If there's a club for people who are over 1m95 tall, no other qualification needed, whose point is to celebrate their height, is it likely to be a good source of information about say the athleticism of tall people?
One would think that the more athletic over-1m95s would spend their time doing sports that require height (basketball, swimming, volleyball, water polo) and therefore that club would end up being about Big&Tall stores & how close together airline seats are.
The problem is that quality-control mechanisms in science have broken down.
There is no professional penalty for publishing garbage studies.
When science was small (before WW2) peer estimations mattered to careers. Since science got big, few scientists, even in a sub-field, know each other personally.
I tend to believe the idea that, at the highest levels of intelligence, mental health declines. I think the effect is definitely not as large as claimed, and I’m not sure if it represents a higher relative risk compared to the overall population or just to those a few IQ points lower. But I buy that there is some reversal.
Among Ashkenazis, for example, there is a hypothesis that some diseases like Tay-Sachs are tradeoffs for their high intelligence. While I am admittedly far from an evolutionary biologist, it makes sense that in equilibrium there should be some reproductive fitness tradeoff of high intelligence.
There is absolutely no valid evidence to suggest a negative impact on mental health by high iq. Again, Ashkenazi Jews are not a representative for all intelligent people.
True; they are a data point, though. I think it's important to be clear that there is more than one way to conceptualize this:
1. IQ has a linear association with negative outcomes, like mental illness.
2. IQ has a curvilinear association with most positive outcomes, such that the ideal level is around +1.5 Sigma(?) and things plateau and eventually get worse from there.
I think #1 is crazy; I think #2 is likely correct; in fact you can detect a hint of it in the US Veteran data graphed above, where the 120 IQ band is lower than the regression line. Yes I know, Emil already told us we weren't supposed to care about that, but I'm mean... there it is, below the line.
I think it is likely that even the non-self-selected samples suffer from selection problems. This is most obvious if you study Nobelists or Field Medalists: highly intelligent but probably high work ethic and motivation as well. Even kids who exhibit interest in mathematics, participate in math competitions and olympiads, are likely to be those of high intelligence who are also well adjusted and have a good work ethic.
The problem is that you may only find an effect at the highest percentiles of intelligence, but it is impractical to sample those people at high enough numbers without running into selection issues.
Use the same arguments about the super high people. If you were to claim that at a certain extreme height, some effect was reversed, it would sound improbable?
I don’t have much evidence of a reversal at very high levels of intelligence. I’m adopting it as a prior mostly because it makes some intuitive sense. It is somewhat of an extraordinary claim after all. As for your example, even in height there are some nonlinearities: super tall people are probably considered less attractive than regular tall people. They are also that tall usually due to hormonal imbalances which causes all sorts of problems. So I don’t think nonlinearities and reversals at the far extreme percentiles are a particularly unlikely pattern
I once had someone share this study, and my short reply was that Mensa selects for people who don’t feel like their high IQ is sufficiently appreciated by others.
I know, there are obviously good reasons to expect positive outcomes will correlate with intelligence across most IQ levels: low genetic load will manifest itself as high function across many traits, and cross-assortative mating for socially valuable traits will encourage everything people like to cluster together. But the *unusual* intelligence --> social isolation --> maladjustment & sadness pathway also seems pretty obvious, and there are some suggestions of it in the literature as well, for instance:
"a significant curvilinear relationship between intelligence and coital status was demonstrated; adolescents at the upper and lower ends of the intelligence distribution were less likely to have sex."
I mean seriously, not having anybody around who can understand your pillow talk is just going to be a real bummer no matter how you look at it.
Another problematic aspect of using “mental illness” as a precise measurement criterion is that doing so trivializes psychotic brain diseases such as Schizophrenia. Research shows that Schizophrenics possess lower IQs before the prodrome phase and diagnosis. Psychosis is almost always associated with lower pre-onset IQ scores. The genius outliers who suffered from severe mental illness -- like John Nash -- probably had high enough IQs to compensate for the noticeable deficits.
But yes, I believe, based on purely anecdotal evidence, that intelligence correlates with neuroticism, the latter of which is always found in OCD patients. However, researchers should partition hyperactivity and rumination from psychosis.
