An interesting tidbit in the nature vs. nurture debate is that nature and nurture interplay in ways you might not expect. For instance, height is approximately 90% heritable in the United States -- but this does not mean that in a vacuum height is mostly genetic. It means that in the United States nutrition has mostly been solved (and yes, even the "food insecure" in the US rarely lack for the actual calories which would impact their height -- food insecurity causes other problems) and therefore the only real differences that can remain are the genetic differences.
It might be useful to look at any twin study through this lens; if we know for sure the genes are the same and nature is off the table, how much variance remains?
Things get even fuzzier when you throw in heritable epigenetics too. We have a balance of these factors at least:
- Genetics (DNA seq)
- Epigentics (Histone acetylation, base methylation etc)
- Brain wiring from experiences
- Chemical impact from experiences, e.g. nutrition, toxins, sunlight, muscle dev etc etc.
> - Chemical impact from experiences, e.g. nutrition, toxins, sunlight, muscle dev etc etc.
Are these not all part of the nurture / environment bucket? Or are we drawing a hard boundary between nurture (eg, parenting) and environment? (eg, lead in the pipes)
They can also simply get picked up in a correlated manner between twins due to the shared womb environment even if those correlated epigenetic traits are completely different from those of the parents.
> Things get even fuzzier when you throw in heritable epigenetics too.
Heritable epigenetics is a nascent field. We have yet to get any real promising research showing it even occurs to a significant degree in human.
I would assume it's not a significant factor until we can find even some minimal evidence to suggest that it's an influence in humans. Even then, my guess is that the impact would be minimal. I know this is contrary to a lot of the "trauma passed down through epigenetics" narratives that are going around, but the actual science is in a very different place than the popular understanding of the topic.
Whether or not epigenetic effects are passed down, epigenetics almost certainly impacts trait expression and complicates the "innate vs. environmental" narrative of hereditarians. I don't believe "inherited trauma" woo either, but epigenetics is much more fundamental to genetic science than that.
Traditional epigenetics is very different than heritable epigenetics.
Epigenetics is most definitely a factor in humans.
Heritable epigenetics is a different story. There is some early evidence suggesting that transgenerational epigenetic effects appear in humans, but it's very early. Some of the first claimed discoveries of transgenerational epigenetic inheritance in humans turned out to be mistakes by researchers. A lot of the theories about human transgenerational epigenetic inheritance revolve around exposure to famines and food crises, which is unfortunately complicated by the way food crises shape dietary behaviors and eating habits which are also passed down via tradition (not genetically).
The difficult part is that journalists often use "epigenetics" to refer to heritable epigenetics, and a lot of commenters make the same mistake.
These are twin studies, you don't need to get into heritable epigenitics. The expression of the shared environment in the womb would be reasonably expected to lead to epigenetic correlations in twins at a crucial stage of development where they would have the highest impact, without them being heritable.
Put simply, the common epigenetics between twins need not be held in common with the parents.
> The expression of the shared environment in the womb would be reasonably expected to lead to epigenetic correlations in twins at a crucial stage of development where they would have the highest impact, without them being heritable.
Sure, but that wouldn't be relevant to twin studies because both twins would be exposed to the same environment.
The pop culture discussion about heritable epigenetics tends to assume influence outside of in utero conditions or crossing multiple generations. It's where the "generational trauma is in your genes" idea came from.
It would be relevant to twin studies. Specifically, separated twin studies, where shared environment is assumed to be negligible. If the developmental impact of epigenetics is significant that won't be true.
In particular, see the third chart, "GREML-WGS heritability estimates," which shows that heritability for height is pretty much the same when they try to adjust for environment in different ways, while the estimates for IQ vary a lot.
Gusev seems convinced that some environmental effect is inflating heritability estimates for IQ that are based on twin studies.
In all these discussions it's helpful to understand the definition of broad-sense heritability (the statistic almost always in play when we're discussing heritability) --- it's a correlation, not a demonstrated causation, and your environment (and gene/environment interactions) are inherited as well.
I am curious about the influence on height by self-limited diets in children who are picky eaters. Is there some self-regulation process that decreases pickiness when nutritional status is at risk?
Is anybody actually picky enough to create a calorie or nutrient deficiency? I would think evolution would have harshly selected against such a behavior.
>and therefore the only real differences that can remain are the genetic differences.
Incidence of undiagnosed digestive and sleep disorders, like celiac disease and sleep apnea, respectively. "Catch-up" growth occurs almost universally after treatment starts, which means that the growth trajectory was lower than the "genetic potential", and likely remains so for children who can't access treatment.
Multiply across any number of silent or untreated disorders, sensitivities, or insufficiencies.
Adult height also isn't 90% heritable in the US. It only reaches that high a percentage of heritability in studies involving closely-related populations. In which case, of course; in populations with low genetic and cultural diversity and moderate or high variability, small differences are magnified.
That’s a good way to put it - eliminating other factors ends up deceptively making it look like genetics are the only difference. On the topic of IQ, the importance of those other factors is implied by things like the Flynn effect (https://en.wikipedia.org/wiki/Flynn_effect#Proposed_explanat...)? It’s also evident in how IQ varies by country. In developing countries, average IQ can measure low due to issues like malnutrition, access to education, etc. Those differences change as those countries develop or across different parts of those countries or when you track immigration populations from those countries.
Oddly, denial of these other variables has become core to scientific racism (https://en.wikipedia.org/wiki/Scientific_racism). For example, a Danish white supremacist named Emil Kirkegaard went so far as to create a fake journal (https://en.wikipedia.org/wiki/OpenPsych) to publish flagrantly incorrect papers on IQ that try to deny these factors and paint the IQ of non-white countries as low, with the only explanation being their “inferior” race/genetics. In fact, he just recently wrote a new “paper” (https://openpsych.net/paper/85/) that has been widely retweeted by supremacists on X.
> denial of these other variables has become core to scientific racism
Scientific racism has always operated this way. When it wasn't fabricating evidence, it was cherry-picking or failing to account for environmental causes. It would be interesting to see how historical claims made in this context hold up (I would expect to see changes that negate the claims; something that comes to mind is that the Dutch were one of the shortest peoples in Europe, but later in the 19th century became one of the tallest).
> flagrantly incorrect papers on IQ that try to deny these factors and paint the IQ of non-white countries as low, with the only explanation being their “inferior” race/genetics
I think the more important point we need to remember is that even if we were to discover that some groups have a genetic predisposition for higher IQ (putting aside the controversial nature of IQ to begin with), those with predisposition for lower IQ are still human and thus still owed what is due by virtue of their basic human dignity. It's also possible that over time, adaptation would change this predisposition. Populations aren't static.
As an anecdote: late 30s male, mother was 5'6 father was 5'9, I'm 6'4.
I have three kids with two different women. All 3 are blonde-haired/blue-eyed. I have brown hair and green eyes, my ex has brown hair and blue eyes, and my spouse has blonde hair and blue eyes. Not as interesting I suppose. Yes, they're all my kids biologically. :)
That's not mostly solved, that's tens of millions of truncated, immiserated lives. Of course calorie quality differences are important to child development.
I think you'll find in comparisons of 2025 to the turn of the 20th century that we lives are, by comparison, neither truncated nor immiserated. Anyways, this has nothing to do with the topic of the thread.
My argument is that we cannot simply say, they're both getting enough calories, therefore we can discount the nutritional component of their IQ differences. Sufficient calories do not remove nutrition as a confound.
You're handwaving away the nutrition hypothesis for the Flynn Effect, and I think losing sight of the timeline for the comparison. It's going to be very difficult to make any case like this if we're looking back to 1900, which is what we're doing when we talk about the nutrition/intelligence shifts we see in stats.
For intelligence? Where "nature" refers to "innateness" of the trait? I think it mostly is off the table, yes. I'm not saying that the only or even the most important environmental trait is nutrition.
(I think it can't possibly be entirely off the table, since we have mechanistic understanding of some gene-mediated cognitive disabilities).
There is something wrong with this article, possibly just copyediting mistakes but it makes me question the whole thing.
For example, check out this mess:
> “Unfortunately, there is one significant issue with the aforementioned data: schooling. Seeing as the majority of work to date includes only aggregate data, it is impossible to account. The first concerns small N: seeing as most publish studies only include a handful of TRA data, there is a lot of room for error and over.
Unfortunately, there is a largely unaccounted for confound in this aggregate data which may make generalized analysis questionable: schooling.”
Good catch. Additionally, one of the authors on this is just a student at UWisc, and the other author is also not a professional researcher but instead an author of popular books.
This is not an ad-hominum, but does put into question the statistical training backgrounds of both of these authors to accurate assess the data.
If not ad-hominum, what is it then? I mean, you did not provide any substantiated reason why would their research be false but you went straight to pin-point their experience, or lack thereof.
FWIW I find this research to align on my thoughts about the IQ - IQ is not a constant but a function of multiple variables, where one of the variables is most likely an education.
For instance, I am pretty sure that drilling through the abstract mathematical and hard engineering problems to some extent during the high-school but much more during and after the University, develops your brain in such a way that you become not only more knowledgeable in terms of memorizing things but develops your brain so that it can reason about and anticipate things that you couldn't possibly do before.
