From the blog “Three-Toed Sloth.” This is just a couple of excerpts, it’s worth it to go and read the whole thing (even though it’s a slog if you don’t have a science background):
Many twin data sets show that the correlations in twins’ IQs actually change with the environment, in a pretty crude way which nonetheless goes beyond what people generally include in their models. I will quote from an old paper by Bronfenbrenner [3, pp. 159–160], because it’s handy and it makes the point:
The importance of degree of environmental variation in influencing the correlation between identical twins reared apart, and hence the estimate of heritability based on this statistic, is revealed by the following examples.a. Among 35 pairs of separated twins for whom information was available about the community in which they lived, the correlation in Binet IQ for those raised in the same town was .83; for those brought up in different towns, the figure was .67.
b. In another sample of 38 separated twins, tested with a combination of verbal and non-verbal intelligence scales, the correlation for those attending the same school in the same town was .87; for those attending schools in different towns, the coefficient was .66. In the same sample, separated twins raised by relatives showed a correlation of .82; for those brought up by unrelated persons, the coefficient was .63.
c. When the communities in the preceding sample were classified as similar vs. dissimilar on the basis of size and economic base (e.g. mining vs. agricultural), the correlation for separated twins living in similar communities was .86; for those residing in dissimilar localities the coefficient was .26.
d. In the Newman, Holzinger, and Freeman study, ratings are reported of the degree of similarity between the environments into which the twins were separated. When these ratings were divided at the median, the twins reared in the more similar environments showed a correlation of .91 between their IQ’s; for those brought up in less similar environments, the coefficient was .42.
(Let us pause a moment to contemplate the sense in which identical twins, growing up in the same town and attending the same school, are “raised apart”.)
By the time you’re done partitioning the twin pairs into classes in this way, n is pretty small, and the sampling errors in the correlations are going to be large, so I wouldn’t give the 0.86 vs 0.26 contrast a lot of credence, but the fact that the differences are all in the same direction, and get pretty big, ought to be hard to ignore. (And it’s worth remembering that n is never very large for identical-twins-raised-apart studies.) The obvious explanation for such results is that the developmentally-relevant environment of twins raised apart, but in similar towns, is much more highly correlated than that of twins raised apart in dissimilar towns. This means that a substantial chunk of the correlation you thought was genetic is actually due to shared environment, and pushes your heritability estimate down. Alternately, you could abandon the lack of correlation between genetic and environmental contributions, or the strictly additive nature of the model by including a very substantial interaction between genes and environment, so that identical genotypes respond very differently to those differences in environment. However you slice it, your estimate of heritability was too high.
And from later in the essay….
There are also randomized studies of interventions with “at risk” families, e.g. ones with unusually low birth-weight children. Depending on the study, the treated groups had IQs 10 to 15 points above the controls. Because of the random assignment, not only is there no problem of endogeneity, it’s also idle to worry about placebo effects — it would be fantastic if a placebo raised IQ by 10 points.
Another nice example (not from Wahlsteen) comes from Heber’s work on Rehabilitation of Families at Risk for Mental Retardation. Rather than summarizing it my own words, I’ll quote someone else’s summary (though no doubt I’ll be told he understood neither genetics nor experimental methods):
It describes an experiment on ghetto children whose mothers had IQs of below 70. Some of these children received special care and training, while others were a control group. Four years after the training period the IQs of the former averaged 127 and those of the latter 90, a spectacular difference of 37 points. The fact that the control children had a 20-point advantage over their mothers is not unexpected [because of regression toward the mean]. [4, pp. 14–15]
At this point, the ritual is for people to begin saying things like “there’s nothing you can do if the environment is already decent”, “the changes didn’t last long after the program” (which would equally show that exercise can’t really change physical fitness; see below), or to raise irrelevancies. (My favorite, among the last, is to point to adoption studies showing that adoptees’ IQs are more correlated with their biological than their adoptive parents’ IQs, conveniently side-stepping which set of parents had scores closer to the adoptees’.)
Please remove the excess manual breaks within the blockquotes, which make the text very hard to read.
Four years after the training period the IQs of the former averaged 127 and those of the latter 90, a spectacular difference of 37 points.
If these results are repeatable, scalable, and even partially sustainable into adulthood (with continued intervention throughout childhood and adolescence), this is beyond fantastic, and even I could get behind a government program to perform these interventions, as it would pay for itself many times over through increased productivity and reduced crime.
