The Transsexual Brain

By Neuroskeptic | November 9, 2011 8:07 am

According to a new paper, the brains of male-to-female transexuals are no more “female” than those of men.

The authors write that “The present data do not support the notion that brains of male-to-female transexuals are feminized” and conclude “The present study does not support the dogma that male-to-female transexuals have atypical sex dimorphism in the brain”.

That last sentence has gained quite a bit of coverage, including a quote on the Wikipedia page for “transgender”.

But is it so simple?

Structural MRI scans were used to compare the size of various brain structures between three groups of volunteers: heterosexual men, heterosexual women and the transexuals (or “MtF”s as I will call them for short) who were diagnosed with gender dysphoria and were “genetically and phenotypically males”.

There were 24 in each group, which makes it a decent sized study. None of the MtFs had started hormone treatment yet, so that wasn’t a factor, and none of the women were on hormonal contraception.

The scans showed that the non-transsexual male and female brains differed in various ways. Male brains were larger overall but women had increases in the relative volumes of various areas. Male brains were also more asymmetrical.

The key finding was that on average, the MtF brains were not like the female ones. There were some significant differences from the male brains, but they weren’t the same differences that distinguished the females from the males.

This is a fairly crude approach. It looks at the groups on average. It’s a finding, but there’s more you could with this data. It would be better perhaps to look at the male and female groups, and then try to work out which group each individual MtF is most similar to. You could do that using a Support Vector Machine such as was previously used to detect autism.

This would also have the advantage that it would integrate the results across different brain areas: maybe the important thing is not just the size of individual areas but the relative size of one area to another area.

My real problem though is with the language used to discuss the data. The authors say that the study doesn’t support “atypical sex dimorphism in the brain” yet this wasn’t a study of “the brain”. It was a study of one specific aspect of the brain, namely the volume of different regions. There could be all kinds of chemical and microstructural differences that don’t show up on these scans.

There are lots of people with severe epilepsy, for example, whose brains clearly differ in some major way from people without epilepsy, yet they look completely normal on MRI. Only using other methods, like EEG, reveals the difference. Because the difference is chemical, not structural.

I have no idea how, or if, the brains of MtF transsexuals are “feminized” but this study doesn’t rule it out. Now I’m sure the authors know all this. And in fact they themselves recently published a paper showing atypical neural responses to smelling “oderous steroids” in transsexual people. But while neuroscientists will know what they meant, I worry that studies like this could be miscontrued by other people (like Wikipedia readers) as a result of overenthusiastic language in papers.

Link: Also blogged at BPS Research Digest. Savic I, & Arver S (2011). Sex dimorphism of the brain in male-to-female transsexuals. Cerebral cortex (New York, N.Y. : 1991), 21 (11), 2525-33 PMID: 21467211

CATEGORIZED UNDER: media, papers
  • petrossa

    Jumping to conclusions is fast becoming a standard. First started with climatology it's now invading serious science.

    I blame education. Since education is open for the masses the standards have dropped significantly to accommodate for mediocre students. Practically very few years standards get reevaluated according to the graduation rate.

    Always downwards.

    With sloppy papers by the 1000's as a result.

    And wikipedia should carry a public health warning.

    • Tao

      ‘publication bias’

  • Bendert Katier

    Interesting Article. We devoted our last journal of social sciences on the topic of gender roles and sexual identity. Thought you might enjoy it!

  • James Cantor

    When in isolation, it is indeed possible to explain the Savic & Arver data by saying that that MtF brains are feminized, but in a way that does not show up on scans.

    However, there are other data (i.e., Rametti et al., 2010) that when considered together, spell out a very different story: There are actually two independent TYPES of male-to-female transsexual, one of which does show a pattern of feminization, and one of which does not. This confirms the neuroanatomical prediction that was published by Blanchard (2008).

    I have recently pointed this out in a Letter-to-the-Editor of the Archives of Sexual Behavior, which is available online as a .pdf, for those interested:

    – James Cantor
    twitter @jamescantorphd

  • practiCalfMRI

    Is it standard procedure in this field to ignore the rest of the body? Surely there might be many, many measurements one might make (hand span, height, weight, hat size, etc.) that should be measured to ensure no systematic difference in brain regions and so on? Isn't this yet another example of correlation science fishing for causation?

