Genes and IQ: An Unsettled Omelet

Without our children, America doesn't mean much. It's our children, the entire population of American children, that make us what we are. Our children are the Americans who will be here tomorrow, and without them this country has little meaning.

The most important thing we do with our children is educate them, and the most important aspect of that process is the ability each child brings to school and what we do with that ability in the classroom.

The question of how much of what we call "ability" is inherited is maybe the most important question in education because the answer to the question influences nearly every decision made about the "educability" of various kinds of students, about segregation and desegregation in schools, about various kinds of "Head Start" programs, about the essence of education -- the education and guidance of our children from tots into adolescence and young adulthood.

Whatever we call it, "educability" is usually defined in terms of IQ tests, and performances on such tests are currently used as the basis for predicting how children will learn or not learn in school.

Let's ignore here any debates about the structure and content of so-called "standardized IQ tests" and simply accept the fact that they exist and are widely used in education as a measure of "learning ability".

In 1969, the psychologist Arthur R. Jensen became famous overnight when he announced in the media that the differences in IQ between Americans was due to genetics and not to environment. When asked how much IQ and scholastic achievement could be boosted in various groups, Jensen's answer was: "Not much."

For nearly forty years now psychologists and geneticists and statisticians have been split into two camps, with hundreds of research studies on the genetics versus environmental basis of IQ, and a great deal of vitriol thrown back and forth between "gene people" and "environment people".

Many affluent political conservatives are "gene people" -- they believe genes reign supreme in human affairs. They like the idea that IQ is fixed by inheritance because it means that affluent- class privileges are based on better genes and that tax money spent on educating poor low-IQ children is wasted. Our cousins the British made a fetish out of these ideas for half a century, a circus of fraud, arrogance, and public gullibility.

It's important to understand that the studies of IQ factors concern how much a factor -- genetics or environment -- contributes to the variation between individuals in performance, and not how much a factor contributes to the performance itself. A 40 percent contribution to variance does not mean that in any performance that factor contributes 40 percent to the biology that underlies that performance. All it means is that a statistical analysis says variation of a particular factor can account for 40 percent of the variance in the population.

In recent years, many biologically oriented psychologists have come to understand that there are many things dubious about certain statistical studies of the heritability of IQ, especially the idea that various contributions to IQ heritability are additive -- the idea that the contributions are independent of each other and simply sum up to comprise the total. We now have enough understanding about the interactions of environment and gene expression to know that this idea of linearity has no merit at all.

Assumptions about linearity are an old problem in science. It often happens that unless linearity is assumed, you cannot say anything at all because treating the full nonlinear relationship is at the moment too difficult. So an approximation is made -- linearity -- which is fine as long as one understands the limitations of the approximation. Unhappily, many psychometricians are too prone to forget the limitations of assumptions of linearity, and by the time the results of the studies reach the media and then the public, there's a rush to base public policy on a pseudoscientific game that in this case may severely handicap the lives of many children.

Until 2003, the general idea among many psychometricians was that genes contributed between 40 percent to 70 percent of the variance in IQ, with environment contributing the remainder. This idea permeated into the media and was touted by major newspapers and magazines.

Then in 2003 a new study, hardly made known to the public, tossed the IQ omelet into the air, and the omelet is still spinning and hasn't come down yet. The study (Psychological Science 2003 14:623-628) was by Eric Turkheimer and a group at the University of Virginia, and the conclusion of the Turkheimer team was that in poor families nearly all the variance in IQ is accounted for by a combination of fetal and other environments, and that the contribution of genes is close to zero. In contrast, in affluent families, the result is almost the reverse.

What does it mean?

One reasonable interpretation of the study by the Turkheimer group is that among poor children genetic differences contribute almost nothing to the measured variance of IQ because environmental damage, both fetal and postnatal, overwhelms all other variables in accounting for IQ variation. In contrast, in the middle and upper classes, in which fetal and postnatal damage to the nervous system is much reduced and hardly variable from one family to the next, genetic differences account for most of the variation in IQ.

In plain English, for middle and upper class children, differences in IQ can be explained mostly by genetic differences, while in lower class children, differences in IQ are explained mostly by non-genetic differences (fetal and postnatal environments).

Here are the words of the authors themselves:

Our study and the ones that preceded it have begun to converge on the hypothesis that the developmental forces at work in poor environments are qualitatively different from those at work in adequate ones.