Lastly, high intelligence confers upon the elderly idiopathic protective effects from developing degenerative diseases like Alzheimer’s. The literature should focus on the healthy advantages of high intelligence — not neuroticism.
The people you meet at their member-only events and on their forums come across as quite angry at the world, they have A LOT of conspiracy theories, all accompanied by a superiority complex despite most of them having not really done anything spectacular.
If there is anything Jordan Peterson gets right, it's how hard it is for many hyper-smart people to come to terms with how useless they are in the world. It's much easier to blame the world than themselves
Plus, I think there are a small-but-significant percentage of Mensans who have hacked the test through repetition just so they can get in and appear smart to others.
Theory A: Mensa attracts those with verbal tilt (the emotionally unstable relative to the same IQ class), and therefore "real smarts" avoid it like the plague. Possibly the bias towards Africans and Middle Easterners vs East Asians can be checked
Theory B: "Mania" as a mental illness is rife in Mensa, and that it, along with some other Cluster-B mental illnesses are inversely correlated with the p-factor (emotional instability). It is likely that the class of illnesses are in the books for reasons of "social inconvenience".
Both your theories don't seem that different, as they both point towards there being more emotional instability at Mensa. Will address what I think you mean the best I can.
Regarding theory A, I think this is somewhat likely, and what Gwern says about self-diagnosing autism comes across as true in my experience. For instance, the first two people I met from there claimed to have autism, but I couldn't help but think that was highly unlikely. Autism is most often characterised by deficits in verbal ability relative to other aspects of intelligence, so having it often makes scoring well on a test specifically testing for verbal intelligence extremely difficult. That being said, their experience of the world is obviously much different than it is for most people so you can empathize with why they self-diagnose (also a lot of them would back their own opinion over any psychiatrist's).
Also, and I can only speak for UK Mensa, the members appear to be overwhelmingly white, so not sure your point about bias holds up. That being said, I think it's right about the top 1%-0.1% of smart people not being that interested in Mensa. I'm willing to bet that many of the people I've met in the mathematics departments of Cambridge (UK) would qualify for Mensa, but they would never apply. If for no other reason than they have better things to do, and that many of them come from families where that level of smarts is the norm, so they never really ponder their intelligence all that much.
As for Theory B, there may be a fair few mental pathologies in there but 'mania' probably isn't one of them, or at least it's not prevalent. In fact, I think many of them could due with being more manic to actually get up off their backside. I'm not too familiar with the literature on the p-factor, but I will say people with some notable form of psychopathology seem to be MORE likely to have others too.
Elaborating on Theory A: Mensa attracts those that are more eloquent, or those who has a heavy bias for the arts rather than STEM. Standard "High Functioning Autism" is likely to NOT have verbal tilt, but bipolar and "Aspergers" are likely to have it.
The measure of verbal tilt can often be proxied through MENA population with higher exposure to verbal tilt vs APAC population with less exposure to verbal tilt (whites as a relative center).
Again, can only speak about UK Mensa, but it is overwhelmingly white male with a lot of individual differences across interests. So you get some with verbal tilt as you say, and some who are much more scientifically-oriented with little interest in the arts. If I had to guess there probably is a bit more verbal tilt than not, but really not sure.
Apologies I wasn't very clear. I agree with you, that most smart people are doing incredibly well in the world with relative ease.
I was referring to the types of people Mensa attracts. People who are smart, and know they're smart, but don't really put their gifts to any meaningful use.
To use an example from a different context, imagine being as attractive and built as Chris Hemsworth is, yet being completely useless when it comes to interacting with women. In a situation like that you would have a massive advantage over at least 98% of men when it comes to attracting the opposite sex. Yet, despite your unearned gifts, you would still struggle to get a girlfriend or even get laid.
This is the equivalent situation for a lot of Mensans, and probably a sizeable chunk of very smart people in general.