    NS, I like your idea of using an SVM or some other way to try to characterize “maleness” or “femaleness.” I'd also want to see how “pretty” males or “masculine” female controls ranked, too. There are so many potential confounds here it boggles my mind.

    Incomplete experiment if you ask me. I would need to see a lot more data before interpreting their “conclusions” at all.

  • James V. Kohl

    I am commenting only on the abstract since I have not yet read the paper in its entirety. Excerpt: “…male groups had smaller hippocampal volumes than HeW.” A clear connection between pheromones, luteinizing hormone (which also is involved in gray matter/white matter ratios)and hippocampal neurogenesis has been detailed. Given the role of the hippocampus in learning and memory during the development of sexual preferences, this might link pheromones to sexual orientation as did other works by Savic and her group.

  • Chris Brand

    I'm curious as to why the study compared gynephilic men and MtF transexuals with androphilic women; it would have made more sense, to me at least, if the comparison had been made to gynephilic women. I haven't read the full article yet, so this might well be explained within.

  • skm

    Thanks for posting this. I was dreading the possible outcomes of conversations where people would tell me that trans people don't exist.

    • Andy

      They *do* exist–it’s just that their condition is (a) mental, and (b) a distortion of what they were intended to be.

  • Charlie

    Thanks for the interesting article.

    FYI, there's a term for people who are not transgender. Cisgender refers to people who experience and present their gender in a way that’s aligned with the sex of their body, as compared to transgender and transsexual people.

    • Tao

      cis is a pejorative. please refrain.

      • danah gaz

        Cis is no more pejorative than trans.

        They are simply latin antonyms.

        • Tao

          And “negro” is just a color. Right.

          • danah gaz

            Not in English it isn’t.

            In Spanish it is. And still is, and always will be.

            And chemists have been using the words cis and trans since long before you were born.

            Also, cisatlantic and transatlantic.

            I’m not sure where you got ideas about language, but they seem more ideological than grounded in any sort of reality.

            Something isn’t a slur just because you say it is.

          • Tao

            Something is a slur if that is how it is used against people. Which it is.

          • danah gaz

            Yeah no.

            Otherwise Jew would be a slur and gay would be a slur.

            Not to mention that the OP you responded to above didn’t use it as pejorative.

            Thanks for playing.

  • Collaterly

    Hi! I'm a researcher who just finished a little over a year in a group using magnetic resonance spectroscopy and imaging to study brain and prostate cancer. I'm also transitioning to female.

    I actually hadn't heard of the study that this blog post is about, and it's not hard to see why. The study is… well, I'm a little confused as to who would fund it, because it's terribly designed. Neuroskeptic is right to say that there could be finer structural differences in trans brains, but is actually being a little too cautious – we have detected these differences! More sophisticated MR techniques, such as DTI, have detected very interesting differences in white matter, neuron count, etc. that do suggest that trans peoples' brains have a number of subtle similarities to the brains of people born in their preferred gender.

    That brings us back to this study. Why bother? Seriously, I don't want to be a spectral chauvinist, but why bother doing a volumetric study at all? It might have been excusable ten years ago, before we had better techniques, but when those techniques exist, and have been used to address this exact question, it's just stupid. Never mind that it is well-established that brain volume is hormonally-mediated. We wouldn't expect to see differences in volume in pre-HRT trans women, for instance. Were the study authors unaware of this? If so, what business did they have investigating trans issues without even a foggy grasp of the field?

    A few resources: basically diffusion tensor imaging (DTI) is one of the most sophisticated functional MRI techniques available and is particularly useful because it shows us axonal tracts – you may have seen some really glitzy 3D images of the brain that look as though they're made up of lots of white cords. DTI and other, similar ways of generating contrast – all MRI images come from bulk changes in the relaxation times and magnetic properties of different kinds of tissue – have identified a few interesting things. There's a summary of one such study here that includes needed discussion of the baggage surrounding much trans research (Blanchard and other idiots)

    Although not originally an MR study (it may since have been confirmed by MR,) the most widely-known brain difference result in trans research (the 1997 BSTc study) is summarized here:

    • Tao

      functional & structural brain differences does not an identity make. correlation =/= causation. brain activity is not (necessarily) consciousness. saying that ‘brain sex’ is responsible for personality and gender identity is sex essentialism, bio determinism, and medical materialism. primary sex traits are extremely dimorphic, and scientific classification is based on immutable gamete size, not genitals, chromosomes, or hormones. legal sex (and therefore gender) are assigned at birth based on apparent sex (how sex organs externally look/measure) and social conditioning into the construct of gender begins there. intersex and ais individuals almost always identify as their assigned gender, without dysphoria.

