Biological differences in MTFs and cisgender males

People continue to misunderstand that there are very real biological differences between the average cisgender male and male-to-female transsexuals. Below is an important image to understand that I extracted from Transgender Chicago: The New Health Frontier.



The image above is the bed nucleus of the stria terminalis (BSTc), in the hypothalamus. Note that heterosexual and homosexual males have nearly identical brain structures in that region. Note that biological females have very different structures in that region from those of the males. And finally note that MTF transsexuals have brain structures that are very close to the female’s and nothing at all like the male’s neurological structures.

There are numerous other studies that highlight the biological differences between MTFs and cisgender males. What this image and many other studies show is that, in the brain where our essential self lies, that MTFs truly are female, not male.

This also shows why both the AMA and the APA regard being transsexual as a medical condition – because it is. And transition, for those that need it, is one of the most successful treatments available.


37 thoughts on “Biological differences in MTFs and cisgender males

  1. Reblogged this on Parker Kierce and commented:
    A great visual, being born “male” doesn’t mean you are actually “male” at all. I’m not sure if anyone has done any studies regarding brain scans of transgender people from earlier on in their lives? Is this something that might be “detectable” to any degree? It’s nice seeing scientific proof of what we all live every day.

    • I saw a youtube video, which I unfortunately failed to bookmark, in which a professor at Stanford stated that they checked (post-mortum) the brains of transsexuals who had transitioned and who therefore had had HRT and the brains of transsexuals who had chosen to not transition and had not experienced HRT. He stated that the brains were identical so HRT isn’t doing this.

      Further, I’ve seen a 2011 presentation to an annual AMA meeting where they discuss how these brain regions are formed early in the development of the fetus, primarily the first 16 weeks. That presentation is over an hour long but I highly recommend it.

  2. It’s also important to note that some of this area of study shows strong evidence for a gender spectrum, for people in-between neurologically too, requiring much more respect for the people with non-binary Gender Identities. Let me quote:

    “Sexual differentiation of the human brain in relation to gender identity and sexual orientation D.Swaab & A.Garcia-Fulgaras Functional Neurology, Jan-Mar 2009:

    One person we studied had untreated male gender dysphoria (S7), took no hormones and kept his transsexual feelings under wraps. He appeared to have a large INAH3 volume – in the male range – but a female INAH3 number of neurons (68) and a female BSTc somatostatin neuron number (95). Hence, this individual’s hypothalamic characteristics were mid-way between male and female values

    • This is absolutely true. One thing that my research in this area shows is that gender is not biologically a binary. There are XY women (from birth) and some have even been fertile and given birth. There are XX men (from birth). Genes are no more destiny than a house blueprint is destiny. They are a plan and during execution variations in the plan may occur. That those variations include people “in between” makes perfect sense since this is all biochemically based and the result of very complex interactions.

  3. Has this data been peer-reviewed and published? If so, where has it been published? I’d be very interested to see the article.

  4. Hmm… not convinced.

    The studies (Zhou et al 1995, Chung et al 2002) suggest that the size and shape differentiation is most likely due to gonadal steroid levels. The MTF transsexual subjects in the 1995 study had all been on feminising hormone treatments, and Chung et al concluded that development of the BSTc continues into adulthood. It’s entirely possible that this effect is caused by hormone treatment, rather than being an innate neural differentiation between cis and trans people. On the other hand, it does to indicate the success hormone therapy has in successfully treating transsexual people. Does the 2011 study include assessment of trans people prior to commencement of hormone treatments? If so, it might clear this up… otherwise it might be a good area for someone’s next paper.

    • You are not convinced because that’s just one study so I suggest you go here: One Stop Trans Brain Research List

      Go through that list. Go through the dozens of different studies about neurobiological differences, including studies that show differences without feminizing hormones having been used.

      I then suggest spending 70 minutes and watching this presentation to the AMA annual meeting in 2011. Because that is the official position of the AMA now – being trans is a medical neurobiological condition. That decision was not based on one single study, but dozens.

      You are, of course, free to continue rejecting that position. I won’t further try to change your mind, but just state that the research is large and growing that being trans is very likely for most trans people a neurobiological medical condition. The medical community accepts this now, which means that in a generation or so, the bulk of citizens in first world nations are likely to accept it too, barring some discovery that disrupts the findings thus far.

