Boiling brains down to one number and ranking them, and saying it comes from genes, nutty. When scientists have no idea why a nematode with 302 neurons turns left or right. You can tell it's religious or political or whatever by how people defend it - no one gets that agitated by how many light years away some star is.
There's more to intelligence than just classroom performance, but there is definitely something being measured by IQ tests and it's related to intelligence. Just because it's not perfect doesn't mean we need to discount it completely.
IQ does measure something, though, and it tracks pretty consistently with that more ephemeral thing that people refer to as "intelligence": if you, personally, found a dozen people you considered "smart" and another dozen you considered "stupid", you'd almost definitely find that the smart ones had high IQs and the dumb ones had low IQs. You might argue about the fine-grained distinction between the smart ones, but you wouldn't find somebody who you considered smart who somehow still had a 50 IQ.
The PhD in Chemistry will have the superior IQ but ask both to survive in the jungle and the tribal wins hands down.
It’s a very facetious assumption that solving a few puzzle questions captures all the facets of a human beings creativity. This is like saying that the best leet coders are the best software engineers.
“ Boiling brains down to one number and ranking them, and saying it comes from genes, nutty.”
Yeah, except that the number correlates with many things we care about in the real world (educational attainment even at the highest levels, crime), even after controlling for many other things we care about in the real world (conscientiousness, SEC).
Not sure the blogpost supports Gould being innumerate or not understanding PCA.. also, an aside, I doubt the author Hsu would agree with your take here, considering his hero Feynman had an underwhelming IQ relative to his success.
BTW what do you mean by IQ measurement is underpinned by PCA (principal component analysis? or the requisite math?)?
IQ is certainly not a perfect metric, but for all the criticisms of it, I haven't seen any reason that psychometrics should not aim to improve.
Yet people don't seem to discuss IQ with the intent to improve the metric. Accepting and rejecting it are not the only two options, we could also discuss ways to do better.
The reality is that its one of the few psychometrics with strong correlations to a number of long term outcomes like income, wellbeing, crime, fertility, etc., and which is responsive to changes in the environment and demographics of a nation.
That said there is a point to be made that IQ is clearly not 1:1 with everything we think of as 'intelligence'. Trivially, most significant accomplishments take place over months or years as opposed to the minutes or hours of an IQ test. However, using it as a useful tool isn't as reductive as dismissing it entirely simply because it isn't a completely perfect or comprehensive metric.
It would be a vast simplification, it would be combining slow and fast twitch muscle fibers, it would be combining heart and legs and torso muscles... you would want to be far more precise when dealing with an individual and evaluating how to benefit or correct their musculature.
But still, you could probably take any group of ten people and quickly arrange them in a line from lowest to highest 's' score. You would find a huge correlation between 's' and various professions.
Equating it with astrology is silly, a rhetorical flourish that nobody sensible could believe.
At the time, Gould's comments about intelligence were widely discussed and accepted in the popular press (NYT, NY Review of Books, etc.,) but subject to a lot of criticism in the science journals which did not get much coverage.
Let's take visuo-spatial ability. Men do better in testing than women generally. So for arguments sake let's say they are correctly measuring a deficiency.
So now the question is, is this for cultural reasons or genetic reasons? Laszlo Polgar was a psychologist who raised his daughter to train on visuo-spatial thinking, mostly with chess. She became the youngest grandmaster ever, the youngest person to break into the FIDE 100. She became the 8th best chess player in the world. It's anecdotal, but if she is capable of this, presumably other women are genetically capable of this level of visuo-spatial thinking, so the tested deficiencies might point to culture over genetics.
As Gould said, it's not just IQ but the blanket if assumptions around it - it is genetic not cultural, it is not remedial etc. As I said, people don't get this worked up over light year measurements.
Here's a monologue from someone with an IQ of 75. Do you expect the problems he describes in his life to be more or less common among people with his IQ than in the general population?
