3 ms·

So a [gigantic meta analysis](https://www.sciencedirect.com/science/article/pii/S0149763421000804 https://www.sciencedirect.com/science/article/pii/S014976342..

So a [gigantic meta analysis](https://www.sciencedirect.com/science/article/pii/S0149763421000804 https://www.sciencedirect.com/science/article/pii/S014976342...) of thirty years of studies on sexed structural and functional differences in human brains found zero evidence of any differences, a completely overlapping distribution, but as soon as "big data AI" is used suddenly not only are there differences, there's literally zero overlap? Count me suspicious. I think I'm going to trust the meta-analysis of 30 years worth of wide-ranging scientific study over the brand new study that's just throwing whatever fad is currently in the vogue at the problem to see what happens.

im3w1l 2y ago
Sounds like a case of "Old measuring instrument cannot tell two things apart. New more precise instrument can tell them apart"
- logicprog 2y ago
  They're using the same measuring instruments, though. They're just feeding them through AI. Maybe, though. Let's see if it replicates.
courseofaction 2y ago
Without making any claims about gender or non-binary people (not my wheelhouse, I simply don't know), there's ample evidence to suggest statistically significant population-level differences between males and females on a many cognitive measures. I don't see how it's surprising that an new generation of signal-detection tool finds population-level differences in the brains that produce these cognitions.
derbOac 2y ago
I think the linked news article is a little misleading, although I share your skepticism. I'd like to see these results replicated rigorously on still new sets of data by independent researchers; I wouldn't be surprised either way, if the results did or did not replicate. However, the news article seems to spin this as "male and female brains are totally different entities that bear no relationship with one another." Although I haven't reread it carefully, it seems like the article is saying something more like "you can identify gender-specific patterns, and those gender-specific components relate to things like cognitive ability gender-specifically". It's not that you can't find overlap — that that wasn't the focus of the study — it's that if you go looking for differences, you can find them.
- RoyalHenOil 2y ago
  It seems to me that in order for male and female brains to be functionally the same, they would need to be physically different to account for the extreme hormonal differences. When you give a man a female dose of hormones or a woman a male dose of hormones, it has a very big effect on their mood, behavior, and mental wellbeing. This change is much, much bigger than the average diffences we see between men and women. For example, an average man with an average woman's level of testosterone will experience a MUCH higher level depression, listlessness, and sexual disinterest than the average woman experiences. This strongly implies that human brains must correct for these huge hormonal differences. Basically, in order for male and female behavior to be similar, their brains must differ. If their brains are the same, then hormones will have a much, much bigger influence on male and female behavior than what we actually see in reality. Hormone-correcting brain differences would also imply that it's possible for people to be born with some type of intersex brain condition, and that these individuals would benefit greatly from receiving hormone therapy to bring their hormone levels in line with their brains. And this, indeed, seems to be something we see occasionally. (In case anyone cares or thinks it is relevant, I wish to note that I am a cisgender woman and I do not think that there are huge innate differences in men's and women's mentality — certainly nothing like on the level that testosterone/estrogen/etc levels would predict. I think most of the differences we do see are environmental, which is why these gaps have been closing in recent history — or widening in some cases/locations. Based on these trajectories, I suspect that men and women are actually FAR more similar than anyone natively groks, and that we exaggerate or invent small differences due to a biological hyperfixation with sex. Note that we don't obsess about the mental differences between, say, male and female cats, even though they have much greater sexual dimorphism than humans do.)
  - rcxdude 2y ago
    > For example, an average man with an average woman's level of testosterone will experience a MUCH higher level depression, listlessness, and sexual disinterest than the average woman experiences. Is this true even if they were to have an average woman's level of estrogen? It may be that the brain needs either set of hormones to work effectively, and doesn't work well when lacking both (of course, gender transition HRT aims for this, but pretty much all undergoing it are trans and so aren't a good indication of average reaction to hormones for their chromosomal sex. And a cis person undergoing the same HRT isn't likely to enjoy the process)
cmcaleer 2y ago