> and the other author is also not a professional researcher but instead an author of popular books.
This makes the awkward wording even more confusing. I don't understand how a professional author who appears to speak English very well would write so poorly and not follow up with edits.
> but does put into question the statistical training backgrounds
This is true of virtually all university research. Statistics is far more nuanced than what a semester course can teach you. And the incentives to publish can cause bad actors to use poorly defined surveys or p hack or whatever.
The language is consistent with ESL writing, in my experience.
The strange thing is that the corresponding author and the co-author appear to be english speakers, as far as I can tell. I googled the primary author and found a YouTube channel where someone by the same name speaks clearly about neuroscience. Maybe I'm looking at another person with the same name and middle initial who also happens to speak about neuroscience and brain development?
As a parent of identical twins watching them develop and grow is fascinating. I do wonder at times how much of it is due to going through every single life stage together but then again there are times where that bond seems to go beyond environment. There was a sobering but very interesting documentary on identical twins called Three Identical Strangers, if you are interested in this type of stuff it's a good watch.
The other side to this is non-identical twins, especially when still very young and have had basically the same experiences (doing everything together), they can be very different.
IQ testing was recently found to be highly driven by response to difficult challenges, and could be influenced significantly by just tuning the rewards for participants doing well. Which suggests that measuring iq is a pretty fraught science, if you are trying to draw conclusions about heritable intelligence...
If you give the same test to different people under completely different reward conditions, nobody is going to be surprised when the results are different.
IQ tests are far from perfect, but when the same test is administered to different groups under the same conditions it can be reasonable to draw some signal out of the differences between groups.
Generally in a study like this the people administering the test wouldn't even know which group each person belonged to.
It's common knowledge that IQ tests aren't perfect, but your linked paper doesn't debunk their usefulness in a study like this. Not unless the researchers unblinded themselves and offered one group a large reward, which would be a much bigger problem than the use of IQ tests.
You see how that's worse, right? That both intrinsic and extrinsic motivation to comply with the test can swing numbers, so choosing exactly one set of conditions necessarily benefits one group over another rather than creating an environment where everyone performs at their local optimum.
You’re really stretching. It’s a test. They’re not dangling marshmallows or $100 bills in front of some participants but not others dependent on results. In a study like this they may not even see the results.
> rather than creating an environment where everyone performs at their local optimum.
There's likely no such thing, so your best bet is to conduct the tests under a variety of conditions and aggregate the results. This has been the case for IQ for almost a century.
This is a restatement of 1990s-vintage science. It almost can't (it very probably isn't, but it's close to mathematically refuted) be true that simple twin study designs are correct about behavioral traits: parents mate assortatively, with large correlations specifically on education, so it's quite unlikely that fraternal twins have 50% shared genes for behavioral traits. Don't see the term "assortative mating" in an article like this? It's trying to sell you something.
(There are other broken assumptions in classic twin studies as well; this is just an easy one to describe).
This is the reason we have modern modalities like RDR (invented by a hereditarian!) and sibling regression.
Noah Carl's Wikipedia page is some interesting background (the author of the Aporia piece you provided). Enjoy!
Of course. IQ tests measures nothing more than the ability to pass an IQ test, which is proxied by a lot of things such as western culture, education, propensity to cram tests, etc.
> IQ tests measures nothing more than the ability to pass an IQ test
Incorrect, IQ is a composite measure correlated with fluid reasoning, crystallized knowledge, working memory, processing speed, and spatial ability. It's true that you can't naively use IQ to compare two diverse groups, but you can correct for this with a large enough sample of any two groups. This idea that it's biased towards western culture or education is vastly overblown.
This study found that adding or removing rewards for performing well can pretty dramatically impact performance.
"it is unclear to what extent the positive manifold reported in intelligence research since Spearman (1904) might be explained not through a shared component of intellectual capacity, but through a shared component of effort or time investment in testing tasks."
So, yes, we don't know, but the ways we don't know should also include "we don't know if iq testing is even measuring intelligence rather than stick-to-it-ness".
Sorry, you're right, I was reading "standardized test" as a rhetorical claim about IQ tests in particular. (I also believe those are trainable, but I'm much less certain of the science).
He's saying that low-class culture does not include quality education or quality influencers. Note that I'm not using influencer in the traditional sense here; I'm referring to anyone - dads, friends, acquaintances - who influences the child.
The weighting of this study is strange. The difference of number of years of education maxes out at 1 point, while being raised in different locations and different school types are each given also 1 point. It seems unreasonable that going to school in London vs New York should be given a point here despite the average educational quality in both cities potentially being the same. This also means that someone with 4 years more education but from the same city is considered educationally similar, and it is impossible to achieve the "educationally dissimilar" metric (ED DIFF > 2) without one of the other two points. I feel therefore like there is some wordplay being done here by the term "educational differences." I think some readers will assume that "educational differences" means "educational quality," but only one metric out of the 3 is directly correlated to this. This said there does seem to be some correlation and that is interesting, as we would expect no difference between location, yet there does seem to be one. In my opinion the different location variable is likely to be measuring something aside from/in addition to education. Some education types would seem to be better than others eg. boarding school. Also worth noting that the "very educationally dissimilar" group is n = 10. This said, the authors do admit that "certain level of inference is involved with comparing pedagogies and curriculum" and "Readers are encouraged to re-score and re-analyze the data in additional ways not done here." I would try weighting location much less and not cap number of years of education at all, instead studying how the differences change as the number of years increases.
I don't know much about IQ. In the most extreme case, of dissimilar education, the different was about 15 points. Is that a lot? What does that mean to laypeople?
On IQ tests, 15 points is a meaningful difference (one standard deviation), or roughly the gap between solidly average and clearly above average. It doesn’t make anyone a genius or a write-off. Still, we’d expect the higher-scoring person to generally find new learning and problem-solving easier, on average, if everything else is equal.
I'm not even sure it's mathematically appropriate to talk about IQ differences as there's no proof IQ is a linear metric. IQ is defined to have a normal distribution, unlike most things which are measured to have a normal distribution.
With IQ tests, the only thing you know that if 2 people fill it out and one of them scores higher, they have a higher IQ. Based on this you can sort people into a list by increasing IQ, but the standard distribution is implied not discovered.
This is like having a double-sided scale and a bunch of weights - you can similarly compare the weights to each other, and sometimes the arm of the scale will lean to the left, sometimes to the right, by a little or lot - you can postulate that the weights are normally distributed are normally distributed, but they are absolutely not required to be (I can choose them any way I like), so your assumption would be wrong. We know this because we have a direct, not just a comparative measure for weight.
I could make up an imaginary 'weight point' scale based on these comparisons, and say weights A is 5 WP heavier than B and C is also 5 WP heavier than A.
But A might be 100g, B might be 1g, and C might be 1kg.
This is what I think of when I see studies clamining the difference between 2 groups was 5 IQ points.
IQ scores are calibrated to be normally distributed with a standard deviation of 15. So 15 is one standard deviation. That's the difference between average, and being in the smartest 16% of the population. Or being in the smartest 16%, and being in the smartest 2% of the population.
Well a sure component of test scores reflects test taking skills. Years ago, I purchased a book of a series of IQ tests, and my numerical result increased with every test. Another component is confidence. And another is ability. It is said by some that among the first big users of IQ tests was the US Army.
It's not nothing, but IQ is already a little squishy. No one's IQ is a single number. But the article also goes into problems with the study and other potential issues.
Basically, they're saying there is this pattern in the data as recorded, but there are multiple confounding factors and issues with collecting the data in the first place.
It's a single number in that if you take an IQ test one time you get one number, but that doesn't mean you'll get that exact number every single time you take an IQ test. Even ignoring more complex questions about them, your score on an IQ test will vary depending on simple things like how tired you are when you take it, so in practice there's some variance and you do not always get the same number every time you take a test.
This is a deeper question than it sounds. The "point" of a modern IQ test is to identify cognitive deficits to target interventions. It's abused widely among non-practitioners as a ranking of intelligence, which it is not.
I don't think we have to hash this out, because "marker of intelligence" and "ranking of intelligence" are not the same thing. A rank implies a reliable scale, which IQ doesn't provide.
(It's also just fine if we disagree about this --- researchers do too!)
I mean, nothing in the human body can be truly represented by a single number.
Even height and weight change throughout the day. People are typically taller and lighter in the morning than in the evening. Weight especially is variable, it can fluctuate up to 5 to 6 pounds.
I think it's inaccurate to call it a dumbing-down. It's more correct to label it as the largest stratification of education we've ever seen.
The smartest kids are smarter than ever before. They're absolutely rocking the house. The problem is that the "middle class has been gutted". Kids who were kinda smart, or kinda dumb, are now lumped in with kids who probably need Individual Education Plans (IEPs). This lowers the educational standard for almost all students - though of course the most well-off among us (educationally, rather than monetarily) are not only not suffering, they're thriving.
Only if they’re wealthy or get extremely lucky and live near a randomly good school. By many metrics I was the smartest in my class, but my family had little money and lived in a rural area with a single underfunded school. I spent my days in class with kids that were still struggling to sound out “cat” in third grade. A few times a week I got to spend an hour in “gifted” class but that was mostly art projects, nothing that would help make up for the rest of the day being wasted.