    • Jem you should have read all the comments on this page, you missed mine and it had the answer you were looking for already given to you right here! I’ll repost the pertinent bit:

      “Sexual differentiation of the human brain in relation to gender identity and sexual orientation D.Swaab & A.Garcia-Fulgaras Functional Neurology, Jan-Mar 2009:

      One person we studied had untreated male gender dysphoria (S7), took no hormones and kept his transsexual feelings under wraps. He appeared to have a large INAH3 volume – in the male range – but a female INAH3 number of neurons (68) and a female BSTc somatostatin neuron number (95). Hence, this individual’s hypothalamic characteristics were mid-way between male and female values

      Untreated. Took no hormones.
      There you go.

      • Thanks, Bayne. I had read that study (or rather the 1999 paper it was based on) separately and I still wasn’t convinced. A sample size of one isn’t really conclusive and the fact that subject’s count was within normal range for both male and female doesn’t strongly support the assertion about numbers of neurons. The SOM result is more interesting, but again we need a lot more data before jumping to any conclusions on the basis of a single subject. There’s lots more to be done in this area because most of the studies re-use data and samples from previous studies.

  5. Reblogged this on Miss Roxie's Blog and commented:
    This is so very important when we try to explain ourselves to our loved ones. They find it very hard to accept that we are other than our birth or physical gender and that our condition is a medical one and so the more we can disseminate this medical and scientific evidence of the differences betwee MTF transgender people and normal and homosexual males the better.

  6. I find these studies very interesting, but ultimately I don’t really agree with the style of the research being performed, which seems profoundly flawed in a number of areas. I’ve read several papers of this type, and the problems seem similar in each case.

    Problem 1: the study groups. First, you need to have large samples, to make sure that your sample group is truly representative of your population. Many of these brain-anatomy studies are based on single figures (e.g. half a dozen individuals), making it uncomfortably likely that selection bias is happening. I accept that it’s very difficult, expensive, time-consuming and ethically uncomfortable to do a study on a large number of people willing to donate their brains to medical science, which is why it so seldom happens (and why science of this kind is seldom reliable).

    Problem 2: Selecting your criteria of comparison. If you just take your samples, and compare them to see what you get, you will inevitably find differences. It is very bad science to say: aha! We found a difference in this tiny little part of the brain! That shows there is a difference between these two groups. It might equally well be that the difference could be explained by one group being coincidentally older than the other; or left-handed. Correlation is not causation.

    You can’t find your hypothesis in your data.

    Problem 3: over-generalising from your results. You can’t take little studies like these and say that they prove there are differences (or similarities) in brain morphology between the groups. At the very best, what this study (and others like it) does is to suggest pathways for future work, e.g. using real-time brain imaging modalities like fMRI to study the same region in living humans.

    Good science would say: we have these two groups, which we define clearly as follows. We hypothesise, in advance, that there is a difference in this little specific corner of the brain, which we think might be there because of hormonal differences, or morphogenetic differences between our groups. We studied these groups using this technique (histology from cadaveric specimens). Allowing for small sample sizes, this is what we found, (which either (1) lends some support to our hypothesis, or (2) doesn’t lend some support to our hypothesis).

    In fairness, over-generalising from results is not something the scientists themselves tend to be guilty of. It tends to be their readers, who use the results to lend credence to one or other point of view.

    I am uncomfortably aware how hard it is to conduct good science, and therefore uncomfortably aware of how common it is to conduct science which does its best to be good but just misses the mark. There is a lot of this out there. There is also a very regrettable quantity of work which doesn’t even try to be good science, other than calling itself that.

    And if you think I am being excessively critical of this paper, ask yourself this. You probably read this paper and thought “hey, that’s pretty interesting”. However, what would you have felt if the conclusion had been that there is NO difference to be found? From a scientific perspective, such a result would be equally valid, but would you be half as interested? Would you even believe it?

    This bias (the desire to publish and read results which are positive rather than negative) results in a powerful bias across the scientific literature. Negative results hardly ever get published, which means academics don’t know about them. If twenty trials show drug A is no better than drug B, it’s quite possible that none of them may be published. If one trial shows drug A is better than drug B, it is much more likely to be published, and read, and therefore doctors, in good faith, may believe that drug A is better, when in fact it isn’t.


