Hacker News new | comments | show | ask | jobs | submit login
A genetic code for genius (wsj.com)
47 points by sonabinu 1617 days ago | hide | past | web | 53 comments | favorite

"Research into the science of intelligence has been used in the past "to target particular racial groups or individuals and delegitimize them,"" says a watchdog group eager for money and public support for its pet cause.


I mean, from what we understand so far of the genetics of IQ, there might be some slight differences based on ethnicity - which is nothing unexpected, as you can see the same variability for blood groups, hair color, HLA, etc.

Basically different groups of people have a different distribution of some genes than other groups of people, and therefore a show a difference in the expression of these genes that can be aggregated at the group level to show... a difference in the visible characteristics between groups. Wow, what a shocking discovery.

The really interesting thing is these genetic differences between ethnic groups are dwarfed by the difference a good upbringing (alimentation and education) can get. And since there are also promising results about the inheritability of IQ, these inter ethnic group differences might also be dwarfed by INTRA group differences.

That is the best argument if this watchdog group really want to disprove that "targetting of particular racial groups"

So instead of crying wolf, they should embrace such research- after all, when we have pinned down some alleles that confers good advantages, one can think a drug could be created, or gene therapy, or whatever - in the end, the whole human population would benefit from it.

Stop thinking about ethnicity. Only the end result matters. It's a shame people in north america and europe are such crybabies about these issues. Thanks China for taking such an endeavour that might benefit humanity - even today naysayers.

(and I say this as someone who does not believe that IQ can be accurately identified or measured with todays techniques, but still there is some noticeable difference between people innate abilities - and trying to understand causes them is a good thing. many genes are likely to be involved, and 10 000 certainly won't be enough, but it's a good start)

I suppose people groups who have had hundreds of years of experience of being oppressed with excuses from the Bible, Social Darwinism or the like might care.

As a species we have considerably more experience with racial/ethnic judgement leading to suffering than leading to something beneficial.

Not to say that it's impossible, but it strikes me as very reasonable for us to argue very thoroughly about the potential implications of this kind of research. I hope we can find a way to improve life for all people. When vaccines first became available there were probably jerks like me moaning about imagined terrible consequences. But that doesn't mean every change made possible by science would necessarily be as equally beneficial.

If we find out that people from East Whateveria have a mean IQ 25 lower than the global target, does that mean we have a right or obligation to "fix" them? What if they don't want to be fixed? I'll attempt stop short of pulling a Godwin, but note that this sort of 'fix defective populations for the benefit of the whole human populations' has a mixed record.

People are free to do what they want. No moral obligation - just freedom.

But if by any chance I found that I had this same mutation and could take advantage of their fix, I would be interested.

If you found out say your son had the same mutation, wouldn't you be interested as well? Please note we are not taking about East Whateveria or how West Whateveria might gloat about not having this mutation. We are just talking about how oneday we might have the freedom to take such a decision for ourselves.

Because it's all about freedom, and unfortunately without having the scientific know-how it's pointless to speculate about what we might do.

This is why I support such initiatives- it will offer more possibilities, possibilities that we can only dream about now.

> People are free to do what they want. No moral obligation - just freedom.

You assume that, I don't. At least not for enormous numbers of humans past and present.

The fact that science has scared off a sufficient amount of idiots in the past doesn't make it a bad thing.

> We are just talking about how oneday we might have the freedom to take such a decision for ourselves.

Unfortunately, watchdog groups like to argue that having that option is detrimental. They usually conflate it with claims of "ethnic eugenics". Btw, to a large extent today you already have the ability to make genetic modifications before insemination or embryo implantation. Screening too..

Allowing ourselves to be ruled by the fear of objective facts is the ultimate level of stupidity.

There is a huge difference based on ethnicity. The average IQ for Ashkenazi Jews is 125, while say the people from West European origin have around 100. That's the difference between somebody capable of doing good job in construction and someone capable of becoming a surgeon.

Wrong, by about a standard deviation. The mean iq of the Ashkenazi is about 113, the average for a very good nurse or less than average accountant. Of course, that mean deviation is still statistically significant enough to show a disproportion of outliers of extrema of a normal distribution. Just get your numbers straight before making statements about the matter.