I have to assume you haven’t bothered reading this, because section 2.3 points out flaws in the methodology of the studies looked at which the methodology in this study kind of tries to address (whether or not it’s a good job of it is left for everyone else to figure out). You shouldn’t dismiss a result out of hand because it doesn’t fit preconceived notions, but it’s absolutely a reason to try to dig in to the methodology of the new study and make sure it’s not flawed. That said, this meta-analysis is also filled with some crazy statements. It seems to imply sexual dimorphism is only really visible in repro organs but women necessarily need to have wider hips to facilitate child birth among other differences. This obvious point should have also been noted when comparing differences in organ mass, since mothers of babies with larger heads are more likely to die so this is selected against. Not an issue with lungs, heart etc., hence larger % differences in sexes there. These aren’t egregious omissions in and of themselves, but it’s certainly useful context I’d like to have were I not familiar with sexual dimorphism. The dismissiveness of a 1.6 fold increase in SDN size of human males compared to human women is bad. That’s enormous! Not something I would prepend with “only” and repeatedly call “small”, even when not comparing the differences between M/F humans and M/F rodents. Bizarre that none of the authors objected to this phrasing, because it’s poisoned reading the rest of this paper for me. How am I meant to trust the authors’ opinion of what a “small” difference is? Some of the points are a bit more compelling, like in section 5.1 where they point out that a difference attributed to M/F was replicated in much smaller size by concentrating on volume instead, or in 5.2 where they point out a few papers that missed crucial nuance. But overall after reading a few thousand words of this, the nicest thing I can say about it is that I agree that it is indeed gigantic.
logicprog 2y ago
Update: had a friend with access send me a PDF of the study and looked through it. It seems that the big breakthrough is only half AI — the other half being looking directly at time series of fMRIs instead of static images with features in them manually selected for relevance, because how the various circuits in the brain operate and circulate over time is important information. Also they got this to replicate well with the same people at different times, and also generalize to two other cohorts, consistently, and also used XAI to check what the AI was keying off, to make sure it wasn't going off something nonsensical, and directly used those features with success as well. It seems like an extremely carefully controlled and designed study tbh.
ars 2y ago
When your study contradicts reality, your study is wrong. That there are major mental difference between men and women is obvious to anyone who interacts with both. If your study can't find those differences it's your study that's wrong, not that the differences don't exist.
- logicprog 2y ago
  That's interesting. So if a study is counter intuitive to "common sense," it's the study that's wrong? Especially since they're talking about structural and functional neurological differences, whereas the "everyday common sense" differences you're gesturing at could be due to other things, so they aren't necessarily in contradiction.
  - ars 2y ago
    > So if a study is counter intuitive to "common sense," it's the study that's wrong? Possibly, yes. But "common sense" is a very weak way of describing things here, because common sense is usually used to describe intuition, not knowledge. We are talking here about knowledge - men and women are different mentally, this isn't something unknown, or something requiring a study. > since they're talking about structural and functional neurological differences Given that there are obvious mental differences, and that most people don't believe in duality, that would imply there have to be neurological differences, if your study didn't find them, your study is faulty. Either that, or it's evidence for dualism.
    - logicprog 2y ago
      > We are talking here about knowledge - men and women are different mentally, this isn't something unknown, or something requiring a study. The whole point of science is to double-check this sort of thing, the stuff "everyone knows" that seems obvious to casual observation, like that the earth is flat. Plus you're simply asserting and re-asserting that this is capital-K Knowledge, without providing any reasons, just banging the podium. That's not very convincing. This is anti-intellectualism. > Given that there are obvious mental differences, and that most people don't believe in duality, that would imply there have to be neurological differences, if your study didn't find them, your study is faulty. Either that, or it's evidence for dualism. Or maybe the differences are very social and contextual — men and women's brains largely operate the same way and are structured the same, but give different outputs because they're given different inputs — and the impacts of those social and contextual inputs are simply too fine-grained to show up.
LudwigNagasena 2y ago
It’s not the first study to find differences. And that’s not the only meta study in existence. I don’t see why one would be surprised given other literature around the subject. BTW, note how the article you cited doesn’t argue against differences between male and female brains but makes a rather pedantic point about “dimorphism”.

