As Razib Khan says in his post on this, "My hunch is that these results will be unsatisfying to many people." I'd go further and say that no-one will be happy with these.
For those who believe that IQ is purely environmental and not genetic, any satisfaction they might feel will be short lived because these authors did replicate the recent finding that genetic variants explain about 50% of the variance in IQ. Looking at all SNPs together, there was a strong correlation between "genetic similarity" and similarity in IQ. That independently confirms what the much-criticized twin studies of IQ said - IQ is about 50% heritable.
But for people who do believe in the genetics of intelligence, this shows us that we have no idea what the genes are, and that everything published so far has been pretty much for naught.
There's another implication. We actually do know of many "IQ genes" in that we know genes that, when mutated, cause mental retardation (very low IQ).
Now many researchers have hoped that if a certain gene causes you to have an IQ of, say, 50 when it's completely deleted by a mutation, then more subtle variants in that gene would have minor effects on IQ. Maybe a variant that reduces expression of the gene by 10% would knock off 5 IQ points.
In other words, if big mutations cause big phenotypes, then small mutations in the same place ought to cause small phenotypes. It seems to make sense - but today's IQ literature shows that it's just not true.
That's not just a problem for IQ though. Take autism or ADHD, we know that there are rare, severe mutations that cause these conditions. Many people are hoping that common variation in the same genes might also be interesting - but if IQ is anything to go by, it won't be.
Perhaps this is not so surprising. Breaking your neck and becoming paraplegic is going to seriously impair your ability to play baseball. That doesn't mean that normal variation in baseballing skill has much to do with minor neck injuries.
Chabris, C. F. et al (2011). Most Reported Genetic Associations with General Intelligence Are Probably False Positives Psychological Science
19 comments:
On the paper about the heritability of intelligence (Davies et al.), I can't see the full paper behind the paywall, but it seems crucial how diverse the environments of the people studied were (the less environmental variation there is, the less impact environment will have, and so the higher the proportional impact of genetic factors will be). How much effort did they make to ensure that their samples came from diverse environments?
@Boyden - That was my first thought too. I also wonder whether the way IQ is tested (important as a scientific measure) could not also relate more to how an individual fits into the society they are in. Be it genetics or environment, some people have a hard time adapting, and over time that tends to factor into their IQ scores. So some of that variation explained by genetics could be genes that affect adaptation to society in general, which then influences IQ, while technically on a biological level they could learn just as well. Of course, that would be interesting too ;) but would require a different perspective on the part of the researchers...
Actually, I'm fine with the observation that those who are more closely genetically related are more likely to have similar IQs. Where I have a problem is drawing any inferences from this finding. It's the same result we'd expect if there were no genetic link. People who are more closely related also share more of their environment.
I applaud this study.
I really believe things are way more complicated than we think they are and we simply don't have the tools to capture whatever there might be. We need to go beyond P-values for publishing threshold/criterion.
These reported P-values may be a combination of confounding factors: it could be plain and simple publication bias, or it could be underlying population stratification poorly corrected for.
Frankly, I think the problem is wide spread and does not apply to "intelligence" genes only. Some genetic disease are the direct consequence of a dysfunctional protein and those are easy to spot. But for the great majority of traits things are way more complicated than that and the current methods we have either only scrape the surface or randomly generated p-values.
I know I'm always all "hey epigenetics!" but I just want to point out that heritability can involve much more than specific genes. I.e. in twin studies they shared a womb, and the mother and fathers childhood environments and toxic exposures etc can alter the gene functioning in the offspring without altering genotype.
But you already know this, right?
Stephanie: Good point, but they excluded anyone who was genetically related to someone else with relatedness over 0.025 (where parent-children is 0.5, and 0.125 is greatgrandparent-greatgrandchild).
So in other words no-one was family to anyone else in any normal sense.
What is possible though is that it's picking up on more subtle sub-population differences. E.g. in the UK, people from the north of England tend to be poorer on average. They are also probably genetically different because they have more Celtic as opposed to Anglo-Saxon ancestry (I assume).
That's not why they're poorer but it's a correlate. I assume.
And I bet the same kind of economic-ancestry differences would be found in most countries; in Italy the poor bit is the south, in Germany it's the east, etc.
So if there's an environmental explanation it may be along those lines...
One implication is that a normal distribution of IQ that includes the entire population is skewed as to the causes of normal variation in intelligence. One really needs a model with a dummy variable which at one value gives a normal distribution of people who are positive for that dummy variable and for the otehr value of the dummy variable gives mental retardation.
This may be slightly off-topic, but I would love to see some of these genetic researchers try the same sort of associations studies on a modern computer CPU chip. They could develop tests that would enable them to see what parts of the CPU are activated (like an fMRI) or look at the types, styles, and composition of the logic gates in the chip and use those tests to predict what the output on a computer screen would be for a given input.
I suspect the results would look very similar to what these correlation studies show. Sometime you get to see a part of the picture but you often can't see the forest because you are too busy looking at a small piece of bark that fell off a single tree.
Neuroskeptic wrote:
"In other words, if big mutations cause big phenotypes, then small mutations in the same place ought to cause small phenotypes. It seems to make sense ..."
I don't think it makes a lick of sense. Until experiment establishes that continuity should be part of a model then it simply should not be.
I disagree, I think the default option should be continuity.
Why would you expect a "break point" in the gene dosage-phenotype curve? I mean it seems as though there is one in many cases, though by no means all, but why would you predict it?
There are thresholds with respect to variables of many types in the universe. Below the threshold? No effect. Above the threshold? Effect. IMO continuity is always an unjustified assumption in the absence of empirical evidence for it.
For any variable in the universe I might grant you that, but we're -in essence- talking biology here; the assumption of continuity is not far fetched at all.
Why would you assume otherwise for biology's sake? Just saying it is so does not make it so.
What about the g-factor as a concept? Perhaps this general factor does not exist?
How meaningful is "x% of the variance is explained by genes" in the presence of gene-by-environment interaction?
I'm not sure this relates, but I remember studying about mosaicism in Down Syndrome, where not all cells were trisomy expressed. http://www.mosaicdownsyndrome.com/faqs.htm
Genome-wide association studies establish that human intelligence is highly heritable and polygenic
-Molecular Psychiatry 16, 996-1005 (October 2011)
" We conducted a genome-wide analysis of 3511 unrelated adults with data on 549 692 single nucleotide polymorphisms (SNPs) and detailed phenotypes on cognitive traits. We estimate that 40% of the variation in crystallized-type intelligence and 51% of the variation in fluid-type intelligence between individuals is accounted for by linkage disequilibrium between genotyped common SNP markers and unknown causal variant"
For those interested in the entire paper :
http://menghusblog.wordpress.com/2012/05/03/intelligence-highly-heritable-and-polygenic-g-davies-et-al/
And for those who are not yet aware :
http://www.newscientist.com/article/dn21705-best-evidence-yet-that-a-single-gene-can-affect-iq.html
Following a brain study on an unprecedented scale, an international collaboration has now managed to tease out a single gene that does have a measurable effect on intelligence. But the effect – although measurable – is small: the gene alters IQ by just 1.29 points. According to some researchers, that essentially proves that intelligence relies on the action of a multitude of genes after all.
"It seems like the biggest single-gene impact we know of that affects IQ," says Paul Thompson of the University of California, Los Angeles, who led the collaboration of 207 researchers. "But it's not a massive effect on IQ overall," he says.
@Meng Hu
if you believe that, I have a bridge I'd like to sell you
Post a Comment