Basically, nobles that failed out of society for whatever reason (drunkards, second borns, etc.) would intermix with the lower classes. This would propagate out "superior" genes that may have aided increases in intelligence over time.
This is quite controversial but it is thought provoking.
It's widely speculated that the demise of the Hapsburg empire was due to centuries of inbreeding eventually producing rather... simple monarchs.
You can argue that today's somewhat-meritocracy is going to sort people by social class in a way that roughly overlaps with IQ if you zoom out very far. I don't see why this would be true at all in an autocratic caste system society.
This sounds like a made up just-so theory that belongs on a neoreactionary Reddit group.
I'm not sure if this is true - I seem to recall studies that can clearly see a genetic impact of just a single particular noble (Gengis Khan), so it seems plausible that the genetic impact of nobles would be oversized as high-status males can have many children from the society in which they have violent power, and their children (even if bastards) have less issues with childhood malnutrition and early death.
Imagine you start a society where nobles have the same "gene quality" as everyone else. Some nobles are pretty smart, others not so much. As nobles would have more children surviving into adulthood (due to better nutrition and living standards) compared to the amount of land/titles they could distribute, some of the children of the nobility would be cast aside. The assumption, and I agree it's an assumption (although a rational one in my opinion), is that the smarter children would tend to inherit, and the less capable children would be dispossessed.
Over time, this constant churn would lead to a concentration of "good genes" among the aristocracy, which would trickle down into the lower classes when every generation some of the nobility lost their position.
I believe that this "trickling down" phenomenon has been proven to have occurred in England during the second millennia by Gregory Clark using census data, although I'll admit I've only heard of his work, not read it.
I am not a geneticist or a historian or even more than barely literate
Since the industrial revolution infant death has thankfully dropped but it seems likely this improvement has had the side-effect of increasing the number of deleterious mutations and therefore also of decreasing the average IQ in Western populations.
Anecdotally, my wife is applying for surgical medical residencies, a field which generally attracts intelligent, ambitious individuals. The whole system seems downright hostile to having children. It obviously wasn't designed with women in mind, and the entire system has been intentionally ossified to prevent it from responding to the desires of residents, for the benefit of attending physicians. I imagine there are a large number of similar industries. This has definitely impacted our plans for children, and it wouldn't be an issue if she were considering a less competitive specialty. In fact, it seems that the very selectivity which ensures that the program is filled with intelligent people is used as justification for intense hours and lack of maternal support.
In my opinion this is the sort of externality that would benefit from government intervention in some way. However, I don't think financial support would be very good at targeting such cases. I think the reason that tech companies have done a better job of providing child care support is that the field doesn't have the same sorts of institutional barriers to responding to employee demands that medicine does - specifically demand by employees in their prime child-rearing ages.
Selection against variants in the genome associated with educational attainment: https://www.pnas.org/content/114/5/E727
Would the results of this study be any different?
I'm not sure how the government could intervene in these situations except to make these fields less competitive such that employees have more bargaining power.
Agree, from a society's point of view (if you believe that intelligence is positive and at least partly hereditary, which I guess is pretty indisputable), one would want to provide incentives to correct this.
Tax reductions per child (scales with income) would be one approach.
Education certainly does seem to decrease family size, particularly rates of female education.
That said I totally agree with your contention that careers need to be more accommodating of human behavior. The cynic in me has little hope anything of this nature will come from capitalist entities or governments. That said the optimist knows these societal constructs are composed of human beings, and are therefore capable of learning and improving.
[0] https://freakonomics.com/2005/05/abortion-and-crime-who-shou...
[1] https://freakonomics.com/podcast/abortion-and-crime-revisite...
According to many even ITT, intelligence is some transcendental category which can only be divined by Nurture and Education.
Let’s say I want to measure the potential of someone to be a software engineer. I ask them to solve algorithm questions under time pressure. Those who do them faster and better have an LQ. The leet code quotient.
