Remember the “Flynn Effect”? It turns out that even James Flynn now agrees with the new mainstream consensus that genotypic intelligence is declining over time. The Flynn Effect was merely capturing enviromental improvements that temporarily boosted average IQ for a few decades.
Author Richard Lynn is most responsible for noticing this more fundamental decline. So I propose we name this the “Lynn Effect” – the fact that world IQ is now declining over time — and the genetic contribution to intelligence appears to have been declining for over 100 years now.
25 Responses to “The “Lynn Effect” – How the world’s IQ is in decline”
February 2
Rethink EvrythingThis is due to high-IQ women typically having fewer babies than the rest of the female population, right?
Intelligence-enhancement seems even more urgent in this situation
February 2
David OrbanThe Lynn effect could be analogous to how the intelligence of domesticated species is lower than that of its wild equivalents (eg. dogs vs wolves), and as such, an indication of our societies becoming less edgy/violent, with lower expectations from the individual, which can still “make it”, even if not as smart or sharp as it would have been necessary previously.
February 2
Eliezer YudkowskyThank you for contradicting what I thought I already knew.
February 2
David OrbanEliezer, one of my favorite words lately is “metanoia” in the sense of “capability of changing one’s mind”, learned from Jason Pontin…
February 2
Jason PontinIt’s from the Greek, and was used in the New Testament to mean repentance; and was adopted by modern psychology. See: http://en.wikipedia.org/wiki/Metanoia_(theology)
February 2
Matt LasaterAt that rate of decline, we’ll be Idiocracy levels in a millenia.
February 2
David OrbanJason, not wanting to hijack Louie’s thread, but yes, I did look that up, and also was pleased to find that etymophilia also exists, as it ought, even if the IQ of my spell checker is not up to it.
February 2
Sarah ConstantinI want to believe in the Flynn effect so bad.:(
February 2
Chris Staffordokay, not going to lie, thought you were talking about the Matt Flynn effect when I first saw the article as it relates to football teams…which makes me aware that I may be contributing into the “LYNN” Effect
February 2
Scott SiskindPeople thought that Flynn effect was genetic? Who?
As I mentioned in Anti-Reactionary FAQ 5.3.2, any dysgenic effect is small, on the order of 4 IQ points per century (more like 3 once we control for immigration). By comparison, Flynn Effect is large – 4 IQ points per *decade*. If Flynn Effect stopped in 2000, then by 2100 dysgenics should at worst take us back to 1990s level of intelligence.
It’s possible that Flynn Effect and dysgenics are not perfectly symmetrical, but AFAIK that would be very speculative. Flynn Effect probably disproportionately works on lower end of the bell curve, but probably so does dysgenics, so hard to say.
February 2
gwernWhy would dysgenics work disproportionately?
February 2
Rethink EvrythingA couple of cheer-ups for those like me who really really want to believe the Flynn-effect:
1. A hundred years ago, we were sligthly over a billion people. As we are 7.3 billion today, there is still a larger amount high-IQ individuals than ever in history.
2. BGI’s Cognitive genomics institute are advancing and researching ways to find out what high-IQ are, so future kids could soon be selected for by selection before birth. See – http://www.epi.msu.edu/seminars/Hsu.pdf, or https://www.cog-genomics.org/
3. Digital technologies extend our cognitive abilities. Computers, iPhones and the like are basically external exocortexes. I remember that I did not enjoy telepathy, instant-knowledge-access and external photographic memory as a kid, and would like to credit that as a large improvement in our capacities. (Watch out for skin-bag bias – the idea that we are only what is inside our skin. A common bias among transhumanists is that we MUST have the technology INSIDE our bodies to enhance ourselves, when were already enhanced by external devices.)
February 2
Louie HelmWhat’s your reason for thinking dysgenics operates primarily on the bottom of the bell curve?
I had hoped to find that data in the literature, but the data published on the matter (unlike the unsourced commentary, which assumes the Flynn effect or that society must be improving on average in some way) doesn’t seem to bear out any trend in the downward pressure on IQ operating on any specific part of the bell curve.
February 2
Misha GurevichLouis I think you misread,
It’s not the genetic effect that’s on the lowest , but the non genetic Flynn effect.
February 2
TaurusDid anyone actually read Flynn’s paper cited above?
“The authors attribute differential patterns of IQ trends to low-IQ non-European immigration, excepting Finland. America is another exception: large immigration of lower-IQ Hispanics has not yet weakened its Wechsler rate of three IQ points per decade. …It is noteworthy that gains have ended in societies all of which are politically progressive. They may have achieved the full modernity that would weaken the triggers of IQ gains (no more progress to be made in education, cognitively demanding jobs and leisure, family size, visual saturation, and so forth).”
And his response to Meisenberg & Woodley http://bit.ly/1fmyfrk
“This study is of outstanding quality. It is the first to systematically test the hypothesis that developing nations are likely to match the mean IQs of developed nations during the 21st century. Some of the former appear to be entering the “first phase” of modernity (massive gains) that the latter enjoyed last century. The predicted dates for a totally closed gap are: 40 years (about 2050) for PISA (math, science, and reading) and 341 years or perhaps never for TIMSS (math and science).