At an IQ of 130, I am near the borderline to qualify for MENSA. I have none of the mental or emotional problems that the "studies" find correlated with MENSA-level intelligence. My principal problem in life has been being hampered by incompetents, be they colleagues, HOA members, attorneys, or physicians. I experience a small degree of stress from such interactions. Perhaps some of the MENSA members' complaints are related to similar interactions, but mostly not being INTJ's they don't realize that getting things done correctly is more important than being sociable. That is, their reported emotional problems are fundamentally problems of dealing with the less intelligent world, but they let their frustration damage other relationships which could be normal.
No I'm with you; Mensa doesn't look too impressive. No attraction there.
But now imagine being so smart that people like you - who describe their experiences in terms of outdated psychological schema like the MBTI, downplay the importance of sociality, and think of themselves as capable, successful, and basically always right - come across as painfully dim. You admit to feeling frustrated with incompetent colleagues and attorneys, but how intelligent do you think they are? One standard deviation below you? Two? Just complete the analogy: how is someone with an IQ of 160 supposed to have a relationship with you that "could be normal?"
The fallacy in your oblique ad hominem is that I have of necessity routinely experienced dealing with and depending upon persons 1 SD below me, and who have far less ability to apprehend the big picture; whereas as far as I know I have never worked with someone who has a 160 IQ; thus I have never impeded a genius's attempt to get a project accomplished. I have typically been able to work with mandatory colleagues for a common goal. Lousy attorneys and physicians I bid so long. The normal relationships that I posited as perhaps affected by frustration are clearly not the professional relations that generated the frustration. Trolls often juxtapose disparate elements of a comment so that they can troll.
Sociability can be p-factor loaded ("we are all anxious of the same thing"), or inversely loaded ("I can be calm enough to work with others"). Maybe there is an environment effect that requires smart people to be sociable in densely populated areas? Or that the amount of opportunities for xNTx straight edge does not match the population size? Will definitely check the Gervais Principle for some guidance on grinders vs leaders.
If there's a club for people who are over 1m95 tall, no other qualification needed, whose point is to celebrate their height, is it likely to be a good source of information about say the athleticism of tall people?
One would think that the more athletic over-1m95s would spend their time doing sports that require height (basketball, swimming, volleyball, water polo) and therefore that club would end up being about Big&Tall stores & how close together airline seats are.
Adverse selection is almost certainly guaranteed.
Dumb people like thinking that there is something wrong with smart people so there will always be demand for studies like this.
The problem is that quality-control mechanisms in science have broken down.
There is no professional penalty for publishing garbage studies.
When science was small (before WW2) peer estimations mattered to careers. Since science got big, few scientists, even in a sub-field, know each other personally.
We need new mechanisms.
I tend to believe the idea that, at the highest levels of intelligence, mental health declines. I think the effect is definitely not as large as claimed, and I’m not sure if it represents a higher relative risk compared to the overall population or just to those a few IQ points lower. But I buy that there is some reversal.
Among Ashkenazis, for example, there is a hypothesis that some diseases like Tay-Sachs are tradeoffs for their high intelligence. While I am admittedly far from an evolutionary biologist, it makes sense that in equilibrium there should be some reproductive fitness tradeoff of high intelligence.
There is absolutely no valid evidence to suggest a negative impact on mental health by high iq. Again, Ashkenazi Jews are not a representative for all intelligent people.
True; they are a data point, though. I think it's important to be clear that there is more than one way to conceptualize this:
1. IQ has a linear association with negative outcomes, like mental illness.
2. IQ has a curvilinear association with most positive outcomes, such that the ideal level is around +1.5 Sigma(?) and things plateau and eventually get worse from there.
I think #1 is crazy; I think #2 is likely correct; in fact you can detect a hint of it in the US Veteran data graphed above, where the 120 IQ band is lower than the regression line. Yes I know, Emil already told us we weren't supposed to care about that, but I'm mean... there it is, below the line.
I think it is likely that even the non-self-selected samples suffer from selection problems. This is most obvious if you study Nobelists or Field Medalists: highly intelligent but probably high work ethic and motivation as well. Even kids who exhibit interest in mathematics, participate in math competitions and olympiads, are likely to be those of high intelligence who are also well adjusted and have a good work ethic.
The problem is that you may only find an effect at the highest percentiles of intelligence, but it is impractical to sample those people at high enough numbers without running into selection issues.