That is provably incorrect, as since Victorian times people lost around 14 to 23 IQ points on average. Notably, the corrected scores have continued on a downward trend for the past century.
People are not getting smarter, as recent events have shown. =3
Sadly, this may be coming to NYC. In principle, I am all for trying to improve the baseline, but you cannot sacrifice better students to do that. Not only is it silly (how would this sacrifice exactly benefit worse students?), but unjust. Furthermore, education begins at home. Parents are the primary educators of children, not necessarily academically, but in the broader "life" sense. If the home environment is not conducive or supportive of education at school, you will be facing a very uphill battle.
The dumbing down of education goes further than what you note, though. Think of classical education and the formation of the human person (I'm not talking about "Dead Poets Society" ersatz, but the real deal). Think of the principles behind the trivium and quadrivium. In the best case, we are producing superficially technically savvy barbarians. Schools are effectively savage factories, and universities are laughable and should be ashamed of calling themselves universities.
There are two different kinds of IQ tests: convergent and divergent. Convergent tests are more common and test either knowledge or pattern matching. These tests are called convergent because they are a center of truth and conformance to that truth is the measured performance criteria.
Divergent tests measure the individual's creativity and abstract reasoning. The source of truth is the quantity of diversity of results submitted by the participant.
The implicit success criteria for convergent testing is reading comprehension. A person with dyslexia, for example, will perform worse on these tests irrespective of their learning speed, learned knowledge, intellectual curiosity, or creativity. This is a form of bias. Other forms of bias include memorization of terms, such as SAT preparation.
To further complicate things these measures typically only account for academic intelligence. Other forms of intelligence include social intelligence, spatial intelligence, creativity, conscientiousness, and so forth. In the concept of multi-dimensional intelligence, which is what is actually addressed in practice in the real world after high school, academic intelligence alone has very little benefit. Its like height in basketball where after 6.5ft all other factors become more important for all participants.
> A person with dyslexia, for example, will perform worse on these tests irrespective of their learning speed, learned knowledge, intellectual curiosity, or creativity. This is a form of bias.
No, it's what they were designed to do - help identify people that may need specialized help because issues like learning disabilities.
> To further complicate things these measures typically only account for academic intelligence. Other forms of intelligence include social intelligence, spatial intelligence, creativity, conscientiousness, and so forth.
The two major IQ tests (Stanford Binet and Wechsler) test visual-spatial reasoning, abstract reasoning, verbal and nonverbal reasoning, working memory, processing speed, inductive/deductive reasoning, attention, concentration, etc.
These tests, if properly administered (individually in person by a professional) work well for figuring out who in school needs extra attention or requires further evaluation. Their use beyond that is questionable.
The way we raise our kids will have major differences in their IQ... Keep in mind there is a 9 point difference between those raised apart vs those raised together. This is not a comparison between those who receive special education efforts vs those who don't, I assume that would be much larger.
This begs the question. How much of "intelligence" is intrinsic? We don't know. It's not zero (we know there are genetically-mediated cognitive deficits), but it's very plausibly not meaningfully intrinsic in most people.
It's always been an insane claim that knowledge can't help you utilize knowledge. It's just an excuse to be racist and classist. Everything around IQ is linear, dumb thinking, suitable for trying to reduce everything to a scalar. IQ is a good tool for figuring out whether your child has a physical issue with their brain development, or maybe a hearing problem.
There's nothing logical about brains; they work on heuristics built on reflexes and associations. Logic is something we learn through physically interacting with the world, and from sitting in class. As you learn it, your thinking about other things gets better. When you read great old Chinese philosophers like Mozi struggling but managing to make good arguments without syllogism, it's a reminder that syllogism is something that is invented. It survives because of its unreasonable effectiveness compared to considering what the ancient heavenly kings had done.
It also helps your soul to remember that a lot of people you're arguing with have no idea what a logical argument looks like. They think that an argument is when you try to shut the other person up. It's not their fault, they just literally don't know. They're not arguing in bad faith, they don't know what good faith looks like because the educational system has failed them.
Outside of a fascination with Intellectual Supremacism, why are people so obsessed with the genetic basis of IQ scores?
Presumably one could do similar identical twin studies on half-marathon race times and SAT test scores. Does no one bother with those, because widespread awareness of half-marathon training regimens and SAT prep courses would spoil the (desired) illusion of some "innate superiority of blood" being measured?
How about the fact that the human mind and genetics are simply fascinating and interesting topics. I would imagine that people don't care as much about running and high school testing because they are fairly niche interests relative to abstract thinking in general, something that almost everyone spends much of their life doing.
Exactly, it is on of the least remarkable results that intelligence is partly heritable. Obviously intelligence is partly genetic, obviously your ability to run a marathon is also partly genetic. Why one has devolved into a bizarre case where so much effort is spent on the minute details of this question and the other is accepted basically as completely uncontroversial (given who is currently competing for top times at marathons it is basically inarguable) is very strange.
Speaking of IQ, when I listen to interviews and podcasts of supposed high IQ people like Mark Zuckerberg and Elon Musk, I struggle to see signs of intelligence (not saying that they don’t have above average IQ). I find that these 2 have a hard time sifting through the noise (competing thoughts) in their head to get to want they want to say (the signal).
When I listen to Steve Jobs, I hear someone who has very strong ability to sift through noise. So Jobs couldn’t see engineering in a new way like Elon does but Elon couldn’t do what Jobs did either.
Regarding Zuckerberg, from what I’ve read he is the Bill Gates type where he has the traditional variant of high IQ, aka raw hardware/horsepower but lower on creativity/imagination side.
So intelligence seems to have different shapes and sizes.
Don't read this as a defense of Zuck or Musk, but some of the most brilliant people I've known sometimes had trouble getting thoughts out. It wasn't that they couldn't see their goal, but that how they thought of it was changing even as the words were leaving their mouths. Have you ever tried to explain while you were still working it out in your head? "See, it's kind of like A..., except that one part is closer to B, and that's interesting because, well, B and C have a relationship that's closer than A and C, so it might be more appropriate to say it's more C-like, except where A is..."
That didn't mean they were unintelligent. It meant that them trying to explain the flurry of ideas in their head was like me trying to type with mittens.
Of course, knuckleheads can also sound like that, but for different reasons. That kind of rambling doesn't imply that the speaker is a dumbass. It definitely doesn't guarantee that they're a genius.
I don't dispute the differences in intellectual capacity between people but IQ tests are like weight lifting contests between people who didn't train to lift weights.
I don't understand why the scientific community keeps using methods that are so flawed. Perhaps it's due to my own lack of information.
The only way I could see IQ tests being a valid measurement of intellectual capacity is if participants were brought to the same level of knowledge and skills, and then were made to train for the IQ test using the exact same means, and even then cultural and language barriers must be accounted for.
Even if these twins have an identical intellctual capacity and they both had the same exact education, and same exact grades, that still doesn't mean they applied and exercised their brains in the same way for the questions of the IQ tests.
IQ tests to measuring intellectual capacity (instead of knowledge level) are like polygraphs to mesasuring truthfulness in terms of accuracy.
For measuring strength, i stand by my earlier correlation with weight lifting. If two people can dead-lift 500lbs, I only know that both participants are have reached that level of strength. What I don't know is how much effort each participant put into getting to that level of strength, which would tell me their natural muscular capacity per effort. IQ tests deceptively seem like they tell you what someone's capacity for intellect is, but they only tell you where that person is right now. Maybe that person worked hard and the score is their max, maybe the person rarely applies their brain to demanding tasks and this is their mid-level capacity.
My point is,it isn't just education or just genetics, it is also personality, effort, motivation. For all I know,someone with double-digit IQ score can work out their brain for a year and hit genius level. Choice. Can a person choose to be a literal idiot and succeed? Certainly, people choose to be incurious and ignorant all the time. Women can body-build and be as strong as many regular men for example. But simply because of a hormone difference, they have to work out a lot more than men to reach the same level of muscular strength. And men who never work out can be as weak as women who never work out.
There is also the question of brain development. Maybe the effort you put has different effect depending on age. small efforts at a young age at applying your brain might have huge impacts, where as if you're a teenager or an adult, applying the same effort might yield less results.
I mean, personally, if I tired, if I ate too much, or too little,if didn't get enough sleep, if I am distracted by something, if I'm unprepared and thus second-guessing myself, these are some of the things that throw me off wildly when taking exams. I've seen huge differences by simply getting enough sleep and calories.
There is no way that asking the same set of questions to to large population (even with control populations in place) can account for all the presumptions of the test.
even if you prepped since you were 5 years old? and what if you're unprepared, will the decrease be significant? I mean of course, even with weight lifting, you can't prepare even a year in advance and show a significant difference over professional competitors. But people do peak. Are you saying Einstien's IQ would be the same if he was tested at 50 yo, and he was working in the coal mines in virginia his since 5 yo? If so, that's indeed enlightening, I'd love to see any research that backs up that claim.
