that

we don't. And having a growth mindset

about the things we do have control over

can be very valuable. This video is from

the very popular YouTube channel

Veritasium. They have almost 20 million

subscribers.

>> Tomorrow I'm going to do an IQ test for

real. Before I do that, I want to try to

improve my score and so I'm going to try

to do a whole bunch of practice tests. I

think this test is trainable, but

tomorrow we're going to see whether

that's true or not.

>> So are IQ tests trainable? It's a really

interesting question. If you take any

specific type of question on an IQ test,

for example, vocabulary questions or

puzzle solving questions, you'll find

that you can actually improve at them

through practice. If you do a whole

bunch of those questions, especially if

you kind of break down the mistakes you

made, you can get better at them. Does

that mean that IQ is trainable, though?

And the research is really interesting

on this. It finds that while you can

improve at any specific type of IQ

question, what you don't find is strong

generalization to other types of IQ

questions. So generally an IQ test will

have a wide variety of types of

intelligence tasks on them. And getting

better at one type of task tends to not

lead to improvements at other types of

tasks. So it's a tricky question. Yes,

you can get better at IQ tests. If you

practice whatever types of questions are

on that test, but that doesn't

necessarily mean you're raising your IQ

because if you go to another IQ test

that just happens to use different types

of questions, it won't generally make

you better at it. There's actually a

really interesting example of this from

the academic literature. A researcher,

Anders Ericson, was studying working

memory and he had people try to remember

digits. And he had one study participant

in particular who started with an

ability to remember about seven digits

at a time, which is very normal, very

typical. But over 2 years, he had this

person do hundreds of sessions where

they had to try to remember digits.

Well, by the end, this person was able

to remember something like 80 digits at

a time, which is way beyond what normal

people are able to do. And it wasn't

because this person had an incredible

IQ. It was because they practiced and

practiced and practiced and their brain

invented strategies for getting better

and better at remembering digits. And I

think this is a really important thing

to understand about intelligence. You

can get better at any aspect of

intelligence that you want that you want

to practice. Go pick that particular

task, practice it, break down your

mistakes, and you'll get better and

better and better. But that's different

than saying you're improving your IQ.

They're simply improving their ability

at that type of task. This is a YouTube

short about IQ that has almost 10

million views. The person with the

highest IQ of all time could at 18

months read through newspapers and at 5

years old calculate the day of any date

in the past 10 years. He got into

Harvard University at 11 years old and

could speak 25 languages. His name is

William James Situs and he is estimated

to have had an IQ of 275.

>> So notice I use the word estimated. So

Situs never took an IQ test as far as

anyone's aware. I don't even know that

IQ test existed at the time. And in

fact, he was born just before 1900 and

IQ testing was very early and there's no

way he took a valid IQ test that could

measure his IQ at that level. In fact,

if today someone claims they took an IQ

test with their IQ greater than 160, you

should be very skeptical because most IQ

tests don't even measure IQ's above 160

reliably. The reason for this is that if

you have extremely high IQ, you will

tend to get all or almost all questions

right on a standard IQ test, which means

that you can't tell how good people are

because if they max out all the answers,

well, you can't tell the difference

between someone with very high IQ and an

extremely extremely high IQ. You have to

actually develop special tests to even

measure IQ's that high. Most tests are

not equipped to do this. And what's so

ridiculous about this claim is that

funnily enough, if IQ were normally

distributed if it follows a bell curve,

as many people believe, then an IQ this

high should actually essentially never

happen in the history of the world. Not

even close. Situs, I'm sure, was a very

smart person. He was a child prodigy,

but he doesn't have an IQ this high. And

in fact, if we think about the highest

IQ person in the world, there's a very

high likelihood that that person never

got an IQ test and that nobody knows who

they are. It's probably some random

person who went about and lived their

life. They may not have ever discovered

the theory of relativity or anything

like that. Maybe they were really good

at their job. Maybe they did really

interesting things, but chances are they

never had an IQ test and nobody's ever

heard of them. This video is from the

Jubilee channel. They have 10 million

subscribers and they like to do a lot of

social experiments. In this case, they

have people rank their own and other

people's intelligence.

>> Why don't you guys rank each other from

1 to six, including yourself, and give

reasons why you want to do what you're

doing?

>> Got to go one.

>> No.

>> Oh, yeah. Two. Three. four, five,

[music] six.

>> So, one thing that's happening here is

that they're ranking each other based on

intelligence. Intelligence is a broad

concept and different people are going

to use that phrase in different ways.

For example, some people might emphasize

social skills more in their definition

of intelligence. Other people might

think, well, social skills are not about

intelligence. Intelligence is about

abstract reasoning or logic or something

like that. So, there's an inherent

ambiguity in these rankings.