im3w1l 2y ago

Sounds like a case of "Old measuring instrument cannot tell two things apart. New more precise instrument can tell them apart"

logicprog 2y ago
They're using the same measuring instruments, though. They're just feeding them through AI. Maybe, though. Let's see if it replicates.

courseofaction 2y ago

Without making any claims about gender or non-binary people (not my wheelhouse, I simply don't know), there's ample evidence to suggest statistically significant population-level differences between males and females on a many cognitive measures. I don't see how it's surprising that an new generation of signal-detection tool finds population-level differences in the brains that produce these cognitions.

derbOac 2y ago

I think the linked news article is a little misleading, although I share your skepticism. I'd like to see these results replicated rigorously on still new sets of data by independent researchers; I wouldn't be surprised either way, if the results did or did not replicate. However, the news article seems to spin this as "male and female brains are totally different entities that bear no relationship with one another." Although I haven't reread it carefully, it seems like the article is saying something more like "you can identify gender-specific patterns, and those gender-specific components relate to things like cognitive ability gender-specifically". It's not that you can't find overlap — that that wasn't the focus of the study — it's that if you go looking for differences, you can find them.

RoyalHenOil 2y ago
It seems to me that in order for male and female brains to be functionally the same, they would need to be physically different to account for the extreme hormonal differences. When you give a man a female dose of hormones or a woman a male dose of hormones, it has a very big effect on their mood, behavior, and mental wellbeing. This change is much, much bigger than the average diffences we see between men and women. For example, an average man with an average woman's level of testosterone will experience a MUCH higher level depression, listlessness, and sexual disinterest than the average woman experiences. This strongly implies that human brains must correct for these huge hormonal differences. Basically, in order for male and female behavior to be similar, their brains must differ. If their brains are the same, then hormones will have a much, much bigger influence on male and female behavior than what we actually see in reality. Hormone-correcting brain differences would also imply that it's possible for people to be born with some type of intersex brain condition, and that these individuals would benefit greatly from receiving hormone therapy to bring their hormone levels in line with their brains. And this, indeed, seems to be something we see occasionally. (In case anyone cares or thinks it is relevant, I wish to note that I am a cisgender woman and I do not think that there are huge innate differences in men's and women's mentality — certainly nothing like on the level that testosterone/estrogen/etc levels would predict. I think most of the differences we do see are environmental, which is why these gaps have been closing in recent history — or widening in some cases/locations. Based on these trajectories, I suspect that men and women are actually FAR more similar than anyone natively groks, and that we exaggerate or invent small differences due to a biological hyperfixation with sex. Note that we don't obsess about the mental differences between, say, male and female cats, even though they have much greater sexual dimorphism than humans do.)
- rcxdude 2y ago
  > For example, an average man with an average woman's level of testosterone will experience a MUCH higher level depression, listlessness, and sexual disinterest than the average woman experiences. Is this true even if they were to have an average woman's level of estrogen? It may be that the brain needs either set of hormones to work effectively, and doesn't work well when lacking both (of course, gender transition HRT aims for this, but pretty much all undergoing it are trans and so aren't a good indication of average reaction to hormones for their chromosomal sex. And a cis person undergoing the same HRT isn't likely to enjoy the process)