But this experiment was done in the early ’70s, which means either that it turned out to be a total bust for all practical purposes (e.g., the results were not sustainable beyond early childhood), that it was not practically scalable (the experiment paired each child with his own supplementary [or substitute?] “mother”), or that this approach was abandoned for no good reason.
I find each of these reasons plausible. It would be interesting to find out which is correct.
Did you say SLOG? Do you have Cliff notes? :P
“Did you say SLOG? Do you have Cliff notes? :P”
Here’s my attempt at such, no guarantees as to accuracy:
They did some twin studies, where they looked at pairs of identical twins who were separated at birth, and checked to see what the correlation was between their intelligences. The twins who were raised together or in the same town had a correlation of somewhere between about .76 and .86. The twins who were raised apart or in different towns had a correlation of about .63-.67.
When researchers look at the different towns and sort them into “types” of communities, the twins raised in similar towns have a correspondence of about .91, and twins raised in REALLY DIFFERENT towns go down to .26.
The author points out that these numbers aren’t as significant as they might seem on first glance, because like most of these studies, the sampling group they’re using is very small. (There’s an argument in developmental psychology that you actually make your results better by using 11-14 children [after removing statistical outliers] in any given study, rather than using more children. I have no idea how that works. It was explained to me. I gaped in puzzlement.)
Nevertheless, it’s clear from the studies that the concept that IQ is entirely heritable is FALSE. It’s obviously affected by environment, and more affected by very different environments than by similar ones.
This makes sense with what we know about the development of most personality traits, in that very little seems to be absolutely genetic. Most personality traits seem to be something like height. Your genes give you an innate sort of tendency to be tallish or shortish or averagish, but that tendency is going to be hugely affected by how much nutrition you have. So I’m 5’4″, on the shortish side of things, but if you plunked me down in the middle of a society with poor nutrition, I’d be on the freakishly tall side of things.
Similarly, twin studies with diseases like Schizophrenia — which is considered to be one of the most biologically associated of the mental illnesses — show that genetic factors only count for (IIRC) about 50% of the tendency to get it. The rest is environment.
In the second quoted section, the author discusses studies that relate to premature infants, and to inner city mothers with IQs of less than 70. When given stimulating environmental factors, such children do much better.
In the case of the inner city mothers with IQs of less than 70, the children in the control group drift to about 90 — which is expected because you would expect the kids to move toward the average intelligence, away from the bottom end of “normal.” When the children are given assistance, howeer, their IQ averages went up to 127, which again demonstrates how fluid and dependent on training IQ can be.
Again, this matches with what we know about IQ, for instance the fact that children who receive poor nutrition and poor attention before age five tend to score much lower than their peers on IQ tests, other factors being held as constant as possible with humans. There’s also a fantastic linguistic anthropology paper that documents the ways in which children are raised in three different communities, and how early play behaviors affect the way children relate to the world and to the school system later. I don’t think that one’s online for free, which is too bad; it’s one of the most striking things I’ve ever read. I can go into more detail if people are interested, but it would be another post about this length.
Standard disclaimer on anything using IQ:
IQ is not reifiable, and our cultural pretense that it is remains damaging. IQ tests are useful in cases like this, where you’re comparing populations, though it’s still necessary to remember that you’re basically measuring performance on a test, not a single, comprehensible, intrinsic quality.
IQ tests were invented for the identification of students whose mental disabilities required that they receive extra help in the classroom. Its inventor specifically warned against using the tests to rank “normal” children — and of course, people did that just about immediately.
Mandolin, I’d be really interested in that linguistic anthro article, even if you just provide reference details, I might want to track it down.
I can remember a large number of details from the article itself, but none that would help you identify it. :( I’ve got a copy of the paper at home in California, but that’s not really useful.
I wrote my linguistic anthropology professor and asked him if he could identify the author or title of the article from the details I gave him. His auto-reply says he’s on sabbatical, so it may be a bit before he responds. I’ll post when he does.
In her study, the anthropologist went into three communities in the United States to investigate how they raised their children with regard to language, and how these skills affected the childrens’ later performance in school.
In the middle class group, she looked at how stay-at-home mothers who had been teachers instructed their toddlers, and found that they centered a lot of their lives around books. Reading was a treat and a reward. Playtime involved a lot of asking questions about the books, both factual and open-ended — e.g. “What is Alfred?” “A duck.” “Do you like this book?” “Yes.” “Why?” — which, of course, mimics school. The children were also asked to bring their knowledge about books into the rest of their lives, for instance looking at ducks on a playground and saying “Where have you seen those before?” (The book about Alfred.) Children were also asked to make comparisons, moving between the text and their real lives to create connections. “Why might Albert have tried to save the ducklings? How is that related to how you feel about your little brother?” Children from these households did very well in school, both in early grades when they were asked to mine texts for factual information, and in later grades when the importance of critical thinking skills and open-ended questions increased.