Use the same arguments about the super high people. If you were to claim that at a certain extreme height, some effect was reversed, it would sound improbable?
I don’t have much evidence of a reversal at very high levels of intelligence. I’m adopting it as a prior mostly because it makes some intuitive sense. It is somewhat of an extraordinary claim after all. As for your example, even in height there are some nonlinearities: super tall people are probably considered less attractive than regular tall people. They are also that tall usually due to hormonal imbalances which causes all sorts of problems. So I don’t think nonlinearities and reversals at the far extreme percentiles are a particularly unlikely pattern
Nice example, thank you
I once had someone share this study, and my short reply was that Mensa selects for people who don’t feel like their high IQ is sufficiently appreciated by others.
First born children have ~3 IQ points advantage over later borns, and are also more prone to allergies, so that correlation could be real.
Hi Emil - You mention Terman's sample more than once. That was the sample popularized at https://prometheussociety.org/wp/articles/the-outsiders/ , right? Looks like there was maladjustment at high IQ there.
I know, there are obviously good reasons to expect positive outcomes will correlate with intelligence across most IQ levels: low genetic load will manifest itself as high function across many traits, and cross-assortative mating for socially valuable traits will encourage everything people like to cluster together. But the *unusual* intelligence --> social isolation --> maladjustment & sadness pathway also seems pretty obvious, and there are some suggestions of it in the literature as well, for instance:
https://www.sciencedirect.com/science/article/abs/pii/S1054139X99000610
"a significant curvilinear relationship between intelligence and coital status was demonstrated; adolescents at the upper and lower ends of the intelligence distribution were less likely to have sex."
I mean seriously, not having anybody around who can understand your pillow talk is just going to be a real bummer no matter how you look at it.
Another problematic aspect of using “mental illness” as a precise measurement criterion is that doing so trivializes psychotic brain diseases such as Schizophrenia. Research shows that Schizophrenics possess lower IQs before the prodrome phase and diagnosis. Psychosis is almost always associated with lower pre-onset IQ scores. The genius outliers who suffered from severe mental illness -- like John Nash -- probably had high enough IQs to compensate for the noticeable deficits.
But yes, I believe, based on purely anecdotal evidence, that intelligence correlates with neuroticism, the latter of which is always found in OCD patients. However, researchers should partition hyperactivity and rumination from psychosis.
Lastly, high intelligence confers upon the elderly idiopathic protective effects from developing degenerative diseases like Alzheimer’s. The literature should focus on the healthy advantages of high intelligence — not neuroticism.
This definitely sounds like the Mensans to me.
The people you meet at their member-only events and on their forums come across as quite angry at the world, they have A LOT of conspiracy theories, all accompanied by a superiority complex despite most of them having not really done anything spectacular.
If there is anything Jordan Peterson gets right, it's how hard it is for many hyper-smart people to come to terms with how useless they are in the world. It's much easier to blame the world than themselves
Plus, I think there are a small-but-significant percentage of Mensans who have hacked the test through repetition just so they can get in and appear smart to others.
Theory A: Mensa attracts those with verbal tilt (the emotionally unstable relative to the same IQ class), and therefore "real smarts" avoid it like the plague. Possibly the bias towards Africans and Middle Easterners vs East Asians can be checked
Theory B: "Mania" as a mental illness is rife in Mensa, and that it, along with some other Cluster-B mental illnesses are inversely correlated with the p-factor (emotional instability). It is likely that the class of illnesses are in the books for reasons of "social inconvenience".
Both your theories don't seem that different, as they both point towards there being more emotional instability at Mensa. Will address what I think you mean the best I can.
Regarding theory A, I think this is somewhat likely, and what Gwern says about self-diagnosing autism comes across as true in my experience. For instance, the first two people I met from there claimed to have autism, but I couldn't help but think that was highly unlikely. Autism is most often characterised by deficits in verbal ability relative to other aspects of intelligence, so having it often makes scoring well on a test specifically testing for verbal intelligence extremely difficult. That being said, their experience of the world is obviously much different than it is for most people so you can empathize with why they self-diagnose (also a lot of them would back their own opinion over any psychiatrist's).