cmcaleer 2y ago

I have to assume you haven’t bothered reading this, because section 2.3 points out flaws in the methodology of the studies looked at which the methodology in this study kind of tries to address (whether or not it’s a good job of it is left for everyone else to figure out). You shouldn’t dismiss a result out of hand because it doesn’t fit preconceived notions, but it’s absolutely a reason to try to dig in to the methodology of the new study and make sure it’s not flawed. That said, this meta-analysis is also filled with some crazy statements. It seems to imply sexual dimorphism is only really visible in repro organs but women necessarily need to have wider hips to facilitate child birth among other differences. This obvious point should have also been noted when comparing differences in organ mass, since mothers of babies with larger heads are more likely to die so this is selected against. Not an issue with lungs, heart etc., hence larger % differences in sexes there. These aren’t egregious omissions in and of themselves, but it’s certainly useful context I’d like to have were I not familiar with sexual dimorphism. The dismissiveness of a 1.6 fold increase in SDN size of human males compared to human women is bad. That’s enormous! Not something I would prepend with “only” and repeatedly call “small”, even when not comparing the differences between M/F humans and M/F rodents. Bizarre that none of the authors objected to this phrasing, because it’s poisoned reading the rest of this paper for me. How am I meant to trust the authors’ opinion of what a “small” difference is? Some of the points are a bit more compelling, like in section 5.1 where they point out that a difference attributed to M/F was replicated in much smaller size by concentrating on volume instead, or in 5.2 where they point out a few papers that missed crucial nuance. But overall after reading a few thousand words of this, the nicest thing I can say about it is that I agree that it is indeed gigantic.

logicprog 2y ago

Update: had a friend with access send me a PDF of the study and looked through it. It seems that the big breakthrough is only half AI — the other half being looking directly at time series of fMRIs instead of static images with features in them manually selected for relevance, because how the various circuits in the brain operate and circulate over time is important information. Also they got this to replicate well with the same people at different times, and also generalize to two other cohorts, consistently, and also used XAI to check what the AI was keying off, to make sure it wasn't going off something nonsensical, and directly used those features with success as well. It seems like an extremely carefully controlled and designed study tbh.

ars 2y ago

When your study contradicts reality, your study is wrong. That there are major mental difference between men and women is obvious to anyone who interacts with both. If your study can't find those differences it's your study that's wrong, not that the differences don't exist.

logicprog 2y ago
That's interesting. So if a study is counter intuitive to "common sense," it's the study that's wrong? Especially since they're talking about structural and functional neurological differences, whereas the "everyday common sense" differences you're gesturing at could be due to other things, so they aren't necessarily in contradiction.
- ars 2y ago
  > So if a study is counter intuitive to "common sense," it's the study that's wrong? Possibly, yes. But "common sense" is a very weak way of describing things here, because common sense is usually used to describe intuition, not knowledge. We are talking here about knowledge - men and women are different mentally, this isn't something unknown, or something requiring a study. > since they're talking about structural and functional neurological differences Given that there are obvious mental differences, and that most people don't believe in duality, that would imply there have to be neurological differences, if your study didn't find them, your study is faulty. Either that, or it's evidence for dualism.
  - logicprog 2y ago
    > We are talking here about knowledge - men and women are different mentally, this isn't something unknown, or something requiring a study. The whole point of science is to double-check this sort of thing, the stuff "everyone knows" that seems obvious to casual observation, like that the earth is flat. Plus you're simply asserting and re-asserting that this is capital-K Knowledge, without providing any reasons, just banging the podium. That's not very convincing. This is anti-intellectualism. > Given that there are obvious mental differences, and that most people don't believe in duality, that would imply there have to be neurological differences, if your study didn't find them, your study is faulty. Either that, or it's evidence for dualism. Or maybe the differences are very social and contextual — men and women's brains largely operate the same way and are structured the same, but give different outputs because they're given different inputs — and the impacts of those social and contextual inputs are simply too fine-grained to show up.

LudwigNagasena 2y ago

It’s not the first study to find differences. And that’s not the only meta study in existence. I don’t see why one would be surprised given other literature around the subject. BTW, note how the article you cited doesn’t argue against differences between male and female brains but makes a rather pedantic point about “dimorphism”.