At the intersection of evolution & intelligence

If you're at ASHG, a session you might want to attend, Scale Effects and Recent Brain Evolution: Theory and Preliminary Evidence. Here's the abstract:

What forces have driven human evolution since the grand human diaspora? In this paper, I argue that the scale effects so central to endogenous growth theory in the field of economics (e.g., Kremer's widely-cited "Population Growth and Technological Change: 1,000,000 B.C. to 1990," Quarterly Journal of Economics, 1993) have been important drivers of human brain development since the diaspora. Scale effects have made prominent appearances in recent explanations of continent-level outcomes. For instance, in Kremer's model, big continents create larger, denser, faster-growing populations. In Diamond's Guns, Germs, and Steel model, wide continents raise the chance that an innovation will arise at a given latitude, an innovation which can then disperse across that latitude, enriching those who live on wider continents. In both models, the Malthusian nature of pre-Industrial Revolution existence imposes strong conditions on the general equilibrium outcome. My model takes those channels as given, and works out the theoretical implications for the divergent evolution of human brains on these continents. Brains are biologically costly, so evolution will only select for larger brains if there is a substantial payoff. And since larger brains tend to have higher levels of intelligence [corr(Brain Size, IQ)= 0.4 in recent in brain-scan studies], larger brains tend to have more processing and memory power. Under certain parameter values, Kremer's and Diamond's models both imply that the payoff to a big brain--a brain that can better adopt someone else's ideas--will be higher on wider, larger continents. Thus, we would expect human populations living on larger, wider continents to develop larger, more powerful brains. I model this relationship formally. This result should only hold on average: intra-group diversity is central to evolutionary theory, and massive intra-group diversity is an important fact of quantitative human genetics. The main purpose of the paper is to set forth the model, but I include some tests of its implications. I discuss whether, as the model predicts, human brain size and average IQ correlate positively with continent size and continent width. Indeed, evidence generally supports this hypothesis. Further empirical testing of the model's predictions will occur as future researchers employ genetic diversity databases. I plan to present the results in a manner intelligible to non-economists.

Here's the info:
Session Title: Evolutionary and Population Genetics Session

Session Location: Exhibit Hall II, Convention Center Session Time: Wed 9:30AM-3:30PM

Program Number: 643/W Poster Board Number:301 Presentation Time: Wed, Oct 21, 2009, 1:00PM-2:00PM

More like this

Hmmm...

I discuss whether, as the model predicts...average IQ correlate[s] positively with continent size and continent width.

Cue the protests.

Nice model, it would be interesting to hear if there are any data. By which I mean genetic data, not phrenology and social science. Given that the story of human population increase, migration, and subsequent self-domestication is basically a story about parasites and diseases, I think it will be a long time before there is really good evidence for selection in recent human evolution for anything other than disease resistance and climate, two obliterating monsters of selection pressure. All forms of evolutionary psychology rest on the rickety premise that our psychology matters very much in our recent evolution compared to other features of our biology. And the evidence that more intelligent individual homo sapiens leave more children? Because in all the other apes this basically comes down to strength or social skills. And look around--IQ ain't social skills. Some people just love this topic, though, and when you've got literally nothing, you model. In ways that would make an economist blush.

I also like the subordinate clause assertion based on nothing that a bigger brain is better at borrowing someone else's ideas. Maybe-maybe-not. A just as cavalier (but also stupid and data-free) assertion might be that intelligence would be more strongly selected where you have to come up with survival skills on your own, rather than borrow them.

Finally, human size in general, which includes hat size, is a highly labile trait with dozens of QTLs and counting. Size selection pressure is almost certainly related to climate (gracile tropicals, thick sturdy vikings) and resources. With evidence.

The correlation between brain size and IQ: is that partialling out body size? What about women, who have smaller brains (but larger for body size) and similar IQs, and our taking over our universities, after we successfully kept them out for 800 years. What about secular changes in body and brain size over the past 100 years, associated with diet changes (and documented changes over the past 4000 years likely associated with the transition between various diets). The author is an economist, so he's good at numbers. Does he have a collaborator who can help with the teensy, technical details of the actual subject matter?

What intrigues me is that humans are so similar in intelligence. The argument can be made that some group, however defined, has a higher average intelligence than some other group, but the differences just are not so great.

The spread in ability between high and low (normal) intelligence people is so large that I would expect the low tail to disappear, but it doesn't seem that it is, in all sorts of different societies and environments. There are factors that balance out the benefit of being smart.

This has the potential to produce a really ugly session. If you go, stay near the exit.

By Bob Sykes (not verified) on 20 Oct 2009 #permalink

Its fun watching the cognitive dissonance of liberals when confronted with inconvenient truths of their own regarding biological group differences. Just like with many conservatives, belief in creationism runs strong.

"Its fun watching the cognitive dissonance of liberals when confronted with inconvenient truths of their own regarding biological group differences."

You should look up the definitions of psychobabble buzzwords before you use them. No one dispute biological differences between groups. The question is which ones you have evidence for, and what the source of the difference is. Biological does not equal genetic, and for group differences even in moderately (but variably) heritable traits like intelligence, the data just isn't there without making a lot of assumptions that make beverages shoot out of the nose of anyone who does real animal genetics (no gene x environment interaction or correlation, strictly additive QTL effects, the list goes on).

What I wonder is why so many people seem invested in the conclusion that there are genetically based between group differences in general intelligence in the absence of any genetic understanding of IQ and little genetic understanding of any group differences. Also interesting is that the people who are most vocal about this issue rarely seem to be the ones actually interesting in doing the work of understanding the genetic basis of cognitive functions.

Unlike economics or sociology, in science you would like to refute your null hypothesis, not just provide evidence for a favored model. Thus: there is no genetically-based difference in IQ that systematically varies by ancestry. Nothing even comes close to refuting this statement, partly because the biology of cognition is in its infancy, partly because understanding the population genetic structure of humans is in its infancy, and partly because there are proven non-genetic proximal causes that explain much and possibly all of the measured IQ differences. Since these differences are already small, whatever small percentage of between-group variability remains for genetics to explain is unlikely to do much explanatory work anyway.

"proven non-genetic proximal causes that explain much and possibly all of the measured IQ differences..."

Wrong. Blacks raised in white homes still only have IQs of 85 when they grow older, according to a famous interracial adoption study.

"Since these differences are already small, whatever small percentage of between-group variability remains for genetics to explain is unlikely to do much explanatory work anyway. "

...And wrong again. Put down the outdated S.J.Gould. Blacks consistently measure a full standard deviation below whites in IQ, no matter the socioeconomic status. Poor whites have higher IQ's than rich blacks.

The intractable black/white IQ difference is most often used to illustrate group differences in IQ because the difference is so pronounced. But similar differences occur, for example East Asians and ethnic Ashkenazi Jews are consistently higher than whites.

The study of the genetic basis of IQ is underdeveloped because of political pressure. Read GNXP's interview with Bruce Lahn, who caused a firestorm not too long ago by uncovering evidence for recent brain evolution in some groups (coincidentally, ones with a history of civilization) but not others (like Africa).

miko,

You're right that the argument in this paper is hardly sufficient, standing by itself, to demonstrate that there are genetic differences across racial lines in IQ. Nor is there yet any convincing argument for any particular explanation as to why, say, Europeans might have higher IQs than Africans.

Yet, evidently, the group of homo sapiens who are the ancestors of Europeans essentially went their separate way from its original African stock 50,000 years ago. That is a LOT of generations ago. It would require only very small differential selection from generation to generation between Europeans and Africans to accumulate a 1 SD change in average IQ, given how much variability already exists within groups of homo sapiens. And does it seem in any way plausible that natural selection does NOT favor the more intelligent? Very likely both Africans and Europeans are constantly being selected for on this trait -- but, again, it is very plausible that they have been so selected at slightly different rates. If Europeans have been more rigorously selected for on the basis of intelligence, perhaps it has to do with colder climates posing more demanding circumstances -- but who knows? What is apriori very plausible, however, is that the rate of selection might have been marginally different.

Given that apriori plausibility, it seems not unreasonable to account for the apparently intractable difference between, say, African-Americans and whites on IQ on a primarily genetic basis. The real problem of course is with the primarily environmental account: no feature in the environment that people have ever pointed to as potential cause of the gap has ever managed to make that gap go away, or even much diminished, when that feature is controlled for or manipulated.

"Blacks raised in white homes still only have IQs of 85 when they grow older, according to a famous interracial adoption study."

Wow, all of them are 85?!? That's amazing. Try for more clarity when you talk about statistics. Regardless, blacks raised in white homes are still black in America, black at school, black when they consume American media and popular culture, and probably underwent embryonic development in a black woman, who was probably poor and under a great deal of stress. No control group exists who is black-but-lives-as-white.

"Black" in the US refers to people who range from 0% to probably ~80% African ancestry, the balance being European, Polynesian, Asian, North or South American native, etc. Africa is the most genetically diverse continent--more diverse than the rest of the world combined, thanks to genetic drift. Skin color is one of the most highly labile human traits. Point being, self identifying as "black" in the US is a terrible proxy for ancestry--it's a junk science category. We won't even get into who counts as white. No study has demonstrated correlations between ancestry (genetically defined, not a questionnaire) and IQ. Garbage in, garbage out.

"Blacks consistently measure a full standard deviation below whites in IQ, no matter the socioeconomic status. Poor whites have higher IQ's than rich blacks."

Did you pull that out of your ass or here it on AM radio?

"The study of the genetic basis of IQ is underdeveloped because of political pressure"

Paranoid fantasy--speaking of AM radio, maybe you should ease off it. For the reasons above and more, links between human genetics and cognitive function is in very early stages. Many people are doing the necessary groundwork in animal models, which is how biology works. Behavioral genetics and cognitive neuroscience are currently two of the most productive and exciting branches of neuroscience. Comparative human genetics is about to explode, and has primarily been impeded by sequencing costs.

