DNA tests can predict intelligence, scientists show for first time
telegraph.co.uk
telegraph.co.uk
> Between 3.64 and 6.84% of phenotypic intelligence (as measured by the VNR Test in UK Biobank) could be predicted (Supplementary Table 10); the upper limit is an improvement of ~43% on the largest reported estimate to date, of 4.8% [16].
Doesn't 7% seem quite small (though better than previous studies)? What am I missing?
Malnutrition and serious physical abuse aren't the only environmental factors with a known significant impact on intelligence, and nutrition (particularly childhood nutrition) patterns with a serious adverse effect on IQ have been identified in research focussed on first world areas of study; they certainly aren't rate enough to be insignificant sources of variation in the first world.
https://www.ncbi.nlm.nih.gov/pubmed/2628311
https://www.sciencedaily.com/releases/2004/11/041117005027.h...
Overall exposure is lower, sure, that doesn't mean it is insignificant in impact on IQ (and other health issues, notably cardiovascular disease.)
No, lead exposure from water systems, unabated older housing, continued use of leaded aviation fuel, and other sources continues to be a nontrivial issue in the US and some other parts of the first world.
"And some of the genetic variants that are linked with an increase in intelligence are also linked with an increase in life expectancy."
Also, what about creativity? Or (pardon me, just an example) emotional intelligence? An over-emphasis of the IQ just feels so 20th century.
> Intelligence, also known as general cognitive function or simply g, describes the shared variance that exists between diverse measures of cognitive ability
You're of course free to come up with a different definition.
It doesn't measure anything well rounded about a person and it makes me pretty uncomfortable to think this kind of thing could actually be used to divide people.
IQ is simply a way to quantify intelligence.
>I was under the impression that the IQ test has a number of biases (e.g., cultural) that leave it flawed.
There is actually no standardized way to measure IQ. You've heard criticisms of a certain test, but that is just one of many. There are tests that do no use any sort of cultural knowledge at all, such as Raven's Progressive Matrices. Theoretically, a cave man and a Harvard professor could reach the same score, assuming all other factors are the same. It only measures visual pattern recognition, but it is generally assumed that that is highly correlated with overall intelligence.
https://en.wikipedia.org/wiki/Raven%27s_Progressive_Matrices
There are also g factor tests that attempt to measure overall intelligence.
https://en.wikipedia.org/wiki/G_factor_(psychometrics)
>or emotional intelligence
Pretty much pseudoscience
Unless we establish laws which compensate for the asymmetrical distribution and power, we will see technology used to amplify it.
Gattaca was truly great for many reasons, including showing us a (benign) version of such a society.
If you just check for two X chromosomes...
--Harrison
The point I guess I'm trying to make is that people are individuals, and should be treated as such.
For example, suppose you have two groups, each one consisting of 1 billion people. Both groups have standard deviation of 15 IQ, but the mean of group 1 is 90, and the mean of group 2 is 100. Then, the IQ of 160 is 4 sigma above the mean of group 2, but 4.6 sigma above mean of group 1. This results in 31670 people above 160 IQ in group 2, but only 1531 people above 160 in group 1. With your cut-off of 180 IQ, only 1 individual out of 1 billion from group 1 will have IQ above 180, but almost 50 in group 2.
Of course, each individual should be treated as such, but you have to take things like above into account if you are wondering for example why so many top sprinters are of certain genetical ancestry.
Intelligence is such an amorphous concept in our society.
Intelligence is a similarly high level word that just encompasses too many things.
Critical thinking and problem solving, for me, fit your criteria. Are these nature or nurture?
And what of drive and ambition?
There are plenty of really smart people on the dole, yes? But they're probably just in the wrong place at the wrong time. How is an advantage an advantage if you're a genius ice sculpturer in say FL? :) Is that not a case of it being relative?
This one seems to do some weird combination of that, plus several smaller datasets which have some kind of test score, all combined & normalized somehow. (Presumably they had some fun conversations with their referees about the statistics...)
It would really help a lot of people if we could segregate children as they enter the school system into more refined groups based on their needs rather than waiting for someone to select you for gifted programs like we do now.
Then each group could focus on its unique challenges in trying to eliminate whatever portion of intelligence is derived from the environment.
Surely this could never happen.
I, for one, hope that this story isn't true.
As is the psychology of intelligence. And the relationship between g and being a good developer.
All of this is called doing science - because it's currently a work in progress - not denying it.
We already do this too an extent, schools have "smart" classes and "dumb" classes, typically for things like math. The biggest problem is that they don't apply this until high school. Aside from any innate ability, kids have different learning speeds and letting them fall behind on a subject they don't understand while the rest of the class moves on does no good for anyone.
http://www.pbs.org/video/frontline-class-divided/
Race is one of the triggers for ignorance and hate, it is so visible. Once we make genetic makeup visible, it will change how we see each other.
I have friends like these, who, if they were to be shoehorned into doing highly abstract creative work, would be downright miserable and would eventually burn out.
