Not really a fan of the title here…something about it just seems too editorialized. Maybe "significant mean differences found between mens' and womens' brains?"
Title using 'very large' does not come from the article, which gives a more subtle image:
>Sex Differences in the Adult Human Brain: Evidence from 5216 UK Biobank Participants
>Males had higher raw volumes, raw surface areas, and white matter fractional anisotropy; females had higher raw cortical thickness and higher white matter tract complexity. There was considerable distributional overlap between the sexes.
Personally I wouldn't call the difference between a mean of 1115.76 with an standard deviation of 89.68 and a mean of 1233.58 with an SD of 98.31 very large (all in cm^3)
> The differences were substantial: in some cases, such as total brain volume, more than a standard deviation. The effect size of d = −1.41 for total brain volume (Table 1) translates to 92.1% of males being above the female mean, and an 84.1% chance that a randomly-chosen male will have a larger total brain volume than a randomly-chosen female.
The height of the average male (178cm) is more than one standard deviation (8.9cm) above the height of the average female (165.1cm).
I don't really know if we should be surprised that the differences between male and female brain volumes are also statistically significant and larger than one standard deviation.
The differences were substantial: in some cases, such as total brain volume, more than a standard deviation. The effect size of d = −1.41 for total brain volume (Table 1) translates to 92.1% of males being above the female mean, and an 84.1% chance that a randomly-chosen male will have a larger total brain volume than a randomly-chosen female.
Yeah, and let's just pretend that your implied argument, that looking at the other values measured finds that they, in toto, mean the brains are equal. Of course that's very far from the truth.
It's funny: your statement is perfectly true, and yet gives completely wrong information ... you ever thought about taking up politics ?
But no worries: the study did mention the forbidden metric of death : cognitive tests.
First, an important note: the cognitive tests were specifically chosen to be tests where the mean between men and women would be the same, this is not a coincidence. It is known that there are tests where the mean is not the same between males and females. For example, males have a slight advantage in 3d reasoning skills and females have somewhat better temporal analysis skills ("males aim, females plan").
It shows the same as all others, by the way. Male brains are more experimental : more variable in nearly every way. The dumbest human alive is very likely male, and so is the smartest.
Let's face facts here: the very basis for tolerance is, flat out, totally, and completely wrong. There are in fact sex differences, including in the brain and resulting ability. (worse: there are race differences too. aiaiai)
Whether that means we should or should not practice tolerance is, of course, an unrelated question. But we should be able to answer it without falsifying and denying scientific results.
There, I think every sacred egg, left and right, racist and tolerant, has been thoroughly stomped on. Can we now have an adult discussion ?
Arguing like this, crossing into personal nastiness while slinging rhetoric like weapons, will get you banned on HN, regardless of how right or wrong your or anyone else's views are. Please don't post like this again.
This paragraph from the discussion is interesting:
> An additional hypothesis—one that is not incompatible with the hypothesis that some of the sex differences seen here are a proximate cause of behavioral differences—is that brain structural differences might sometimes be the result of compensatory mechanisms for differences in sex-specific hormones, and might thus act to reduce behavioral sex differences that would otherwise have been present (De Vries 2004; McCarthy and Arnold 2011). This perspective may in part explain an apparent paradox in human sex difference research: that the (raw) effect sizes found for brain measures such as volume and surface area are so large, whereas most behavioral sex differences are so small (Hyde 2014). Our descriptive results do not directly speak to any causal mechanisms, but it should be borne in mind that they are compatible with these multiple interpretations.
Please note that 'very large' was indeed quoted directly from the paper.
"The average difference for the 14 subcortical volumes was d = −0.70. A set of Bayesian t-tests (see Supplemental Materials and Table 1) confirmed that the mean sex differences were very large, with extremely strong evidence in favor of the hypothesis that males differed from females "on every overall and subcortical volume."
You add that in your personal opinion the difference is 'not very large'. What should be the magnitude of the numbers below to fall into the 'large' category in your assessment?
Quote: "...total brain volume translates to 92.1% of males being above the female mean and an 84% chance that a randomly-chosen male will have a large brain volume than a randomly chosen female" That's pretty significant in terms of means.
When the topic is divisive, doing this is particularly bad. It can easily ruin the thread for discussion. If you want to say what you think is important about an article, please do so in a comment. Then your view is on a level playing field with everyone else's.
Of interest: There was generally greater male variance across the raw structural measures. They also cite Global Sex Differences in Test Score Variability (Science, 2008).
Towards the end of the Introduction, they have a paragraph where they discuss how the study intends to test the "greater male variability hypothesis", something I had not heard of before.
They reference other studies which have show greater variability among males in: cognitive ability, academic achievement test results, personality, athletic performance, birth weight, and adult weight.
Uh-oh this is going to hurt a lot of people.
Even if the article did not state anything about the actual effects of these differences.
Just mentioning or hinting at this nowadays will result in a lot of automatic flack and backlash.
Notice how the incoming comments will be about the title and not the contents and the subsequent derived meaning of this.