I have reams of data that show that the higher your leet code quotient the more you earn in salary , the more successful the companies you work in tend to be.
I’m sure many on HN would rail against “LQ” being used to drive policy. Like saying we need to make sure we hire and train more high LQ engineers. Yet this is exactly the tone on discussions about IQ.
The more we believe in hiring based on Leet code the better it makes it look. The same is happening with our education system. What if we measured peoples eye sight and then discriminated against people with low VQ ? You don’t need to do that you can invent glasses so everyone has the same VQ.
All of IQ testing is based on a comically simple assumption. Your score on an IQ test is G + S + N. Where G is the general factor , S is the subject specific factor and N is noise.
Who even said there are only 3 factors ? Modern models add a few more but intelligence may not even be a scalar or a low dimensional vector. It might be a vector with thousands of quantities.
And to those who think that IQ is “real” ponder the following - I give an IQ test to Albert Einstein and a Bedouin from the Sahara desert. Afterwards I drop both of them alone in the desert to survive. Who is more likely to live ?
People need to stop putting faith in crude statistical models that have very weak causal basis. I’m not denying that some people are better at some things than others. But assuming that solving a few puzzles and doing crude statistics allows you to predict that is comical.
IQ is a very crude general prediction. It’s completely possible for a 100 IQ to beat a 140 IQ on many different tasks. Like seducing women or playing football or rapping. We have a very narrow definition of intelligence.
We are barely able to manage our own economies without destroying the entire ecosystem, every country is a mess but we believe that we are able to shape evolution itself towards positive goals when such a process requires tens of thousands of years to even work. The last time this was tried we got one of the worst things in human history. Some people thought they were the “master race” only to get steamrolled by the “inferior” slavs.
The reality is that psychology has been used, is being used, as a vehicle for racism rather than legitimate science. In lieu of real science, nothing at all is better.
the good news is that China is doubling down on research into the genetic basis of intelligence. The West will either adjust their worldview or be left behind
> Over several generations, there is no doubt that dysgenic selection lowered the quality of genes for intelligence and increased migration had an adverse effect.
> Looming over all is their message that the pool of those who reach the top level of cognitive performance is being decimated: fewer and fewer people attain the formal level at which they can think in terms of abstractions and develop their capacity for deductive logic and systematic planning.
So, as often as we hear the joke, is Idiocracy actually happening?
On top of this the ratios change as the environment changes. Studies after WWII found a positive correlation between reproduction and intelligence though that seems to have swapped more recently.
[0] https://www.worldviewweekend.com/worldviewpedia/transcript/j...
[1] https://steelonsteel.com/quotes-from-progressive-elites/
[2] https://nwvstore.com/product/weapons-of-mass-instruction/
---
Some quotes for the curious:
From [0]:
John Dewey, Karl Marx, Aldous Huxley, B. F. Skinner, and Benjamin Bloom were interested in a student’s academic achievement only if it would in some way benefit the State. Before a student's cognitive knowledge could be used to its full potential, however, the student's attitudes, values, feelings, and beliefs must conform to that of the State. In his book, My Pedagogic Creed, John Dewey explains:
"I believe the true center of correlation on the school subjects is not science, nor literature, nor history, nor geography, but the child's social activities. . . . I believe that the school is primarily a social institution. . . . The teacher's business is simply to determine, on the basis of the larger experience and riper wisdom, how the discipline of life shall come to the child. . . . All these questions of the grading of the child and his promotion should be determined by reference to the same standard. Examinations are of use only so far as they test the child's fitness for social life."
Education is thus a most powerful ally of humanism, and every American public school is a school of humanism. What can the theistic Sunday schools, meeting for an hour once a week, teaching only a fraction of the children, do to stem the tide of a five-day program of humanistic teaching?