The area where I see IQ failing miserably is in accounting for different types of intelligences. Like the parent commenter said, people solve the same problem differently. I'm keenly aware of this because I'm neurodivergent and I was continuously judged as poor in subjects that I liked the most and was very good at. I had to reformulate every problem and solution in a way that I understood (mostly based on spatial intelligence), leaving me at a significant disadvantage of time. And even then, those solutions were sometimes rejected, despite being objectively correct and clear in the dumbest possible way - just because it didn't follow the textbook pattern. That continued until we were in a situation where we had to solve the problems ourselves.
The reason I mention this is because I see the exact same problem with IQ tests. It emphasizes certain types of intelligences and ignores others. Human intelligence is extremely multidimensional and a single number is simply incapable of representing its overall quality. For example, there are people who score poorly in IQ, but are superhuman in remembering places and navigating their way around complex routes. Meanwhile, many high scoring ones become hopelessly lost, failing to institute or follow even simple mitigation strategies for that problem. There are situations where this determines whether you live or die, and your IQ score becomes worthless indicator of even your survival.
Afaik iq correlates highly with g-factor (scientific name for general intelligence). We don't have much better ways to measure it.
Overall, afaik, again, problem solving often needed in our society correlates with this: "general intelligence". When people say: "But there are many ways to approach a problem, many types of intelligence".
This very well may be, but data consistently shows that people having high IQ scores solve different problems better. So, contrary to popular belief, people having higher iq in general have better talent for languages, same with understanding others emotions and using it to your advantage ("emotional intelligence").
As for reframing an issue and solving it in a different way. This may be valid approach (to teach people etc). But IQ is also time-measured. If your new approach does not help you to solve previously unseen problems quickly, it is not noticably increase your intelligence.
Thus, we see consistently that people cannot really prepare for iq tests much. You only get a few points more if you prepare. Same difference as if you are sleep-deprived.
G isn't so much the "scientific name for general intelligence" so much as one explanation for the positive manifold of intelligence. There are others: mutualism and sampling.
I'm not sure you're right about the trainability of tests, either.
> The area where I see IQ failing miserably is in accounting for different types of intelligences.
We don't have any real evidence of "different types of intelligences". People with high IQ just tend to be better at all cognitive tasks than people with lower IQ. That's why it's considered a reasonably good measure of the g factor. People with higher IQ also have better memory recall.
While you're right that that IQ isn't the full story of a person's abilities, it's use for diagnosing people that may face challenges in a traditional learning environment well motivated. All of the examples you mention of people scoring poorly on IQ and facing challenges in school supports the use of IQ, it doesn't discount it.
> We don't have any real evidence of "different types of intelligences". People with high IQ just tend to be better at all cognitive tasks than people with lower IQ.
That would imply that a person with a high IQ score is uniformly better at every task compared to a person of lower IQ. But that's absolutely not what's observed. People show better skill levels at different tasks. A person who's bad at math may instead be a natural-born singing sensation. This is why I said that a single number is just incapable of representing intelligence. The mathematical dimensionality of the quantity called intelligence is just too high. You'll need a bunch of numbers - a vector, at the minimum. And even that can't be used for comparing people's intelligence. It can only be used for assessing someone's suitability for a particular task. Basically, IQ score is a scalar that is used to represent a vector, discarding very important information in the process.
> People with higher IQ also have better memory recall.
There are people who recite entire Shakespeare plays without understanding a sentence in it. I have also seen people who recall long derivation sequences using Maxwell's equations exactly and score well on exams, without having any clue as to what the del (∇) operator even means in practice. Needless to say, they have difficulty adapting that information to a novel situation. Memory recall is just one aspect of intelligence. There are other cognitive skills required to make that memory useful. This again goes back to my previous conclusion. A single number is quite meaningless at best and utterly misleading at worst.
> it's use for diagnosing people that may face challenges in a traditional learning environment well motivated. All of the examples you mention of people scoring poorly on IQ and facing challenges in school supports the use of IQ, it doesn't discount it.
On the contrary, it just misleads the educators to the student's real abilities. It ignores and hides the atypical talents that students possess. All it can do is to predict if the student will do well or not in generalized education and standardized testing, because they follow the equally illogical concept of using standardized pedagogy on everyone, neglecting their uniqueness. So, even if we accept your claim that the IQ score predicts their performance in schools, it means that the score has the very narrow scope of testing someone's capability to learn under a standardized curriculum. That's far too narrow to be identified as an indicator of intelligence. And even after identifying the students who are going to struggle, IQ gives no clue as to how to remedy that. All it does is put a dunce label on these students who might otherwise have done better. It does more harm than good.
> That would imply that a person with a high IQ score is uniformly better at every task compared to a person of lower IQ.
No it doesn't, that's why I said "tend to".
> On the contrary, it just misleads the educators to the student's real abilities. It ignores and hides the atypical talents that students possess.
The test simply doesn't measure atypical talents, so there's no "hiding". That's why I said the test is used to identify people who will face challenges. What happens next has nothing to do with the IQ test and everything to do with other methods of assessment that should be used. Every test has limitations that those applying the test should understand.
> That's far too narrow to be identified as an indicator of intelligence
That's why it's a (strong) correlation and not a strict equality.
Look, jumping height has been empirically demonstrated to be a good predictor of overall athletic performance, and you're coming here and saying that there exist some people without legs that can do more pull-ups than most people with legs. That's all well and good, but that doesn't somehow refute the idea that jumping height is a good predictor of overall athletic performance. It's just a complete red herring. If you happen to encounter someone without legs, then use a different test.
People with legs who don't exhibit good jumping heights but are very good at some specific athletic challenge could exist, but they tend to be good only at that specific challenge (and ones that are very, very similar). This is a completely different thing than the general factor of fitness, which is correlated with all sorts of physical abilities, just like general factor g is correlated with all sorts of cognitive abilities.
I think you are being a little bit conclusory here. There are in fact serious theories of intelligence that admit to multiple distinct intelligence factors. Mutualism is the obvious counterexample. You're arguing from what are arguably exploratory statistical correlations as if the correlation demonstrated the underlying cause. You might be right, but you're not presenting evidence for that position.
On the contrary, it is so easy to increase your IQ score by prepping that test administrators ask the test takers not to do it because it invalidates the results.
I don't know how many more nails in the coffin of heritable intelligence are needed before public policy capitalizes on the opportunity here. We have seen huge shifts in intelligence in places like Korea after armistice, in China after economic and cultural reforms, and in the U.S. after COVID and smartphones. We have seen that individual tutoring reliably creates extreme outliers from the examples of Avant-garde, Williams, Tao, Polgár, the Hungarian Martians, etc. Aside from disorders, any heritable differences can be easily dwarfed by environmental effects, and some disorders, like OCD, only magnify environmental effects.
> Of the 87 pairs, 52 experienced a similar type and duration of schooling within a similar location. In fact, 25 of these pairs attended the same school for some period of time. Analysis revealed that ‘Educationally Similar’ TRA pairs have an ICC of 0.87 ± 0.02 (n = 52)
So this study has 87-52-25=10 data points? Am I reading this correctly? Quite the reach to conclude what the article claims, if so.
> With this said, it is important to note that the ‘very dissimilar education’ group consisted of only 10 TRA pairs. This small N is not a shortcoming of this analysis, per se. Rather, this is a shortcoming of the TRA field; these 10 pairs represent the entirety of individual data published over the last century.
An interesting tidbit in the nature vs. nurture debate is that nature and nurture interplay in ways you might not expect. For instance, height is approximately 90% heritable in the United States -- but this does not mean that in a vacuum height is mostly genetic. It means that in the United States nutrition has mostly been solved (and yes, even the "food insecure" in the US rarely lack for the actual calories which would impact their height -- food insecurity causes other problems) and therefore the only real differences that can remain are the genetic differences.
It might be useful to look at any twin study through this lens; if we know for sure the genes are the same and nature is off the table, how much variance remains?
Things get even fuzzier when you throw in heritable epigenetics too. We have a balance of these factors at least:
> - Brain wiring from experiences
> - Chemical impact from experiences, e.g. nutrition, toxins, sunlight, muscle dev etc etc.
Are these not all part of the nurture / environment bucket? Or are we drawing a hard boundary between nurture (eg, parenting) and environment? (eg, lead in the pipes)
I'm splitting up both "nature" and "nurture" into slightly less-broad categories. In the case of what you highlighted, yes.
For example, epigentics is sort of both "nature" and "nurture", in that you can pick up these traits, and pass them on/get them passed on.
They can also simply get picked up in a correlated manner between twins due to the shared womb environment even if those correlated epigenetic traits are completely different from those of the parents.
Genes interact with the environment. There aren't hard boundaries the way you've phrased them.
> Things get even fuzzier when you throw in heritable epigenetics too.
Heritable epigenetics is a nascent field. We have yet to get any real promising research showing it even occurs to a significant degree in human.
I would assume it's not a significant factor until we can find even some minimal evidence to suggest that it's an influence in humans. Even then, my guess is that the impact would be minimal. I know this is contrary to a lot of the "trauma passed down through epigenetics" narratives that are going around, but the actual science is in a very different place than the popular understanding of the topic.
Whether or not epigenetic effects are passed down, epigenetics almost certainly impacts trait expression and complicates the "innate vs. environmental" narrative of hereditarians. I don't believe "inherited trauma" woo either, but epigenetics is much more fundamental to genetic science than that.
