IQ gets a pretty bad rap.
I'll be the first to say that IQ isn't a perfect measure of intelligence. IQ kept me out of gifted education in high school. IQ once gave me heartburn. And I still blame my first IQ test taking experience for my current "IQ flashbacks" (much worse than acid flashbacks) every time I have to take a timed, standardized, multiple choice test.
So whenever I read a bit of IQ bashing rhetoric-from paleontologist Stephen J. Gould's The Mismeasure of Man to Journalist Stephen Murdoch's recent book IQ: A Smart History of a Failed Idea-my gut reaction is "Yeah, IQ sucks!"
Over time though, I've gotten a lot better about inhibiting such emotionally-driven reactions.
While the IQ test certainly has a less than pretty history of abuse and misuse, I actually think the test can sometimes—when administered correctly and properly interpreted—be useful. Unfortunately, IQ is often misunderstood, misinterpreted, over interpreted, and over hyped by a number of test administrators, journalists, scientists, and laypeople.
So let's have a little heart to heart here and briefly touch on a commonly used pawn amongst IQ bashers: The Flynn Effect.
A caption in Malcolm Gladwell's book review of James R.Flynn's book What is Intelligence?: Beyond the Flynn Effect reads:
"If what I.Q. tests measure is immutable and innate, what explains the Flynn effect—the steady rise in scores across generations?"
This would be a reasonable question if it weren't for one fact: the Flynn effect—the rise in IQ across generations—doesn't tell us much about the heritability of IQ—the extent to which genes explain differences in IQ within a generation in a particular population.
To better understand why this is, we have to understand the idea of a "social multiplier."
I spoke about "individual multipliers" in an earlier post. Remember Bo and Mo? Bo started with a slight genetic advantage that led him to self-select stimulating environments which fostered further gains in his ability. The environment thus took his tiny advantage and "multiplied" it again and again as such interactions were reiterated through the course of his development. All this, while the discrepancy in ability between Bo and Mo got greater and greater.
Well, just as individual multipliers can operate within a generation to stretch and exaggerate what might begin as small genetic differences between one person and another, "social multipliers" can operate across generations.
This is the basic gist of Dickens and Flynn's social multiplier model . Culture decides what manifestations of intelligence it wishes to educate, spotlight, and nurture. To see how this could operate, let's take my favorite example-breakdancing.
Within a generation, we may find that people who seem naturally inclined to dunk a basketball, run 100 yards, and dive into a pool also tend to make good breakdancers. In other words, within a generation, researchers may find a statistical tendency for these four skills to positively correlate with one another.
But imagine that breakdancing all of a sudden becomes an Olympic sport. We shift our interest in watching basketball to watching breakdancers perform. We drop more money into educating everyone in the fine art of the baby-freeze
, the windmill
, and the headstand
. Breakdancing makes a triumphant return, appearing everywhere--on stamps, in grade school classrooms, and on the streets.
What happens? Well for one, the ante is upped. People who are more naturally inclined toward breakdancing will still have an advantage, but the average standard of performance will be greatly increased. Aspiring breakdancers will have to learn increasingly complex moves in order to stay competitive.
In a couple of generations, breakdancing researchers test everyone, and find that people who are generally athletic still tend to have greater breakdancing potential, but that breakdancing ability on average is on the rise. Well, duh. This is all society's doing!
This example can be translated to the rise in IQ. Within a generation, various tests of cognitive ability correlate with each other—people who do well on one test tend to do well on the others. But across generations, tests involving abstract generalizations have increased the most. Performance on tests involving memory, vocabulary, and general knowledge have hardly budged.
Various social multipliers have been proposed to account for this increase-factors such as better nutrition, increased test familiarity, heterosis, increased scientific education (see prior post), video games, TV show complexity, modernization, etc.
Most certainly, a combination of all of these factors contributed to the rise.
But what does all of this have to do about whether IQ is innate and immutable?
Let's re-visit Gladwell's review. At one point he says:
"If whatever the thing is that I.Q. tests measure can jump so much in a generation, it can't be all that immutable and it doesn't look all that innate."
This would be correct, if only the Flynn effect indeed shows that the IQ of a single person jumps like a jelly bean within a generation. Instead, the IQ gains have only been documented on populations across generations. Within a generation, the rank ordering of IQ remains fairly stable.
So what does the Flynn effect tell us? It does teach us a very important lesson: The average IQ of a population can be influenced by environmental factors. When we give more opportunity for people to prosper, more people prosper.
But does this bear on the issue of whether IQ is "immutable" and "innate"? Not really. Because the Flynn effect controls for genes. It takes genetic differences out of the picture, and only looks at environmental effects that impact a population across generations.
So does the Flynn effect bear at all on the nature-nurture issue? Surely. The Flynn effect suggests that IQ isn't determined just by genes alone; environment matters.
But then again, we already knew that. Extreme genetic determinists, or as Gladwell calls them, 'IQ Fundamentalists' (I admit they do exist) shouldn't be trusted. But Behavioral geneticists will be the first to tell you that roughly half of the variability in IQ can be explained by environmental factors. And that whatever genetic effects there are could be indirect-such as in the case of individual multipliers discussed above.
And any good geneticist worth their weight in DNA will admit that genetics only determines the range of our potential; no single test, administered at one point in time, can perfectly test the absolute limits of human potential.
To be fair to Gladwell, his review is mostly thoughtful and I appreciate the fact that he is increasing awareness of important issues relating to intelligence, race, and society. He is quite right that the Flynn effect has implications for closing IQ gaps.
But the Flynn effect doesn't prove that genes don't matter in determining IQ. The effect just shows the power culture has in shaping the average ability of its inhabitants. While this doesn't render IQ useless, this is nonetheless a valuable lesson.
© 2008 Scott Barry Kaufman, All Rights Reserved
 Dickens, W.T., & Flynn, J.R. (2001). Heritability estimates versus large environmental effects: The IQ paradox resolved. Psychological Review, 2, 346-369. [pdf]