    • I’m wondering what you read. Your “good science” is exactly what was done in several of these studies. A hypothesis was made, tests were done, and these discoveries were found. And the differences are not small. Heterosexual adult females have BSTc over 40% smaller on average than adult heterosexual males. And MtFs have BSTcs on the lower end of the female range. All of them. That’s 40% smaller than heterosexual adult males. That’s not a variation within statistical boundaries of normal ranges. That’s a variation of large scope.

      I suggest you visit this other entry on my blog: One Stop Trans Brain Research List

      It has dozens of formal and peer reviewed scientific studies. So these studies met the criteria of “good science”. They had to in order to pass peer review. It appears to me that you are cherry picking and not fully informed about the studies and the extent of neurobiological research that has been done in this regard over the last 20 years. The AMA didn’t make a decision that this is a neurobiological medical condition based on a handful of samples.

      You are, of course, free to believe what you wish but to tell professional scientists whose work has been peer reviewed that it is not “good science” seems, to me, to be a bit arrogant.

  7. Hi Cara. Thanks for your reply.
    Am I cherry-picking? Maybe. Am I not fully informed? Surely. Am I arrogant? Almost certainly!
    I am not suggesting for one moment that the authors of this paper, and similar ones, are villains trying to hoodwink us (or anyone else). I am merely saying that there are methodological flaws in their work (for very good and understandable reasons) which mean that their conclusions simply don’t carry very much weight, and that there is a tendency by others towards over-reliance on these flimsy conclusions.
    I believe that the over-reliance on publishing (“publish or perish”) as a surrogate measure of academic success pushes academics toward publishing just for the sake of publishing. Humans respond to incentives.
    I also believe that the publication bias inherent in most journals (bias toward publishing positive results rather than negative ones) skews our awareness of the work which is being done. That means that I am indeed not fully informed, but neither, with respect, are you or anyone else.
    Right at the top of my wish-list for good science in this field is good robust definitions of terms, which everyone accepts and everyone uses. For example, is someone who transitions at 25 scientifically comparable to someone who transitions at 55? (I think probably not). So how do we define “transsexual”? How do we define “homosexual” (behaviour? identitiy?) How do we even define “female” or “male” (phenotypical, hormonal, chromosomal)?
    In my own blog I touch on the difficulty of creating definitons here:
    Kind regards,

    • I think we also have to be honest with ourselves about the very small size of the trans community. I saw another study recently (about HRT safety) that followed 2000 transsexuals in Europe and the US. Normalized against the general population, that is equivalent to a study against 800,000 individuals. In other words, those 2000 individuals represented such a large sample compared to the general size of the trans community overall that the conclusions simply could not be ignored.

      There will always be issues with studying humans. But the various studies, starting with Zhou and moving forward to today, have tried to account for these issues. Moreover, the consistency in different studies finding differences between different sample sets of trans individuals in terms of neurobiology provide even more credence to this position.

      In short, the lack of evidence to the contrary, and the lack of evidence for other positions now puts the burden of further proof on those who want to claim this is not what’s happening.

      Extraordinary claims require extraordinary proof. I’ve yet to see the evidence that the neurobiological position is wrong, only semantic and philosophic arguments to the contrary.

      So my position to those who disagree is simply show me your evidence. Show me your data. And right now, today, those who oppose this interpretation lack such evidence, in spades.

      • I am not sure there is a “burden of proof” on anyone. True science simply seeks to find out what’s demonstrable and reproducible, without caring about who is “right” or “wrong”.

        I haven’t seen evidence that the neurobiological evidence is wrong either; I am just very difficult to convince that something as subtle and nuanced as human gender identity or behaviour can be clearly delineated by our current investigative modalities. I accept that the evidence is mounting, but I think it has a while to go before it is unequivocal.

        You allude to the trans community being small. I think it’s probably reasonably large; certainly large enough to be studied robustly.

        As well as problems of definition of terms, another one of my problems about trans studies is that they are often retrospective, observational or sometimes longitudinal. What I am really after is something really good and prospective. Take a large sample of gender non-conforming people. Scan them top to toe to establish a baseline. Randomise them. Provide one group with hormones, or surgery, and the other group with placebo (or have 3 or more arms; but one needs to be placebo). Follow them up regularly using blinded investigators to study their behaviour, their responses, their adjustment to life, their medical and psychological morbidity. And run your study for a long time (ideally 10 years) to see what happens.