Traditional epigenetics is very different than heritable epigenetics.
Epigenetics is most definitely a factor in humans.
Heritable epigenetics is a different story. There is some early evidence suggesting that transgenerational epigenetic effects appear in humans, but it's very early. Some of the first claimed discoveries of transgenerational epigenetic inheritance in humans turned out to be mistakes by researchers. A lot of the theories about human transgenerational epigenetic inheritance revolve around exposure to famines and food crises, which is unfortunately complicated by the way food crises shape dietary behaviors and eating habits which are also passed down via tradition (not genetically).
The difficult part is that journalists often use "epigenetics" to refer to heritable epigenetics, and a lot of commenters make the same mistake.
Right, I'm not pushing back on you, just sticking up for the idea of paying attention to epigenetic factors in this broader question of innateness.
These are twin studies, you don't need to get into heritable epigenitics. The expression of the shared environment in the womb would be reasonably expected to lead to epigenetic correlations in twins at a crucial stage of development where they would have the highest impact, without them being heritable.
Put simply, the common epigenetics between twins need not be held in common with the parents.
> The expression of the shared environment in the womb would be reasonably expected to lead to epigenetic correlations in twins at a crucial stage of development where they would have the highest impact, without them being heritable.
Sure, but that wouldn't be relevant to twin studies because both twins would be exposed to the same environment.
The pop culture discussion about heritable epigenetics tends to assume influence outside of in utero conditions or crossing multiple generations. It's where the "generational trauma is in your genes" idea came from.
We're discussing twins raised apart. They're not necessarily in the same environment (except during gestation).
It would be relevant to twin studies. Specifically, separated twin studies, where shared environment is assumed to be negligible. If the developmental impact of epigenetics is significant that won't be true.
I don't fully understand this blog post, but possibly of interest:
https://theinfinitesimal.substack.com/p/the-missing-heritabi...
In particular, see the third chart, "GREML-WGS heritability estimates," which shows that heritability for height is pretty much the same when they try to adjust for environment in different ways, while the estimates for IQ vary a lot.
Gusev seems convinced that some environmental effect is inflating heritability estimates for IQ that are based on twin studies.
In all these discussions it's helpful to understand the definition of broad-sense heritability (the statistic almost always in play when we're discussing heritability) --- it's a correlation, not a demonstrated causation, and your environment (and gene/environment interactions) are inherited as well.
Re: nutrition
I am curious about the influence on height by self-limited diets in children who are picky eaters. Is there some self-regulation process that decreases pickiness when nutritional status is at risk?
Is anybody actually picky enough to create a calorie or nutrient deficiency? I would think evolution would have harshly selected against such a behavior.
Yes. See e.g. anorexia nervosa.
>and therefore the only real differences that can remain are the genetic differences.
Incidence of undiagnosed digestive and sleep disorders, like celiac disease and sleep apnea, respectively. "Catch-up" growth occurs almost universally after treatment starts, which means that the growth trajectory was lower than the "genetic potential", and likely remains so for children who can't access treatment.
Multiply across any number of silent or untreated disorders, sensitivities, or insufficiencies.
Adult height also isn't 90% heritable in the US. It only reaches that high a percentage of heritability in studies involving closely-related populations. In which case, of course; in populations with low genetic and cultural diversity and moderate or high variability, small differences are magnified.
That’s a good way to put it - eliminating other factors ends up deceptively making it look like genetics are the only difference. On the topic of IQ, the importance of those other factors is implied by things like the Flynn effect (https://en.wikipedia.org/wiki/Flynn_effect#Proposed_explanat...)? It’s also evident in how IQ varies by country. In developing countries, average IQ can measure low due to issues like malnutrition, access to education, etc. Those differences change as those countries develop or across different parts of those countries or when you track immigration populations from those countries.
Oddly, denial of these other variables has become core to scientific racism (https://en.wikipedia.org/wiki/Scientific_racism). For example, a Danish white supremacist named Emil Kirkegaard went so far as to create a fake journal (https://en.wikipedia.org/wiki/OpenPsych) to publish flagrantly incorrect papers on IQ that try to deny these factors and paint the IQ of non-white countries as low, with the only explanation being their “inferior” race/genetics. In fact, he just recently wrote a new “paper” (https://openpsych.net/paper/85/) that has been widely retweeted by supremacists on X.
> denial of these other variables has become core to scientific racism
Scientific racism has always operated this way. When it wasn't fabricating evidence, it was cherry-picking or failing to account for environmental causes. It would be interesting to see how historical claims made in this context hold up (I would expect to see changes that negate the claims; something that comes to mind is that the Dutch were one of the shortest peoples in Europe, but later in the 19th century became one of the tallest).
> flagrantly incorrect papers on IQ that try to deny these factors and paint the IQ of non-white countries as low, with the only explanation being their “inferior” race/genetics
I think the more important point we need to remember is that even if we were to discover that some groups have a genetic predisposition for higher IQ (putting aside the controversial nature of IQ to begin with), those with predisposition for lower IQ are still human and thus still owed what is due by virtue of their basic human dignity. It's also possible that over time, adaptation would change this predisposition. Populations aren't static.
As an anecdote: late 30s male, mother was 5'6 father was 5'9, I'm 6'4.
I have three kids with two different women. All 3 are blonde-haired/blue-eyed. I have brown hair and green eyes, my ex has brown hair and blue eyes, and my spouse has blonde hair and blue eyes. Not as interesting I suppose. Yes, they're all my kids biologically. :)
[dead]
> It means that in the United States nutrition has mostly been solved
Nature is not off of the table. We've just traded problems with calorie quantity to quality.
That's not mostly solved, that's tens of millions of truncated, immiserated lives. Of course calorie quality differences are important to child development.It's solved in the narrow "you'll be 6 inches too short" sense, but not the wider "you'll avoid diabetes and heart disease" sense.
I think you'll find in comparisons of 2025 to the turn of the 20th century that we lives are, by comparison, neither truncated nor immiserated. Anyways, this has nothing to do with the topic of the thread.
My argument is that we cannot simply say, they're both getting enough calories, therefore we can discount the nutritional component of their IQ differences. Sufficient calories do not remove nutrition as a confound.
You're handwaving away the nutrition hypothesis for the Flynn Effect, and I think losing sight of the timeline for the comparison. It's going to be very difficult to make any case like this if we're looking back to 1900, which is what we're doing when we talk about the nutrition/intelligence shifts we see in stats.
We seem to be having different conversations. I'm responding to
... and saying that nature isn't off the table at all. Are you saying that it is?For intelligence? Where "nature" refers to "innateness" of the trait? I think it mostly is off the table, yes. I'm not saying that the only or even the most important environmental trait is nutrition.
(I think it can't possibly be entirely off the table, since we have mechanistic understanding of some gene-mediated cognitive disabilities).
There is something wrong with this article, possibly just copyediting mistakes but it makes me question the whole thing.
For example, check out this mess:
> “Unfortunately, there is one significant issue with the aforementioned data: schooling. Seeing as the majority of work to date includes only aggregate data, it is impossible to account. The first concerns small N: seeing as most publish studies only include a handful of TRA data, there is a lot of room for error and over.
Unfortunately, there is a largely unaccounted for confound in this aggregate data which may make generalized analysis questionable: schooling.”
Good catch. Additionally, one of the authors on this is just a student at UWisc, and the other author is also not a professional researcher but instead an author of popular books.
This is not an ad-hominum, but does put into question the statistical training backgrounds of both of these authors to accurate assess the data.
If not ad-hominum, what is it then? I mean, you did not provide any substantiated reason why would their research be false but you went straight to pin-point their experience, or lack thereof.
FWIW I find this research to align on my thoughts about the IQ - IQ is not a constant but a function of multiple variables, where one of the variables is most likely an education.
For instance, I am pretty sure that drilling through the abstract mathematical and hard engineering problems to some extent during the high-school but much more during and after the University, develops your brain in such a way that you become not only more knowledgeable in terms of memorizing things but develops your brain so that it can reason about and anticipate things that you couldn't possibly do before.
> and the other author is also not a professional researcher but instead an author of popular books.
This makes the awkward wording even more confusing. I don't understand how a professional author who appears to speak English very well would write so poorly and not follow up with edits.
> but does put into question the statistical training backgrounds
This is true of virtually all university research. Statistics is far more nuanced than what a semester course can teach you. And the incentives to publish can cause bad actors to use poorly defined surveys or p hack or whatever.
The language is consistent with ESL writing, in my experience.
The strange thing is that the corresponding author and the co-author appear to be english speakers, as far as I can tell. I googled the primary author and found a YouTube channel where someone by the same name speaks clearly about neuroscience. Maybe I'm looking at another person with the same name and middle initial who also happens to speak about neuroscience and brain development?
As a parent of identical twins watching them develop and grow is fascinating. I do wonder at times how much of it is due to going through every single life stage together but then again there are times where that bond seems to go beyond environment. There was a sobering but very interesting documentary on identical twins called Three Identical Strangers, if you are interested in this type of stuff it's a good watch.
The other side to this is non-identical twins, especially when still very young and have had basically the same experiences (doing everything together), they can be very different.