    • blacksnake35

      Well, there are differences between the brains of schizophrenics and normal people too. So we treat such people with drugs to try to normalize their brains, we don’t indulge their psychotic fantasies.

  • glitterary

    Hey, I really enjoy this blog and really appreciate the critical look it gave this study. As Collaterly pointed out, it seems a really blunt tool was used to try to look at a far more complex issue which has already been considered through more appropriate means.

    I'm not a scientist, and looking at the linked articles there's a lot of terminology that seems to be used as standard in research which coming from a non-scientific background sounds offensive. Terms like “Male to female” have given way to “Male assigned at birth”, and (as Charlie pointed out) it would be good to refer to non-trans people specifically as being cisgender–otherwise just using “women” when what is meant is “cisgender women” is exclusionary. While I accept I'm unfamiliar with the convention, would it be possible to indicate more clearly that the scientific terminology used is different to what is politically appropriate?

    • PrincessOfTheCrystal

      The problem is that the science community tends to be made up, primarily, of straight, middle class white males, like many other fields and they often refuse to acknowledge any bias because science and science is unbiased. But everything a scientist says or does is not “science”. Everything must be critically examined, including the language used.

      • Alex König

        No, the problem is that political correctness overweights facts. In other words: The problem are people like you.

  • Richard

    I pretty much agree with everything Collaterly said (and pretty much nothing petrossa said…), except that DTI isn't functional 😉

  • petrossa

    I have that effect on people. :)
    Unfortunately never with well founded arguments

  • Reneta Scian

    The brain is the most complex organ in nature, point simple. This study appears to have sound data, but the an Office Space grade of “Jumping to Conclusions”. Their data doesn't discuss or differentiate anything relating to sexual orientation which is demonstrated to have differences. Furthermore, MRIs on multiple occasions have shown transsexual brains to be morphologically distinct from both male and female brains, sometimes in between.

    The study also doesn't compensate for cranial volume in it's data, in a comparative manner, and it fails to calculate in the effects of current sexual hormones into the equation. The study sounds dismissive and one-sided, though the data tells a story of it's own. That is what happens in science when you go into experimental testing with bias as a premise to experimentation.

    To me, their conclusion appears to be invalid. Cranial volume is correlated to testosterone exposure of the bone structures (especially in puberty), therefore in structural analysis, you have to compensate for it. For a study like this to be truly revealing it would have to correlate with multiple types of scans, and include pre- and post-op transsexuals, both with and without hormones, and transsexuals who transitioned at puberty.

    As a further amplification to understand the effects of hormones and brain structures, include intersexed people of both male and female spectrums and multiple histories to give the steroidal effects on CAH patients, and the estrogenic effects of HRT on cAIS patients. This study's conclusion is anecdotal, though the data isn't necessarily invalidated, just incomplete. I am in general agreement with Collaterly, and I think the methods and conclusions are woefully inadequate considering current methods.

    Generally speaking, the contradicting the body of evidence for brain differences in transsexuals, plus the study's own data indicates reason enough to believe the conclusion is false. The commentary about not supporting “dogma” indicates to me a very biased opinion on the matter. Gender differences in the brain with transsexuality is a reasonable conclusion given the body of evidence the study is trying to refute.

    Dogma, indicates a belief that someone is refusing to believe otherwise on principle, asserting the larger body of evidence was biased (or conspiracy). I personally wouldn't mind a scientific explanation for the phenomenon, but that study isn't it. Sound scientific method isn't eroded by bad studies, so long as you view everything with skepticism and refuse to take people at their word alone.

  • Anonymous

    My goodness James Cantor must be paranoid that someone might think gay people are not born gay.

    The MRI suggests AGP transsexuals are not born that way due to the plasticity of the brain, how ever DR Cantor knows for a fact chemically castrating an autogynephilie at 25 does not stop the dysphoria at 39, he has seen it for himself.

    Savic and Arver say repetitive thinking may be responsible but when you take the agp away via castration from 25 to 39 and you still end up dysphoric like me at 39 its unlikely the change in the area of body perception is due to repetitive thinking.