        I can’t begin to count the reasons why a study of this kind will never be performed, but unwillingness of people to be recruited to it is probably the biggest. And the farther that studies come from this gold standard (using surrogate outcomes and markers and guesswork) the farther they come from solid evidence.


      • That study will never be done, just like that sort of study would never be done for a host of other human conditions either. What you’re suggesting is fundamentally immoral in denying a certain group actual health care in order to get the “double blind” effect. And it’s not necessary. Studies like those I’ve cited are post-mortem and don’t rely on changes after having HRT and/or surgery, as some of the trans identifying people never did transition, yet still had the same brain structure differences.

        And you are now changing the entire discussion. The original discussion was not about therapies, which is what you are now targeting, but about observed biological differences, something that is done routinely all the time, in everything from frogs to elephants as well. So you’re moving the goalposts to support your idealized “double blind” study and ignoring perfectly valid results derived from observations made from dissected biological remains.

        Biology has, for centuries, used dissection and observation as perfectly valid ways to examine physical body structures. Not everything has to be a double blind study. In fact, the National Institute of Health only recommends double blind studies in relationship to drug studies. We don’t conduct double blind studies of particle acceleration in physics. We don’t conduct double blind studies in geology of subterranean geological structures. A double blind study is wholly useless and irrelevant in many cases of scientific study and by itself a double blind study doesn’t constitute any sort of “gold standard” at all.

        The core principles of the scientific method do not demand a double blind study at all. Those principles are:

      • Ask a Question
      • Do Background Research
      • Construct a Hypothesis
      • Test Your Hypothesis by Doing an Experiment
      • Analyze Your Data and Draw a Conclusion
      • Communicate Your Results
      • Repeat the entire cycle
      • As for the trans community, excluding cross dressers (because they are not usually gender dysphoric nor wishing to change their gender totally), the total number of transsexual people in Europe and the US is probably between 1.5 and 2 million. The total for the US alone is estimated at just 750,000. So a study of even hundreds represents a large study compared to the overall population of transsexual persons. And these are the ones showing the neurological differences.

        I find it amusing that organizations as conservative as the AMA have already accepted this conclusion yet members of the greater transgender community continue to resist accepting this.

        And as for my burden of proof comment, if someone lacks a better hypothesis and the data to support it, then I find their criticisms especially hollow. When someone takes a position directly in contravention to observed data, in dozens and dozens of different studies, yet offers no alternative hypothesis and no data to support an alternative hypothesis, I can only question their motives in resisting the clear evidence to date.

  8. I’m not intentionally shifting the goalposts, but you’re right: I have muddied the waters, for which I apologise.

    I guess this study is designed to be observational: do trans brains and cis brains differ in a measurable and reproducible way? (And I wholly agree that empirical observation is a cornerstone of scientific enquiry). However, I guess my interpretation of it was post-hoc: does transition cause changes in people’s brains which are measurable and reproducible?

    I am, however, deeply interested in this question: Suppose I am gender-dysphoric, and I am a rational individual seeking to maximise my outcome. What is the course of action which is likely to produce the greatest good to me? It may be that hormonal therapy, as well as full sex reassignment, is likely to produce the greater good and the lesser harm. On the other hand, it may be that, while it produces some benefits, HRT and GRS is overall more harmful than beneficial. And it may be that I could be stratified into a subgroup which would demonstrate clear benefit from one course of action versus another. (All medical treatments carry harms as well as benefits, and it can certainly be shown that some individuals come to regret the process of transition). For the moment, we simply cannot answer this question with certainty.

    Doing randomised, double-blinded, controlled trials isn’t unethical or immoral, but essential (paediatric oncologists have managed it for years, and as a result have made huge strides in dealing with dreadful childhood cancers). Placebo controls are not always necessary; one can compare a new treatment against the accepted gold standard, provided randomisation and blinding are carried out. We don’t need to do double-blind studies on (say) electrons, because electrons are all identical and behave identically, whether humans care about them or not. But human beings are affected by the presence of other human beings, and therefore studies on human beings should observe rigorous measures to prevent expectancy, observer bias and other human tendencies from muddying the result. (This is an area in which I claim some direct expertise).

