This week’s links from Jason Collins are interesting, as usual. For example, Alison Gopnik writes,
When psychologists first started studying twins, they found identical twins much more likely to have similar IQs than fraternal ones. They concluded that IQ was highly “heritable”—that is, due to genetic differences. But those were all high SES twins. Erik Turkheimer of the University of Virginia and his colleagues discovered that the picture was very different for poor, low-SES twins. For these children, there was very little difference between identical and fraternal twins: IQ was hardly heritable at all. Differences in the environment, like whether you lucked out with a good teacher, seemed to be much more important.
If you read this paragraph, you may pick up inferences that I suspect are not supported by the data. One inference is that IQ is not heritable among low-SES children. I do not know much about genetics, but it is hard for me to see how a characteristic can be heritable at one SES level but not at another. Yes, I can see how a characteristic can be affected by the environment at one income level but not another.
The other inference is that what accounts for the difference in IQ between two low-SES children could be having “lucked out with a good teacher.” There is no evidence that teachers have anything to do with this.
We (almost all of us) need to pay way more attention to “necessary but not sufficient” and “more than fully adequate is a waste”
So “achieving full potential” might require min(genes) * min(family love) * min(education) * min(luck)
Having more than min of any of those things is perhaps of very little import. Having a shortage in one area can be offset in others, but only to a limited degree.
[Sort of like how regular exercise probably doesn’t extend your life, but rather, prevents lack of exercise from making it shorter or more miserable than it natively would be.]
And of course there’s the correlation elephant in the room – to what extent do “poor genes” lead to “low SES” such that being born into a low SES circumstance is a double wammy of poor genetics and poor environment?
“[It] is hard for me to see how a characteristic can be heritable at one SES level but not at another. Yes, I can see how a characteristic can be affected by the environment at one income level but not another.” I’m not a geneticist either, but I think I can answer this one. In fact, you answered it yourself in your second sentence:
Heritability within a given population is (technically) the proportion of phenotypic variation in that population (variations in IQ in this case, or variations in height, or other things) that is explained by genetic variation in the population. So if there are environmental conditions that significantly impact IQ (or height, or whatever) in a given population (e.g., malnutrition, exposure to lead, terrible schools, stress caused by living in areas marked by frequent violence, etc., to name some of the typical candidates) then the measured heritability of IQ in that population will be lower. As I understand it, this is basically what the paper is claiming is the case.
This is an ongoing theme of Razib Khan’s, incidentally: That as populations become wealthier overall and more and more people benefit from good nutrition, lower exposure to environmental toxins, etc., we will find the measured heritability of height, IQ, etc., to be higher. This also relates to Bryan Willman’s comment about requiring a minimum of certain environment inputs (in addition to whatever genetic cards you’ve been dealt) to achieve full potential.
There was also a comment on Tyler’s blog about this as well when he was debating with Bryan on whether or not parents affected parental outcomes or not.
It seems like the wealthier you are the more genes determine outcomes because they’re the deciding factor when all the other prerequisities for success have been satisfied. Less so when you’re poor and you have other obstacles in your path.
A good example: Magnus Carlsen.
If he and his sister were born into a poor Bangladeshi farming village, neither one would end up being very good at chess. Since they were born into a wealthy Norwegian suburb they had the free time to explore various interests, not worry about sickness or deprivation, and have loving parents who helped them foster their talents.
Then Magnus’ remarkable chess talents are allowed to bloom and the rest is history.
This result could be because there is more Christmas treeing among the low SES twins. If one twin just randomly fills in the circles and the other ties the differences in the scores can be large even though the IQ is the same.
The important thing is that heritability has nothing to do with how much causal impact there is from genes ON THE WHOLE. It’s only about explained variance given all the other factors.
Think about reading. If you look at a group of kids who all have parents who read to them and who all have good reading teachers, their reading environment may be fairly equal. Then the only thing left to explain variance is genes. So reading ability will look more heritable in that group.
But think of another group of kids (say, poor kids) who have highly variable environments re: reading. Some of the poor kids have parents who read to them regularly, some do not. Some of them have good reading teachers, some do not. With environment being highly variable, then there’s a lot more room for environment to explain some of the variance. Therefore reading ability will be less heritable.
Adding to Frank Hecker’s comment above about this being a theme of Razib Khan. Here’s the link to Razib’s post called “The end of environmental inequality means the rise of genetic inequality”
http://blogs.discovermagazine.com/gnxp/2011/08/the-end-of-environmental-inequality-means-the-rise-of-genetic-inequality
I find Razib’s take a bit scary in it’s implications, so did a related post myself
http://praxtime.tumblr.com/post/35528392746/perfect-meritocracy-replaces-class-with-caste