= 27d567bc2d48d9f40e9ccc20ea370b15)
Lumifer comments on Open Thread - Aug 24 - Aug 30 - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (318)
That's almost two standard deviations and looks iffy to me. Yes, I followed the link, the main study resulting in this number is behind the paywall, but I suspect that the sample wasn't very representative.
Basically, for a sub-population with a recessive trait that leads to mental retardation the outcomes are going to be massively different from the outcomes for a sub-population without such a recessive trait.
Available the usual places: https://www.dropbox.com/s/8gv0el0anfyymed/1971-seemanova.pdf / http://moscow.sci-hub.bz/092ee3d082e9cffd04b7064c36ba808a/10.1159%40000152391.pdf (Personally, I'm impressed she managed to get that big a sample in just Czechoslovakia. Creepy. Also, note that's only one of the studies being meta-analyzed.)
It's very clear that inbreeding is really bad: to give some examples, it drives species extinct within generations, its effects are long-term and underestimated, inbreeding can result in 20+ IQ points loss in other populations such as in India and of course we all know about the Habsburgs (which was so extreme that "Charles II was moderately more inbred than the average among the offspring from brother-sister matings").
What makes you think there are any human populations none of whose recessive and mutation loads affect cognition?
Her sample covers 37 years (1933-1970) and, looking at the sources, she utilized official reports including maternity homes, district courts, etc. so the sample is all Czechoslovakia had officially.
I don't think this. Intelligence is strongly polygenic, as far as I know, so for everyone it's a mix of something good and something bad. But that gene mix is uneven in populations, so some people (low-IQ) get more bad and less good genes, while some people (high-IQ) get more good and less bad.
The thing is, I would expect incest to strongly correlate with very low IQ (your basic drives are still there, but social norms are... less binding). Besides, it's easier for smarter people to not be caught. So if you select a population which engages in incest (and is detected), you are co-selecting for low IQ and for a larger proportion of bad-for-IQ genes. And the larger that proportion, the worse are the chances (growing superlinearly, too) for the child to have normal IQ.
Do note that in the study sample only 4 females and 2 males among parents attended secondary school, the rest didn't have any education beyond elementary school (out of 141 mothers and 138 fathers).
P.S. Thanks for the link to the study.
Poor countries are like that. The people in the Indian studies and elsewhere won't be too highly educated either.
Shouldn't affect within-population comparisons... Although since prevalence of cousin-marriage differs drastically from country to country, the inbreeding effect could be driving a nontrivial amount of between-population differences in intelligence. (And of course, it's not like intelligence is unrelated to national wealth either.)
This is Europe, though, and Communist governments tend to be big on education.
Within which population? The control group involves one parent "from the outside", so regression to the mean kicks in and the chance of the recessives finding a pair falls dramatically.
I am not arguing that incest has no significant consequences. I am arguing that if you take children of incestuous unions where both parents have reasonable IQ (say, >85), the mean IQ of children would NOT drop by 28 points.
It is one of the poorest parts of Europe, and Communist governments tended to be big on a lot of things they couldn't deliver.
If you're comparing within an Indian population, then the much higher rates of inbreeding aren't the confound; all you have is the remaining selection effect, and that's must be small because anything else would drastically contradict the animal and other breeding experiments, and the estimates from genomic methods.
It would probably drop by more like 25 points, looking at the weighted averages. (For the surviving children, that is.)
Communist governments delivered on that one.
Take a look here, specifically pages 21 and 23. The secondary education in Eastern Europe was more prevalent than in Mediterranean countries and Great Britain + Ireland (but less than in Nordic countries and Central Europe). And Czechoslovakia was one of the better Eastern European countries.
I went and looked at the Jammu & Kashmir study and it is more convincing than the Czechoslovak study. Hm. It seems my scepticism about the 20+ IQ points drop was unfounded, <screech of rusty metal> changing my mind... </screech of rusty metal> :-)
But why did my intuition didn't like the large magnitude of IQ drop? I think because it implies that intelligence is very fragile and very easy to genetically screw up. But if the IQ drop is valid, then intelligence is very fragile. Hmm...