I’m just curious how accurate that is. There’s likely dozens of cases that contradict that - how many polynesian islands were settled by very small numbers of initial immigrants? Or other isolated locations? Newfoundland, Tierra del Fuego, the Andamans, Madgascar etc. - I suspect that the settlement of Easter Island was from a single canoeload or two of explorers, although there is no record of the details; but there is certainly no record that there was any back and forth after the original settlement.
you would think that especially with Easter Island, the initial pool that the settlers came from was at the extreme end of the polynesian expansion and so a very limitied gene selection. Howver there’s no indication that they suffered from problems that prevented them from growing to possibly 13,000 people and a thriving (ultimately self-destructive) civilization.
Schizophrenia/schizotypy has also been conceptualized along these lines. Except that instead of analytical thinking, creativity is the trait that is believed to be selected for.
So in the Utopia, no math geeks or artsy/fartsy people should be allowed.
The ‘controlled’ environment of the studies is not the same as the environment in which this colony would exist, therefore any heritability seen in those controlled experiments cannot be extrapolated to this theoretical colony;
Really, that’s the only point that matters.
Heritability is not determinism. Heritability is not a genetic guarantee. Heritability is a correlation, nothing more. Further, given that the precise genetics of the actual underlying genome aren’t even fully known, it is simply impossible to state anything about the relative intelligence of this proposed sequestered population over time. Not even 100% heritability is a guarantee of anything.
Yes, that’s true. Epigenetics can have a very great effect, as well as just general environmental influences.
However, there’s nothing in any scientific study that would lead one to believe there’s any real possibility that environment could pull the average IQ of the hypothetical group back to 100 within, say, 20 generations, without there being serious health consequences. IOW, it’s pretty difficult to lower someone’s IQ through environmental and epigenetic influences without also seriously harming their physical health. We need adequate nutrition and social and intellectual stimulation. Take those away, and, while you might be able to lower the average IQ, you’re also going to mess up a lot of other stuff too.
There was a plethora of truly bad studies of the heritablilty of intelligence, largely from the “tabula rasa” crowd in the mid 20th century.
More recent carefully controlled studies included the following kinds of studies to discern genetic vs. environmentally heritability:
twins reared apart
fraternal twins
natural siblings vs. adopted siblings
According to Steven Pinker’s “The Blank Slate”, genetics accounted for 50% of the variation. Parental/family enviroment accounted for up to 10%. The other 40-50% was unaccountable, and probably due to chance factors in fetal development.
These studies also tended to show that while parents can increase the IQ scores of their kids, these differences don’t tend to persist; they fade over a decade or two.
If we agree that intelligence is 50% genetic, then the average intelligence of the next generation would be lower, but would stabilize for successive generations.
The book I cited above is dated 2002. It is consistent with similar facts cited in Paul Erlich’s “Human Natures”. Research since then might have yielded different results, but would need to be carefully scrutinized for methodology. It’s terribly easy to do misleading research in heritability.
Nope. On this MB, it is never acceptable to do that, except maybe in the Pit. In this forum, particularly, you should expect to be able to back up any assertion you make with a cite.
The messy part is recognizing that in a lot of studies IQ tests are often (and perhaps controversially?) used as a measure of intelligence. You could argue that a correlation is revealed in how well IQ tests are taken and then argue whether that has much of anything to do with a broader idea of intelligence, but that would be a different thread.
TriPolar seems to be confusing population level effects with individual level effects. There is nothing mutually exclusive about 1)the mean of a trait (height, intelligence, or whatever) being higher in the descendants of a cherry picked founder group and 2)some members of that F1 or subsequent generation being below the old mean. You don’t have to know the genetic cause of the trait in question for this to be so.
Suppose you selected all professional basketball players from the NBA (average height 6’7") and WNBA (average height 6’0") and isolated them to start a new population. Take as a given that the average height of these individuals is taller than the average height of the origin population as a whole. Suppose for the sake of example that the average height of all adults in the original source population is 5’10" for males and 5’5" for females.
You then come back in 500 years to measure the height of the descendants. The average height of all adults is likely to form a bell curve distribution, with a mean somewhat higher than 5’10" and 5’5" male and female heights of the source population. And that can be true even though there may be some significant part of that population that is shorter than those heights. This expectation is based upon the observed heritability of height being about 0.6 to 0.8 and thus substantially genetically influenced.
Genetic heritability of height can be inferred from twin studies with no need to identify specific responsible genes.
The answer to this question is that it depends on which selection pressures drive successful reproduction. In general, of course, if you want to breed smarter people you have to keep breeding every succeeding generation selected out for intelligence. That way, you get the smartest genes.
However, the original group contains a vast selection of genes, so there is going to be a natural regression toward a less-smart average than the parent pool for the first generation. After that, and for subsequent generations, it depends on what the selection pressure is. At least you have some high-performing genes in the pool, sure, but what if the smart people are best able to take advantage of everyone else and become rich and lazy, and decide kids are too much work, and the dumber people are poorer and harder working, and decide large families are a greater blessing than poverty is a curse? You’ll regress toward a much lower average than if there is selective breeding with high reproduction rates for those with high-intelligence genes.