People look at BNW as some type of dystopia, but I felt like the novel handled it rather well in the sense than there is a place for outliers to be free as well, without coercion (Iceland).
The people who thought they didn't fit in their 'caste' so to speak weren't forced to work at gunpoint.
The point of BNW was that there could be effective “coercion” based on molding our desires instead of stamping them out with fear and force. The gunpoint wasn’t necessary; your prison walls were built in your head by chemistry and psychological conditioning.
We all have our prison walls. To pretend that the biological walls are somehow more horrifying and cruel than sociological or economical is, in my opinion, lunacy.
Success in academics is almost entirely derived from motivation and the amount of work and interest that the student puts into it. ESPECIALLY at the elementary school level.
This makes me distinctively uncomfortable.
Also, you've made the conclusion that relative intelligence, if even very accurate, implies specific, categorical needs. I'm very confused by this.
What is wrong with having a more foolproof and efficient way?
Where it would be diffrent is very young disadvantaged students that never made it into the gifted program because they didn’t get any attention when they were young will have a much harder time getting missed.
Top tier schools could also start accepting the most genetically superior applicants. Given two students with equal test scores, one might already be 'maxed out' due to environmental advantages. We can start screening for the most potential and get rid of AA while we're at it.
They do not stream the education system and the quicker children are expected to help explain what they have learned to those still getting to grips with the material. Integrating this method within the lesson structure has the effect of raising the standard of both the slower and quicker kids, for fairly obvious reasons.
They have many theoretically equal voluntary high school subjects and tests. More able students take many more, known harder ones. This allows self segregation, much like AP classes in the US will have few people with reading difficulties. No one particularly cares because if you want to go to a third level institution after high school you can.
“The prevailing attitude at Oxford at that time was very anti-work. You were supposed to either be brilliant without effort or accept your limitations and get a fourth-class degree. To work hard to get a better class of degree was the mark of a ‘grey man’, the worst epithet in the Oxford vocabulary.”
You could stream by behavioural attributes to achieve that if you wanted, but then you would have to accept that the classes would be very wide in ability, unless you stream by both behaviour and ability and then end up with a massive logistical nightmare.
edit - also, from my observation, concentrating all the nutters under one roof does not really help, if your aim is educational. It just takes a lot of little problems across a lot of schools and turns them into one big unmanageable problem that can be politely ignored.
----
John Donne - Devotions upon Emergent Occasions - Excerpt from Meditation 17
A genetic test that can predict a single-digit percent of the variation in a value which can reasonably inexpensively be measured more directly isn't even a step toward useful tech for that purpose.
In the largest ever study looking at the genetic basis for intelligence, researchers at the University of Edinburgh and Harvard University discovered hundreds of new genes linked to brain power.
Previous studies have suggested that between 50 per cent and 75 per cent of intelligence is inherited, and the rest comes through upbringing, friendship groups and education. That figure was calculated by studying identical twins who share the same DNA, therefore any differences in IQ between them must be non-genetic.
But nobody knew which were the ‘smart genes.’
Now by studying the genetic data from more than 240,000 people, scientists have found 538 genes which are linked to intelligence.
Researchers were even able to predict intelligence just based on a person’s DNA, a breakthrough which could potentially help doctors to diagnose impaired cognitive ability, or allow children to be given an tailor-made education based on their innate abilities.
Scientists knew that a large part of intelligence was inherited but did not know the genes responsible
Dr David Hill, of the University of Edinburgh’s Centre for Cognitive Ageing and Cognitive Epidemiology (CCACE) who led the research, said: “Our study identified a large number of genes linked to intelligence. “We were also able to predict intelligence in another group using only their DNA.”
The study also showed that the same genes which influence intelligence are also linked to other biological processes such as length of life.
Although it is known that intelligent people live longer it was generally assumed that the link was due to social causes, such as a better education, leading to a more well-paid job, which brings a higher standard of living and a healthier life.
But the new research suggests that intelligent people are biologically fitter.
The team also found that genes linked with problem-solving powers were associated with the process by which neurons carry signals from one place to another in the brain.
A biological intelligence test could help create an individual curriculum for children
“We have shown is that two biological processes neurogenesis, the process by which new brain cells are created, and myelination of the central nervous system are associated with intelligence differences,” added Dr Hill.
“And some of the genetic variants that are linked with an increase in intelligence are also linked with an increase in life expectancy.”
The study’s principal investigator, Professor Ian Deary, also from CCACE, said: “We know that environments and genes both contribute to the differences we observe in people’s intelligence.
“This study adds to what we know about which genes influence intelligence, and suggests that health and intelligence are related in part because some of the same genes influence them.”
Previous studies by King’s College London discovered that up to 65 per cent of the difference in pupil’s GCSE grades was down to genetics, after analysing genetic data from, 12,500 twins. They found that all exam results were highly heritable, demonstrating that genes explain a larger proportion of the differences between children.