They will probably have to change the title to not offend anyone.
Sadly this is our current state of approach on science.
The title is not the title of the paper, but it is a direct quote from the paper. The paper does say differences are very large. (Search for "very large".)
Among other things, it can mean, “dear God, just from the title I can tell that the comments on this are going to be 90% people staying their preconceptions of how other people will feel about it and/or imputing motives to thread or comment moderation, 5% people reacting to the title, and 5% people quoting the article in response to the previous 5%.”
Quite sad, indeed, when people are offended by untrue things being presented as true! How will we ever get science done in such an environment of easily offended people?
I was replying to the sentence about offense, namely, "They will probably have to change the title to not offend anyone."
The title of this HN post, "Mean differences in men's and women's brains are very large," is in fact untrue as the top-voted comment points out. They're statistically significant but not that large, and there's lots of overlap in the distributions.
There is a positive correlation between cranial capacity and height, some correlation between brain volume and height, and some correlation between brain volume and intelligence within gender.
Women score as well as men on general intelligence tests, although men have higher levels of variance and score slightly higher on certain subsets (e.g., spatial reasoning). This study shows the former very well - although men scored slightly better on a verbal-numerical reasoning test, the results were statistically insignificant (p=.45).
The relationship between body size, brain volume, and general intelligence only appears to apply within a given gender, and not between the two genders.
As far as I can tell, all of this has been known for a long, long time, so I don't think anyone's feelings are going to get hurt from a study that looked at brain volumes in older (44-77 year old) adults in the United Kingdom.
> Even if the article did not state anything about the actual effects of these differences.
Yes it did.
> We linked the structural brain differences to scores on two cognitive tests taken at the time of the imaging visit: verbal-numerical reasoning and reaction time (see Method). Descriptive statistics for the cognitive tests are shown in Table 1. Note that we coded both tests (reflecting the reaction time variable) so that higher scores indicated better performance. The test scores correlated positively, but weakly (r = .12). Males had a slightly higher mean score than females on verbal-numerical reasoning (d = −0.18) and slightly faster mean reaction time (d = −0.22); there was no significant variance difference for verbal-numerical reasoning (VR = 0.97, P = .45), though males had marginally more variance in reaction time (VR = 0.92, P = .03).
I don't know why they didn't include any graphs for those results.
The article itself is not so much interesting to me that the reactions to it on HN. The articles says nothing on how these differences translate into psychological traits/abilities etc. Nevertheless, even a hint that there might be differences, and that they're might be quite pronounced, somehow cannot be a subject of civilized discussion on HN and people will flag it straight away. Why? What are you afraid of? Understanding these results might help us to understand ourselves better. Or do you prefer not to have such research in the name of political correctness?
HN is divided on divisive issues. Anthropomorphizing it into something that "likes" what you dislike and "dislikes" what you like is just an illusion, but has become a rhetorical trick, as cheap as it is common. Please don't do that here.
For the same reason that Galilei and Copernicus came into conflict with the Roman Catholic church authorities, why Socrates was forced to drink poison for 'corrupting the youth': some ideas are not tolerated within certain societies no matter their scientific underpinning. In other words, those ideas go against the dogmata [1] of such societies.
Because (a) you violated the site guidelines by editorializing the title (see https://news.ycombinator.com/item?id=17147115), and (b) to do that when the subject matter is divisive is flamebait, violating the guidelines in a second and worse way. HN's rules are the way they are for good reason. Please (re-)read and follow them from now on: https://news.ycombinator.com/newsguidelines.html.
Threads are acutely sensitive to initial conditions. What you did made a very large increase in the probability that this discussion would be ruined before it began.
43 comments
[ 0.26 ms ] story [ 68.7 ms ] thread>Sex Differences in the Adult Human Brain: Evidence from 5216 UK Biobank Participants
>Males had higher raw volumes, raw surface areas, and white matter fractional anisotropy; females had higher raw cortical thickness and higher white matter tract complexity. There was considerable distributional overlap between the sexes.
Personally I wouldn't call the difference between a mean of 1115.76 with an standard deviation of 89.68 and a mean of 1233.58 with an SD of 98.31 very large (all in cm^3)
> The differences were substantial: in some cases, such as total brain volume, more than a standard deviation. The effect size of d = −1.41 for total brain volume (Table 1) translates to 92.1% of males being above the female mean, and an 84.1% chance that a randomly-chosen male will have a larger total brain volume than a randomly-chosen female.
I don't really know if we should be surprised that the differences between male and female brain volumes are also statistically significant and larger than one standard deviation.
The differences were substantial: in some cases, such as total brain volume, more than a standard deviation. The effect size of d = −1.41 for total brain volume (Table 1) translates to 92.1% of males being above the female mean, and an 84.1% chance that a randomly-chosen male will have a larger total brain volume than a randomly-chosen female.
It's funny: your statement is perfectly true, and yet gives completely wrong information ... you ever thought about taking up politics ?
But no worries: the study did mention the forbidden metric of death : cognitive tests.