A second community was working class. Reading books was incorporated as a rigid, necessary part of the daily routine. Questions and interruptions from the children during reading time were not allowable — IIRC, the author compared reading the book to sitting still for a church sermon, the text was seen as unbreakable instead of interactive. Afterward, the children might be asked factual questions — “What is Albert?” “A duck.” — but were unlikely to be asked open-ended questions or questions that required critical thinking or comparison between one situation and another. Children from these households performed very well in the early grades when they were asked to look at texts and identify factual information, but floundered when later grades required open-ended responses.
A third group of children was, I believe, raised in a poor, nonwhite neighborhood. People here viewed language acquisition as something that would happen on its own, if you let the children run around and learn things at their own speeds, and indeed this turned out to be correct. Children don’t really need word by word oversight. The children learned from intuiting and overhearing instead of direct, deliberate instruction (and they learned fine). There was an emphasis on learning to tell stories, I think, but I’m worried that I”m confusing this part of the study with numerous other studies I read on the acquisition of narrative. If I recall correctly, this community had a tradition of storytelling which encouraged and rewarded creative comparisons (I don’t remember if it was gendered; a lot of studies on storytelling, inside and outside non-white communities, found that female children were rebuked for failing to tell the truth, while male children were rewarded for being creative. After a while, this seems to create a situation in which female children tell fewer overall narratives). Children from this community struggled when they were asked to look at a text and identify factual information, which (again IIRC) the author explained as being because the children had been taught to look at the world hollistically and had not been given experience with transforming symbols into easily digestible soundbites. For instance, a representation of a duck on the page doesn’t look like a real duck. Children in groups one and two were trained to identify “yellow and cartoony representation of duck” with “those real things that wander around and flap at me and are many different colors.” These children hadn’t been. They did very well when asked open-ended questions, however, and demonstrated at a very young age an advanced understanding of how to create metaphors and manipulate symbols.
I dunno. The author seems to spend a lot of time going after the genetic question, but at the end in an almost-footnote concedes the similarity to other biological functions is pretty important.
It seems fairly obvious that some authors have vastly overstated the genetic component of I.Q. OTOH there’s a big difference between the Bell Curve and a claim that genetics don’t matter at all, or that there is no evidence to suggest they do, which is what this author is implying.
After all, it’s an exception claim, and those should be taken with a grain of salt. Genetics are implicated in a lot of interesting things–including other psychological traits like addiction behavior, for example. to suggest that intelligence just happens to be one of the comparatively rare exceptions that avoids a genetic influence is fine… but even without a “gun to your head” it isn’t so difficult to think it’s less likely.
And part of what makes the entire debate so frustrating is the apparent bias of players on both sides. People didn’t jump to respond to the Bell Curve just because they thought it was bad science. they did so because they thought it was racist (it was), or inappropriate, or… When it came out, I ran into plenty of people who condemned it before having read it.
Obviously this also happens on the pro-genetics side as well; there are probably a lot of people invested in genetic IQ transmission for a variety of reasons. But when you have people doing science to prove a point, rather than doing it to see what happens, the science becomes suspect.
I don’t think the claim is that intelligence has no genetic basis, Sailorman, only that the genetic component is highly susceptible to environmental effects. Study after study after study has shown that cognitive development, especially language development, takes place only in the presence of stimulation. (These studies were for the most part not trying to establish the degree of heritability of intelligence.) It’s fairly well established that there’s a window for language acquisition that opens soon after birth and closes around age 15, or puberty. I say “fairly” because in the rare cases of “wild children” who had a chance to learn language only after the window had closed, it was impossible to determine if they might have been abandoned in the first place due to mental impairments that might have kept them from learning language anyway, and obviously you couldn’t do a study like this on purpose. Language is so intertwined with intelligence and with any kind of intelligence test that could be devised that you can’t really tease the two apart. It’s well known, though, that it’s much, much easier to become fluent in a second language before that window closes. I find it almost a no-brainer (as it were) that different environments would stimulate similar brains in different ways and to different degrees.
(In connection with language development, I’d also note that my severely mentally disabled son understands and uses simple syntax [“I want to watch Toy Story“; “It’s not Halloween today”] and has a working vocabulary of maybe 150 words; I wouldn’t call him fluent, but he is way, way ahead of someone who doesn’t use language at all, or who understands only a few words.)