But some people who have a preferred answer are willing to skip the hard, informative, interesting science part and jump to the "cherry-pick some poorly designed social science studies and run every statistical test until I get the answer I want" and then scream about it at the top of their lungs. These people are debunked and dismissed, because they either produce garbage or purposefully juke their stats when they should know better, not to name names.

If you knew what Gould's arguments were (mainly about the statistical concept of "g" itself), you would know that the ones I have been making aren't them. I don't have a preferred answer regarding the relationship between IQ and ancestry, and I don't really think it's all that interesting. I do, however, have an understanding of what data are necessary to reach a reasonable conclusion, and they don't exist yet. What is interesting to me is why someone would invest so much personal conviction in the idea that racial differences in IQ are genetic.

"Black" in the US refers to people who range from 0% to probably ~80% African ancestry, the balance being European, Polynesian, Asian, North or South American native, etc.

no, the median is 80% african for african americans. 10% of self-identified african americans are more than 50% white. see here. ~60% are 12.5% or less white.

it looks like in the near future brazil might be the best test case where total genome content is decoupled from socially agreed upon racial phenotype. i.e., there are people who are more than 50% white who look black, and identify as such, and people who are 50% african who look white, and identify as such (since race in brazil is more a matter of phenotype than ancestry).

Thanks for the correction...that remains a striking smear of points for African Americans relative to the other groups. Failure to have genetic data as a factor is still a major weakness of IQ studies, although ones that stick to strict heritability are at least often methodologically sound. Agreed that Brazil could be a good test case, better yet will be datasets like the personal genome project, that will simply look at real allele frequency correlates of traits... even haplotypes have issues. I don't want to get in to "is race a biological category" - clearly it depends on how you define it. For studying genotype-phenotype relationships, race is just irrelevant once you have the genotypes.

.that remains a striking smear of points for African Americans relative to the other groups.

latinos are the same, the axis would just be perpendicular. at least mexicans. puerto ricans have enough amerindian that their distribution is less linear. american whites are overwhelmingly european, less than 5% non-european generally.

For studying genotype-phenotype relationships, race is just irrelevant once you have the genotypes.

yes, but a lot of the time you don't understand interaction effects, etc., of gentoypes assuming you have a full sequence. IOW, are you *always* comfortable just throwing population assignment out and assuming you have a rock-hard model of the genetic background? i agree that total sequences are going to obviate the need to population information (e.g., "are you black or white?") in many or most cases, but i think there can be valuable information there.

"no feature in the environment that people have ever pointed to as potential cause of the gap has ever managed to make that gap go away.."

A problem here that behavioral geneticists constantly confront is that there is no theory of the environment. It is poorly defined, difficult to quantify, and there is always a likelihood that "controls" turn into "more bias." (You are assuming you know what's important when you choose what to control for). The bottom line is you can't do human studies of this kind with the kind of rigor that biologists would find satisfying.

I don't think there is evidence that there is strong, persistent selection for intelligence. Do you know of any? Undoubtedly it varies and a good chunk of the variance is due to genetics (particularly under stable and healthy conditions). It does not follow that is recently selected for. I would guess that cognitive selection, if there has been any of significance in recent human history, would tend more toward those abilities useful in increasingly socialized environments. Fitness would be related to the ability to take advantage of others' abilities and charisma. I don't know how you could get this data, but I don't think there is a correlation between IQ and charisma, and there is currently (and no evidence that it's new) a negative correlation between IQ and reproductive fitness.

Miko, are you the same person as the Miko on MetaFilter?

By Jason Malloy (not verified) on 21 Oct 2009 #permalink

"Miko, are you the same person as the Miko on MetaFilter?"
No.

"yes, but a lot of the time you don't understand interaction effects, etc., of gentoypes assuming you have a full sequence."

I'm not sure how population assignments help with that--is that what you meant? Analyzing groups is useful, but finding significant differences between groups does not mean your grouping criteria is biologically causal to the between group difference of whatever you're measuring. Group differences are also informative in looking at heritability estimates, because it can give insight into gene x environment interaction... e.g. IQ is about twice as heritable a trait among affluent populations compared to the poor.

Large n for genomes + good trait measurements + (and this one's critical for avoiding spurious correlations) population structure data is going to make humans a respectable model organism soon. But to your point, what to do when the traits are morasses of dozens of small-effect QTLs with complex epistatic interactions? There was a recent spate of articles in Nature about the problems behavioral genetics faces...the kind of weird disciplinary navel gazing you usually get in other departments.

I'm not sure how population assignments help with that--is that what you meant? Analyzing groups is useful, but finding significant differences between groups does not mean your grouping criteria is biologically causal to the between group difference of whatever you're measuring.

let's assume that the group has a common demographic history. what i'm getting at is that common history may result in evolutionary genetic phenomena which apply to the common genetic background of this group. i mean, you still want to check to see if there are any outliers in that group, discard them, but wouldn't you find it more convenient to just start with the group itself first? in a world of near perfect omniscience this might not be important as you can reduce everything down to the fundamentals, but that's a long way off.

But to your point, what to do when the traits are morasses of dozens of small-effect QTLs with complex epistatic interactions?

you gotta solution? i'm curious. i'm feeling pessimistic, but perhaps that's cuz the days are getting shorter.

I see your point, and genome biology will reveal interesting group structures that will correlate variably well with geography and demography. But I don't think complex cognitive traits are likely to vary systematically among these groups, all things being equal. As I said on another thread, recent human evolution may be characterized by fairly brute selection pressures like climate and parasites. With regard to the theory in the original post...our cultures have evolved to fit the capacities of a general hominid brain, I doubt culture is much of a selection pressure on the brain, because the brain is already the most fantastically plastic biological entity there is. You've got Terence Deacon listed to the left here...basically his argument about language, generalized.

Yeah the pessimism cuts deep in human behavioral genetics... someone called it "the gloomy prospect." Fortunately for me, I don't think IQ is a particularly interesting trait--it's going on 100 with still a merely operational definition--so I don't sweat that one. Though I'm fascinated by the genetics of personality and hope someday there will be results that are reproducible and non-trivial. Anyway, a good reason to study worms or flies: awesome results.

As I said on another thread, recent human evolution may be characterized by fairly brute selection pressures like climate and parasites

but there might be correlated responses to these! though i grant you, i can't see a set of systematic biased correlated responses shifting a quantitative trait's mean value.

i think measures of IQ have great utility, so we probably disagree on that, though i've become more skeptical of general theoretical modes of IQ over the years. nevertheless, just to be clear the researcher above is also doing work on personality and its effect on economics.

Brains are biologically costly, so evolution will only select for larger brains if there is a substantial payoff. And since larger brains tend to have higher levels of intelligence [corr(Brain Size, IQ)= 0.4 in recent in brain-scan studies], larger brains tend to have more processing and memory power.

Mark Changizi has an interesting idea relating cities and brains. As brains and cities become larger, both need to maintain a certain level of interconnectedness (highways for cities and pyramidal neurons for brains) to maintain proper functioning. Based on that, a more interesting relationship to examine could be population density, neuronal density and IQ. The famous study of increased hippocampus size of taxi drivers in cities like London and New York, already suggests a link between spatial layout of cities and brain morphology.

A problem here that behavioral geneticists constantly confront is that there is no theory of the environment. It is poorly defined, difficult to quantify, and there is always a likelihood that "controls" turn into "more bias.

Are you familiar with twin (identical [MZ] and fraternal {DZ]) studies and gene-environment correlations? Here's a thought experiment that essentially summarizes what the problem is with IQ research using twin studies and the concepts of gene-environment correlations:

MZ twins reared in different environments: Twin 1 in "bad" environment (IQ = 100), Twin 2 in "good" environment (IQ = 90)

DZ twins reared in different environments: Twin 1 in "bad" environment (IQ = 90), Twin 2 in "good" environment (IQ = 100)

Given this hypothetical data set, is it possible for claims to be made about the influence of genes and environment on IQ?

I don't think there is evidence that there is strong, persistent selection for intelligence. Do you know of any?

For such a question to be answered, one most likely also has to: (a) assume there were qualitatively distinct forms of intelligence from which to select, or, there was a particular type of intelligence that got "better" via evolutionary mechanisms; (b) explain why that particular form of intelligence selected was adaptive (and whether this implies non-human animals don't have this intelligence); (c) how modern IQ tests [and particularly, the questions contained within them] are connected to evolution; and (d) assuming only one type of intelligence was selected for [doubtful], how observed average magnitude differences in this particular intelligence could exist given all anthropological evidence pointing to human origins in Africa.

I don't know how you could get this data, but I don't think there is a correlation between IQ and charisma,

From what I recall in the abnormal psych lit, socio/psychopaths have social intelligence (and above average standard IQs as I recall) that could be considered charismatic.

and there is currently (and no evidence that it's new) a negative correlation between IQ and reproductive fitness.

I've seen this in the developmental psych literature as well--negative correlation between mom's IQ and family size. However, this correlation has been shown to disappear with increased income.

By Tony Jeremiah (not verified) on 21 Oct 2009 #permalink

miko, you say,

"I don't think there is evidence that there is strong, persistent selection for intelligence. Do you know of any? Undoubtedly it varies and a good chunk of the variance is due to genetics (particularly under stable and healthy conditions). It does not follow that is recently selected for."

Do you really mean to offer that argument up seriously?

Now, you are certainly right that it's possible that intelligence has not recently been selected for. But how likely is that possibility?