IQ testing was recently found to be highly driven by response to difficult challenges, and could be influenced significantly by just tuning the rewards for participants doing well. Which suggests that measuring iq is a pretty fraught science, if you are trying to draw conclusions about heritable intelligence...
https://pmc.ncbi.nlm.nih.gov/articles/PMC10990577/
If you give the same test to different people under completely different reward conditions, nobody is going to be surprised when the results are different.
IQ tests are far from perfect, but when the same test is administered to different groups under the same conditions it can be reasonable to draw some signal out of the differences between groups.
Generally in a study like this the people administering the test wouldn't even know which group each person belonged to.
It's common knowledge that IQ tests aren't perfect, but your linked paper doesn't debunk their usefulness in a study like this. Not unless the researchers unblinded themselves and offered one group a large reward, which would be a much bigger problem than the use of IQ tests.
You see how that's worse, right? That both intrinsic and extrinsic motivation to comply with the test can swing numbers, so choosing exactly one set of conditions necessarily benefits one group over another rather than creating an environment where everyone performs at their local optimum.
You’re really stretching. It’s a test. They’re not dangling marshmallows or $100 bills in front of some participants but not others dependent on results. In a study like this they may not even see the results.
> rather than creating an environment where everyone performs at their local optimum.
There's likely no such thing, so your best bet is to conduct the tests under a variety of conditions and aggregate the results. This has been the case for IQ for almost a century.
I would guess that people will do better on just about any test when given a reward for good performance.
The highest IQ man in history chose to be a plumber. =3
which will bias all correlations towards zero, while height measurement is very accurate.
yet iq inheritability is only a bit lower, around 80%
The most generous twin study numbers get you to 80%; modern study designs get you nowhere close to that. This is the "missing heritability" crisis.
[dead]
This is a restatement of 1990s-vintage science. It almost can't (it very probably isn't, but it's close to mathematically refuted) be true that simple twin study designs are correct about behavioral traits: parents mate assortatively, with large correlations specifically on education, so it's quite unlikely that fraternal twins have 50% shared genes for behavioral traits. Don't see the term "assortative mating" in an article like this? It's trying to sell you something.
(There are other broken assumptions in classic twin studies as well; this is just an easy one to describe).
This is the reason we have modern modalities like RDR (invented by a hereditarian!) and sibling regression.
Noah Carl's Wikipedia page is some interesting background (the author of the Aporia piece you provided). Enjoy!
[flagged]
[flagged]
Is there a 'teaching the test' element to this? Does more exposure to education increase your ability to take a standardized test?
Of course. IQ tests measures nothing more than the ability to pass an IQ test, which is proxied by a lot of things such as western culture, education, propensity to cram tests, etc.
> IQ tests measures nothing more than the ability to pass an IQ test
Incorrect, IQ is a composite measure correlated with fluid reasoning, crystallized knowledge, working memory, processing speed, and spatial ability. It's true that you can't naively use IQ to compare two diverse groups, but you can correct for this with a large enough sample of any two groups. This idea that it's biased towards western culture or education is vastly overblown.
We don't know.
We do know that response to difficulty influences performance in iq testing: https://pmc.ncbi.nlm.nih.gov/articles/PMC10990577/
This study found that adding or removing rewards for performing well can pretty dramatically impact performance.
"it is unclear to what extent the positive manifold reported in intelligence research since Spearman (1904) might be explained not through a shared component of intellectual capacity, but through a shared component of effort or time investment in testing tasks."
So, yes, we don't know, but the ways we don't know should also include "we don't know if iq testing is even measuring intelligence rather than stick-to-it-ness".
I mean I agree with that too! I would just bucket all this under "we don't know". :)
We do know that practice taking tests makes you better at taking tests. Kids get tutored in that manner every day, and it works.
Sorry, you're right, I was reading "standardized test" as a rhetorical claim about IQ tests in particular. (I also believe those are trainable, but I'm much less certain of the science).
Define "more exposure to education". ;)
Also, NCBLA standardized testing has demonstrably ruined education in America.
https://en.wikipedia.org/wiki/National_Children%27s_Book_and...
More exposure to upper middle-class culture increases ones ability to take IQ tests.
I don't think golf and tennis increases IQ
He's saying that low-class culture does not include quality education or quality influencers. Note that I'm not using influencer in the traditional sense here; I'm referring to anyone - dads, friends, acquaintances - who influences the child.
Gluten reduces IQ (jk)
Lots of good reason to believe nutrition does.
The weighting of this study is strange. The difference of number of years of education maxes out at 1 point, while being raised in different locations and different school types are each given also 1 point. It seems unreasonable that going to school in London vs New York should be given a point here despite the average educational quality in both cities potentially being the same. This also means that someone with 4 years more education but from the same city is considered educationally similar, and it is impossible to achieve the "educationally dissimilar" metric (ED DIFF > 2) without one of the other two points. I feel therefore like there is some wordplay being done here by the term "educational differences." I think some readers will assume that "educational differences" means "educational quality," but only one metric out of the 3 is directly correlated to this. This said there does seem to be some correlation and that is interesting, as we would expect no difference between location, yet there does seem to be one. In my opinion the different location variable is likely to be measuring something aside from/in addition to education. Some education types would seem to be better than others eg. boarding school. Also worth noting that the "very educationally dissimilar" group is n = 10. This said, the authors do admit that "certain level of inference is involved with comparing pedagogies and curriculum" and "Readers are encouraged to re-score and re-analyze the data in additional ways not done here." I would try weighting location much less and not cap number of years of education at all, instead studying how the differences change as the number of years increases.
I don't know much about IQ. In the most extreme case, of dissimilar education, the different was about 15 points. Is that a lot? What does that mean to laypeople?
On IQ tests, 15 points is a meaningful difference (one standard deviation), or roughly the gap between solidly average and clearly above average. It doesn’t make anyone a genius or a write-off. Still, we’d expect the higher-scoring person to generally find new learning and problem-solving easier, on average, if everything else is equal.
I'm not even sure it's mathematically appropriate to talk about IQ differences as there's no proof IQ is a linear metric. IQ is defined to have a normal distribution, unlike most things which are measured to have a normal distribution.
With IQ tests, the only thing you know that if 2 people fill it out and one of them scores higher, they have a higher IQ. Based on this you can sort people into a list by increasing IQ, but the standard distribution is implied not discovered.
This is like having a double-sided scale and a bunch of weights - you can similarly compare the weights to each other, and sometimes the arm of the scale will lean to the left, sometimes to the right, by a little or lot - you can postulate that the weights are normally distributed are normally distributed, but they are absolutely not required to be (I can choose them any way I like), so your assumption would be wrong. We know this because we have a direct, not just a comparative measure for weight.
I could make up an imaginary 'weight point' scale based on these comparisons, and say weights A is 5 WP heavier than B and C is also 5 WP heavier than A.
But A might be 100g, B might be 1g, and C might be 1kg.
This is what I think of when I see studies clamining the difference between 2 groups was 5 IQ points.
IQ scores are calibrated to be normally distributed with a standard deviation of 15. So 15 is one standard deviation. That's the difference between average, and being in the smartest 16% of the population. Or being in the smartest 16%, and being in the smartest 2% of the population.
15 points is significant difference.
If someone is 15 points above average, they are in 84th percentile, or in top 16%.
Well a sure component of test scores reflects test taking skills. Years ago, I purchased a book of a series of IQ tests, and my numerical result increased with every test. Another component is confidence. And another is ability. It is said by some that among the first big users of IQ tests was the US Army.
Well 100 would be your average person and 85 would be a person that has difficulty in school and at work with some tasks.
15 points is right around one standard deviation.
It's not nothing, but IQ is already a little squishy. No one's IQ is a single number. But the article also goes into problems with the study and other potential issues.
Basically, they're saying there is this pattern in the data as recorded, but there are multiple confounding factors and issues with collecting the data in the first place.
Isn't the whole point of IQ that it is a single number? Or I suppose potentially two numbers if the quotient was expressed as a fraction.
It's a single number in that if you take an IQ test one time you get one number, but that doesn't mean you'll get that exact number every single time you take an IQ test. Even ignoring more complex questions about them, your score on an IQ test will vary depending on simple things like how tired you are when you take it, so in practice there's some variance and you do not always get the same number every time you take a test.
This is a deeper question than it sounds. The "point" of a modern IQ test is to identify cognitive deficits to target interventions. It's abused widely among non-practitioners as a ranking of intelligence, which it is not.
I think it's widely used in research as a marker of intelligence (and for good reason - correlates really well with cognitive abilities).
I don't think we have to hash this out, because "marker of intelligence" and "ranking of intelligence" are not the same thing. A rank implies a reliable scale, which IQ doesn't provide.
(It's also just fine if we disagree about this --- researchers do too!)
I mean, nothing in the human body can be truly represented by a single number.
Even height and weight change throughout the day. People are typically taller and lighter in the morning than in the evening. Weight especially is variable, it can fluctuate up to 5 to 6 pounds.
Corollary: the “dumbing” down of public school (elimination of gifted programs, delay of algebra, etc) has a permanent impact on our society’s IQ.
I think it's inaccurate to call it a dumbing-down. It's more correct to label it as the largest stratification of education we've ever seen.