I did hear of a study showing that the average welfare mother (I believe this was before welfare reform) had a smaller working vocabulary than the average two-year-old (!) child of upper-middle-class professionals being raised by her/his parents. Again, this doesn’t much surprise me, and it’s easy to see how this disparity would propagate from one generation to the next.
Brandon raises an interesting point: why was no attempt made to replicate the Heber study?
I too would be interested in reading that paper, Mandolin.
Sailorman,
I don’t know if it’s as much an issue of “bias on both sides” as it is what happens when non-scientists with an agenda try to jump into the issue. After all, it’s one thing for a scientist to have ongoing work on various aspects of cognition that may or may not be heritable. It is another thing entirely for people who have a desperate need to be smarter than black people to grab hold of science they understand poorly and go making all kinds of claims that don’t really stand up.
The problem is that folks like Steve Sailer and Charles Murray draw an entire area of study into disrepute with their racist garbage.
“… a claim that genetics don’t matter at all, or that there is no evidence to suggest they do”
Can you point to the location of this claim?
Mandolin,
“Here, my honest answer would be that I presently have no evidence one way or the other” is the shortest bit of quote that really fits the bill, though it’s fairly obvious that the entire tone of the article supports that position.
Any opinion on the rest of my response?
Nope.
Sailorman wrote:
That’s not obvious at all; in fact, it’s obvious the piece read fairly doesn’t contain the position you attribute to it.
Let’s take that quote you took out of context, and put it back into context, with some emphasis added by me:
There is no reasonable reading of the above passage which supports your interpretation, SM.
In the introduction, she states her thesis quite clearly (I’ve added bold):
She’s not saying that there is no genetic influence on intelligence; she’s saying that, especially when comparing people without developmental disabilities, people tend to overstate the importance of heritability to IQ.
I’m a little curious on how identical twins can end up being raised apart, and how this affects one or the other of them in a non-symmetrical fashion.
Wikipedia has an article on twin studies, but it doesn’t explain how twins come to be separated. I think they tend to be adopted out to different homes, but I’m not sure if I’m remembering correctly.
Time was when twins and other siblings put up for adoption would be adopted out to different homes, on the undoubtedly correct theory that a single child is easier to place than two or more who come as a set, and on the theory that it was better for siblings to have stable adoptive homes than to stay together, if those were the only options.
As for non-symmetrical effects — suppose hypthetically that twins were adopted out to different homes; one twin got proper nutrition and medical care and good schooling, while the other was locked in a dungeon and given only crusts of bread to eat and water to drink, with no one but the rats and roaches to talk to. Would you be surprised if, after ten years, the unfortunate twin was both shorter and stupider than her sister?
Now suppose that twins were born with a remarkable aptitude for mechanical engineering but only mediocre lingusitic talent. One twin was placed with parents who ran an auto-repair shop and tinkered with inventions in their spare time, the other with a pair of professors of romance languages, brilliant linguists but dangerous with so much as a pair of pliers; both twins received roughly the same nutrition, medical care and schooling. Would you be surprised at the end of the same ten years if the first twin, her natural aptitude recognized, stimulated and lovingly nurtured, was both happier and smarter than the second?
Lu, I’d like to see a copy of the study you mentioned on welfare mothers, because the result seems utterly preposterous on its face. Part of Sailer/Murray/Lynn’s act depends on their audiences’ substantial ignorance and dearth of first-hand experience regarding the groups that they claim are borderline retarded. Anyone who’s had extensive contact with low-income people knows that accusations like the aforementioned study levies would be laughable if people out there didn’t take them at face value.
You know what, Banana Danna, I gotta thank you for calling me on that, because I don’t know what I was thinking. I hadn’t seen the study or even read anything about it myself, only heard about it. Just now I did a quick google and found this commentary, which links to the study in question. The study did not in fact say any such thing; it compared three-year-olds (not two-year-olds — that was my bad memory) from different backgrounds and found that the richer children had bigger vocabularies than the poorer ones and that the richer parents used more words in talking to their children; but all of the parents used many more words to talk to other adults than they did to the children. As the commentary points out, that just proves that adults use simple language when talking to young children, which any parent knows.
I stand corrected and extremely red-faced.
Mandolin, if you’re interested in linguistic evolution of meaning and the connection between thought and language, I have a book recommendation for you, my father’s an evolutionary linguist. Email me if you’d like the title/author, putting it in the comments would be a breach of the nod I attempt to make towards anonymity.