Look, again, the group of homo sapiens that were the ancestors of Europeans broke off from their African forebears 50,000 years ago. Do you really imagine that a greater ability to solve problems, be it in hunting, or storing of food for the winter, or in agriculture, would not have had an effect on survival? The evolution of the homo sapiens species itself came about through a series of hominid species with larger and larger brains. Why on earth might that stop, as if the survival problem was simply completely solved, when, at the same time, virtually all homo sapiens groups were regularly dying from famine or starvation? (And a related point. You earlier claimed that the only important force of selection currently operating on human beings is that which selects for resistance to disease. Yet all animals, hominids included, are always subject to disease, and are always being selected to resist it. Nonetheless, animals develop other adaptive features simultaneously -- and hominids have so developed larger, more intelligent brains.)

I should think that the default position would be that intelligence has continued to be selected for among human beings throughout virtually all of the 50,000 years in which the European group has lived and bred separately from African groups. There is simply no reason to believe otherwise.

And what is the best evidence that the rates of selection for intelligence has, in fact, been different between Europeans and Africans? The intractable and quite substantial gap in IQ between African-Americans and whites. Basically, nothing -- nothing -- environmental has made any real dent on that gap, despite many decades of trying, and countless experiments to contrive some feature or set of environmental features that would diminish that gap appreciably.

Given that the genetic theory is both very plausible apriori, and that it explains perfectly the gap and its intractability, why not simply accept it as the scientifically favored hypothesis?

"Wow, all of them are 85?!? That's amazing. Try for more clarity when you talk about statistics. Regardless, blacks raised in white homes are still black in America, black at school, black when they consume American media and popular culture, and probably underwent embryonic development in a black woman, who was probably poor and under a great deal of stress. No control group exists who is black-but-lives-as-white."

True, but those who were of mixed ancestry but "still black in America" averaged between the BB adoptees and the WW adoptees.

http://www.gnxp.com/blog/2007/10/james-watson-tells-inconvenient-truth_…

Also see the various asian adoption studies here.

http://www.gnxp.com/MT2/archives/004064.html

"and there is currently (and no evidence that it's new) a negative correlation between IQ and reproductive fitness."

Have you seen Greg Clark's papers following his book 'A Farewell to Alms'?

"âBig men achieve their positions because they excel
in the things that matter in life, they are good talkers, they are courageous, they are skillful in exchanges of wealth.â (Sillitoe, 1978, 253). Success in these societies - status, pigs and wives â comes not from skill in production or innovation, but from success in war, social intercourse, and
social negotiations.

So the move to societies where violence is centralized and limited seems to be an important social evolution distinct from the achievement of settled agriculture. It was this that changed the nature of competition in societies for reproductive success towards strictly economic means. This evolution took place in some parts of the world, but notably not in others such as the Highlands of Papua New Guinea...

But the evidence from pre-industrial England suggests that economic success was highly hereditable. Given how hard it is to change by social policy traits that are acquired within families, in terms of modern social impact it does not matter whether the mechanism of inheritance was genetic or not.

But evidence from the modern world establishes that economic success has a very important genetic component. Putting these together we can establish that there must have been significant"

"A Response to Critics (European Review of Economic History, August, 2008)"

http://www.econ.ucdavis.edu/faculty/gclark/Farewell%20to%20Alms/EREH%20…

Surnames and Survival of the Richest (January, 2009)

http://www.econ.ucdavis.edu/faculty/gclark/Farewell%20to%20Alms/Clark%2…

Miko, did you see this discussion earlier in the year? Steve Hsu set out the argument quite well here I thought:

"Nicely summarized. I'd like to ask Mike which assumption he disagrees with -- if any, since it should now be clear that the way that between group difference would arise is not necessarily what he had in mind:

0) intelligence is partially heritable

1) there is plenty of extant genetic variation, probably due to a large number of genes of individually small effect -- no additional mutations are required

2) selection can act if reproductive rates are impacted by these genes (i.e., by intelligence)

3) simple estimates suggest that 50,000 *could have* been enough time to produce .5 SD (genetic) group differences

See here for some elaboration on (2) and (3) using Greg Clark's (economist at UC Davis) data on English inheritances and family size:
http://infoproc.blogspot.com/2008/12/recent-natural-selection-in-humans…

The simplest model would be that time since development of agriculture varies between groups, and this variation leads to different levels of selection for traits which might be more useful for agriculturalists than for hunter gatherers. As Razib pointed out, there is a lot of evidence for group differences; the question is only how much is due to genetic factors.

Finally, everyone here is (I think) reasonable and realizes this matter is not yet *resolved* scientifically. The question is whether one can *exclude* possibilities based on current evidence. Mike is (almost) asserting that you can exclude the possibility of (genetic) group differences in intelligence -- e.g., he might be 95% or 99% confident that they cannot exist. We are really only arguing about what the appropriate probability estimate (confidence level) should be. For Mike to be so sure he must be able to exclude some of the assumptions (0)-(3) with relatively high confidence. Care to elaborate?

Posted by: steve hsu | December 19, 2008 10:49 AM"

http://scienceblogs.com/gnxp/2008/12/complex_traits_evolution_1.php

Miko doesn't think there has been recent strong selective pressure for increased cognition. He doesn't know why, he just doesn't think so. Such an idea is just inexplicably offensive.

That is hilarious.

"Black" in the US refers to people who range from 0% to probably ~80% African ancestry, the balance being European, Polynesian, Asian, North or South American native, etc"

Oh please.

The average white admixture with American Blacks is about 14% (explanation for why American blacks have higher IQ's than Africans?). Other races are mostly negligible in the population.

"What I wonder is why so many people seem invested in the conclusion that there are genetically based between group differences in general intelligence ..."

Because the social policy implications are so important. We are not going to be able to make the rocket scientists of the future with the peasants of the 3rd world, no matter how much money we throw at education.

"Did you pull that out of your ass or here[sic] it on AM radio?"

The information is out there if you look for it. Poor whites (under 20k per year) outperform rich blacks in cognitive tests such as SAT scores. Look it up I'm not your research assistant. The old GNXP.com is a good place to start.

"I don't have a preferred answer regarding the relationship between IQ and ancestry, and I don't really think it's all that interesting. "

That explains why you are shocked! SHOCKED! to learn what people who DO find it interesting have to say about the matter.

"What is interesting to me is why someone would invest so much personal conviction in the idea that racial differences in IQ are genetic. "

Would you want to import millions of Mexicans with low IQ's if you knew about the heredity of IQ? That's just one social policy implication but there are others. I know your bigot radar probably goes off when racial differences are discussed, but truth be told, bigots do not like the subject either. It drives white supremacists crazy that certain Asian groups have higher IQ's than whites.

"What is interesting to me is why someone would invest so much personal conviction in the idea that racial differences in IQ are genetic."

The only one who seems to have strong conviction here is you. The extent of your evidence has been "well there is a lot we don't know" as if that is much of an argument. Actually it reminds me of those Christian fundies in their attempts to poke holes in evolution. There is a lot we don't know in every scientific field, and a lot we do know.

You can't honestly believe that intelligence is the only human trait that is distributed exactly evenly across every single human group. The Chinese do not make as good sprinters as West Africans, on average. This is genetic (different ratios of fast twitch muscle fibers). What is the hangup with intelligence? Only an egalitarian sense of political correctness.

I don't think, judging by your response, that you ever did read Bruce Lahn's interview with GNXP. Here you go
http://www.gnxp.com/blog/2006/10/10-questions-for-bruce-lahn_10.php

"Do you really mean to offer that argument up seriously?" [that there might not be persistent selection for intelligence]

Yes, for two reasons. The joke-but-plausible reason is that almost all domesticated animals are stupider than the wild species they are derived from. I happen to believe, and am willing to discuss, that humans are in some sense (but a real sense) self-domesticated. The arc of this domestication is decreasing selection for many abilities. Whatever level of stabilizing selection for intelligence may exist has certainly been declining for a long time. I've never bought the idea that you have to be smarter to be in an agricultural society than hunter-gatherers. In fact, I'd suggest the division of labor and loss of inter-personal accountability in large agricultural societies might favor "cheaters" in the evolutionary sense more than in small group hunter-gatherer cultures. Again, not that I think that there is much selection. Arguable, but just as supportable an assertion as the other.

The second is the plasticity argument I alluded to above. Nothing in biology comes close to the mammalian brain in general and human brains in particular in terms of plasticity. Once you have a broad phenotypic reaction norm, selection on variation with this range becomes difficult unless conditions are very stable over a long period.

"Basically, nothing -- nothing -- environmental has made any real dent on that gap" [between whites and blacks]

There is a double standard here. If a QTL that explains 5% of the variance in IQ (or any behavioral trait) is found, we break out the champagne and Newsweek has "Scientists Find Harvard Gene" on the cover. If a candidate environment effect is similarly "small" it is taken as evidence that the environment doesn't matter!

Let's remember what we're dealing with. QTLs are usually functionally distinct alleles of a gene. They are physical entities that are discretely measurable. We have a pretty good understanding of their mode of action (i.e. we can guess they will somehow affect neural development or physiology, or perhaps some metabolic process that indirectly affects the brain). Whatever: there is a very good theory of how genes work.

Environmental variables are themselves continuous, fluctuating, and hard to measure, making even establishing correlations difficult. We don't have general theories about how they work on brains or any part of our biology--generalities might not exist. Most biologists are trained to not be interested in them, and a good chunk of our time is spent unsuccessfully trying to remove them from our experiments.

Environmental effects are going to be like QTLs but much harder. There will be many, they will be of small individual effect, and they will interact in complex ways. Furthermore the interactions between environment and genotype will be complex. A hypothetical example: black Americans have a higher propensity to diabetes, probably genetic. The American diet, particulary for poor people, is also great at causing diabetes. Let's say this same metabolic difference had a small negative affect on cognitive development, but only when these adverse dietary conditions are experienced within a certain window of development. Is this a genetic difference, or an environmental one? The science answer is "both" as it is for any phenotype that exhibits a reaction norm. However, as a member of society, I'd argue the latter, because the difference--again, a small QTL-like difference--requires a strange and specific set of cultural circumstances to manifest.