Can be, just seen the number of 125 in some study. By surgeon I meant a PhD doing operations on humans, but I guest somewhere it means a a nurse.

People used to use an IQ test scale that had 24 instead of the current 15 IQ points as the standard deviation. An 125 IQ in that scale would correspond to 116 IQ in the currently used one.

No, a different standard deviation doesn't change the scores, only the predicted population distribution of the scores.

If you sit down to a standardized IQ test, your score won't change because someone thinks the distribution of scores within the population is wider than someone else does.

It's not about what you guess the distribution in the population to be, it's how you represent distribution you get from the actual test results. Someone who does the IQ test and gets a result 1 standard deviation above the mean of all the test-takers gets an IQ score of 124 from a test-giver using the old scale and an IQ score of 115 from a test-giver using the current scale.

No, it doesn't work that way. The test delivers a specific IQ score for a given performance regardless of the traits of the population of test-takers. One's score doesn't rely on the scores of other test-takers or the statistics of the population as a whole.

If this were not true, people's IQ scores would change depending on the makeup of the tested population, but in fact, it's the other way around -- the statistical makeup of the population, the mean and standard deviation, depends on assessing many individual scores, each of which is immutable and unrelated to the population's overall statistics.

If your position were valid, if one person took one IQ test on a desert island, he would not be able to get a score at all, for lack of a population to give a context to the test result. But this is not how IQ testing works.

Let's simplify this. Let's say it's an arithmetic test of 200 questions of gradually increasing difficulty. An average test-taker can answer 100 of the 200 questions. A very smart person can answer 150 questions. Do you really think an individual's score depends on the average score and distribution of the population of which he is a part?

I must add that, if IQ scores really depended on the population's traits, then IQ testing would really deserve its present terrible reputation. Apropos:


I'm talking about the initial standardization of the test, where they determine which results correspond to which test scores by taking the results of an initial population and fitting it on a mean 100, SD 15 normal distribution: http://www.americanscientist.org/issues/id.881,y.0,no.,conte...

Some scoring systems use an initial standardization where the standard deviation is 15 points, others use 24 points, so the same test performance can get you IQ 115 or IQ 124 depending on whose test you take.

Yes, I know. But that debate has been settled, and AFAIK the result is population mean 100, standard deviation 15.

Unless, of course, different groups of testers are using different assumptions, but without being driven by the analysis of the largest possible collection of standardized test scores. If so, it casts IQ testing into doubt as a reliable tool.

Evidence for this being a settled issue is the fact that workers in this field report a gradual increase in IQ over the decades:


If mean IQ really was adjusted to agree with current test scores, the mean would always be 100, regardless of test score changes over time.

> Some scoring systems use an initial standardization where the standard deviation is 15 points, others use 24 points, so the same test performance can get you IQ 115 or IQ 124 depending on whose test you take. [ephasis added]

The conclusion is still false -- the tests itself doesn't change, only the scoring assumptions. Those who assume σ = 15 could acquire the tests from those who assume σ = 24 and add them to their own dataset, and vice versa. Also, I have to say, either the standard deviation can't change the test scores, or the test scores have no meaning.

One more thing -- the standard deviation shouldn't be an assumption, with one group arbitrary choosing 15 and another choosing 24. The value should be derived from a large set of test scores, not a committee casting a vote.

Your argument seems to be that one's IQ score depends on the population result, along with some arbitrary assumptions like σ = 15 or σ = 24. But that's the reverse of normal statistical practice, in which the mean and standard deviation derive from test scores, not the other way around.

Obviously I'm not doubting that what you say may be so, only that it shouldn't be so -- the standard deviation shouldn't be based on anything but the analysis of a large set of standardized test scores.

My argument is that there are two* different units for IQ points, like inches and centimeters, which are both called "IQ points" by psychologists because psychologists suck at units.

I'm not making an argument about problems with the actual process of measurement, I'm making an argument that the confusion between two reported values sounds quite a lot like a confusion between two reported lengths would sound if they were of the same object, but one had been made with a centimeter ruler and another with an inch ruler, but both had been labeled just "length" in the report.

You can obviously trivially convert between the scales and convert things to the modern scale, once you know that the value you got uses the different SD value. But when the values get just thrown around as "x IQ", you don't know if they are on the old scale.

I'm not entirely sure what you think I'm arguing, but so far you've been talking about something quite different the entire time.

