How Much Does IQ Matter? - Spencer Greenberg - #978

**Chris Williamson** (0:00)
What was this study you did on intelligence?

**Spencer Greenberg** (0:04)
Yeah. So some people say that IQ is a pseudoscientific swindle, and other people think it fully captures everything about your intelligence and says who you are as a person fundamentally. And as you probably know, there was a big replication crisis in social science where many studies failed to replicate. And so we thought, hey, we can do our part by trying to replicate a lot of claims about IQ and intelligence. So we recruited over 3,000 people. We implemented 62 distinct intelligence tasks, which includes everything from like memorization, puzzle solving, math problems, spelling, reaction time. So everything you could possibly think about, you could do online automatically. And then we tested about 40 different claims that are made about intelligence. So we checked if they held up.

**Chris Williamson** (0:50)
Right. Yeah, it is. Intelligence is kind of like the barbell chalk and cheese. It's the, how would you say? It's like the cognitive Rorschach test. It's like, what do you, how you see it kind of tells me probably quite a lot about your priors coming into it. You're right. It's either the fundamental underpinning that explains all of the outcomes that you're going to get in life, or it is a Nazi-ish policy that never had any basis in science and should be totally disbanded.

**Spencer Greenberg** (1:21)
Yeah, it's pretty wild. And so the question is, yeah, what's really true about it? To what extent is academia right about it? To what extent are the lay public right about it? And so we explored a bunch of those questions.

**Chris Williamson** (1:31)
Why do you think, even before we get into that, why do you think it's so contentious?

**Spencer Greenberg** (1:35)
Yeah, I think it's a few things. One, it does have a horrendous history. It was used, for example, like the idea of IQ, or measuring intelligence, was used for forced sterilization, so that's pretty horrible. I mean, the Nazis got into really some evil stuff around thinking about how intelligent people are and murdering people on that basis. So I think that's part of it. I think another thing is, everyone kind of thinks, it's like, yeah, it's good to be organized, but if someone says you're disorganized, it doesn't feel like it cuts to the bone, like of who you are as a human, or it doesn't make you subhuman. But I think some people feel this about intelligence. If you're told you're unintelligent, that it sort of like says you're bad and have a fundamental way, or you're lesser. You're less of a person, yes. And I think that's, you know, and I think, so I think that it feels different.

**Chris Williamson** (2:20)
That's interesting. Intelligence is very close to our sense of self in a way that the cleanliness of our cupboards or like our obsessiveness or something isn't quite so much.

**Spencer Greenberg** (2:32)
Yeah, and it's really interesting because having run many different studies on human psychology, we find there are some things people can report accurately about themselves and there's some things they can't. For example, if you ask people how organized they are, they're pretty accurate actually. But if you ask people, are you intelligent?
Not so accurate. Like it's hard for people to self-assess because everyone has to believe that they're intelligent, right? Or if you ask people how rational they are, it's not something they can self-assess. And so actually one kind of just funny finding, we asked people to estimate their own IQ and then our study was able to get an accurate estimate of their IQ. And guess how related those two things are?

**Chris Williamson** (3:09)
Were people 50% right?

**Spencer Greenberg** (3:14)
Yeah, it turns out not even, right? The correlation was about 0.23. So yeah, people don't have a very accurate assessment. It's better than total random guessing, but not a very accurate assessment of their own intelligences, at least as far as IQ is concerned.

**Chris Williamson** (3:25)
Okay, let's get into the different claims about IQ. Which ones held up? Which ones totally fell apart under scrutiny?

**Spencer Greenberg** (3:33)
Totally. Well, should we talk about what IQ is maybe?

**Chris Williamson** (3:35)
Yeah, yeah, yeah. Let's explain that.

**Spencer Greenberg** (3:38)
What is this thing, IQ, right? Because if you really want to say, IQ is intelligence, well, is that really the case? Theoretically. And so we implemented these 62 distinct intelligence tasks. Each one is totally different. And then one simple question we can ask is, suppose you do well at one of the tasks, does it make you more likely to do well at the others? And it turns out it's almost always the case that being better at one makes you more likely at the others. Almost throughout all of the different 62 intelligence tasks. And this is a very strange finding. This was a finding that we replicated that occurred early on in the history of IQ. And it's not obvious that it would be the case. For example, you can imagine a world where being good at math, this says nothing about how good you are at, let's say, vocabulary or spelling, right? But that's not the world we live in. There's something funny that's going on about humans, where being good at one task makes you more likely to be good at the other tasks. And then IQ is kind of built on this observation.