MK...those assumptions are all correct. How likely one thinks systematic group differences are depends on some other assumptions about recent human selection pressures, as discussed above.

Razib...you make a good point that selection need not be "for" intelligence for drift/bottlenecks to create differences between groups. This likelihood probably decreases with increases in QTL number, but hard to say. *If* the differences hold up, that's probably a better explanation, since as you can tell I don't think selection particularly looks at IQ. Someday, everyone on this planet will be a descendant of the Octo-mom.

@ liberal biorealist

Now, you are certainly right that it's possible that intelligence has not recently been selected for. But how likely is that possibility?

Assuming there's a relationship between intelligence and brain size, is this even a topic of debate? From what I recall, hominid encephalization started about ~1.8 million years ago with H. erectus, with no significant encephalization until archaic H. sapien neanderthalensis ~400,000 years ago. Anatomically modern humans appeared ~100,000 ya. Interestingly, there's some evidence showing neanderthals skulls are larger than modern humans. So there should be some skepticism about the link between brain size and intelligence.

Look, again, the group of homo sapiens that were the ancestors of Europeans broke off from their African forebears 50,000 years ago.

There are two competing models for this idea: the recent out of Africa and multiregional hypotheses. The former assumes no gene flow between African and European populations; the latter assumes gene flow. Assuming the current IQ differences are real and are genetically based, this would support the recent out of Africa hypothesis. Also, it's important to remember that for a comprehensive evolutionary theory of IQ differences, one also has to consider the Asians (whose IQ, reportedly, is higher on average than Europeans) and Native Americans.

If we go with the out-of-Africa hypothesis, the current observed IQ differences could be considered, technically, to be not strictly due to the influence of genes on IQ (which presumably means brain functions associated with whatever IQ tests measure). Going with the story of the migration of a group of individuals from Africa to places outside of Africa, sounds more like there were personality differences in the original African family, such that one part of the family wanted to stay and another wanted to leave. (And presumably during that time, there weren't any significant phenotypical differences during that time that we make such a big deal about now).

There's a personality theory (Five Factor Model) asserting that humans differ in personality based on five basic traits (openness to experience, conscientiousness, extraversion, agreeableness, neuroticism). What would be interesting, is to see
whether there are racial differences in personality associated with IQ differences.

Do you really imagine that a greater ability to solve problems, be it in hunting, or storing of food for the winter, or in agriculture, would not have had an effect on survival? The evolution of the homo sapiens species itself came about through a series of hominid species with larger and larger brains.

Well, there's the old expression that necessity is the mother of invention. However, if one group stays in a particular location with which they are familiar, and another group moves toward unknown territory, it's possible that moving to unknown locations (especially involving significant changes in temperature towards the colder variety), are the foundations of invention (and brain size changes).

Why on earth might that stop, as if the survival problem was simply completely solved, when, at the same time, virtually all homo sapiens groups were regularly dying from famine or starvation?

It's been awhile since anthro class, but I don't believe this is an accurate portrayal of prehistoric events, based on the out-of-Africa model. Especially given there are still people living in Africa.

I should think that the default position would be that intelligence has continued to be selected for among human beings throughout virtually all of the 50,000 years in which the European group has lived and bred separately from African groups. There is simply no reason to believe otherwise.

Let's assume that to be true based on the out of Africa model. A fundamental question that needs to be addressed, is the process by which this intelligence was selected. Was it due to the need to adapt to environmental pressures (as suggested by the comment about ability to solve problems), or, could it be due to personality differences that motivates one to move away from a particular location that forces one to adapt. Additionally, might those personality differences also cause a change in the social environment of a particular group, that could lead to social differences in how members of a group select mates. According to Geoff Miller's hypothesis , the rapid change in brain size is due to runaway sexual selection, which is more likely due to changes in the social environment of a group rather than an adaptation to changes in the physical environment. So the question remains whether racial differences in brain and body morphology according to this study are due to runaway sexual selection or environmental selection pressures. The former is likely more likely related to social influences, the latter more to an actual need to adapt to the environment. Either way, both technically point to environmental factors (social and physical) influencing changes in genetic makeup.

It's puzzling that the underlying assumption of racial differences in IQ seems to be genetic determinism, despite the fact that evolutionary processes seem to indicate anything but genetic determinism (especially given that evolution implies changes in genetic makeup over time, usually associated with environmental factors).

And if we really want to consider intelligence from a more practical point of view, we should really consider the question of what happened to the dinosaurs, in the context of assuming global warming is true, what presumably is the cause of global warming, and whether ancient human civilizations that still remain to this day (as well as cockroaches) are more likely to survive than presumably advanced civilizations.

By Tony Jeremiah (not verified) on 22 Oct 2009 #permalink

According to Geoff Miller's hypothesis , the rapid change in brain size is due to runaway sexual selection,

miller rejects this hypothesis now last i checked. the math doesn't work because encephalization is too gradual to have been runaway (says so in mating mind).

miller rejects this hypothesis now last i checked. the math doesn't work because encephalization is too gradual to have been runaway (says so in mating mind).

Thanks for this info. I had come across that hypothesis in a class somewhere in the mid 90s, so not too surprised of the change.

From the info at this site, it looks like Miller has changed the hypothesis from runaway sexual selection to mutual mate selection. Looks like runaway sexual selection was based on the idea of females primarily doing the choosing. He's also rejecting the hypothesis based on its prediction of gender differences in brain size, but research is apparently showing no statistically significant differences. (I guess that depends on what aspect of the brain is being measured. I recall some study suggesting significant gender differences in corpus callosum size--larger in women).

At any rate there's still a lot of questions to be answered: In what way are IQ and brain size related (larger size = more neurons?). If so, it seems that neuronal density might be a better measure; assuming brain and body morphological differences among the racial groups, at what point in evolutionary history would that have arisen and why? Assuming brain size has something to do with IQ, the likely place to look for the onset of racial differences in IQ probably has something to do with the evolutionary progression from H. erectus to neandertalis to anatomically modern humans, whose evolution is currently best explained by the out-of-Africa hypothesis; assuming the out-of-Africa hypothesis is the place to look for an IQ divergence, what is the cause of this divergence (individual differences in personality, social factors, environmental factors)?; Is it possible to operationally define prehistorical intelligence based on modern IQ tests?

The only reasonable modern assessment I could see related to prehistoric intelligence, is the Five Factor model of personality, which has one trait (openness to experience) known to correlate (~0.3) significantly with two types of intelligence (g factor and crystallized). Persons high on Openness to Experience usually rank high on imagination and creativity. And it's personality traits that would seem to be the most obvious observable trait in a prehistoric environment.

So it's still a tough call to answer whether social, environmental, or both pressures were present during prehistory, which explains (the presumed) observed racial IQ differences, or whether racial IQ differences are in fact a modern phenomenon linked to a history of racial discrimnation.

Given one can identify at least four types of factors that can impact intelligence: psychologically (i.e., stereotype threat), sociological (i.e., the Pygmalion/self-fulfiling prophecy effect), environmental (i.e., nutritional), and genetics (i.e., the impact of genes on brain functioning), I'd be interested in seeing if a model that adds all of those factors together, closes the (presumed) racial differences in IQ, and what % of that model is actually accounted for by the direct influence of genes.

By Tony Jeremiah (not verified) on 22 Oct 2009 #permalink

"what way are IQ and brain size related (larger size = more neurons?). If so, it seems that neuronal density might be a better measure;"

There is a good summary of the neurobiological basis of intelligence here by neuroscientisst Paul Thompson & psychologist Jeremy Gray.

www.loni.ucla.edu/~thompson/PDF/nrn0604-GrayThompson.pdf

A recent twin study of Thompson's was also reported earlier in the year showing the importance of myelin quality.

"The UCLA researchers took the study a step further by comparing the white matter architecture of identical twins, who share almost all their DNA, and fraternal twins, who share only half. Results showed that the quality of the white matter is highly genetically determined, although the influence of genetics varies by brain area. According to the findings, about 85 percent of the variation in white matter in the parietal lobe, which is involved in mathematics, logic, and visual-spatial skills, can be attributed to genetics. But only about 45 percent of the variation in the temporal lobe, which plays a central role in learning and memory, appears to be inherited.

Thompson and his collaborators also analyzed the twins' DNA, and they are now looking for specific genetic variations that are linked to the quality of the brain's white matter. The researchers have already found a candidate--the gene for a protein called BDNF, which promotes cell growth. "People with one variation have more intact fibers," says Thompson."

http://www.technologyreview.com/biomedicine/22333/page2/

There seem to have been some adaptations including central nervous system genes, which may have increased fitness in a hierarchical agricultural society.

For example new versions of serotonin transporters in both Europe and east Asia (such as SLC6A4). There are other neurotransmitter-related changes, also changes in genes that affect brain development. East Asians have a new version of DAB1, a gene involved in the development of the layers of the cerebral cortex, while there is a fairly common new version of NKX2-2 (a brain homeobox gene) in Europeans.

http://biology.plosjournals.org/perlserv/?request=get-document&doi=10.1…%

http://www.plosgenetics.org/article/info:doi/10.1371/journal.pgen.00300…

"...which may have increased fitness in a hierarchical agricultural society."

That's exactly the kind of out-of-ass speculation that drives me crazy about evopsych. I'm getting repetitive, but as far as I know the only cases where an actual selection pressure can be correlated with a selective sweep is in the case of disease resistance mechanisms, and possibly food metabolism.

There is a good summary of the neurobiological basis of intelligence here by neuroscientisst Paul Thompson & psychologist Jeremy Gray.

