[captainmuon]

The way I see it is that, as often when it comes to differences due to sex, you have two wide Gaussian distributions that overlap a lot. Sure, there are behavioral differences (due to testosterone or other reasons), but the variation in behavior within one sex is probably larger than the distance between the means.

The problem is, that in casual speech, we can only easily express black-and-white: either men and women are different due to testosterone, or they are not. Neither statement is really correct.

[chroma]

With mental rotation, there is practically no overlap. See the graph from the original study.[1]

It's a similar story for physical strength. Nobody denies that men tend to be stronger than women, but most people vastly underestimate the disparity. In a study that measured grip strength in 2,000 people[2], the weakest 5% of men were comparable to the top decile of women. Even elite female athletes were only as strong as the median untrained male.

There are more examples (propensity for violence is another big one), but it takes time to find citations. Suffice it to say that most people think the sexes are more similar than they actually are.

1. https://brothersdiamond.files.wordpress.com/2015/02/menwoman... (For more on mental rotation and testosterone's effects, see my comment downthread: https://news.ycombinator.com/item?id=10443021)

2. http://egitim.judo.gov.tr/Dosyalar/makaleler/-ENG-Hand-grip-...

[esailija]

There's plenty of studies that debunk the mental rotation myth:

http://www.sciencedaily.com/releases/2010/09/100915080431.ht...

http://www.researchgate.net/profile/Angelica_Moe/publication...

There's also a study showing the field of neuroscience having a tendency to make unsupported and untested claims to confirm stereotypes http://link.springer.com/article/10.1007%2Fs12152-012-9169-1

[no1ne]

http://www.northwestern.edu/newscenter/stories/2015/08/scien... Is this also stereotypical? Read what the researcher(a woman by the way)had to say. For 20 years, she actively avoided studying sex differences in the brain until her own data showed her that differences between females and males were real. Let science be about cold-hard facts and let it not be muddied by "feelings".

[oldmanjay]

there's a pretty big difference between "debunk[ing] the ... myth" and showing the effect is real and can be reduced via training. in particular, your second link says the differences are probably real in its conclusion!

[esailija]

The myth is that the differences (in mental rotation ability) are innate or "biological", not that there is differences (in mental rotation ability) in the first place.

[oldmanjay]

If one gender displays innate ability and the other can be trained to match it, then the difference is not shown to be a myth. There is a very dishonest redefinition of a common word involved to make the meaning apply.

Political fictions are comfortable but don't really change facts, they just color interpretations. Mischaracterizing observation does not serve to make an argument appear stronger.

Note that none of this means women are inferior, and if you are construing my point to mean I am arguing for that position, you have imposed that on me.

[no1ne]

>> myth is that the differences are innate or "biological"

erm...please read

1. Genomic differences between developing male and female brains in the womb-- http://www.sciencedaily.com/releases/2015/02/150203190223.ht...

and if you still feel that somehow humans try to gender stereotype whilst the fetus are still in the wombs, then read this also-

2.Sex differences in rhesus monkey toy preferences parallel those of children-- http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2583786/

[esailija]

I was referring to mental rotation ability, I clarified it in an edit.

I also don't have time to go into the studies you linked about general differences but similar studies have been debunked before. For example with 1.5 day old infants where the experimenters knew about the sex and their bias subtly corrupted the experiment and it did not reproduce when the bias was truly accounted for.