The smartest kids are smarter than ever before. They're absolutely rocking the house. The problem is that the "middle class has been gutted". Kids who were kinda smart, or kinda dumb, are now lumped in with kids who probably need Individual Education Plans (IEPs). This lowers the educational standard for almost all students - though of course the most well-off among us (educationally, rather than monetarily) are not only not suffering, they're thriving.
How are they rocking the house?
The standard public Chinese education turns out better 90th percentile than our 99.9th percentile house rockers.
How to confirm this? Talk to PIs and others hiring researchers. Home grown talent doesn’t even come close. It’s very upsetting.
> The smartest kids are smarter than ever before
Only if they’re wealthy or get extremely lucky and live near a randomly good school. By many metrics I was the smartest in my class, but my family had little money and lived in a rural area with a single underfunded school. I spent my days in class with kids that were still struggling to sound out “cat” in third grade. A few times a week I got to spend an hour in “gifted” class but that was mostly art projects, nothing that would help make up for the rest of the day being wasted.
>The smartest kids are smarter than ever before.
That is provably incorrect, as since Victorian times people lost around 14 to 23 IQ points on average. Notably, the corrected scores have continued on a downward trend for the past century.
People are not getting smarter, as recent events have shown. =3
https://www.youtube.com/watch?v=ZMHfBobgLSI
Sadly, this may be coming to NYC. In principle, I am all for trying to improve the baseline, but you cannot sacrifice better students to do that. Not only is it silly (how would this sacrifice exactly benefit worse students?), but unjust. Furthermore, education begins at home. Parents are the primary educators of children, not necessarily academically, but in the broader "life" sense. If the home environment is not conducive or supportive of education at school, you will be facing a very uphill battle.
The dumbing down of education goes further than what you note, though. Think of classical education and the formation of the human person (I'm not talking about "Dead Poets Society" ersatz, but the real deal). Think of the principles behind the trivium and quadrivium. In the best case, we are producing superficially technically savvy barbarians. Schools are effectively savage factories, and universities are laughable and should be ashamed of calling themselves universities.
Sooo... yeah, but its not what you think.
There are two different kinds of IQ tests: convergent and divergent. Convergent tests are more common and test either knowledge or pattern matching. These tests are called convergent because they are a center of truth and conformance to that truth is the measured performance criteria.
Divergent tests measure the individual's creativity and abstract reasoning. The source of truth is the quantity of diversity of results submitted by the participant.
The implicit success criteria for convergent testing is reading comprehension. A person with dyslexia, for example, will perform worse on these tests irrespective of their learning speed, learned knowledge, intellectual curiosity, or creativity. This is a form of bias. Other forms of bias include memorization of terms, such as SAT preparation.
To further complicate things these measures typically only account for academic intelligence. Other forms of intelligence include social intelligence, spatial intelligence, creativity, conscientiousness, and so forth. In the concept of multi-dimensional intelligence, which is what is actually addressed in practice in the real world after high school, academic intelligence alone has very little benefit. Its like height in basketball where after 6.5ft all other factors become more important for all participants.
> A person with dyslexia, for example, will perform worse on these tests irrespective of their learning speed, learned knowledge, intellectual curiosity, or creativity. This is a form of bias.
No, it's what they were designed to do - help identify people that may need specialized help because issues like learning disabilities.
> To further complicate things these measures typically only account for academic intelligence. Other forms of intelligence include social intelligence, spatial intelligence, creativity, conscientiousness, and so forth.
The two major IQ tests (Stanford Binet and Wechsler) test visual-spatial reasoning, abstract reasoning, verbal and nonverbal reasoning, working memory, processing speed, inductive/deductive reasoning, attention, concentration, etc.
These tests, if properly administered (individually in person by a professional) work well for figuring out who in school needs extra attention or requires further evaluation. Their use beyond that is questionable.
The way we raise our kids will have major differences in their IQ... Keep in mind there is a 9 point difference between those raised apart vs those raised together. This is not a comparison between those who receive special education efforts vs those who don't, I assume that would be much larger.
If results are influenced by education or upbringing then the test does not measure "intelligence" as intrinsic ability.
That's actually probably a very difficuly topic: cam we devise a test that measures intelligence irrespective of education (i.e. training)?
This begs the question. How much of "intelligence" is intrinsic? We don't know. It's not zero (we know there are genetically-mediated cognitive deficits), but it's very plausibly not meaningfully intrinsic in most people.
It's always been an insane claim that knowledge can't help you utilize knowledge. It's just an excuse to be racist and classist. Everything around IQ is linear, dumb thinking, suitable for trying to reduce everything to a scalar. IQ is a good tool for figuring out whether your child has a physical issue with their brain development, or maybe a hearing problem.
There's nothing logical about brains; they work on heuristics built on reflexes and associations. Logic is something we learn through physically interacting with the world, and from sitting in class. As you learn it, your thinking about other things gets better. When you read great old Chinese philosophers like Mozi struggling but managing to make good arguments without syllogism, it's a reminder that syllogism is something that is invented. It survives because of its unreasonable effectiveness compared to considering what the ancient heavenly kings had done.
It also helps your soul to remember that a lot of people you're arguing with have no idea what a logical argument looks like. They think that an argument is when you try to shut the other person up. It's not their fault, they just literally don't know. They're not arguing in bad faith, they don't know what good faith looks like because the educational system has failed them.
Outside of a fascination with Intellectual Supremacism, why are people so obsessed with the genetic basis of IQ scores?
Presumably one could do similar identical twin studies on half-marathon race times and SAT test scores. Does no one bother with those, because widespread awareness of half-marathon training regimens and SAT prep courses would spoil the (desired) illusion of some "innate superiority of blood" being measured?
How about the fact that the human mind and genetics are simply fascinating and interesting topics. I would imagine that people don't care as much about running and high school testing because they are fairly niche interests relative to abstract thinking in general, something that almost everyone spends much of their life doing.
There are many reasons. For example, there are large policy implications for schooling and education.
> Presumably one could do similar identical twin studies on half-marathon race times and SAT test scores.
Yes, and some of those have been done: SAT scores correlate very strongly with IQ.
Exactly, it is on of the least remarkable results that intelligence is partly heritable. Obviously intelligence is partly genetic, obviously your ability to run a marathon is also partly genetic. Why one has devolved into a bizarre case where so much effort is spent on the minute details of this question and the other is accepted basically as completely uncontroversial (given who is currently competing for top times at marathons it is basically inarguable) is very strange.
[flagged]
Speaking of IQ, when I listen to interviews and podcasts of supposed high IQ people like Mark Zuckerberg and Elon Musk, I struggle to see signs of intelligence (not saying that they don’t have above average IQ). I find that these 2 have a hard time sifting through the noise (competing thoughts) in their head to get to want they want to say (the signal).
When I listen to Steve Jobs, I hear someone who has very strong ability to sift through noise. So Jobs couldn’t see engineering in a new way like Elon does but Elon couldn’t do what Jobs did either.
Regarding Zuckerberg, from what I’ve read he is the Bill Gates type where he has the traditional variant of high IQ, aka raw hardware/horsepower but lower on creativity/imagination side.
So intelligence seems to have different shapes and sizes.
Don't read this as a defense of Zuck or Musk, but some of the most brilliant people I've known sometimes had trouble getting thoughts out. It wasn't that they couldn't see their goal, but that how they thought of it was changing even as the words were leaving their mouths. Have you ever tried to explain while you were still working it out in your head? "See, it's kind of like A..., except that one part is closer to B, and that's interesting because, well, B and C have a relationship that's closer than A and C, so it might be more appropriate to say it's more C-like, except where A is..."
That didn't mean they were unintelligent. It meant that them trying to explain the flurry of ideas in their head was like me trying to type with mittens.
Of course, knuckleheads can also sound like that, but for different reasons. That kind of rambling doesn't imply that the speaker is a dumbass. It definitely doesn't guarantee that they're a genius.
Interesting perspective, thanks
IQ tests are very deceptive and often misused.
I don't dispute the differences in intellectual capacity between people but IQ tests are like weight lifting contests between people who didn't train to lift weights.
I don't understand why the scientific community keeps using methods that are so flawed. Perhaps it's due to my own lack of information.
The only way I could see IQ tests being a valid measurement of intellectual capacity is if participants were brought to the same level of knowledge and skills, and then were made to train for the IQ test using the exact same means, and even then cultural and language barriers must be accounted for.
Even if these twins have an identical intellctual capacity and they both had the same exact education, and same exact grades, that still doesn't mean they applied and exercised their brains in the same way for the questions of the IQ tests.
IQ tests to measuring intellectual capacity (instead of knowledge level) are like polygraphs to mesasuring truthfulness in terms of accuracy.
For measuring strength, i stand by my earlier correlation with weight lifting. If two people can dead-lift 500lbs, I only know that both participants are have reached that level of strength. What I don't know is how much effort each participant put into getting to that level of strength, which would tell me their natural muscular capacity per effort. IQ tests deceptively seem like they tell you what someone's capacity for intellect is, but they only tell you where that person is right now. Maybe that person worked hard and the score is their max, maybe the person rarely applies their brain to demanding tasks and this is their mid-level capacity.