*www.loni.ucla.edu/~thompson/PDF/nrn0604-GrayThompson.pdf

*http://www.technologyreview.com/biomedicine/22333/page2/

Interesting article and website. The main issues I have with them include:

(a) intelligence/IQ [which should not be used interchangeably]) actually refers to a specific type of thinking involving "mathematics, logic, and visual-spatial skills", and associated with a specific brain region (i.e., the parietal lobe). There are additional forms of intelligence/thinking according to Howard Gardner and Robert Sternberg, presumably involving additional brain regions.

(b) h2(x) = 2(rMZ(x)ârDZ(x))

* I see this heritability estimate is still being used. The problem here, is that due to genetic-environmental correlations, the heritability estimate should be viewed with skepticism. This is because measurements like rMZ and rDZ cannot be decoupled from environmental effects. For example, research reveals that MZ twins become increasingly similar in IQ over time, while DZ twins become increasingly dissimilar. If one were to use the heritability measurement, it would show IQ heritability increases over time. This is an active genetic-environmental correlation better known as niche-picking, with the explanation being that people choose environments fitting their genetic makeup over time. (And this would have to assume that one has the ability to choose their environments over time.)

A better measurement would be more along the lines of (T= raised Together, A = raised Apart):

*rMZ[T] - rMZ[A] = eMZ (environmental effect for identical twins)

*rDZ[T] - rDZ[A] = eDZ (environmental effect for fraternal twins)

*1.0 - eMZ = gMZ (genetic effect for MZ twins)

*0.5 - eDZ = gDZ (genetic effect for DZ twins)

*h(2) = 2[gMZ - gDZ]

Thompson and his collaborators also analyzed the twins' DNA, and they are now looking for specific genetic variations that are linked to the quality of the brain's white matter. The researchers have already found a candidate--the gene for a protein called BDNF, which promotes cell growth. "People with one variation have more intact fibers," says Thompson

Seems as though this would be quite useful for understanding the mechanisms of Alzheimers disease, given the white matter breakdown associated with it. Might be interesting (if possible) to see if MZ twins with and without Alzheimers exist.

For example new versions of serotonin transporters in both Europe and east Asia (such as SLC6A4). There are other neurotransmitter-related changes, also changes in genes that affect brain development. East Asians have a new version of DAB1, a gene involved in the development of the layers of the cerebral cortex, while there is a fairly common new version of NKX2-2 (a brain homeobox gene) in Europeans.

I'll have to look into this further. It's currently beyond my basic background in biology to make an informed comment. But my first hunch, is that given the current trend toward epigentic research, I'm guessing the link between allelic influences and brain functioning won't be too straightforward.

By Tony Jeremiah (not verified) on 23 Oct 2009 #permalink

I have yet to hear any argument that justifies a statement like:
"I am 95 or 99% confident that group differences in IQ cannot be due to genetics"

So anyone holding that opinion is, in my opinion, relying on priors (biases) and not data.

The most justifiable scientific position is, in my opinion, that we *don't know* and group genetic differences in intelligence are a distinct possibility. (The "gloomy prospect" that Miko mentions may turn out to be the case; in which case skeptics will never be convinced at 99% confidence level one way or another.)

Regarding the utility of the construct "IQ" (Miko hinted that he doesn't believe in that), I think the evidence is *overwhelmingly* in favor, and in fact society uses IQ to allocate scarce resources like access to higher education (SAT), advanced training (GRE, LSAT), etc.

Miko, you seem like a smart guy -- I suggest you dig into the literature on psychometrics.

http://infoproc.blogspot.com/2009/01/psychometrics-links.html
http://infoproc.blogspot.com/search/label/iq

To my taste, one of the most convincing things in psychometrica is the relationship of /g/ to the complexity, and speed of resolution, of EEG waveforms evoked by banal stimuli such as repeated clicking sounds. As discussed in Jensen's book. As I recall, they do something like 120 clicks over two minutes. You can stop consciously reacting to the clicks, of course, but obviously your brain continues to react every time, in order to provide (do?) the aural qualia, and decide subconsciously that the salience of the stimulus is low.

Higher /g/ people have a more complex waveform that resolves faster. This finding is very like the /g/-related morphological white matter differences on MRI - but cooler.

Awfully tempting is the proposal that there is noise and error, on a neural level, when you try to think, which is amplified as the thought unfolds by structural imperfections detracting from /g/ (mildly deleterious mutations?). Some wrong neurons fire, and the error propagates. And if you are not smart enough to think that particular thought, the noise overwhelms the signal while you are trying to go through with it.

By Eric Johnson (not verified) on 25 Oct 2009 #permalink

Maybe I'm extending a moribund thread, but Steve raises a couple things I'd like to clarify.

I agree IQ predicts a lot of social outcomes and performance in various things, and there will be physiological correlates of it--it measures a difference of some kind. Also not at issue is that it varies, and that some of this variation is genetic. The question is: does the genetic component vary systematically and broadly according to ancestry.

One of my arguments--which is only an argument about a lazy assumption--is that IQ is not now, and there is no evidence that it has been in recent human history, correlated with reproductive fitness. This is not an argument about the existence of IQ or its modern socioeconomic correlations.

So all I'm saying is that don't have the data to answer this question, for all the reasons that have been raised in this thread. I think someday we might, though this makes the large assumption that intelligence is genetically tractable.

There are screamers on both sides of the debate who really think it HAS to be one way or another--well, it doesn't. Most of these opinions appear to me to be emotional or political. People are very easily swayed by "evidence" that supports their prior opinions, and easily reject that which does not, which creates the illusion of increasing certainty for both sides. As a preoccupation, I admit I find the "innatist" position distasteful as it is often used as a lazy justification for social inequity. I also think the outcome of this research is irrelevant for both policy and how we treat people individually.

As for confidence %s and where the burden of evidence lies, we are obliged by method to prefer the null hypothesis.

"I also think the outcome of this research is irrelevant for both policy and how we treat people individually."

Right, so countries with low birth rates can import groups from anywhere and can expect the same outcomes? If you assume groups are interchangeable & abilities are evenly distributed amongst them then that would be a sensible policy no?

"The question is: does the genetic component vary systematically and broadly according to ancestry."

Every measurement taken indicates it does. Next to someone's immediate family members, ethnicity is the single most accurate predictor of IQ.

"Every measurement taken indicates it does."

Tim, read the effing thread, no one's debating that it varies by ethnicity, it's over how good the evidence is and the extent to which that that variation is genetic.

"Right, so countries with low birth rates can import groups from anywhere and can expect the same outcomes?"

No, though if they import zygotes you might get close. However, in almost all cultures people who look different from expected norms are treated according to provincial prejudice, positive and negative.

***Right, so countries with low birth rates can import groups from anywhere and can expect the same outcomes?***

No, though if they import zygotes you might get close. However, in almost all cultures people who look different from expected norms are treated according to provincial prejudice, positive and negative."

That may be true, but it's pretty hard to keep people with ability in a certain area down. In terms of the Jews, Charles Murray comments:

"As soon as Jewish children born under legal emancipation had time to grow to adulthood, they started appearing in the first ranks of the arts and sciences. During the four decades from 1830 to 1870, when the first Jews to live under emancipation reached their forties, 16 significant Jewish figures appear. In the next four decades, from 1870 to 1910, the number jumps to 40. During the next four decades, 1910â1950, despite the contemporaneous devastation of European Jewry, the number of significant figures almost triples, to 114."

Similarly, Japanese brought over as indentured labour to work in Brazil, or Chinese and Indians in the Caribbean countries now excel despite being minorities. Also the Chinese in Malaysia, despite being a minority perform above average.

Then you have transracial adoption studies which also suggest hereditary differences are a major factor.

http://www.gnxp.com/MT2/archives/004064.html

Contra you, Miko, this issue makes a huge difference for the economic future of the USA and Europe. Not so much for northeast asia, where people seem to think a little more clearly than you. What they may realize is that it doesn't really matter, economically, whether some redoubtably obscure environmental influences create the entire black-white or white-hispanic difference (though that is very unlikely anyway). What matters is that you can't fix the difference.

By blue anonymous (not verified) on 27 Oct 2009 #permalink

I really don't follow what your argumenta are, Fang or blue. You guys throw out historical anecdotes as if they are data. One of the points I've been trying to make in this thread is that although we often have to lower the bar for what we can experimentally accomplish in human behavioral genetics, we shouldn't lower the bar for interpretation, in fact we should probably raise it. I'm a biologist, this a biological question, and there are standards for evidence and inference. You're jabbering about a dozen different peculiar and parochial sociopolitical situations that have little in common with each other, and nothing to do with human genetics. And by the way, all of those success stories you mentioned have in particular times and places been inferior races whose mental capacities were scientifically "proven" to be deficient.

"Contra you, Miko, this issue makes a huge difference for the economic future of the USA and Europe."
Why? Do you have any evidence IQ has anything to do with the economy? Anywhere at any time? Are you for real?

The environmental influences acting on IQ differences between ethnic groups are "obscure"? Name one genetic one.

Adoption studies are notoriously assumption-laden and problematic, and a classic case of bar-lowering for human studies, especially when interpreted by people who don't understand heretability estimates or their attendant assumptions. Your homework is to teach yourself why.

"What matters is that you can't fix the difference."

So, you're saying, whatever the source of the differences, because not being racist is hard, we embrace apathy and social inequity? Charming.

I was wondering how long it would take before miko threw the "racist!" slur. Yeah sure miko, one day we are commenting on GNXP and the next we are joining the Klan.

Let me ask you this. Would you marry and have children with someone with a low IQ? Probably not. Does ruling out a disproportionate amount of certain minorities from this pool of prospective brides make you a racist? Of course not. You just have standards. You don't want dumb kids.

So why would you want to import massive amounts of low IQ's to your own country's gene pool? Psychometry has shown IQ is mostly hereditary, and that education doesn't increase IQ - you can't fix stupid. So that makes accepting our 3rd world neighbor's extremely fecund elemenary school dropouts a bit of a bad idea, doesn't it?

