Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Most Reported Genetic Associations with Intelligence are Probably False

dash.harvard.edu

Post reply on HN
Print view View original

11–20 of 57 posts

Re: Most Reported Genetic Associations with Intelligence are Probably False

#11

Post by rch »

tehwalrus wrote: post #9

I would assume that many behaviours (including language itself) are inherited - and yet have nothing whatever to do with genetics. How would you control for this? Only last week, we had an article on here about how vocabulary in very young children was a function of how much parents talked to their children - with marked differences between economic groups[1]. There are many many systematic errors in this kind of dat…

Check out 'The Blank Slate' by Steven Pinker. It presents a fairly thorough accounting of how these sort of conclusions may be reached.

Re: Most Reported Genetic Associations with Intelligence are Probably False

#12

Post by tokenadult »

tehwalrus wrote: post #9

I would assume that many behaviours (including language itself) are inherited - and yet have nothing whatever to do with genetics. How would you control for this? Only last week, we had an article on here about how vocabulary in very young children was a function of how much parents talked to their children - with marked differences between economic groups[1]. There are many many systematic errors in this kind of dat…

How would you control for this?

The classic way to control for cultural and in-family influences in studies of human behavior genetics is to look for the rare cases of monozygotic ("identical") twins brought up in separate households after adoption. Those studies of monozygotic twins reared apart (MZA twins) set a MAXIMUM figure for the heritability of any trait of interest--because the genes are as similar as they can be between two different human individuals, but the environments, even in separate households, may not be as maximally different as they could be in influence on the trait in question. There is a whole book about the most famous study of identical twins reared apart, which took place at my alma mater while I was an undergraduate and postgraduate student there (in other subjects).

http://www.hup.harvard.edu/catalog.php?isbn=9780674055469

As some people will remember from the press reports at the time, the most culturally variant upbringing two of the separated twins, Oskar and Jack, studied by the Minnesota researchers received was one twin being raised as a Jew, and the other twin being brought up in Nazi Germany and becoming a member of the Hitler Youth.

http://online.wsj.com/article/SB1000142405270230355210457743...

Re: Most Reported Genetic Associations with Intelligence are Probably False

#13

Post by bayesianhorse »

No surprise there...

Knowing a lot about genomic animal breeding made me wonder how people can be so sure about genetic basis of intelligence, while we understand neither intelligence nor the genetic basis behind it. For that matter, even plain body height in humans is barely understood...

Re: Most Reported Genetic Associations with Intelligence are Probably False

#14

Post by bayesianhorse »

Xcelerate wrote: post #5

Note that the article is not saying that intelligence is not heritable. In fact it makes it quite clear that despite the controversy, the opposite is true: > Although the exact figures have been the topic of much debate, the claim that IQ is at least moderately heritable is widely accepted. IQ may in fact be similar in heritability to the physical trait of height What the article is saying is that intelligence cannot…

For all practical purposes, "intelligence is largely not heritable" is actually closer to the truth than the opposite.

As long as you can't tie intelligence to particular genes, as long as you can't tell what amount of intelligence in an individual is inherited, and how much isn't, there is no use for this in education, politics or social welfare.

What I see the argument being used for, mostly, is to justify racism and entitlement issues in wealthy people...

Re: Most Reported Genetic Associations with Intelligence are Probably False

#15

Post by tehwalrus »

tokenadult wrote: post #12
tehwalrus wrote: post #9

I would assume that many behaviours (including language itself) are inherited - and yet have nothing whatever to do with genetics. How would you control for this? Only last week, we had an article on here about how vocabulary in very young children was a function of how much parents talked to their children - with marked differences between economic groups[1]. There are many many systematic errors in this kind of dat…

How would you control for this? The classic way to control for cultural and in-family influences in studies of human behavior genetics is to look for the rare cases of monozygotic ("identical") twins brought up in separate households after adoption. Those studies of monozygotic twins reared apart (MZA twins) set a MAXIMUM figure for the heritability of any trait of interest--because the genes are as similar as they c…

I guess I was saying that the rarity of this happening in the real world makes good statistics difficult (which are essential in biological systems,) and of course intervening to force better statistics is a rather horrific proposition.

Thank you for the links though, I'll have a read!