(*Wikipedia says there are actually three common IQ scale conventions, two psychologists had some sort of feud and one of them picked SD=16 to piss the SD=15 guy off.)

> My argument is that there are two* different units for IQ points, like inches and centimeters, which are both called "IQ points" by psychologists because psychologists suck at units.

Yes, I'm not doubting that this is so, only that it shouldn't be so in a scientific endeavor. If IQ testing were purely scientific (as opposed to being partly political), all those involved in IQ testing would allow a large set of test scores in a standardized test to produce the mean and sigma values on which everyone would need to agree. In other words, an empirical outcome.

> I'm not entirely sure what you think I'm arguing, but so far you've been talking about something quite different the entire time.

Apparently so. My point is that IQ test scores must be collected on an absolute scale based on testing results, before any of the adjustments you're describing. If this weren't the case, if test outcomes depended on something other than the direct performance of the subjects measured in a uniform, reliable way, the testing procedure would be fatally undermined.

Bottom line: I doubt that changes in mean and sigma can produce two different IQ scores in a standardized test as you're claiming. For this to be true, the relationship between the population statistics and the analysis result (mean, sigma) would have to be reversed -- it would put the cart before the horse.

Imagine this conversation:

Q. How do the statistical results derive from the test scores?

A. By a straightforward procedure -- the test scores are subjected to a classical statistical analysis, resulting in a mean and standard deviation.

Q. How are the original test scores arrived at?

A. They're derived from (a) the test results, but (b) adjusted by the the mean and standard deviation values of the population created above.

Q. (after a long pause) But ... but ... doesn't that create an example of circular reasoning, in which the scores rely on the stats and the stats rely on the scores?

A. What? I'm not following you. Can you draw a picture?

Q. http://i.imgur.com/ovRvYQE.jpg

A. Okay, I get it. So the statistic analysis depends on the test scores and the test scores depend on the statistical analysis. I don't see a problem with that.

Q. Have a nice day, doctor.

I give up.

can you provide a citation for the claim that quality of education has a significant effect on IQ?

Is anyone at all concerned with a Chinese eugenics program? We already know the extent to which China has gone to produce high-caliber athletes, especially for the Olympics. (Yao Ming is basically their Kwisatz Haderach[1]) Let's say for the sake of argument that there is an identifiable genetic component to genius/IQ/intellectual achievement. Is the following sequence of events particularly outlandish?

1) This study and others like it discover genetic components to intelligence.

2) China conducts large scale genetic analysis of the intelligentsia / upper classes etc. to find carriers

3) Strong incentives provided for couples to use IVF/embryo selection/whatever to produce children with the maximum number of these intelligence genes

4) 25+ years from now China all of a sudden has a lot more smart people than anywhere else does

[1] http://www.time.com/time/magazine/article/0,9171,1126765,00....

Yes - (1) and (4) are extremely unlikely, though (2) and (3) are certainly plausible :)

More specifically, this study may publish some marginal results about genetics of extreme IQ, but any single effect will likely be so small that nothing actionable will come from it.

If (4) is the goal, I suspect they'll have much more success exploring drugs and other cognitive enhancements. What genetics can do is help inform how to discover those drugs - ie. what genes should you study.

(Shameless plug: if you like genetics and writing python, drop me an email if you want to work on studies like this!)

(4) is very plausible even if the study fails to find genetic variants associated with intelligence. Assuming intelligence is heritable at all, you don't need to understand the underlying genetic architecture to breed for it any more than neolithic farmers needed to understand the genetics of docile sheep to breed them.

This has been known for a long time, but there has been no large scale, serious attempt to implement a eugenics program breeding for intelligence. I don't see how this research would change anything, successful or not, other than by getting the subject some attention in the news.

Why do you think discovering genetic components to intelligence is extremely unlikely?

I expect it to be easier to find genetic components for intelligence than for other polygenetic traits like height -- the brain has fewer components influencing development.

> the brain has fewer components influencing development

What do you mean by that? I'd argue the opposite. In a sense, we already have discovered genes for high height with Marfan Syndrome.