My point is,it isn't just education or just genetics, it is also personality, effort, motivation. For all I know,someone with double-digit IQ score can work out their brain for a year and hit genius level. Choice. Can a person choose to be a literal idiot and succeed? Certainly, people choose to be incurious and ignorant all the time. Women can body-build and be as strong as many regular men for example. But simply because of a hormone difference, they have to work out a lot more than men to reach the same level of muscular strength. And men who never work out can be as weak as women who never work out.
There is also the question of brain development. Maybe the effort you put has different effect depending on age. small efforts at a young age at applying your brain might have huge impacts, where as if you're a teenager or an adult, applying the same effort might yield less results.
I mean, personally, if I tired, if I ate too much, or too little,if didn't get enough sleep, if I am distracted by something, if I'm unprepared and thus second-guessing myself, these are some of the things that throw me off wildly when taking exams. I've seen huge differences by simply getting enough sleep and calories.
There is no way that asking the same set of questions to to large population (even with control populations in place) can account for all the presumptions of the test.
It is very hard to increase your own IQ score meaningfully by prepping to the test. That is why its a good measure.
This just isn't true. It's definitely possible to train for them if you want.
even if you prepped since you were 5 years old? and what if you're unprepared, will the decrease be significant? I mean of course, even with weight lifting, you can't prepare even a year in advance and show a significant difference over professional competitors. But people do peak. Are you saying Einstien's IQ would be the same if he was tested at 50 yo, and he was working in the coal mines in virginia his since 5 yo? If so, that's indeed enlightening, I'd love to see any research that backs up that claim.
The area where I see IQ failing miserably is in accounting for different types of intelligences. Like the parent commenter said, people solve the same problem differently. I'm keenly aware of this because I'm neurodivergent and I was continuously judged as poor in subjects that I liked the most and was very good at. I had to reformulate every problem and solution in a way that I understood (mostly based on spatial intelligence), leaving me at a significant disadvantage of time. And even then, those solutions were sometimes rejected, despite being objectively correct and clear in the dumbest possible way - just because it didn't follow the textbook pattern. That continued until we were in a situation where we had to solve the problems ourselves.
The reason I mention this is because I see the exact same problem with IQ tests. It emphasizes certain types of intelligences and ignores others. Human intelligence is extremely multidimensional and a single number is simply incapable of representing its overall quality. For example, there are people who score poorly in IQ, but are superhuman in remembering places and navigating their way around complex routes. Meanwhile, many high scoring ones become hopelessly lost, failing to institute or follow even simple mitigation strategies for that problem. There are situations where this determines whether you live or die, and your IQ score becomes worthless indicator of even your survival.
Afaik iq correlates highly with g-factor (scientific name for general intelligence). We don't have much better ways to measure it.
Overall, afaik, again, problem solving often needed in our society correlates with this: "general intelligence". When people say: "But there are many ways to approach a problem, many types of intelligence".
This very well may be, but data consistently shows that people having high IQ scores solve different problems better. So, contrary to popular belief, people having higher iq in general have better talent for languages, same with understanding others emotions and using it to your advantage ("emotional intelligence").
As for reframing an issue and solving it in a different way. This may be valid approach (to teach people etc). But IQ is also time-measured. If your new approach does not help you to solve previously unseen problems quickly, it is not noticably increase your intelligence.
Thus, we see consistently that people cannot really prepare for iq tests much. You only get a few points more if you prepare. Same difference as if you are sleep-deprived.
G isn't so much the "scientific name for general intelligence" so much as one explanation for the positive manifold of intelligence. There are others: mutualism and sampling.
I'm not sure you're right about the trainability of tests, either.
> The area where I see IQ failing miserably is in accounting for different types of intelligences.
We don't have any real evidence of "different types of intelligences". People with high IQ just tend to be better at all cognitive tasks than people with lower IQ. That's why it's considered a reasonably good measure of the g factor. People with higher IQ also have better memory recall.
While you're right that that IQ isn't the full story of a person's abilities, it's use for diagnosing people that may face challenges in a traditional learning environment well motivated. All of the examples you mention of people scoring poorly on IQ and facing challenges in school supports the use of IQ, it doesn't discount it.
> We don't have any real evidence of "different types of intelligences". People with high IQ just tend to be better at all cognitive tasks than people with lower IQ.
That would imply that a person with a high IQ score is uniformly better at every task compared to a person of lower IQ. But that's absolutely not what's observed. People show better skill levels at different tasks. A person who's bad at math may instead be a natural-born singing sensation. This is why I said that a single number is just incapable of representing intelligence. The mathematical dimensionality of the quantity called intelligence is just too high. You'll need a bunch of numbers - a vector, at the minimum. And even that can't be used for comparing people's intelligence. It can only be used for assessing someone's suitability for a particular task. Basically, IQ score is a scalar that is used to represent a vector, discarding very important information in the process.
> People with higher IQ also have better memory recall.
There are people who recite entire Shakespeare plays without understanding a sentence in it. I have also seen people who recall long derivation sequences using Maxwell's equations exactly and score well on exams, without having any clue as to what the del (∇) operator even means in practice. Needless to say, they have difficulty adapting that information to a novel situation. Memory recall is just one aspect of intelligence. There are other cognitive skills required to make that memory useful. This again goes back to my previous conclusion. A single number is quite meaningless at best and utterly misleading at worst.
> it's use for diagnosing people that may face challenges in a traditional learning environment well motivated. All of the examples you mention of people scoring poorly on IQ and facing challenges in school supports the use of IQ, it doesn't discount it.
On the contrary, it just misleads the educators to the student's real abilities. It ignores and hides the atypical talents that students possess. All it can do is to predict if the student will do well or not in generalized education and standardized testing, because they follow the equally illogical concept of using standardized pedagogy on everyone, neglecting their uniqueness. So, even if we accept your claim that the IQ score predicts their performance in schools, it means that the score has the very narrow scope of testing someone's capability to learn under a standardized curriculum. That's far too narrow to be identified as an indicator of intelligence. And even after identifying the students who are going to struggle, IQ gives no clue as to how to remedy that. All it does is put a dunce label on these students who might otherwise have done better. It does more harm than good.
> That would imply that a person with a high IQ score is uniformly better at every task compared to a person of lower IQ.
No it doesn't, that's why I said "tend to".
> On the contrary, it just misleads the educators to the student's real abilities. It ignores and hides the atypical talents that students possess.
The test simply doesn't measure atypical talents, so there's no "hiding". That's why I said the test is used to identify people who will face challenges. What happens next has nothing to do with the IQ test and everything to do with other methods of assessment that should be used. Every test has limitations that those applying the test should understand.
> That's far too narrow to be identified as an indicator of intelligence
That's why it's a (strong) correlation and not a strict equality.
Look, jumping height has been empirically demonstrated to be a good predictor of overall athletic performance, and you're coming here and saying that there exist some people without legs that can do more pull-ups than most people with legs. That's all well and good, but that doesn't somehow refute the idea that jumping height is a good predictor of overall athletic performance. It's just a complete red herring. If you happen to encounter someone without legs, then use a different test.
People with legs who don't exhibit good jumping heights but are very good at some specific athletic challenge could exist, but they tend to be good only at that specific challenge (and ones that are very, very similar). This is a completely different thing than the general factor of fitness, which is correlated with all sorts of physical abilities, just like general factor g is correlated with all sorts of cognitive abilities.
I think you are being a little bit conclusory here. There are in fact serious theories of intelligence that admit to multiple distinct intelligence factors. Mutualism is the obvious counterexample. You're arguing from what are arguably exploratory statistical correlations as if the correlation demonstrated the underlying cause. You might be right, but you're not presenting evidence for that position.
On the contrary, it is so easy to increase your IQ score by prepping that test administrators ask the test takers not to do it because it invalidates the results.
[dead]
I don't know how many more nails in the coffin of heritable intelligence are needed before public policy capitalizes on the opportunity here. We have seen huge shifts in intelligence in places like Korea after armistice, in China after economic and cultural reforms, and in the U.S. after COVID and smartphones. We have seen that individual tutoring reliably creates extreme outliers from the examples of Avant-garde, Williams, Tao, Polgár, the Hungarian Martians, etc. Aside from disorders, any heritable differences can be easily dwarfed by environmental effects, and some disorders, like OCD, only magnify environmental effects.
> Of the 87 pairs, 52 experienced a similar type and duration of schooling within a similar location. In fact, 25 of these pairs attended the same school for some period of time. Analysis revealed that ‘Educationally Similar’ TRA pairs have an ICC of 0.87 ± 0.02 (n = 52)
So this study has 87-52-25=10 data points? Am I reading this correctly? Quite the reach to conclude what the article claims, if so.
> With this said, it is important to note that the ‘very dissimilar education’ group consisted of only 10 TRA pairs. This small N is not a shortcoming of this analysis, per se. Rather, this is a shortcoming of the TRA field; these 10 pairs represent the entirety of individual data published over the last century.
Authors plead innocence!
I read the quoted sentence as "25 of the 52" -- not an additional 25 of the 87 -- because of the word "these".
52 pairs similar schooling 25 of these had very similar schooling
Thus the math becomes "87-52=35". 35 data points. Not spectacularly high regardless.