First, an important note: the cognitive tests were specifically chosen to be tests where the mean between men and women would be the same, this is not a coincidence. It is known that there are tests where the mean is not the same between males and females. For example, males have a slight advantage in 3d reasoning skills and females have somewhat better temporal analysis skills ("males aim, females plan").
It shows the same as all others, by the way. Male brains are more experimental : more variable in nearly every way. The dumbest human alive is very likely male, and so is the smartest.
Let's face facts here: the very basis for tolerance is, flat out, totally, and completely wrong. There are in fact sex differences, including in the brain and resulting ability. (worse: there are race differences too. aiaiai)
Whether that means we should or should not practice tolerance is, of course, an unrelated question. But we should be able to answer it without falsifying and denying scientific results.
There, I think every sacred egg, left and right, racist and tolerant, has been thoroughly stomped on. Can we now have an adult discussion ?
I think we should be able to acknowledge and accept our differences but also be tolerant of them.
https://news.ycombinator.com/newsguidelines.html
This paragraph from the discussion is interesting:
> An additional hypothesis—one that is not incompatible with the hypothesis that some of the sex differences seen here are a proximate cause of behavioral differences—is that brain structural differences might sometimes be the result of compensatory mechanisms for differences in sex-specific hormones, and might thus act to reduce behavioral sex differences that would otherwise have been present (De Vries 2004; McCarthy and Arnold 2011). This perspective may in part explain an apparent paradox in human sex difference research: that the (raw) effect sizes found for brain measures such as volume and surface area are so large, whereas most behavioral sex differences are so small (Hyde 2014). Our descriptive results do not directly speak to any causal mechanisms, but it should be borne in mind that they are compatible with these multiple interpretations.
"The average difference for the 14 subcortical volumes was d = −0.70. A set of Bayesian t-tests (see Supplemental Materials and Table 1) confirmed that the mean sex differences were very large, with extremely strong evidence in favor of the hypothesis that males differed from females "on every overall and subcortical volume."
You add that in your personal opinion the difference is 'not very large'. What should be the magnitude of the numbers below to fall into the 'large' category in your assessment?
Quote: "...total brain volume translates to 92.1% of males being above the female mean and an 84% chance that a randomly-chosen male will have a large brain volume than a randomly chosen female" That's pretty significant in terms of means.
https://news.ycombinator.com/newsguidelines.html
When the topic is divisive, doing this is particularly bad. It can easily ruin the thread for discussion. If you want to say what you think is important about an article, please do so in a comment. Then your view is on a level playing field with everyone else's.
They reference other studies which have show greater variability among males in: cognitive ability, academic achievement test results, personality, athletic performance, birth weight, and adult weight.
People getting upset from the actual result: Zero.
Sadly this is our current state of approach on science.
"Title using 'very large' does not come from the article, which gives a more subtle image:"
which sounds to me like someone taking offense to the title?
Among other things, it can mean, “dear God, just from the title I can tell that the comments on this are going to be 90% people staying their preconceptions of how other people will feel about it and/or imputing motives to thread or comment moderation, 5% people reacting to the title, and 5% people quoting the article in response to the previous 5%.”
You could have stopped typing there.
The title of this HN post, "Mean differences in men's and women's brains are very large," is in fact untrue as the top-voted comment points out. They're statistically significant but not that large, and there's lots of overlap in the distributions.
Women score as well as men on general intelligence tests, although men have higher levels of variance and score slightly higher on certain subsets (e.g., spatial reasoning). This study shows the former very well - although men scored slightly better on a verbal-numerical reasoning test, the results were statistically insignificant (p=.45).
The relationship between body size, brain volume, and general intelligence only appears to apply within a given gender, and not between the two genders.
As far as I can tell, all of this has been known for a long, long time, so I don't think anyone's feelings are going to get hurt from a study that looked at brain volumes in older (44-77 year old) adults in the United Kingdom.
Yes it did.
> We linked the structural brain differences to scores on two cognitive tests taken at the time of the imaging visit: verbal-numerical reasoning and reaction time (see Method). Descriptive statistics for the cognitive tests are shown in Table 1. Note that we coded both tests (reflecting the reaction time variable) so that higher scores indicated better performance. The test scores correlated positively, but weakly (r = .12). Males had a slightly higher mean score than females on verbal-numerical reasoning (d = −0.18) and slightly faster mean reaction time (d = −0.22); there was no significant variance difference for verbal-numerical reasoning (VR = 0.97, P = .45), though males had marginally more variance in reaction time (VR = 0.92, P = .03).
I don't know why they didn't include any graphs for those results.
HN is divided on divisive issues. Anthropomorphizing it into something that "likes" what you dislike and "dislikes" what you like is just an illusion, but has become a rhetorical trick, as cheap as it is common. Please don't do that here.
[1] https://en.wikipedia.org/wiki/Dogma
Threads are acutely sensitive to initial conditions. What you did made a very large increase in the probability that this discussion would be ruined before it began.