>> 1 2 3

4 5 6. So, one thing that's interesting

here is they're actually having to rate

their own intelligence. And we actually

did some research on this and we found

some really fascinating findings. So,

this is our site personality map where

we make available over a million

correlations about humans. And you can

actually go sign up and you can search

it. You can find all the different kinds

of things about how human traits are

related to other human traits. We

actually collected data on people's

overestimation of their own IQ. We had

them estimate what IQ percentile they

are and then we actually measure their

IQ percentile and we look at the

difference between those. And it's

really interesting to look at what

predicts people overestimating their IQ.

We first have a bunch of variables

related to just performance. And that

makes a lot of sense because of what's

known as the Dunning Krueger effect.

It's well known that people with higher

IQs tend to underestimate their IQ and

people with lower IQs tend to

overestimate it. And so that's kind of

the first factor we find here. But it's

funny because we also find another

factor here. We also find this other

interesting factor which is narcissism.

The more narcissistic people are, the

more they tend to overestimate their IQ.

This video is a clip from the show Adam

Ruins Everything by Adam Conover.

>> We think IQ tests are objective

evaluations of our mental abilities, but

in reality they are deeply biased and

controversial tools that might not

predict intelligence at all.

>> You know, it's interesting might not

predict intelligence at all. I mean that

statement might is doing a lot of work

there. But I think almost any way you

measure intelligence, you'll find that

IQ is at least correlated with it. That

being said, people can define

intelligence differently. Okay. So, not

everyone's going to agree on what is

intelligence exactly. And the

correlation between IQ and intelligence

will depend on how you choose to define

intelligence.

>> Nope. The only thing IQ tests tell you

is how good you are at IQ test.

>> That's definitely not true. I mean,

people have studied what IQ tests

predict and they predict a wide range of

different things. They to some extent

predict income. They predict education,

grades in school. So, yeah, IQ tests

definitely don't just measure what's on

an IQ test. That's the whole point. If

they just did that, they would be

useless. If you look at any research,

you'll find immediately that they do

predict a wide range of things. And in

fact, the whole idea of an IQ test is

that it's designed to be one number that

predicts performance across a wide range

of tasks. That is literally how you

measure IQ. You give people a range of

intelligence tasks that are different

from each other and you look for the

single number that predicts across them

as well as possible.

>> Early tests focus almost entirely on

concepts that only rich white folks

would know.

>> That can't be true. Well, I'm not rich

and I can answer these questions.

Describe everything required by a riata.

>> Well, there absolutely are ways of

measuring IQ that are extremely biased.

Right? If you ask for factual knowledge

that tends to be cultural, then you're

going to find that some people will

perform better because of their cultural

background. There have been attempts to

try to make tests that are less

culturally based. For example, the

Ravens Progressive Matrices test. What

it does is it shows you nine boxes and

eight of them are filled in with symbols

and you have to try to guess what the

next symbol is, what the ninth symbol

is, essentially looking for the pattern.

And the idea here is well that doesn't

depend on knowledge about the world.

That might be cultural. However, it

could still be a bit problematic because

if you grew up in a culture where you

tend to do tasks of that type, even if

it's not exactly the same, you'll have

more practice where if you come from a

culture where you've never seen anything

like that before, then you'll have more

of a disadvantage of doing that kind of

task. This video is from Brainy Dose.

They have 2.4 million subscribers.

>> Have you ever encountered someone who

really wanted everyone to [music] think

they were smart? And whether they were

using big words or just flaunting their

nerdy hobby, it was quite obvious that

they weren't as bright as they tried to

appear. While smart individuals are more

likely to believe that they aren't

particularly smart.

>> So that's actually false. people who are

well at least if we think of IQ as a

measure of intelligence. People with

higher IQs do tend to believe they have

high IQs. So there is a link between

having a high IQ and believing you have

a high IQ. I think a lot of times people

get confused about this because they

think about things like the Dunning

Krueger effect which is that people with

higher IQs tend to underestimate their

IQ which is true. So higher IQ people do

tend to think they have higher IQs but

they also tend to underestimate their

IQ. Both of those things are true

simultaneously. So as we can see in this

chart for example the blue line here the

trend line does go up. People with

higher measured IQ do assess their IQ as

being higher. Not by a lot, but they do

assess it as higher. So higher IQ people

do think they have higher IQs on

average, but they also tend to

underestimate their IQs. I hope you

found today's discussion informative. If

you're interested in understanding your

intelligence better, we have a cognitive

assessment we make. You can find it on

our website clearthinking.org. It

measures seven different aspects of your

intelligence and gives you a full report

to help you understand yourself better.

You might enjoy our video where we ran a

giant study in intelligence and we

tested many different claims about it.

If you found this video interesting, I'd

really appreciate it if you'd subscribe

to our channel.