> So, you're saying, whatever the source of the differences, because not being racist is hard, we embrace apathy and social inequity? Charming.

I'm saying massive Latin immigration into the US will probably its reduce per capita productivity, ceteris paribus. I am staunchly opposed to recognizing race in US law applying to citizens.

You will find that IQ is the single strongest correlate of individual job performance, although I acknowledge that it does not explain all that much of the variance. Things like corruption, which discourages entrepreneurship, probably contribute much to economic disparities between the US and Mexico. There may also be group-level IQ effects on productivity. In other words group performance need not be additively determined by individual differences because there could also be, as it were, dominance or epistasis; synergy.

I call environmental factors responsible for group differences in the USA "redoubtably obscure" because there are not many plausible ideas; in particular, controlling for income does not get rid of the differences. I welcome any efforts you or anyone may make in your career to uncover such factors so we can improve the world. I would not be in favor of stopping such research because of hopelessness at any point. However, I still doubt you will find anything. Obviously it's a different story outside the wealthy world where macro- and micronutrition have very large effects, but I am talking about the West.

Again, what matters is that so far you cannot alter the IQ differences or their consequences - or other differences. I take your emotional response to that as a tacit acknowledgment. You may be aware that 6% of fourth-generation latins in the US complete college, vs 35% for whites, a sixfold difference. Again, these are not recent immigrants. I doubt you can do much about that. I suspect the American economy will see decline. High-productivity persons of all races will see more and more of their income appropriated, their other choice being to not participate in the economy to their fullest capability. I don't really care that much. It's not going to affect me. I was simply born without much status aspiration or love of wealth, I just love having my time. As I read "Walden" at a fairly young age I thought on every page, "wow, this dude is making 50 kinds of sense."

Since you ask, I don't hate anyone. As a kid, I worked labor with latins and military-caste whites for two summers, some of the latins non-anglophone and probably illegal - and I like them.

By blue anonymous (not verified) on 27 Oct 2009 #permalink

"one day we are commenting on GNXP and the next we are joining the Klan."
OK, this time with more patience. I didn't call anyone on this thread racist...do you claim there is not systematic racism in the United States, or that combating it has been difficult? Or that we should stop trying to address it? Have you ever taken an implicit association test and are you familiar with the results regarding race in the US?

"Psychometry has shown IQ is mostly hereditary"
You are demonstrating that you don't know what HERITABILITY is. Royal titles and Harvard admissions are hereditary. Traits can be heritable, but this is a population measure and has nothing to do with the relative developmental contributions of genes and environment in an individual. It is also highly dependent on who, where, and when it is measured, which is why heritability estimates for IQ range from 0.3-0.8. Once you understand the concept, you will see why it is informative that heritability for IQ is much higher in the upper socioecomic strata than in the lower.

I am repeating myself, but as a measure, heritability is still a bit limited as it ignores epistasis, gene-environment interaction and correlation, maternal effecs, etc etc. All of which are the norm, not exceptions: I can promise you this as someone who is obliged to address these complications in animals every day. For an unbiased and thoroughly scientific introduction to these concepts for laypeople, I recommend Michael Rutter's "Genes and Behavior."

Blue: Italians, Irish, Poles, Germans, Japanese, Chinese... all at one point or another were inferior races that could never succeed based on exactly the same kind of "data" you cite. Why should I be convinced that 2009 is a stable state that captures the end of the trajectory for particular social groups?

Full circle: this thread and my arguments have been about the question of whether there is sufficient evidence to conclude whether or not between group differences in IQ A) are genetic, and B) vary systematically by ancestry. The answer is no. I promise. Many people who don't understand genetics think that the fact that IQ is heritable is support for the position that ethnic differences in IQ must be genetic. It is not.

What will be required?
1. Some understanding of the genetics of cognitive traits. This requires undertanding neural development and plasticity, which are fairly mature but far from complete fields. There is almost no understanding about the links between neuroanatomy/physiology and cognitive traits, but this is going to be an exciting field.
2. A large sample of human genomic data. The Hapmap project and Personal Genome Projects will provide this in the near future.
3. Understanding environmental influences, gene x environment interactions, and epistasis. This is the doozy, and has been a huge problem for behavioral genetics. As I mentioned above, see the recent special section in Nature about the struggles of this field. In the 50s, everyone thought protein structure would be easy and the genetic code would be hard--whoops. Relatively ignored early on in the 80s and early 90s, these issues are turning out to be the hard problem of behavioral genetics. Model organisms suggest there is cause for hope.

You can't fix ALL social inequity, Miko, because people with 85 IQ's and head circumferences of 18 inches just can not be systems analysts, computer programmers, or scientists. They can make good janitors, construction workers, and landscapers. I wish there was great potential in all of us but sadly, the evidence is in and it is just not true. Nature is harsh like that.

By commenter (not verified) on 28 Oct 2009 #permalink

Commenter, your comment has nothing to do with this discussion. First, social equality refers to barriers of opportunity, it does not deny that there are differences between individuals or imply that we should not be meritocratic. Second, I am saying that there is insufficient biological evidence for genetically-based between-group differences in cognitive abilites to claim that these differences are a cause and not an effect of socioeconomic differences between groups.

Pay attention so you don't look stupid or irrelevant. Blogs are harsh like that.

"Traits can be heritable, but this is a population measure and has nothing to do with the relative developmental contributions of genes and environment in an individual."

Blacks raised in rich white homes have the same avg IQ when they grow up as those raised in the ghetto, according to the Minnesota interracial adoption study. Biracial kids scored right in between blacks and whites in the same study. Did the white adoptive parents apply only half of their racism to the biracial kids? Even when they didn't know they were biracial? Blaming low IQ entirely, or even mostly, on the environment is specious reasoning not supported by the data.

Is there absolute proof that group differences in intelligence are completely genetic? No. But the genetic evidence found for recent brain evolution indicates it has not happened equally across all populations. And why would it? People have had to adapt to a huge range of environments, some more friendly to us than others. Genes associated with high IQs have been identified, they are just more common in some groups than others. There is a DAB1 variant that is unique to the Chinese and is thought to enhance cognition. See also http://www.sciencemag.org/cgi/content/abstract/309/5741/1717?rbfvrToken…

The idea we have to explore every biochemical pathway before recognizing that there are significant genetic components to intelligence is absurd, and it is not a standard that other significantly genetic traits such as height and athleticism are held to.

"Genes associated with high IQs have been identified, they are just more common in some groups than others. "

This is false. Name one.

And again--yawn--see above for the litany of assumptions employed in adoption studies and heritability estimates. I understand that these are the best studies people can do with humans at the moment, I am pointing out the simple fact that they do not rise to the level of convincing evidence by modern biological standards.

The microcephalin paper shows some evidence for selection, however there are no phenotypic correlations in brain size or cognitive abilities with the genotypes at these loci (and people have checked thoroughly). These genes are widely expressed and have multiple functions, as there are no cognitive phenotypes they may have been selected for one of these other functions. What constitutes evidence for positive selection is itself a dynamic field, and there are many ways to get false positives...for example in the HAR (human-accelerated) loci.

I can't find the DAB1 paper you refer to.

The Ashkenazi paper is a hypothesis by anthropologists that presents no experimental data. Again--BIG YAWN--I didn't argue anywhere that it is impossible for cognitive abilities to vary genetically, only that there is insufficient evidence that it does so systematically by ancestry.

I'm not going to read Slate or whatever that other web site is, but doubt that either contains primary data.

***Genes associated with high IQs have been identified, they are just more common in some groups than others.***

Some of the candidate genes are looked here, although they're probably of small effect.

http://congenialtimes.blogspot.com/2009/02/evolution-didnt-stop-at-neck…

***Blue: Italians, Irish, Poles, Germans, Japanese, Chinese... all at one point or another were inferior races that could never succeed based on exactly the same kind of "data" you cite. Why should I be convinced that 2009 is a stable state that captures the end of the trajectory for particular social groups?***

Because their trajectory is different.

"They're not just like the Irish--or the Italians or the Poles, for that matter. The large influx of Hispanic immigrants after 1965 represents a unique assimilation challenge for the United States. Many optimistic observers have assumed--incorrectly, it turns out--that Hispanic immigrants will follow the same economic trajectory European immigrants did in the early part of the last century....

These results do not depend on the time period considered. Economists Jeffrey Grogger and Stephen Trejo reached the same conclusions when they used CPS data from the mid-1990s for a similar analysis of Mexican Americans. And other datasets tell the same story. One study reported results from the Latino National Political Survey, conducted in 1989 and 1990. Among its striking findings was that the percentage of Mexican-American households with incomes higher than $50,000 rose from 7 percent in the first generation to 11 percent in the second. But the same statistic in the third and fourth generations stayed at 11 percent, at a time when the national rate was 24 percent. Another study, the National Longitudinal Survey of Youth, began tracking a representative sample of young Americans in 1979. By 1993, the Hispanic 3+ generations in that sample had, if anything, slightly worse outcomes than the second generation in terms of wages, educational attainment, and cognitive test scores."

http://www.aei.org/article/100860

http://www.latimes.com/news/local/la-me-toddlers21-2009oct21,0,200059.s…

"I can't find the DAB1 paper you refer to."

Here is the DAB1 paper.

http://www.plosgenetics.org/article/info:doi/10.1371/journal.pgen.00300…

Here is one showing differences in serotonin:

"Recent articles have proposed that genes involved in brain development and function may have been important targets of selection in recent human evolution [8,9]. While we do not find evidence for selection in the two genes reported in those studies (MCPH1 and ASPM), we do find signals in two other microcephaly genes, namely, CDK5RAP2 in Yoruba, and CENPJ in Europeans and East Asians [46]. Though there is not an overall enrichment for neurological genes in our gene ontology analysis, several other important brain genes also have signals of selection, including the primary inhibitory neurotransmitter GABRA4, an Alzheimer's susceptibility gene PSEN1, and SYT1 in Yoruba; the serotonin transporter SLC6A4 in Europeans and East Asians; and the dystrophin binding gene SNTG1 in all populations."

http://biology.plosjournals.org/perlserv/?request=get-document&doi=10.1…

"The Ashkenazi paper is a hypothesis by anthropologists that presents no experimental data."

And the study won't happen because of political correctness apparently.

What do you think of Jensen's argument here in Chapter 12 of the g factor (the chapter also discusses the various twin studies & environmental arguments)?

"The behavioral capacities or traits that demonstrate genetic variation can also be viewed from an evolutionary perspective. Given the variation in allele frequencies between populations for virtually every known polymorphic gene, it is exceedingly improbable that populations do not differ in the alleles that affect the structural and functional basis of heritable behavioral traits. The empirical generalization that every polygenic physical characteristic that shows differences between individuals also shows mean differences between populations applies to behavioral as well as physical characteristics. Given the relative genetic distances between the major racial populations, one might expect some behavioral differences between Asians and Europeans to be of lesser magnitude than those between these groups and sub-Saharan Africans.

The behavioral, psychological, or mental characteristics that show the highest g loadings are the most heritable and have the most biological correlates (see Chapter 6) and are therefore the most likely to show genetic population differences. Because of the relative genetic distances, they are also the most likely to show such differences between Africans (including predominantly African descendants) and Caucasians or Asians.

Of the approximately 100,000 human polymorphic genes, about 50,000 are functional in the brain and about 30,000 are unique to brain functions. The brain is by far the structurally and functionally most complex organ in the human body and the greater part of this complexity resides in the neural structures of the cerebral hemispheres, which, in humans, are much larger relative to total brain size than in any other species. A general principle of neural organization states that, within a given species, the size and complexity of a structure reflect the behavioral importance of that structure. The reason, again, is that structure and function have evolved conjointly as an integrated adaptive mechanism. But as there are only some 50,000 genes involved in the brain's development and there are at least 200 billion neurons and trillions of synaptic connections in the brain, it is clear that any single gene must influence some huge number of neurons-not just any neurons selected at random, but complex systems of neurons organized to serve special functions related to behavioral capacities.

It is extremely improbable that the evolution of racial differences since the advent of Homo sapiens excluded allelic changes only in those 50,000 genes that are involved with the brain."

http://www.prometheism.net/articles/chap12.html

"What do you think of Jensen's argument here in Chapter 12 of the g factor (the chapter also discusses the various twin studies & environmental arguments)?"

I think it's an exceedingly long-winded way of saying that people differ genetically according to ancestry. So what? It does not provide any support for the idea that "g" has been a differentially selected trait in recent human evolution. The brain does more than produce "g." A lot more. The presence of allelic differences between genes expressed in the brain is not evidence that the genes affect higher cognitive capacities. It's not even close.

There are lot of spurious and unreplicated results in the field of detecting positive selection, as I seem to never tire of pointing out. See work by Pardis Sabeti on the continuing improvements and refinements to these techniques, as well as information about what kind of positive selection events HAVE been replicated. Hint: they tend to be about whether you live to puberty, not how you do on your SATs.

Please stop linking to blogs as evidence. I appreciate the references to primary literature, though I would like to point out that your claim ("There is a DAB1 variant that is unique to the Chinese and is thought to enhance cognition") is nowhere in that paper. There are no structural or cognitive differences shown between people with these alleles. Mice with loss of function mutants in this gene have severe brain patterning defects. It does not follow that different human alleles affect cognition in any way, let alone "enhance" it in Chinese people.

"think it's an exceedingly long-winded way of saying that people differ genetically according to ancestry. So what? It does not provide any support for the idea that "g" has been a differentially selected trait in recent human evolution. The brain does more than produce "g." A lot more. The presence of allelic differences between genes expressed in the brain is not evidence that the genes affect higher cognitive capacities. It's not even close."

Jensen is saying that in light of the totality of evidence, including the brain size/shape differences which seem to have a moderate relationship with intelligence (www.loni.ucla.edu/~thompson/PDF/nrn0604-GrayThompson.pdf).

Pleiotropy could lead to cognitive side effects in some of the genes that have undergone recent selection?

Regarding DAB1 I think another commentator suggested that enhanced cognition - I just provided the link to the paper!

"Regarding DAB1 I think another commentator suggested that enhanced cognition - I just provided the link to the paper!"

Oh yeah, sorry.

libo? I'm really excited to read the paper about how a DAB1 variant "enhances cognition" in Chinese people. But I still can't find it. Darn! I'll settle for how any gene variant does that in a population subgroup--I'm not picky!

Miko,
It was you who opened the political question. You said in comment 35 that you felt the innatist vs nuturist debate was irrelevant for policy. Before you said that, I was following the thread - but I was just going to do my breathing exercises, try to relax, make sure to take my blood pressure meds, and just leave you alone. Anyway, the nature-nurture debate does not necessarily have a boolean answer of course, but it matters a great deal for policy. Moreover, the standard of scientific "certainty" or such like that you keep referring to, is not the correct standard for policy questions; one hardly needs to be a Burkean to agree with that. One just needs to be conscious. *Some* level of Burkeanism is adhered to by every person, including explicit anti-Burkeans. I think you know I am right about this. Only very modest levels of confidence are necessary. Further, anyone can see that Morocco, Turkey, Mexico, and El Salvador do not have societies that you or I want to live in, even though we wish those peoples the best and admire many of their traits. (I admit, some of them, probably most of them, are probably a lot more fun to hang out with than I am.) Do you have robust confidence that these economic and political outcomes, and crime levels, have nothing to do with the nature of those peoples? Of course you don't.

> Blue: Italians, Irish, Poles, Germans, Japanese, Chinese... all at one point or another were inferior races that could never succeed based on exactly the same kind of "data" you cite. Why should I be convinced that 2009 is a stable state that captures the end of the trajectory for particular social groups?

Unless you can cite data - not doxa - I will assume you are fantasizing, or just made an honest mistake in your thought. Or, are you referring only to the per capita economic productivities of those countries in old times? If so, that is not really a strong point, because other countries like Germany and England were also comparably poor prior to industrialization. The failure by old-time American nativists and English to give Ireland a few decades to catch up vis-a-vis industrial inventions, and get educated, was not reasonably charitable. You may also know that Ireland is rather amazingly little-populous and may have a slightly lower IQ than England. It was natural for it to lag England. But eventually it became advanced - just like all six (or so) ethnically NE Asian nations skyrocketed to prosperity from nothing in the 20th century, but often had a little lag time first. (Except China which started off only when Deng rose c. 1982, and of course good ol' North Korea.)

You are probably aware that modern IQ measurements in all the countries you mentioned are around 100+ and they demonstrate modest violence and corruption. That's what I call data. Nonscientific doxa held by old-time nativists are not my responsibility. Also, I would not necessarily assume that the USA *was not* in fact made less prosperous, more corruptable, or more violent by the addition of each of those groups, though any negative impacts were probably modest, because those countries do not show terribly strong group differences compared to old Anglo Yankee people.

> Full circle: this thread and my arguments have been about the question of whether there is sufficient evidence to conclude whether or not between group differences in IQ A) are genetic, and B) vary systematically by ancestry. The answer is no. I promise. Many people who don't understand genetics think that the fact that IQ is heritable is support for the position that ethnic differences in IQ must be genetic. It is not.

Again, that is all well and good for the standard of scientific certainty, which is not an appropriate one for determining our political future. I think even you admit that evidence tends toward demonstrating innate between-group differences even though it does not reach the standard you would want using mice of flies - something I acknowledge. And again, you have no answer to the very important point that there is essentially zero evidence, to date, that group differences can be mediated, regardless of their cause. Despite vast effort. That's what matters for our political future: whether these things are remediable, not whether they are necessarily genetic.

By blue anonymous (not verified) on 29 Oct 2009 #permalink

blue,

I said that because equality of opportunity in the US is an important part of our national identity, and for the most part a universally admired one. We know people vary in their abilities and personality traits in many ways, and there are many sources of this variation. I stand by my claim that we don't know that any of between group variation in cognitive abilites is genetic. By "don't know," I mean the evidence is weak and there are plausible alternatives.

There is not a "standard" for flies and one for humans. There are standards for reaching conclusions. What is different between humans and flies are the types and quality of experimental data we are able to obtain with which to support or contradict particular conclusions.

Do you think there is equality of opportunity among groups in the US--just to pick a country at random? I think that anyone who claims so is delusional or lying, and I doubt you do. Does opportunity vary systematically by group for historical reasons and cultural norms? It does. Can it be remedied? I claim that quite a bit of it has been remedied.

If you want to argue that there are no remaining remediable differences between groups you have to argue one (or some mixture) of the following:
1. The sociocultural institutions or forces that create the difference are insurmountable. I'd say this is demonstrably falsified by history, unless you want to make some baseless special pleading for 2009 being the endpoint of history.

2. The source of between-group difference is genetic. Which is what the thread has been about. I return to the double standard. IQ measurements are dynamic in time and space, requiring renormalization--that's statistically kosher, I have not argued that IQ doesn't measure something. But clearly environment has a strong effect even assuming group difference, no one could seriously argue otherwise. When we look at the relatively low levels of heritability in low SES groups and high levels of of heritability in high SES groups, it suggests to me that a heterogeneous set of unknown, low SES-associated environmental factors interacting in complex ways with one another contribute significantly the difference.

Some candidate environmental factors are shown to have measurable effects. No genetic factor(i.e. allelic difference) has been shown to contribute to between group differences, or even within group differences. (There are candidate QTLs for the latter, none substantiated). Maybe there there will be at some point, maybe not due to the "gloomy prospect" mentioned above.