-Charles.F. Potter: (1930)
In the New Order, Socialism will triumph by first capturing the culture via infiltration of schools, universities, churches and the media by transforming the consciousness of society…
-Antonio Gramsci (1915)
John Taylor Gatto’s Weapons of Mass Instruction focuses on mechanisms of familiar schooling which cripple imagination, discourage critical thinking, and create a false view of learning as a by-product of rote-memorization drills. Gatto’s earlier book, Dumbing Us Down, put the now-famous expression of the title in common use worldwide. Weapons of Mass Instruction promises to add another chilling metaphor to the brief against schooling.And I thought they were nuts the whole time. Until the left started to try to get rid of advanced math classes.
Also, according to the abstract, the US actually sustained positive gains longer than other OECD nations, except South Korea.
The way in which things are trending, an Idiocracy is unlikely because of an AGI singularity.
That even if we might all descend into idiots, we won't be running the world much before that happens.
To quote Douglas Adams - "There is another theory which states that this has already happened."
As discussed in this forum many times, there has been no obvious progress towards AGI.
The way in which things are trending, an AGI singularity is unlikely because of the IQ decline over time.
My recommendation would be to post some quoted passages from the text to add some momentum to a productive discussion.
I think one of the interesting passages is
> From 1990 to 2012: the “working class” share of jobs has steadily de- clined from 31.4% to 20.5%; the “creative class” share has been pretty stable, rising from 29.3 to 32.0; the “service class” has risen sharply from 39.3 to 48.5. He calls the latter “low-skill jobs”, seeing increases in occupations like nursing assistants, personal care aids (whether at home or in nursing homes), retail sales, and food prep workers (McDonalds). Note that the ratio between low skill and creative has risen in favor of the former: 1.35 to one in 1990; and 1.57 to one in 2012.
I'd really be interested in changes in the resulting economies for where those working class jobs ended up.
Reading a paper is hard, having a discussion about a paper is even harder. And especially when it has aspects of class and intelligence. Everyone is an expert in their own opinions on those two subjects.
How much confidence should I have in the correctness of the author's analyses, given that the authors can't even make it through the first column of the introduction of a camera-ready version without glaring grammatical mistakes?
I admit, "messy writing portends poor execution" a very general heuristic that definitely isn't causal and might be wrong more often than it's correct. But the authors want me to take "g" seriously, so shrugs.
(in any case, if editing were completed on this article, I'd blame the editors and not the authors for leaving these mistakes in).
While I respect good writing, bad writing doesn't reflect on the quality of the methodology or analyses in my opinion — as long as it doesn't get in the way, which I don't think it did in this case. I know many smart, careful people who can't string a complete sentence together to save their lives.
But the author of the quoted article doesn't bother to share their methodology or data, so who the hell really knows shrugs.
Par for the course in sociology, really :)
> Correlation is 80% between test and retest, meaning you being you explains less than 64% of your test results.
As if that's bad? 80% is a very good reliability.
> If you renamed IQ, from "Intelligent Quotient" to FQ "Functionary Quotient" or SQ "Salaryperson Quotient", then some of the stuff will be true. It measures best the ability to be a good slave confined to linear tasks. "IQ" is good for @davidgraeber's "BS jobs".
I get it, you have negative valence towards IQ. Still, even if IQ measures "the ability to be a good slave" (it doesn't, but let's assume it does), that's an outcome we care about, proving IQ's validity. It dismisses IQ's correlation with academic success as "just another test", but no, it doesn't work that way. IQ also correlates with research success and research is not test.
We should define research success and test. Research is definitely a fitness function that is intertwined with lots of dependent and independent variables no?
Most of the Taleb argument is that you can't compress a high dimensional space down to one dimension is it not?
In your link he even states IQ is a weak measure for some tasks. That the trend is reversing - people are now getting worse at "test taking" - is notable and interesting.
Whether IQ should be used as a proxy for expected real world success doesn't seem relevant to this study.