…I predict that the 21st century will reflect both of these results. Some nations like Brazil, Turkey, and Kenya appear to have begun the road to modernity, while others like Sudan have not, and others like Saudi Arabia and Dominica have made some steps along the way but face daunting obstacles. My prediction is conditional on two things: continued progress uninterrupted by climate disaster and the trend toward nil IQ gains in developing nations becoming general. As to whether the developed versus developing gap will ever close, Lynn has speculated that there is a genetic hierarchy, one that entails that other nations will never rival the mean IQs of East Asians. This posits an evolutionary scenario (Chinese north of the Himalayas during the Ice Ages) that has been falsified (Flynn, 2012, pp. 33–35; Flynn, 2013, pp. 52–54).”
Conclusion:
“I believe that a consensus is emerging that massive IQ gains over time have real-world consequences. Since 1900, Americans have gone from a people with a median of less than six years of schooling, through the mid-century high school revolution, into the tertiary revolution (12% of Americans had some tertiary experience in 1950 as compared to 52% today). Since 1900, Americans have gone from 3% in professional or sub-professional jobs to 35% in jobs that demand cognitive creativity, ranging from highly paid professionals (15%) to teachers, lower management, computer programmers and technicians (20%). It is inconceivable that their habits of mind did not alter, as the new educational and occupational demands and their cognitive abilities interacted with one another.
Second, between 1995 and 2006, American adults made gains on theWAIS whose subtest magnitudes had a correlation with subtest g loadings somewhere between positive 0.540 and 0.621 (adjusted for restriction of range). Duringmuch the same period, American children made gains on the WISC whose subtest magnitudes had a correlation with subtest g loadings somewhere between negative 0.302 and 0.409. So one tallies with g and the other does not. I maintain that correlations with g are largely irrelevant to social significance. Today adults communicate better and have wider general knowledge; children do not and have not.”
February 2
Louie HelmSo Flynn agrees that g goes down… but says it doesn’t matter for society since the non-g component of IQ is going up? All I can say is that he’s welcome to his commentary on the data as much as anyone.
I’m not saying there’s only one way to explain this data. I’m just pointing out that the phenomenon of genotypic intelligence declining over time is a robust consensus in the field by those who study intelligence. And it seems worth mentioning because most people don’t realize that this is a central result in the field that must be explained by any successful theory of what is happening and not a fringe hypothesis that’s challenged in all corners (as most outside of the field believe).
February 2
Louie HelmMisha, I was responding to Scott’s speculation that dysgenics and the Flynn effect both operate on the bottom end of the bell curve.
February 2
Hayim Prometheus DarOK, what? Firstly, isn’t IQ an incredibly imprecise and problematic measure of abilities?; secondly, if the dysgenic thing is an anticorrelation between fertility and IQ scores, exactly what is the hypothesised causation, or are we just ignoring the entire reality of social, cultural, economic interactions between the two?; thirdly, “new eugenics of biotechnology” are you kidding me? If you want evidence that intelligence has decreased, it’s that anyone thinks that’s gonna work out well..!
Eugenics’ paradox: we’re not enlightened enough to administer a such a program, and if we were, we wouldn’t need it.
February 2
Louie HelmAnd I do agree with Scott that the decline is slow. We’re not in any danger of slipping down to a 95 IQ generation anytime soon. And having 10x as many people around to create the same sized scientific community kinda works.
The only problem is that a society that doesn’t notice it’s brightest people are becoming more rare over time may simply push 10x as many people into scientific fields when the population grows 10x (or more, since all the more “basic” jobs are filled quicker).
February 2
Daniel PowellI don’t see the problem with moving the bar on entry into scientific fields down. Sure, the average rate of return per man-hour will go down, but the total returns should go up.
February 2
Michael Caohttp://www.youtube.com/watch?v=BBvIweCIgwk
February 2
Ceri StaggDaniel, that depends if the important stuff is grind-work or moments of inspiration/insight.
February 2
Jason SheltzerSo there’s some sleight-of-hand stuff going on here. You state that Flynn now believes that genotypic intelligence is decreasing, and then move from that to make the claim that “the world IQ is now declining over time.” Note that the two claims are not synonymous, and if you read Flynn’s article, he makes it clear that the vast phenotypic gains in intelligence outweigh any genotypic losses.
February 2
Louie HelmJason, no one disputes there was a Flynn Effect for awhile. Flynn and others also don’t dispute that it has ended. Flynn thinks the gains outweighs the declines and this is certainly true in the past if you count all causes of higher IQ equally (like removing lead from gasoline… which seems like it should count for something to the degree that people actually got smarter for it).
The reason I mentioned this is because many outside the field of intelligence studies think the idea of genotypic intelligence decreasing over time is either completely disproven (by the Flynn effect) or a totally marginal, fringe belief that’s only advanced by cranks. I thought that before looking at the literature. I wanted to point out that this is in fact the strict consensus within the field.
February 2
Scott SiskindLouie, I may be wrong about dysgenics being mostly a bottom-of-bell-curve effect. My reasoning was that, assuming assortative mating, then smart couples have X kids and dumb couples have Y kids, where Y > X. If X is greater than or equal to 2, then there remain the same number of smart people, they’re just increasingly outnumbered by dumb people. If X is less than two, there are fewer smart people, but the decrease is probably not as large as the increase in dumb people, so the dysgenic effect is mostly on the bottom of the bell curve. Assuming assortative mating, dysgenics is consistent with the preservation of a large very smart overclass indefinitely, although as the absolute size grows smaller it does have lower chance of producing the occasional freak genius.