The research was published in the journal Molecular Psychiatry. """
https://www.wired.com/2007/12/ps-dna/ (2007)
When will be comfortable accepting that "race" is biological? We need to test drugs on multiple races for this reason.
However, "race" is absolutely not biological because so little of race is physical. Nobody has ever denied that skin colour and facial features are biological, its the accompanying traits that round out what makes a "race" that have no significant biological basis
Yes, people who share similar melanin probably have similar genes providing them that attribute, but there are significant genetic differences between a black Jamaican and a black Ethiopian; or, a white Spaniard and a white Russian (not the drink). Saying they're just black or white dismisses the substantial differences that exist.
>Uncertain about the science, the police asked Frudakis to take a blind test: They sent him DNA swabs from 20 people to see if he could identify their races. He nailed every one.
>On a conference call a few weeks later, Frudakis reported his results on their killer. "Your guy could be African-American or Afro-Caribbean, but there is no chance that this is a Caucasian."
So we can see that not only are the classes used more precise than just "black" or "white", but there is empirical evidence that they can be predicted from DNA.
The latter. The claim is pretty unambiguous. There are genetic markers that are only found in certain populations. His statements imply that he observed no markers that are only found in Caucasian populations, and several that are only found in African or Afro-Caribbean populations. If a person were mix raced, they would likely have a combination.
I read the article and responded to a comment overemphasizing the importance of determining race from DNA.
> So we can see that not only are the classes used more precise than just "black" or "white", but there is empirical evidence that they can be predicted from DNA.
There is no implication from my comment denying the ability to derive an individual's race from DNA. Nor would I ever suggest that you could not predict race using DNA, I am simply questioning the value of doing so. Information from DNA is extremely precise, so diluting the information to interpret race is degrading its potential. As pointed out by others, the human categorization of race is an inaccurate, archaic taxonomy.
Even using the law enforcement aspect of the Wired article, you can clearly demonstrate why race is an antiquated distinction to make when evaluating DNA. If you have more information from DNA, why stop at race? Technology is developing to render facial models from DNA. IMO that is a far superior use of DNA than determining, "Your guy could be African-American or Afro-Caribbean, but there is no chance that this is a Caucasian."
What is biological is that different people from around the world are genetically more similar to their neighbours than people more distant and people with a similar genetic background tend to look similar.
What is more interesting is not race, but that different human populations around the world are different species that can produce fertile offspring. There are four known living human species; sub-Saharan Africans, Euro/Asians, Melanesian/Australian Aborigines, and African pygmies (there might be some more human species in Africa, but we lack the fossil DNA data to know this). Each one of these different species is a hybrid between ancient sub-Saharan African’s and one of the other homo species that lived on the planet before 50,000 years ago [0].
As for races vs species, that well has been so thoroughly poisoned that I for one refuse to drink from it. But, you know... you do you.
I think where people get worked up is thinking because there are different species of humans living on the planet that one must be better (or worse) than the other. Nobody with a clue would say that lions are better cats than tigers, just different. The different humans species are the same.
The other thing that makes things more complex is all living humans are predominantly sub-Saharan African (90% to 100%). If you had a large cat population that was 95% lion and 5% tiger you would most likely not even notice it was not a pride of lions. It would still be a different species, and the average member of this population would be different to the average lion, but the differences would be relatively subtle.
These groups are not considered different species or subspecies of human (homo sapiens sapiens). There are homo sapiens subspecies, but all except sapiens are extinct.
When people connect race and talents/abilities is when people get uncomfortable though for fairly obvious reasons. Which is likely how many will interpret this result (Race related to intelligence) and probably what you really wanted to say.
People that observe that “race” is a highly malleable concept (there isn't even a broad agreement on how many races there are) that evolves with social, demographic, and economic shifts in the culture whose views of race are examined.
Race and ethnicity are simply larger and smaller group categorizations of humans, both with reasonably accurate biological tests.
No, people who recognize that despite the conceit that race is a “biological concept”, the identification of race (both ascription to others and self-identified) has always been driven by the same sociological phenomon that drive ethnic identity, and that, in practice, the concept of race operates as simply a broader set of ethnic categories, which is why identification of what races exist and how people are identified as being in one race or another vary not only between cultures, but within the same cultures over time.
Of course it is biological. Doesn't stop it being an appallingly crap system for human taxonomy. Especially given that there are a very large quantity of people out there whom you would really have to give a long list for in this taxonomy, rather than a word, as people have a habit of not sticking to these neat little groupings when they have sex, even if you force them to with law.
I'm more interested in finding the genes responsible for racism. Racism being such a giant human flaw that can lead to the collapse of societies it would be great if we could identify individuals prone to racism early.
Yes the frequency of response and the side effects will vary by an individual's genetic background, but there is to my knowledge no drug or side effect found only in one broad population grouping.
We don't want unrelated controversies, generic tangents, or flamewars on HN. Racewar is right out. If that's what you want to foment, please foment somewhere else.