    I think (at least I hope) what the AMA has tried to do is to synthesise the best evidence it can find in order to reach its viewpoint. I like to think that if the evidence changes or evolves, so will the AMA. I also think there are alternative viewpoints which are acceptable to hold with our current limits of knowledge and technique.

    Your viewpoint seems to be that we have it all sewn up. I personally think we are still some way off, and I think there are good grounds for uncertainty, even scepticism in some areas. Nonetheless, I am looking forward to the developments which the future promises.


    • Actually, my viewpoint is not that it’s all sewn up, just that there is now a preponderance of evidence in one direction, and none of the competing theories have evidence that (a) either supports their position to this level or (b) answers the existence of the neurobiological differences. Those must also be accounted for now by any competing theory. They are hard data and cannot be thrown away. Any competing theory will also have to account for those neurobiological differences. Any attempt to just sweep them under the rug would itself be bad science.

      As to your question about course of action, that has been studied extensively but not double blind, because to do a double blind leaves transsexuals at extreme risk of suicide and that is simply immoral. The famous Swedish study that looked at HRT and GRS followed 324 sex-reassigned patients over a 30 year period – 1973 to 2003. This was 133 FtMs with 1330 birth sex matched controls for comparison, and 191 MtFs with 1910 birth sex matched controls for comparison.

      During that time they saw the transsexual patients break into two distinct groups, the pre-1989 group and the post-1989 group. The entire group saw 29 suicide attempts post-GRS total but the vast majority of these were in the pre-1989 group. Adjusted hazard ratios for the controls for suicide attempts was 1.0 outcome incidence rate per 1000 person years (range 0.8-1.4, 95% confidence interval). The rate for the pre-1989 group was 7.9 and the rate for the post-1989 group was 2.0, nearly as low as the background population.

      Now what we have to pay attention to is that the suicide attempt rate of the entire sex-reassigned group after GRS was 29 attempts among 324 participants tracked. That is not quite 9% for the total group versus 44 incidents among 3240 controls, for an attempt rate of 1.4% (which matches most other first world nations for typical suicide attempt rates among the general population). But, what the study did not try to account for was pre-GRS suicide attempt rates, which for transsexuals we know to be in excess of 46% (Williams Institute at UCLA). The more popular 41% transgender suicide attempt rate often quoted is lowered by the inclusion of cross dressers who have much lower rates than the transsexual community, but still higher than the background population.

      Further, when we look at the incidence rate, we find that for the post-1989 group, which had access to additional medical techniques above and beyond just GRS (such as FFS), the incidence rate is only double the normal population, or under 4%. This is a 90% reduction in suicide attempt rates. It’s still not as good as the general population, so the conclusion of the study was not that transition and GRS are bad (in fact the study supports them!) but that additional psychiatric care is likely to be needed by the transsexual community post-GRS.

      So, to answer your question, this has been studied and the results have been known for over a decade now – HRT, transition, and GRS produce definite and clear improvements for transsexual individuals. And I’m really unsure how you can perform a better study among humans than what the Swedes did without creating ethical dilemmas and denying outright care to a group that has a history of epidemic suicide rates.

  9. Pingback: Interesting Read On Trans Brain Structures! | TransHope United

    • There are references in the comments. Also, on the right sidebar, find my own page One Stop Trans Brain Research List. I actively maintain that list. It is long. Some of the links are studies. Some are articles about studies.

      The process of identifying the parts of the human brain that relate to gender identity are still ongoing but there are solid leads at this point. My reference page has articles about both FTM and MTF individuals, because it appears the root cause in both cases lies with in utero hormone ratios not being in the expected range for a given gender.

      Please also watch this 2011 AMA Annual Meeting web cast about hormonal ratios during pregnancy and how they cause people to be transgender (as well as gay, lesbian, bisexual, etc.).

      AMA 2011 Annual Meeting Webcast About In Utero Hormonal Ratios

  10. Pingback: A General Education Concerning the Biologic Basis of Gender Identity (unfinished/working title) | kittysbelle

  11. Pingback: a general education-transgender (unfinished) | kittysbelle

  12. Pingback: Interesting Read On Trans Brain Structures! – TransHope United

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s