The old question of whether IQ performance of lower class children can be substantially improved is now answered by a resounding YES, in contradiction to Arthur Jensen's famous "not very much" -- a sorry media-exploited comment by Jensen based on insufficient data.

People cannot be manipulated like laboratory rats, which means we're severely limited in the kinds of studies and experiments that are possible. The consequence is a limitation in the validity of many of our conclusions about human behavior -- which in turn means we need to be extra careful in how we apply such conclusions to public policy.

Given a six-year old lower-class child standing in front of us, it's not possible to change the fetal environment of that child, and we usually cannot do much to immediately change the home environment of that child. But if we give that child at least a middle-class educational environment, we will be doing our best to allow that child to fulfill its potential as a human being. Anything less is a disservice to that child, a disservice to humanity, and a contradiction to our American ideals.

Comments

[HeevenSteven]

Oops, I posted the last comment on the wrong blog.
Sorry!

[HeevenSteven]

Interesting post. I've had "Junk Science" on my Amazon wish list for a while; I just added it to my cart.

This post just sparked memories of a couple things I’ve read over the last few years; specifically Richard Rhodes’ "The Making of the Atomic Bomb": In a chapter titled "Men From Mars" Rhodes discusses the Ashkenazi Jews in late 19th Century Europe producing such a mind boggling collection of geniuses; and Nicholas Wade’s, “Before the Dawn” if I remember correctly (maybe not) discussed theories of evolution selecting for high intelligence in genetically isolated groups of people -i.e. the Ashkenazim.

I understand that these theories are controversial at best, but they are widely disseminated, as my two examples demonstrate. Rhodes’ book won a Pulitzer, and Wades was on some best-seller lists for a time.

Is it even possible to study this stuff objectively without shooting your foot off?

[DanAgin]

In the text above, "fetal and shared environments" should be "fetal and other environments".

Dan Agin

[SamThornton]

IQ tests themselves are extremely indirect methods of measuring intelligence, if intelligence is a biological characteristic.

So far as I know, there is no scientific methodology that can correlate physical brain structure or activity to what we refer to as "intelligence."

Absent those methods, IQ testing by current means is as much witchcraft as it is science.

[mrcontinental]

Rich children go to the university and poor children go into the penal system and that will not change.

[nainam97]

It's funny---today an (admittedly racist) man at my gym was commenting on why some people say "ax" instead of "ask."

I have had some extremely intelligent people "ax" me a question and I have had some really dumb people "ask" me a question.

I don't know how this fits into the school environment vs. the home environment. Maybe it's simply a matter of how we hear and learn a language, which is usually at home before we ever go to school.

Just to use an entirely different kind of comparison, I happened to live in a Thai village when I was a child. Thai is a tonal language which if you don't learn very early on, you'll never really "hear" the tones and pronounciation.

I know a lot of people with extremely high IQs who have studied Thai for years and lived in Thailand for years, but because they never "heard" it early enough, their vocabulary and grammer are perfect but their pronounciation is awful.

Go figure. I don't have any answers to all these interesting question but I do wholeheartedly agree that all children should get the best education possible. The more you learn, the better, no matter how you may pronounce the things you learn.

[Kansas Evans]

White southerners used to use 'ax' until it became associated with Black speech. So, 'ax' isn't an indicator of anything.

There's no such thing as proper english, only standard english.

Also, Black english isn't 'lazy' english, but an English dialect all of its own. It has grammatical rules and everything. That's why we look at white people strangely when they try to speak black english. It's not because it's funny to hear black speech coming from a white face, it's because they don't speak proper black english.

Whatever you learn at home is hard to unlearn. White students have a great benefit that they're home language is also their school language while black students have to suffer through being shamed by their home languauge when it has nothing to do with IQ. In European countries, they tell students their home dialect is fine, but for success, they need to learn 'academic' dialect. In the US, it's called code-switching, and where code-switching is used as opposed the embarrasment, students succeed.

Lastly, even though black english is generally bad mouthed by mainstream culture, a lot of words and expression have been borrowed by mainstream culture from black english, like 'bad mouth.'

Hope that helped, and thanks for the opportunity to talk about a great interest of mine.

[Kansas Evans]

Really great article. Though, I'm disappointed that a study had to be done to prove the point. It seems like common sense. It's very obvious to me that a lower quality of prenatal healthcare and a toxic postnatal environment added to a poor education would equal lower IQ scores.

Maybe it IS common sense, and people don't wanna admit that they're not better, just more fortunate, than everyone else.

Anyway, great article.