Also, and I can only speak for UK Mensa, the members appear to be overwhelmingly white, so not sure your point about bias holds up. That being said, I think it's right about the top 1%-0.1% of smart people not being that interested in Mensa. I'm willing to bet that many of the people I've met in the mathematics departments of Cambridge (UK) would qualify for Mensa, but they would never apply. If for no other reason than they have better things to do, and that many of them come from families where that level of smarts is the norm, so they never really ponder their intelligence all that much.
As for Theory B, there may be a fair few mental pathologies in there but 'mania' probably isn't one of them, or at least it's not prevalent. In fact, I think many of them could due with being more manic to actually get up off their backside. I'm not too familiar with the literature on the p-factor, but I will say people with some notable form of psychopathology seem to be MORE likely to have others too.
,
Elaborating on Theory A: Mensa attracts those that are more eloquent, or those who has a heavy bias for the arts rather than STEM. Standard "High Functioning Autism" is likely to NOT have verbal tilt, but bipolar and "Aspergers" are likely to have it.
The measure of verbal tilt can often be proxied through MENA population with higher exposure to verbal tilt vs APAC population with less exposure to verbal tilt (whites as a relative center).
Again, can only speak about UK Mensa, but it is overwhelmingly white male with a lot of individual differences across interests. So you get some with verbal tilt as you say, and some who are much more scientifically-oriented with little interest in the arts. If I had to guess there probably is a bit more verbal tilt than not, but really not sure.
Most super smart people are doing great socially and in their jobs. Why do you say they are useless?
Apologies I wasn't very clear. I agree with you, that most smart people are doing incredibly well in the world with relative ease.
I was referring to the types of people Mensa attracts. People who are smart, and know they're smart, but don't really put their gifts to any meaningful use.
To use an example from a different context, imagine being as attractive and built as Chris Hemsworth is, yet being completely useless when it comes to interacting with women. In a situation like that you would have a massive advantage over at least 98% of men when it comes to attracting the opposite sex. Yet, despite your unearned gifts, you would still struggle to get a girlfriend or even get laid.
This is the equivalent situation for a lot of Mensans, and probably a sizeable chunk of very smart people in general.
At an IQ of 130, I am near the borderline to qualify for MENSA. I have none of the mental or emotional problems that the "studies" find correlated with MENSA-level intelligence. My principal problem in life has been being hampered by incompetents, be they colleagues, HOA members, attorneys, or physicians. I experience a small degree of stress from such interactions. Perhaps some of the MENSA members' complaints are related to similar interactions, but mostly not being INTJ's they don't realize that getting things done correctly is more important than being sociable. That is, their reported emotional problems are fundamentally problems of dealing with the less intelligent world, but they let their frustration damage other relationships which could be normal.
No I'm with you; Mensa doesn't look too impressive. No attraction there.
But now imagine being so smart that people like you - who describe their experiences in terms of outdated psychological schema like the MBTI, downplay the importance of sociality, and think of themselves as capable, successful, and basically always right - come across as painfully dim. You admit to feeling frustrated with incompetent colleagues and attorneys, but how intelligent do you think they are? One standard deviation below you? Two? Just complete the analogy: how is someone with an IQ of 160 supposed to have a relationship with you that "could be normal?"
The fallacy in your oblique ad hominem is that I have of necessity routinely experienced dealing with and depending upon persons 1 SD below me, and who have far less ability to apprehend the big picture; whereas as far as I know I have never worked with someone who has a 160 IQ; thus I have never impeded a genius's attempt to get a project accomplished. I have typically been able to work with mandatory colleagues for a common goal. Lousy attorneys and physicians I bid so long. The normal relationships that I posited as perhaps affected by frustration are clearly not the professional relations that generated the frustration. Trolls often juxtapose disparate elements of a comment so that they can troll.
Sociability can be p-factor loaded ("we are all anxious of the same thing"), or inversely loaded ("I can be calm enough to work with others"). Maybe there is an environment effect that requires smart people to be sociable in densely populated areas? Or that the amount of opportunities for xNTx straight edge does not match the population size? Will definitely check the Gervais Principle for some guidance on grinders vs leaders.