Douglas_Knight comments on Open Thread, September 1-15, 2012 - Less Wrong

6 Post author: OpenThreadGuy 01 September 2012 08:13AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (353)

You are viewing a single comment's thread. Show more comments above.

Comment author: see 01 September 2012 09:55:06PM 2 points [-]

Was reading up on the Flynn effect, and saw the claim it's too fast to reflect evolution. Is that really true? Yes, it's too fast, given the pressures, for what Darwin called natural selection, given the lack of anything coming along and dramatically killing off the less intelligent before they can reproduce. But that's not the only force of evolution; there's also sexual selection.

If it's become easier in the last 150 years for women to have surviving children by high-desirability mates, then we should, in fact, see a proportionate increase in the high-desirability characteristics. And since IQ and socioeconomic status are correlated, and SES is a known high-desirability characteristic, we would expect an increase in IQ accordingly, insofar as IQ is heritable.

And, in fact, there is a change in society that would do that — increasing urbanization. Not only have cities become healthy enough to have non-negative population RNIs for the first time in history, but they've also become the home of the majority of the human species for the first time in history. Studies of infidelity rates show it does, in fact, correlate fairly strongly with urbanization (probably for the logical reasons that increased population density increases opportunities and urban anonymity makes it easier to conceal from a mate).

So, the urbanization of the last 150 years increased successful infidelity. The usual models of sexual selection indicate that successful infidelity by women should result in high SES men having more children. IQ is correlated with high SES. IQ seems to be heritable in large part. And the period where we would expect high SES men to have more kids is matched by an increase in the general population's performance on tests of IQ.

I'm currently operating without good access to scientific journals to see if this has been considered and debunked, or not considered, or considered and put forward. But, at least sitting here just thinking about it without the resources to test it (or even model it effectively mathematically), it seems an increase in the genes that increase IQ as a result of sexual selection could be a plausible explanation of the Flynn Effect.

Comment author: Douglas_Knight 02 September 2012 04:40:07AM 4 points [-]

When people say that it's "too fast," they are making a quantitative claim. The Flynn effect is a standard deviation per generation. Under your scenario of no selection on women, this would require that the bottom half of the bell curve to have no biological children. 50% cuckoldry, perfectly correlated with IQ? Even men who think they've been cuckolded don't have that high a rate.