gillygeewhiz:

Every now and then I see a spike in people presenting a particular study on brain sex and debating it as if it has equal validity to the rest of the research on the subject. It very much doesn’t, so here’s a PSA:

‘Sex beyond the genitalia: The human brain mosaic’ by Joel, et al. (2015) does not disprove neurological sexual dimorphism. It came to dishonest conclusions about its results and has been refuted multiple times by other researchers.

This study was ridiculously sensationalized and (mis)reported in many news outlets when it came out, so you may have seen it presented with a headline like:

• “Scans prove there’s no such thing as a ‘male’ or ‘female’ brain”
• “Men are from Mars, women are from Venus? New brain study says not”
• “The brains of men and women aren’t really that different, study finds”
• “Male brain vs. female brain? Research says they’re unisex“
• and plenty, plenty more

If you’re reading anything claiming to debunk the idea of brain sex and see a reference to “Daphna Joel” or “Tel-Aviv University” or “brain mosaic”, that’s a huge red flag.

Joel’s claim is that no brain is truly male or female, and we all have a “mosaic” of masculine and feminine traits. There’s a huge level of variability and overlap in individual male and female brains, and overall we are more alike than we are different – that much is true. But the results of Joel’s study actually did show notable sex differences between male and female brains. In order to make it look like their results did not demonstrate this, they fudged their threshold for measuring sex differences (”internal consistency”) in a way that almost no amount of sex differentiation would be able to meet, allowing them to come to the conclusion that there are no meaningful sex differences in the brain at all.

After this study was published it was quickly refuted not just once, but four times by other scientists in the same journal:

• ‘Multivariate revisit to “sex beyond the genitalia”’ by Rosenblatt used a multivariate analysis and was able to demonstrate a clear dividing line in Joel’s data between male and female neurology despite the variability across individuals. Fig. 1. has a helpful visual.
• ’Patterns in the human brain mosaic discriminate males from females’ by Chekroud, et al. used a regression model which, using patterns in Joel’s proposed “mosaic”, was more than 90% accurate in predicting the sexes in a set of 1500 brain scans.
• ’Joel et al.’s method systematically fails to detect large, consistent sex differences’ by Del Giudice, et al. criticized Joel’s narrow definition of “internal consistency” for sex differences, and pointed out that the findings did remark on notable sex differences that were dismissed by this definition. They followed up Joel’s study with a demonstration of how flawed their methods were, concluding that “under realistic conditions, Joel et al.’s method almost always returns the same pattern of results”.
• ‘Yes, there is a female and a male brain: Morphology versus functionality’ by Glezerman took a different approach and rather than challenging Joel’s assertions with other data, he criticized the conclusion with logical reasoning. Logically, based on the already-proven effects of sex hormones on the brain and sex differences in gonadal function (and differences in other organs), the brain is necessarily sex dimorphic regardless of if those differences are visible on a brain scan.

Researchers took issue with it outside of journals, too, allowing them to explain in more plain-English:

• An author of one of the above refutations

  also wrote a Psychology Today article about how Joel’s “statistical abracadabra” led to a misleading conclusion.

• One researcher from Dublin wrote two analyses of it on his neuroscience blog, the first specifically comparing Joel’s study to one from 2013 that had similar results but the opposite conclusion, and the second again exploring the flaw of Joel’s methods and how their conclusion was not an accurate representation of their research.

In more plain-English, but with less weight as they were written by non-scientists:

• This trans woman contacted Daphna Joel personally about whether this study disproves the sexual dimorphism that has already been observed in the BSTc in studies on gender & sexual orientation, and what that means about the brains of trans people. Assuming this woman was being honest about their correspondence, Joel backpedalled in her response, claiming that their study did not have certain enough results to have implications for trans people without (their grossly flawed definition of) “internal consistency” also being applied to the trans research. Joel conceded the possibility of gender identity being controlled by sex-dimorphic brain structures despite claiming that there is no male or female brain.
• This article points out the inaccuracy of the science reporting on this study and explains how the actual results do not show that there is “no male or female brain” as claimed, but that the conclusion was intentionally written in a way that would play into journalists’ biases and be easy to sensationalize. This one’s much more straightforward in suggesting that Joel published this study with an agenda – I personally think it’s likely but it’s nonetheless a pretty serious accusation to make, so I’ll leave it up to you to take it with a grain of salt and make your own conclusions about Joel’s intentions.

TLDR this study’s conclusion reflects neither its own results nor the general scientific opinion on the subject. It does not disprove the existence of male and female brains. It should not be treated with equal weight as the rest of the research showing sex differences in the brain.