Re: Most Reported Genetic Associations with Intelligence are Probably False

#17

Post by tehwalrus »

rch wrote: post #11
tehwalrus wrote: post #9

I would assume that many behaviours (including language itself) are inherited - and yet have nothing whatever to do with genetics. How would you control for this? Only last week, we had an article on here about how vocabulary in very young children was a function of how much parents talked to their children - with marked differences between economic groups[1]. There are many many systematic errors in this kind of dat…

Check out 'The Blank Slate' by Steven Pinker. It presents a fairly thorough accounting of how these sort of conclusions may be reached.

I have that on a shelf somewhere, but I don't think I persevered to the end. I'll have another look, thanks!

Re: Most Reported Genetic Associations with Intelligence are Probably False

#18

Post by bayesianhorse »

tokenadult wrote: post #3
JoeAltmaier wrote: post #2

Nature vs nurture is hard to separate, especially when genes and culture tend to be 'handed down' to the same individuals. However I'm curious that they chose to use the very-powerfully-determined-by-genes features of Alzheimers and BodyMassIndex as their control. Surely there are other genetic traits that are not 'disorders' which could have been referenced? Breakdowns in physical function would naturally have large…

Surely there are other genetic traits that are not 'disorders' which could have been referenced? Studies of the issue of "missing heritability" have often investigated height (bodily stature), which has a very high calculated heritability (resemblance among close relatives) but which must be influenced by hundreds of genes, many of very small effect. Even very extensive genome-wide association studies (GWAS) in large…

One example for the trouble with genetics in intelligence is the issue of Malaria: If a child gets Malaria, it will have a lower intelligence quotient. Resistance/resilience of Malaria sure has genetic factors. Would those factors show up in genetic studies about intelligence? Betcha.

Should we accept this variance in IQ as genetic and unavoidable? Or should we just buckle down, eradicate Malaria and stop worrying about these factors?

Re: Most Reported Genetic Associations with Intelligence are Probably False

#19

Post by scott_s »

tehwalrus wrote: post #9

I would assume that many behaviours (including language itself) are inherited - and yet have nothing whatever to do with genetics. How would you control for this? Only last week, we had an article on here about how vocabulary in very young children was a function of how much parents talked to their children - with marked differences between economic groups[1]. There are many many systematic errors in this kind of dat…

I think you're using "inherited" to mean "comes from your parents in any way," but that's not what it means in this context. If something is learned from your parents, then it would not be consider an "inherited trait".

Re: Most Reported Genetic Associations with Intelligence are Probably False

#20

Post by rayiner »

bayesianhorse wrote: post #14
Xcelerate wrote: post #5

Note that the article is not saying that intelligence is not heritable. In fact it makes it quite clear that despite the controversy, the opposite is true: > Although the exact figures have been the topic of much debate, the claim that IQ is at least moderately heritable is widely accepted. IQ may in fact be similar in heritability to the physical trait of height What the article is saying is that intelligence cannot…

For all practical purposes, "intelligence is largely not heritable" is actually closer to the truth than the opposite. As long as you can't tie intelligence to particular genes, as long as you can't tell what amount of intelligence in an individual is inherited, and how much isn't, there is no use for this in education, politics or social welfare. What I see the argument being used for, mostly, is to justify racism a…

IQ is highly heritable: http://en.wikipedia.org/wiki/Heritability_of_IQ#Estimates_of... (estimates of adulthood heritability are 0.7-0.8, similar to height).

We can reject the existence of an observed natural phenomenon because we can't explain it. We still haven't unified gravity with the standard model, but that doesn't mean we reject that gravity exists (for practical purposes or otherwise).

The fact that a natural phenomenon might be socially inconvenient is even less of a good reason to reject its validity.

Finally, I don't even see why it should even be so socially inconvenient. Say the experiments are right, and intelligence is mostly heritable rather than a result of education, upbringing, etc. Why should that justify entitlement on the part of the wealthy? Indeed, it should cut across that grain. If its mostly heritable, being intelligent shouldn't be seen as intrinsically virtuous any more than is being tall. Moreover, the heritability of intelligence may very well imply a need for a change in our priorities when it comes to social welfare. E.g. maybe we need to de-emphasize education, and divert resources to building up social and family networks.

Post reply on HN