I think IQ is orders of magnitude more complex. Awesome conversation but probably hard in an online forum, so I'll just cite my favorite example: Dickens and Flynn [1]

I absolutely think this study has a chance of revealing some signal, my skepticism is just that it will have any actionable conclusions. Here's a litmus test: what results would allow you to choose from a group of 10 men and 10 women, based only on genotype, which pair is most likely to have a high IQ child. And how would your power compare to just reading all their resumes?

[1] http://en.wikipedia.org/wiki/Heritability_of_IQ#Dickens_and_...

Barring specific diseases like Marfan Syndrome, we can't yet explain most of height -- "Height is a highly heritable trait, with estimates of heritability as high as 90%. Recent genome-wide association studies of height have discovered over 180 common variants associated with height. These variants have small effect sizes and collectively explain approximately 10% of the heritability." [1]

Growth involves many systems with many types of cells interacting -- digestive, skeletal, endocrine, etc. There's a lot of complexity when you have so many systems involved.

By comparison, there are relatively few distinct types of cells in the brain, so the effects of different alleles should be more apparent. In particular, genes controlling sensitivity to and production of neurotransmitters directly impact brain function.

If they find many specific mutations that cause better performance on specific tasks (working memory, pattern matching, verbal ability, ...) it should be possible to predict high IQs.

I don't expect 100% accurate predictions, but to not find _anything_ would be surprising. Maybe that's your point-- if they only get 10% heritability, it's not really enough for breeding decisions.

[1] http://www.plosgenetics.org/article/info%3Adoi%2F10.1371%2Fj...

I recognize that your litmus test was rhetorical, but it's a question that does seem answerable, so I'll try to answer it.

I would imagine there are a number of genes that are associated with lowered intelligence. For example, one might start by looking at genes such as the Dopamine D4 receptor gene, for which the 7-repeat allele seems to be a pretty good indicator of ADHD [1][2], which in turn is associated with significantly lower IQ. Of course, chains of correlation don't always work, and that is why it would be important to have tens or preferably hundreds to thousand of things like that to look for. While it would be nice and easy to look for an "intelligence gene", that may not be feasible. However, my guess is that looking for "mild cognitive impairment" genes and then selecting the pair with the lowest number of those genes could give a rate of successful guessing well above random.

So this study, alone, would not be likely to allow you to make a prediction that's much better than random, and resumes are going to be hard to beat period (though with the resumes and the genetic data, you might be able to make better predictions). And if Zhao and his team do find a gene that correlates with increased intelligence, that will be big news (though I would not give good odds on that actually happening).

In any case, I agree that IQ is complex, but I don't think that either the "intelligence gene" hypothesis or Dickens and Flynn's hypothesis is the whole story, because identical twins raised apart still have a significant and fairly high correlation in IQ scores [2].

[1] http://duepublico.uni-duisburg-essen.de/servlets/DerivateSer...

[2]http://onlinelibrary.wiley.com/doi/10.1111/j.2044-8295.1966.... (unfortunately paywalled, though I'm sure someone will accidentally drop it if you need it).

Change is the only way forward. If the Western civilization refuses to accept the facts of life and change itself, it will become just an interesting topic in the history books.

I'm not really interested in getting into the conspiracy minutiae, but let me just play devils advocate on #4:

What, exactly, would be the harm to humanity of a coming generation of hyper-intelligent Chinese nationals?

The assumption is that more hyper-intelligent Chinese nationals will mean more world influence for PRC. Whether that's good, bad, or neutral depends on your POV.

What, exactly, would be the harm to humanity of a coming generation of hyper-intelligent Chinese nationals?

technology drives military power. Therefore, one's POV will be influenced by nationality.

Or in plain English, an American probably wouldn't want to be on the wrong side of Chinese drone airforce.

The evolutionary psychologist Franz Herbert had written a piece about Chinese eugenics some time back


eugenics and the military/economic impact down the road was my first thought.

One of the fun facts about IQ is that it seems to be highly genetic, but not particularly hereditary. There's a .76 correlation for identical twins separated at a young age, but only .22 between a parent and child, when the child was lived apart from the parent[1]. Compare this to height, which is much more heritable[2]. So they'll probably need more than the 10,000 samples they needed to find height-related genes. It'll be interesting to see the findings.

[1]: http://en.wikipedia.org/wiki/Heritability_of_IQ#Correlations...