> Note the online magazine Quillette seems to be a cover for a sinister eugenics program (with tendencies I’ve called “neo-Nazi” under the cover of “free thought”.)
I'd love to see Taleb and Zizek spar, or riff, either one.
The Flynn effect, which showed widespread gains in IQ scores without any alteration in population genetics is the clearest indication that very significant variations in IQ can occur as a consequence of environmental effects.
The most logical conclusion of the data presented in paper is that the test results is unstable between dissimilar environments to the extent we shouldn't really bother using it for comparisons between populations (as opposed to an extremely crude measure of relative abilities within a cohort, where it is generally better than other crude proxies for actual human capability), still less mourning real world losses of reasoning ability. Tbh I thought doubting the tests was the page Flynn was already on.
Nevertheless, from other reports in the past few years, it appears that the rises in IQ widely observed during the 20th century (the Flynn effect) have not continued into the 21st. This would possibly argue that improvements in childhood and general health were large contributors to the Flynn effect.
On the other hand, if you take Flynn effect seriously, you also should take loss of Flynn effect seriously.
For example, the WAIS-IV measures "working memory", in part, via orally administered arithmetic word problems. An imminently reasonable thing to associate with intelligence in the pre-pocket-calculator economy!
But today? Kids spend a lot less time drilling arithmetic now than N decades ago. Those drills are more often administered on computer screens than orally in front of a classroom. And almost no one does a non-trivial amount of mental arithmetic in their work life. So this is mostly a useless skill. And, because it's a useless skill, we devote a lot less time to mastery. Even once mastered, the skill atrophies because it's just not very useful.
Question: if scores on these oral arithmetic word problems only barely decline despite significantly less practice -- both in the classroom and in everyday life -- what should we conclude?
Stated differently: if someone with 10K hours of tetris grind drops 2% more blocks than a total newbie, who's likely to be more adept at other forms of spatial reasoning?
The more general critique is that "intelligence research" has ossified around Cargo Cult psychometrics. At one point, "orally administered arithmetic word problems" were a fantastic proxy for economically useful mental faculties. That is no longer true. As the set of useful capacities change, so too should psychometric evaluations. Both because treatment effects are going to make cross-generational comparisons worse than useless, and also because the thing being measured has become irrelevant.
The humble pocket calculator should have taken Sociology by storm half a century ago. But the field hasn't kept up, and the problem is about to get exponentially worse as generative AI tools and other types of advanced automation start to augment every aspect of mental labor.
What it means to be "good at solving hard problems and getting things done" when assisted by GPT-614830 might look entirely different from what it means to be "good at solving hard problems and getting things done" today.
Flynn effect and its reversal are both environmentally caused: https://www.pnas.org/content/pnas/115/26/6674.full.pdf
For example, a distribution that becomes increasingly bimodal tells a very different story than a distribution shifting to the left.
Do what you're told, go home, come back next day. Repeat.
That is not 100% correct but good enough to summarize modern life.
Why learning abstract stuff in school if your career is stuck in call center jobs or you're the warehouse guy?
You don't really need math at home. There is few demand for intelligent problem solvers. There is a very high demand for "just do what you're told".
I see that everyday in my corporate setup. It is sad but I see it.
My own experience (in the UK) is that younger people perhaps seem less bothered about grammar, history, philosophy (the classics?) perhaps because they have learned that all information is on the internet and have subconciously assumed therefore that they do not need to learn the application of such knowledge. When I was young I remember my schoolmates lamenting that learning maths was pointless because "calculators", maybe that is coming home to roost now.
We could suggest that IQ is a measure of intelligence but I would argue that IQ only matters if you're applying to university which also gives it this correlative value of "wealthy" and "parents that care" if you're cherry picking economic outcomes which makes it hard to pick apart as a vacuumed characteristic. It also is a very narrow measure given that the breadth of human endeavour requires a lot more types of intelligence than just shit-hot pattern matching.