    As well, if the part of the mind responsible for body image and meditation is changed due to the mind being plastic then I would aslo expect the part of the mind that deals with shame and humiliation and fear to also be different, but its not.

    Lets not forget autogynephilies are the only ones who are proven to be different via IQ scores, autogynephilies are for some reason highly intelligent. That my friend Dr Cantor is the only tangible data anyone has that is non negotiable that points to anything that's innate convulsively.

    Jenna McCarthy

    • PrincessOfTheCrystal

      Can we please not use the word “Autogynephile” in a science blog? It is essentially a slur. Just like you had scientific racism back in the day, you have scientific transphobia now. It is Blanchard’s “pet term” and not one that has much in the way of real scientific backing despite being widely accepted in some fields. If we were dealing with hard science, sure, it might be time to suck it up and accept reality. But it’s not a term that belongs in the field of neurology.

  • Anonymous

    Sometimes, people just respond to environmental stimuli and do what they do. Maybe a little boy was born rather small, was picked on & treated like a girl. Maybe that had a psychological impact and resulted in insecurity about it's sexuality and worth as a man. And maybe it identified more with the weaker sex. And maybe it resulted in him surgically altering his body to look like a female. I believe the answer here has always been mental.

    Sometimes, it cannot be explained with neurology, biology or physiology. The brain is a powerful thing…and I strongly believe the chemistry and physiology of the brain can change with different stimuli. I do not believe some men are born gay or with a female brain. I think science is reaching…in order to be politically correct.

    • Bryan Leger

      There is a hormonal element. MtF transexual bodies have been found to be unable to make full use of the testosterone produced. Could be same with estrogen on the other side. Also CAH in women has been shown to bring about masculine behavior and self-image.

  • Anonymous

    Not to be an ass, but 'transsexual' is misspelled in your title. (Spelled correctly in the text itself, though.)

    Just thought you should know. Feel free to delete this comment after reading. :)

  • Neuroskeptic

    Oops. Thanks. Fixed!

  • PrincessOfTheCrystal

    There are about 8-9 other studies that do not reach the same conclusion as this one… as far as I can tell this only tackles one of them? There has been a lot more research into this in recent years, it would take a lot more to overturn all of those findings.

    I also doubt it accounts for neuroplasticity vs hardwiring, if you are raised as male(and exposed to testosterone for at least a decade) then obviously there are going to be some differences, it doesn’t discount the idea that gender identity is hardwired.

    People should be more responsible with studies like this because a lot of “unbiased” individuals in the sciences fail to acknowledge that like any other field, science is full of cultural biases. Having a sound scientific method doesn’t guarantee it’s implementation or the conclusions drawn from various results will be as well. There is a terrible history of scientific racism for example and we must examine these biases constantly.

    Studies like this *will* be reposted by trans-exclusive radical feminists, right wingers and transphobes and often taken out of context in of itself. While you complain about over-enthusiastic language in the original studies(which frankly, were a lot more scientifically worded than this article) whereas you are using very decisive language when you say things like “According to a new paper, the brains of male-to-female transexuals are no more “female” than those of men.”

    It is very easy from a sheltered podium to say things like this and not be considerate of (or how bad you may look in a few months when this study is, in turn, debunked) how your wording may affect the test subjects and marginalised group you are dehumanising.

  • George Davis

    I think the other studies of gender dysphoria and the brain have used the same approach as this one; they compare the group averages and look for areas where they are different.

    One thing that was different about this study is that they only looked at trans women (born male) who were attracted to women.

    Many of the studies of gender dysphoria compare the brains of trans women who are attracted to men to cis women who are attracted to men and cis men who are attracted to women (cis=not trans). This means that if you find that the trans women’s brains are more like the cis women’s brains, you don’t know if this is because of gender identity or sexual orientation.

    This study avoids that problem, but it leaves out half of all trans women. We need more studies with trans women attracted to men and trans women attracted to women; we also need control groups that include cis lesbians and gay men.

    What I find most interesting about this study, though, is the ways the trans women’s brains were different from cis people’s brains. It looks like it has something to do with perception of your body.

  • Blood__Raven

    Jesus, mention transgenders and the comments get flooded with crap.

  • Christine Annette LaBeach

    BS transgender people have female brains from birth. I know because I am MTF.

    • Solar Surfer

      That explains a lot.

      Did you read this article before posting?




No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.


See More

@Neuro_Skeptic on Twitter


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar