Hacker News Reader

Comment by quantdev | Hacker News Reader

Or maybe it's because, somewhat depressingly, IQ is highly hereditary, as research has shown for decades and new gene research is currently confirming, and people with lower IQs tend to have more children.

View on HN

Why would this effect suddenly present itself in 2004? Why hasn't it caused average IQ to decline for all of human history?

"A recurrent fear during the past century is that the mean IQ level of populations will decline because persons with lower IQ scores have above-average fertility. Most microlevel data demonstrate such fertility differentials, but population IQ levels have risen rather than fallen. In this article, a simple two-sex model shows that negative fertility differentials are consistent with falling, rising, or constant IQ distributions. Under a wide variety of conditions, a constant pattern of fertility differentials will produce an unchanging, equilibrium distribution of IQ scores in the population. What matters for IQ trends is how the IQ distribution in one generation relates to the equilibrium distribution implied by that generation's fertility differentials. Intuition fails in this important area because it does not account for the macro structure within which micro results must be interpreted."

- http://www.journals.uchicago.edu/doi/10.1086/230135

Also note that there is "a strong inverse correlation between early life intelligence and mortality".

- https://en.wikipedia.org/wiki/Cognitive_epidemiology#Overall...

Because birth control is a recent invention. Before that, it took smarts to survive in a crapsack world. We have this stereotype of peasants and medieval people as being stupid in general but all the evidence shows otherwise. You had to be smart and capable of making robust plans to eke out a living on hardscrabble.

Also, until recently mortality rates used to be a lot higher.

The most likely explanation is the prior to perhaps 200 years ago the differentials were different. (Also having lots of children may not result in a lot of descendents, so is an insufficient measure). This would be due to the fact that in an era before universal rights and universal economic security there were reproductive consequences to intelligence (or lack of).

So there is no evolutionary pressure. Evolution is a (brutal) system of unfairness that represents everything that modern progressive values work against. This is something that will need reconciling (or many technologists will hope for genetic engineering I guess) or in the long term will not be sustainable.

Consider your children are on average genetically inferior to you. This is mathematically true. A society which uniformly has 2 children per person + 100% survival will regress. In fact it may take something closer to mediaval or even prehistoric reproductive outcomes to have positive evolutionary pressure. This could be many people having 4-5 children, with many other people who are 'unfit' not having children at all (either remaining childless or dying early).

The subsequent improvements in education,environment and nutrition have likely provided a dividend that masked a fall in underlying genetic intelligence. Naturally as those gains have been fully realised the underlying trend will reassert itself.

> there is no evolutionary pressure

Please take another look at my second link: https://en.wikipedia.org/wiki/Cognitive_epidemiology#Overall...

If there's some analysis of evolutionary pressure over time I'd be interested to see it, although I'm not sure it can really be quantified like that.

Whatever the truth, that section is poorly and contradictory.

'A strong link between cognitive ability and the risk of death.'

'A 15-point increase in intelligence was associated with a decreased risk of mortality of 32%.'

Wikipedia suffers a bit from not trying to be too neutral!!

Local cases of intelligence not helping with mortality/reproduction in todays society are not contradictory. This is the (dysgenic) thesis that intelligence does not help. Although increased mortality is perhaps a little harder to explain, but again hardly irreconcilable - I could speculate many plausible explanations.

To address your point, I feel it is intuitively true that we do not have positive evolutionary pressure. If indeed IQs are now falling despite the flynn effect then for intelligence at least here is the evidence for insufficient evolutionary pressure.

I agree that it feels intuitively true, but intuition is often wrong, as that wikipedia section shows. Pay more attention to the sources than the wording of the article. All of the sources demonstrate an inverse correlation between intelligence and mortality, meaning that intelligence does help.

People confuse the normative and the descriptive. Everything you say is basicly correct. But it's horrible so it can't possibly be true. But why should evolution not be horrible?

Evolution does not care about our ideology. If we don't account for it and nerf its edge, though forms of selection that don't cause suffering to conscious creatures, like embryo selection or genetic engineering, then we will have to wait for Malthus to provide sufficient selection pressure to ensue our survival or extinction. This fingers-in-ears approach to biological reality would be concerning if countries like Korea and China weren't poised to exploit embryo selection and genetic modification for increased health and IQ, thus likely forcing more reactionary countries like us in the west to adopt them, too.

"if countries like Korea and China weren't poised to exploit embryo selection and genetic modification for increased health and IQ" Increasing IQ could only make sense for me if the society would know how to integrate people with High IQ. But for now we are far from it with 50% of high IQ people failing their education cursus and having issues interacting socially

We will soon have the selection power to select for many traits simultaneously. Also, much of the problem super high IQ people have is the result of being surrounded by people much less capable than themselves. If we seize the high-IQ alleles and distribute them to the proletariat this would not be an issue.

being parent of high-IQ children is quite tiresome compare to average IQ children. So you would need a big bang in term of school reorganization and support to parents.

Actually, reading James Flynn's book What Is Intelligence?, I got the impression that environment and culture plays a huge factor in IQ. Genetics is a factor, obviously, too. But accepting IQ as a mostly function of race or sex, which some people are always itching to do whenever this is brought up, is reductivist.

In his book, Flynn attributes part of the mysterious increase in IQ during the 20th century to the spread of general education and inculcation of scientific and logical modes of reasoning more broadly.

He recounts a story from Soviet anthropologists who travelled through Russia studying peasants at the beginning of the 20th century. An anthropologist presented a villager with a simple syllogism:

All swans are white. If I told you there was a swan in the next village over and sent you over to report its color for me. What color do you expect it to be?

Peasant's response: I don't know. I'd have to go check.

Which makes sense when you exist in a world in which you're operating mostly on the basis of personal relationships and trust is based on a person's word and status and not on abstract and institutionalized forms of knowledge. You'd be an idiot to trust the word of some stranger you just met.

But this guy probably did not do so well on the IQ test he was given.

In fact, check out this statement from Flynn's Wikipedia page:

Women, he argued, caught up to men in these nations as a result of exposure to modernity by entering the professions and being allowed greater educational access.

https://en.wikipedia.org/wiki/James_Flynn_(academic)

>James Flynn, namesake of the secular increase, finally agrees that these IQ gains are not g factor gains [1], but believes that they still represent economically relevant skill gains. This is doubtful for two reasons: first, because measures of aca- demic ability over time do not reveal any concomitant Flynn-like gains [2, p. 8; 3]. Second, because we know from large-scale military re- search that almost all the predictive power of IQ tests in training contexts comes through g alone [4, p. 129].

[1] Flynn, JR. The Flynn Effect: Rethinking intelligence and what affects it

[2] Flynn JR. Searching for justice: the discovery of IQ gains over time. Am Psychol 1999;54:5–20.

[3] Flynn JR, Weiss LG. American IQ Gains From 1932 to 2002: The WISC Subtests and Educational Progress. Int J Test 2007;7:209–24.

[4] Ree MJ, Earles JA. The ubiquitous productiveness of g. In: Rumsey MG, Walker CB, Harris JH, editors. Personnel selection and classification. Hillsdale, NJ: Lawrence Erl- baum; 1994. p. 127–36.

And connecting the two dots too loudly makes you into a monster.

Just lazy.

Well drawing strict conclusions in a field that's still highly debated by researchers does, yes.

psychometricians largely aren’t debating the existence of intelligence, its variability, or the fact that it is heritable.

Actually...

http://www.apa.org/research/action/intelligence-testing.aspx

http://www.independent.co.uk/news/science/iq-tests-are-funda...

https://blogs.scientificamerican.com/beautiful-minds/iq-test...

If intelligence were simple we'd know how to test for it. It isn't, and we don't. Not reliably, and certainly not in ways that are absolutely independent of culture and background.

The UK used to have a concept called "mental agility", which seems to be closer to the abilities measured by IQ than the much more complex quality psychologists call general intelligence.

You can have exceptional mental agility - the ability to remember and manipulate symbols and parse symbolic relationships quickly - and still be an utter idiot when it comes to dealing with some or all of reality.

A high IQ will not protect you from ideological beliefs, especially if you grow up with no exposure to other viewpoints, from a lack of mental openness (not to be confused with a lack of curiosity), from "moral" perspectives that are self-reinforcing prejudices, from poor socialisation, and so on.

The book outliers says that research shows roughly half of IQ dictated by genetics. Do you have research that contradicts this?

50% would be "highly hereditary".

It's interesting to visualize what % heritable actually means. Here is a graph from a recent study trying to predict height from human genes. They found a correlation of 0.64

https://2.bp.blogspot.com/-4qSTryry0so/WcAhygytvTI/AAAAAAAAS...

That means, in practice, your height outcome tends strongly toward a fairly narrow range around the predicted mean. I found it illuminating to visualize the pattern.

(The full article is http://infoproc.blogspot.ca/2017/09/accurate-genomic-predict...)

50% is a lot...

So you're saying the movie Idiocracy is where we will end up? Guess I had better start investing in Gatorade.

Or maybe it's not so simple, and IQ can also be changed by environment. It's not a crazy idea that people with more money have better access to education.

IQ being highly hereditary doesn't mean that it isn't also influenced by environment.

And people with more money have less kids. So more intelligent people have less kids, more intelligent people make more money, can spend more to develop those fewer kids' educations, so you have an ever decreasing amount of 'smart' kids and thus the average IQ goes down.

Reply on news.ycombinator.com