So, yes, I conclude that we should not have policies that assume irremediable biological differences in abilities between *groups* of people. It is a false assumption, and is counter to equality of opportunity, and counter to the notion of individual liberty.

As for setting policies based on non-biological differences. Whatever: we do that all the time, particularly with immigration policy. Not the point here, except to say that I don't see any reason to conclude that these differences between groups within the US are irremediable--history suggests otherwise. This is what I meant when I caricatured the "irremediable" position as "not being racist is hard, so let's give up." Not being racist IS hard--even for black people, if we take IATs at face value. So I feel that we are morally and rationally obliged by our political ideals to attempt to remedy social inequality, even if it's hard.

"When we look at the relatively low levels of heritability in low SES groups and high levels of of heritability in high SES groups, it suggests to me that a heterogeneous set of unknown, low SES-associated environmental factors interacting in complex ways with one another contribute significantly the difference."

Here's Jensen & Rushton's comment in their working paper regarding Nisbett's book:

"Most studies find IQ scores to be just as heritable in non-White samples as in White samples [72]. There is no indication of any extreme deprivation, or special cultural influenceâsuch as being raised as a visible minorityâat work in one group and not in others. If there was, Black heritabilities would be consistently lower than White heritabilities and we would indeed conclude that poverty, the legacy of slavery, or White racism had operated to suppress the level of intelligence in Blacks. However, Osborne [73] tested this hypothesis empirically by comparing several hundred pairs of Black and White twins aged 12- to 18-years on the Basic Test Battery, the Primary Mental Abilities test, and the Cattell Culture Fair Intelligence test, and found heritabilities of about 50% in each group. (The heritabilities in the Basic, Primary, and Cattell tests were, respectively: Whitesâ.61, .37, and .71; Blacksâ.75, .42, and .19.)

The Turkheimer et al. [70] study that Nisbett cites is an outlier. In Britain, the exact opposite of Turkheimerâs result was found in over 2,000 pairs of 4-year-old twins (N = 4,446 children), with greater heritability observed in high-risk environments [74]. A re-analysis of the Hawaii Family Study of Cognition also found contrary results to Turkeimerâs. Nagoshi and Johnson [75] found no reduction in the relationship between parental cognitive ability and offspring performance in families of lower as opposed to upper levels of socioeconomic status. In the 1,349 families they studied, the relationship remained the same across tests, ethnicity, and sex of offspring."

Finer grained analyses show the same within- and between-race heritabilities for Blacks and for Whites. Rowe and Cleveland [76] examined full- and half-siblings (106 pairs of Black half-sibs, 53 pairs of White half-sibs; 161 pairs of Black full-sibs, and 314 pairs of White full-sibs) from the National Longitudinal Survey of Youth on three Peabody Achievement Tests (Mathematics, Reading Comprehension, and Reading Recognition). The best fitting model was one where the sources of the differences between individuals within race and the differences between races were the sameâ50% genetic and 50% environmental. Similarly, Jensen [25, p. 465] tested four alternative models on IQ data from 123 Black and 304 White pairs of 12- to 18-year-old twinsâonly genetic factors, only environmental factors, neither genes nor environment, and genes plus environment. The model of both genetic and environmental factors best explained the observed Black-White IQ differences, while both the genetic-only and the environmental-only explanations were inadequate." (page 18)

The paper discusses a number of the explanations that you suggest above. What is your take on their argument regarding inbreeding depression scores?

"The history of the false claim that inbreeding depression scores correlate as highly with the secular rise as they do with Black-White differences is worth recounting. The story begins with a 1989 study by Rushton [79], which used Jensenâs [25] method of correlated vectors (Section 3.2) to find that inbreeding depression scores from 1,854 cousin marriages in Japan correlated r = .48 (P < .05) with Black-White differences in America on the same 11 tests. No-one has proposed a non-genetic explanation for inbreeding depression scores so the fact that they predict the mean Black-White differences clearly implies that the Black-White differences are genetic in origin. Nor do Rushtonâs results stand alone. Nichols [80] calculated heritabilities for 13 tests and found a .67 correlation between them and the magnitude of Black-White differences on the same tests. Jensen [24, pp. 103-119) calculated the environmentality of a test (defined as the degree to which sibling correlations departed from the pure genetic expectation of .50) and found it inversely related to the magnitude of Black-White differences (r = -.70), i.e., the more environmentally-influenced a test, the less pronounced the Black-White difference."

Also, in terms of x factor variables such as stereotype threat, lower expectations, or being a caste type minority:

"A series of studies on large samples by David Rowe [52] found no evidence of X factors acting to lower the IQ scores of Blacks. For example, Rowe et al. [53] examined test scores for 8,528 Whites, 3,392 Blacks, 1,766 Hispanics, and 906 Asians and found the exact same relation between background variables and outcome measures in each race. There was no evidence that any special Black minority-specific process was affecting academic achievement or lowering IQ scores.

Another study examined longitudinal data on academic achievement (Rowe et al. [54]). Correlations were computed between academic achievement and family environment measures in 565 full-sibling pairs (White N = 296 pairs; Black = 149; Hispanic = 120) from the National Longitudinal Survey of Youth tested at ages 6.6 and 9.0 years. Including age as a variable yielded three 8 x 8 correlation matrices. Analysis showed the matrices were equivalent across the three groups with no evidence of any Factor X affecting academic achievement or developmental changes of any group."

They also discuss the brain size difference which shows up early on:

"Nisbett attributed any brain size differences found between Blacks and Whites to perinatal factors and prematurity, as well as to poorer post-natal nutrition. But the evidence has never supported these long standing and reasonable hypotheses. For example, race differences in brain size can be observed before birth. In a study of fetuses, Schultz [128] found that from the 9th week of intrauterine life, 165 Black fetuses averaged a smaller brain case than did 455 White fetuses. The difference became more prominent over the course of fetal development.

Similarly, perinatal factors were ruled out as important for the development of IQ on the basis of a longitudinal study by the U. S. National Collaborative Perinatal Project of 40,000 children followed from birth to seven years [129,130]. It showed that at birth, 4 months, 1 year, and 7 years, East Asian children averaged larger cranial volumes (based on head circumference measures) than White children, who averaged larger cranial volumes than Black children (Figure 3). Within each group, the children with larger head sizes obtained higher IQ scores (mean r = .20).

Moreover, the differences in brain size were not mere correlates of body size, and unlikely to be due to nutrition, since the East Asian children were the shortest in stature and lightest in weight, while the Black children were the tallest in stature and heaviest in weight."

http://www.ssc.uwo.ca/psychology/faculty/rushtonpdfs/Intelligence%20and…

Thanks Fang, I can't remember the last time I was so quoted at. If how-many-published-claims-you-can-cite decided issues of fact, there would be a tie about everything in the social sciences (within a certain confidence interval).

I think I've made clear on my opinion of the role (loud) and worth (limited) of quantitative social science in regard to addressing this ***biological*** question. BUT ANYWAY, the Turkheimer study is not an "outlier." It would be if they tried to precisely repeat someone else's finding and gotten a way-different result. Their result is novel because they used different--they would say better--methods and made different assumptions. Like all research of this kind, your result depends heavily on assumptions and the parameterization of your model. I'm not saying they have discovered Truth--we're talking about psychology and statistics here--but it underscores the fact that there are a variety of plausible models that give very different results.

Unfortunately, any model complex enough to be biologically realistic and explanatory cannot be addressed with QSS. Simply can't. And that is my whole friggin' point. So if you've got a personal axe to grind or feel like you need academic citations to support your political proclivities, you're welcome to them. Just don't pretend its science.

"I think I've made clear on my opinion of the role (loud) and worth (limited) of quantitative social science in regard to addressing this ***biological*** question."

Just in relation to the Turkheimer study, it would be interesting to do a later follow up:

1 - The study included only young children and does not make any attempt to extrapolate that all other findings of significant increases in h^2 by age 17 are in any way invalid. The effects of the shared environment vanish at around age 12.

2 - Turkheimer began his paper by recognizing that the heritability of cognitive ability in childhood is well established.

3 - Turkheimer made no attempt whatsoever to determine what components of SES he was measuring. There are three obvious items to consider: macro environmental, micro environmental, and genetic. All work to date indicates that the first of these can be found in children, but that it is absent in late adolescents; by late adolescence, all of the environmental component is of the second type; and that genetic intelligence is the largest determinant of SES.

4 - Turkheimer says that the effect he observed was related to the homes in which the children were raised. This is interesting, since it relates to the adoption studies which show that after childhood there is no adult IQ correlation between biologically unrelated children who were reared together in the same home.

5 - Turkheimer discusses in some detail that SES is not strictly an environmental variable, since it is known to be (statistically) caused by the intelligence of the parents. He points out that the models he used "cannot determine which aspect of SES is responsible for the interactions" observed.

As for the reason the shared environmental component vanishes there are a number of possible reasons that are not yet sufficiently investigated to establish proof of causation.

Yeah, I think you put your finger on a number of the ambiguities that plague these studies, particularly gene/environment correlations. There is an interesting study -- Duyme 1999, sorry don't have the link -- tracking low IQ kids from abusive/unhealthy situations who are adopted into low/middle/high SES families. The greatest IQ gains are seen in the high SES environments--strikingly so. Equally interesting, though is that the initial measurement for individuals is still predictive of rank within a given SES environment at the end.

All this reinforces what we know and aren't arguing about as far as I know: IQ measures some-thing(s), and is influenced by genes and environment in complex, non-additive ways. We don't know what any of the genes are, what most of the environmental factors are (after all, as you point out, "SES" is not really a factor at all--just a poor proxy for a number of possible and heterogeneous influences), or how any of them interact. Thus my position that the causes for between group differences are unknown.