[2]: http://en.wikipedia.org/wiki/Twin_study

To me, this suggests that the trait is highly combinatorial, involving potentially large combinations of polymorphisms instead of individual loci acting relatively independently (in an "additive" fashion).

Of course, such combinatorial traits are much harder to tease out of data.

There has been plenty of controversy over the meaning of IQ tests in the first place. One complaint is that they test cultural knowledge more than intelligence itself. Naturally, there will be some correlation between the two, but my point is that a great deal of what IQ tests test may be learned knowledge. While your ability to learn that knowledge may be related to intelligence, it does not follow that all intelligent people will score highly on a particular IQ test. Intelligence may be heritable even if IQ test performance is not.

I'll be curious to see if they pull this off. Finding genetic associations reliably is notoriously tricky, simply because of the enormous potential for false positives.

When there are thousands of possible genes to consider, there are thousands of opportunities for a gene to be "lucky" and have a coincidental correlation with intelligence in your sample. Previously published efforts have produced almost entirely false positives:

"Most Reported Genetic Associations With General Intelligence Are Probably False Positives" - http://pss.sagepub.com/content/23/11/1314

Let's hope that BGI doesn't merely torture the data until it confesses, as so many have.

There are those (generally average people) that wish to democratize IQ insisting that unlike beauty and athletic ability that IQ is nurture and not nature. But like beauty and athletic ability high IQ and giftedness in science and the arts is mostly genetic. It should also be noted that IQ tests frequently do not indicate true intelligence -- at least among the truly gifted. For example, Genius Richard Feynman was reported to have an IQ of 125 -- clearly not a true measure of one of the greatest minds of the 20th century.

Regarding Ashkenazic Jews (disclosure, I am one) and their high IQ, Jews instituted universal education until 6th grade for all males beginning 2,000 years ago -- about 1,750 years before any other ethnic group. It was required of all Jewish males to study the Old Testament and The Talmud and Jewish law since that time. It was around that time that Jewish males were encouraged to move away from agriculture to occupations that required reading, writing, and math.

The Chosen Few: How Education Shaped Jewish History, 70-1492 http://www.amazon.com/The-Chosen-Few-Education-Princeton/dp/...

How can you disentangle upbringing from genetics? They are hopelessly intertwined, especially in your example of Ashkenazic Jews.

Athletic ability is especially sensitive to nurture as well.

Talent is fine; most of accomplishment is application and practice. I can completely believe that Feynman had an IQ of 125. Applied with a depression-generation self-discipline, it can (did) accomplish amazing things. It's disingenious to take a counter-example to an argument and pretend its supporting that argument.

This article is not genius - it is nonsense. There is no one "genetic code" for genius - this is polygenic. IQs measured by the stanford-binet or wechsler are actually variable - one can work to bring these scores up dramatically - they are completely game-abale. Furthermore there are many types of intelligence, people with high general intelligences can consciously hone specialty intelligences. Or, people with low general intelligences may possess high specific intelligences. This depends on the structure of the brain from macro to a micro perspectives. Also, intangibles govern the way we interact and as the environment changes around us, as we near a "singularity," "intelligences" we value today may not be the same "intelligences" we value tomorrow. The definition of intelligence will change over time. This article is nonsensical and loserish.

Citation needed for the claim that IQ tests are highly game-able. And nobody denies that there are different types of mental abiltiy, so I'm not sure who or what you are trying to argue against.

When someone reduces human intelligence to one number, that is really a good indicator of that someone's intelligence. We don't know much about intelligence except that it is diverse. There is a bunch of excellent ideas about intelligence, talent and creativity in [1]. I particularly liked a bit where current model of education is compared to strip mining of Earth for commodities i.e. only small fraction of human talents is developed while others are ruthlessly squandered.

[1] TED talk by Sir Ken Robinson, "Do schools kill creativity".

When someone trots out the same old "multiple intelligences" line, that is a pretty good indicator of their intelligence. or at least their intellectual laziness.

If they are successful I assume this will likely not result in a way to improve the intelligence of existing individuals though gene therapy because their brains are already wired a certain way and unlike other cells of the body the neurons in the brain do not regenerate rapidly. Is this a correct assumption?

If they find specific genes or non-coding RNAs with relatively strong associations (a big if), drugs could be designed to alter the expression levels of those transcripts.