Evolution is about descent with modification, and the descent part is pretty darn important. It is definitely not tied just to intelligence. One can imagine a world where a species gets so smart they manage to wipe themselves out (nukes; AGW;modified viruses…) while the cockroaches still hang out just fine after several hundred million years.
So let me see if I understand what seems to actually be the consensus here:
A wide variety of cognitive and physical abilities have some degree of genetic basis, some basis in cultural factors from family factors to broader community and societal ones, some other physical environment basis, and some who knows random variation. How much of each for any particular trait or ability is up for some debate.
All agreed so far?
To the degree that any factor has a genetic contributor the founder effect would mean that the population would be score “higher” in that trait than the originator population. Everything else being equal to the degree that the trait is not genetically based there would be some degree of regression towards (but not completely to) the mean.
It of course is improbable that everything else would be equal. A population that consisted of the sigma outliers of abilities and accomplishments would likely be providing an environment more likely to facilitate each child reaching their potential than the originator society did. These are not just kids whose parents are smart and who have a better chance at having the genes associated with intelligence; they are being raised in a family with smart parents and surrounded by people smarter than average, possibly including those better skilled at teaching as well. They are being exposed to the work of the best artists, coached by the best athletes … “Would the total population remain smarter, stronger, etc than the average of the people they left behind?” Well sure.
Does that seem reasonable to all?
Although I still wonder about adverse effects … here, for example, is an article going into detail about the association monstro mentioned.
The parallel with Bar-Cohen’s autism hypothesis are indeed strong. A little is good; too big a load is deleterious. (Similarly with sickle cell, one allele protects from malaria, two causes a bad disease). Concentrating those genes may have unforeseen adverse consequences.
Off-topic, but did they run out of their 100,000 condoms (or did the organizers forget to order magnums), or what?
ETA: your article says they did, but I think the ESPN article might have gotten it more correct…
I understand what you are saying. And I think anybody would assume that the most likely outcome of selection for traits is to see those traits remain or enhanced. But intelligence may be more complex than height as a trait, and there are a variety of reasons to consider in the OPs scenario that there may be no enhancement of intelligence over time. I think it requires more specific indicators that there would be significant increase in intelligence in this group in the end. I certainly didn’t say anything about it being impossible.
I don’t even know what al27052 is arguing about. My responses to him were directed at his statements that never seemed to relate to my original post.
Perhaps I was not clear there and someone thinks I meant that intelligence would regress, but I only meant there might be no increase in intelligence on average as an effect. That would apply to my use of individual examples as well, to explain why intelligence might not increase over time for individuals, not to say that it would decrease across the population over time.
But I also think that selecting the most measurably intelligent people on earth would tend to pick up outliers who wouldn’t necessarily pass the genes for such super intelligence to their children. 100,000 people is a tiny percentage of the world population who’s intelligence level may be so high and have occurred as a result of a variety of factors that may not reproduce easily, and it’s entirely possible there would be an average decrease in the first generation. This is the difference between picking from NBA players, and the tallest people on earth.
There’s also the issue of “what kind of ‘smarter’”? A rocket scientist guy who couldn’t spot a mating opportunity if a girl gave him her phone number written on a no-tell motel business card? A gal who could build a rocket out of a box of scraps in a cave and then break it down and mail it to somebody in Nigeria who is offering the former dictator’s bank account in trade?
That is a problem in this discussion. We have to assume the measure is consistent, and there should be some strong evidence that the trait can be inherited.
This is a good point. We know a random pattern of reproduction would produce a partial regression to the mean.
The question is, what would evolution, given a very similar environment as the rest of humanity, do to the bell curve for this special population? Move it even closer to the mean, farther from the mean, or have little noticeable effect?
How long are you talking about? And is this population growing? There should be some idea of what happens to general characteristics over time.
Also, you should be happy to know that after doing a little math which I should have done before, I see that the top 100,000 people would be 1 in 70,000 people, which wouldn’t be a bunch of extreme outliers. So you’re right that this population should expect to follow a predictable pattern based on the general heritability of the trait. But do you know of something that would indicate the level of intelligence would increase in a group already selected for high intelligence?
How long? Maybe 60-70 generations. For the sake of the example, we’ll say no, the population isn’t growing or shrinking (although that is a good question, and I agree that it might have multiple effects on different levels).
No, I doubt they would become even more intelligent. I am open to the possibility that there might be a “tipping point”, though, above which the average person is so intelligent that…well, I don’t know, but I refuse to discount the possibility, anyway, at this point. I might later, though.
And I believe the original number was 1000, not 100,000.
Well if it was 1000 I think you be selecting a lot of extreme outliers. But the OP said “a few hundred thousand people from the top 1% of humanity” so that’s an even better selected group than I thought the second time around.