Since it is unclear how much of IQ is determined by the "macro" structure of neurons and synapses and how much is caused by intracellular events, it is really impossible to say how much an intervention could do to increase intelligence. The current state of the nootropics field, however, is not very encouraging.

In any case, I would not pin my hopes on gene therapy. The hurdles that would be required for regulatory approval of "cosmetic" gene therapy (as this would be classified) are nearly insurmountable.

>"cosmetic" gene therapy (as this would be classified)

Since higher IQ (at least up to a point) correlates positively with good health and other desirable characteristics including overall happiness [1] would this really be considered "cosmetic"?

[1] https://journals.cambridge.org/action//displayAbstract?fromP...

Personally, I agree with you, but the NIH does not. So far, "stupidity" is not a disease (nor is aging, for that matter).

How did he get that position if he is a dropout? Especially in China where it's metricopia?

In future, there won't be any need for interviews and resumes. just send your genome for admission to harvard or a job a google!

From the article, which I read in the print edition this afternoon: "The majority of the DNA samples come from people with IQs of 160 or higher." I call baloney on that, as there is no currently normed, well validated IQ test that yields a score of higher than 160. From farther along in the article: "Those samples were obtained through a U.S. project known as the Study of Mathematically Precocious Youth, now in its fourth decade." There is nothing so amazing about the genes of the participants in the Study of Mathematically Precocious Youth (one of whom is my immediate descendant, also a participant here on HN) compared to what is amazing about their upbringing, including having the opportunity to take the SAT college admission test before age thirteen as part of regional Talent Search


to qualify for inclusion in the Study of Exceptional Talent.


As another few replies have already noted, IQ is polygenic, and no one can identify any genes with a strong effect on IQ. But there are plenty of educational practices that spare a minority of United States young people the lousy elementary mathematics instruction that most receive, and the pupils who got a good start in mathematics stand out when they take a mathematics test at an unusual age, if the parents think to arrange the test.

A really good bibliography about IQ and human intelligence topics can be found at


as part of Wikipedia userspace, references to reliable sources that ought to be used by more Wikipedians to update the articles there.

Prediction: the study reported on in the article kindly submitted here will not find any genes with strong effect on IQ. The review article Johnson, W. (2010). Understanding the Genetics of Intelligence: Can Height Help? Can Corn Oil?. Current Directions in Psychological Science, 19(3), 177-182


looks at some famous genetic experiments to show how little is explained by gene frequencies even in thoroughly studied populations defined by artificial selection.

"Together, however, the developmental natures of GCA and height, the likely influences of gene-environment correlations and interactions on their developmental processes, and the potential for genetic background and environmental circumstances to release previously unexpressed genetic variation suggest that very different combinations of genes may produce identical IQs or heights or levels of any other psychological trait. And the same genes may produce very different IQs and heights against different genetic backgrounds and in different environmental circumstances. This would be especially the case if height and GCA and other psychological traits are only single facets of multifaceted traits actually under more systematic genetic regulation, such as overall body size and balance between processing capacity and stimulus reactivity. Genetic influences on individual differences in psychological characteristics are real and important but are unlikely to be straightforward and deterministic. We will understand them best through investigation of their manifestation in biological and social developmental processes."

>I call baloney on that, as there is no currently normed, well validated IQ test that yields a score of higher than 160.

I know the WAIS caps at 155, but the Stanford-Binet test accomodates 160+, does it not?

> I know the WAIS caps at 155, but the Stanford-Binet test accomodates 160+, does it not?

Maybe, maybe not. Know why there's so much uncertainty? The population of possible subjects is too small. For a population with a mean IQ of 100 and a standard deviation of 15, the proportion above 160 IQ is only 0.003%. For a population of 300 million, that's about 9500 people. And remember that choosing the subjects would be a self-referential quandary -- you need an accurate IQ test even to select the population to submit to the IQ test.

So no, there's no basis for asserting accurate measurement in that part of the IQ distribution.

Extra credit -- Marilyn Vos Savant is supposed to have an IQ of 228. Oh, really? I'm not disputing that it's possible, only that there's no way to accurately measure an IQ like that. It's not reliable science or statistics.


I definitely understand your argument and agree with it. Thanks for clarifying; didn't quite get the gist of where you were going with that.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact