Neuroscience has become a battleground in debates over gender identity. The findings, when examined carefully, don’t say what advocates claim they do.
Key points at a glance:
- Headlines often claim neuroscience has found a biological cause for gender identity, leading to ‘male’ and ‘female’ brains.
- However, scientific studies show only small differences and severe limitations in sample size.
- Brains also change continuously in response to medication, stress, and life experience.
- A single brain scan, or even multiple such scans, therefore cannot prove that a gender identity exists from birth.
- Sound public policy in Singapore requires clear scientific facts rather than media slogans.
Every so often, a headline asks if scientists have found the “transgender brain.” The implication is usually simple: that the brain of a person who identifies as “transgender”, more closely resembles average patterns associated with the opposite sex than with that person’s biological sex.
In public debates over gender identity, that claim has a tactical objective. If neuroscience has identified a stable, inborn neural signature, then transgender identity may appear to have been settled by biology.
But that is not what the current evidence establishes.
Neuroimaging research has reported some average differences between transgender-identifying and non-transgender groups. In some structural, functional, and metabolic measures, participants who identify as transgender have shown patterns more similar to those associated with the sex with which they identify. But in other measures, they have shown patterns more similar to those associated with their biological sex.
The evidence is therefore mixed.
How a Small Finding Became a Big Story
A systematic review of the neuroimaging literature found some average brain-measure differences between transgender and non-transgender groups. But it did not identify a single, consistent neural pattern that could straightforwardly be called a “transgender brain.”
More importantly, the same review reported ‘conflicting results,’ small-to-moderate samples, and methodological variation extensive enough to prevent a full meta-analysis. By no means therefore, do we arrive at the grand conclusion of a transgender brain.
The reality is that not a single brain scan has been clinically validated as a reliable way to determine whether a particular individual identifies as transgender. Nor have existing studies established that observed brain differences were innate, rather than reflecting development, hormone exposure, sexual orientation, health, medication, stress, or years of lived experience.
Brains Change, and So Does the Evidence
This causal uncertainty matters especially because most studies in the field examine adults. By adulthood, brain measures may reflect puberty, ageing, hormone exposure, medication, social experience, psychological distress, illness, and chronic stress.
A scan taken at one point in adult life can identify an association, but it cannot ordinarily establish its timing or cause. The difference may have been present early in development, emerged during puberty, or been shaped later by these factors. A cross-sectional design cannot reliably distinguish among those possibilities.
This is not a hypothetical concern. A 2025 scoping review of 52 studies logged 340 significant brain-related changes across 63 regions in transgender and gender-diverse participants, including 274 reported after hormone treatment. Yet only 12 findings could be directly compared across studies. The review called for larger samples, longer follow-up, and standardised methods before stronger conclusions could be drawn.
Hormone treatment creates a further interpretive problem. Because estrogen and testosterone can reshape brain structure, comparing a treated adult’s brain with an untreated control cannot, by itself, establish what that brain looked like before treatment began. Also, it cannot isolate the effects of treatment from pre-existing differences.
Ironically, hormone treatment can obscure the very evidence that gender activists want to see.
The Limits of Structural Claims
The human BSTc is frequently cited as evidence of an innate neurological basis for transgender identity. In post-mortem research, the region in males who identified as women was reported to resemble the female pattern more closely than the male one.
But the foundational evidence was extremely small. The famous 1995 study examined only six transgender-identifying males, and the investigation was exploratory: researchers examined several hypothalamic regions rather than beginning with a specific prediction about the BSTc.
There is also a timing problem. A developmental study found that male-female differences in BSTc volume and neuron number became statistically significant only in adulthood, later than the age at which gender identity typically emerges.
This does not exclude every possible developmental influence. But an adult BSTc difference cannot, by itself, establish that the feature existed before gender identity emerged or caused it.
Recent MRI work has also questioned whether total BNST volume is reliably sexually dimorphic in humans, reinforcing the need for caution when generalising from older post-mortem studies of particular subregions.
Hormone exposure is not the only confounding factor. Depression, anxiety, chronic stress, medication, and physical illness can also correlate with changes in brain structure or function. These factors do not explain away anyone’s identity. They do, however, limit what an observational adult scan can establish about the origin of that identity.
Sexual Orientation Complicates the Picture
Reported sexual orientation may also affect the results.
Some early studies compared transgender participants with control groups that did not have similar sexual orientations. This is important to account for because some brain findings change when researchers group people by sexual orientation.
This does not prove that sexual orientation is hardwired into a specific brain structure. It means researchers cannot be sure that every brain difference they find is caused by transgender identity, especially when another reported characteristic is associated with variation in the same measures.
In fact, in some studies, transgender participants were more similar to control participants with the same reported sexual orientation than to people of the opposite sex as a whole.
That weakens the claim that transgender identity has a unique brain signature. If the results change when researchers sort people into different self-reported groups, it becomes harder to say that one clear biological pattern explains transgender identity.
Sex Differences Are Real
None of this requires denying that male and female brains differ.
Large neuroimaging studies find reproducible average sex differences in brain structure and function. Biological sex influences human development, including the brain.
But population differences are not the same as two perfectly uniform types of brain. Men and women overlap on many neurological measures, just as they overlap in height despite large average differences between the sexes.
A male possessing a particular feature more common among females therefore does not mean he possesses a “female brain”. Nor does finding several sex-atypical features establish that the brain itself belongs to the opposite sex.
The relevant question concerns the strength of the claim. Sex plainly affects the brain, but whether neuroscience has discovered an opposite-sex neurological identity within transgender people is a much stronger proposition.
What Machine Learning Actually Shows
Newer studies use machine-learning programs to analyze many brain measurements at once instead of focusing on one brain region. These programs are trained to distinguish male and female brain scans. In one study using MRI scans matched for brain volume, the program classified participants’ biological sex with more than 92% accuracy in the main group studied.
This supports only a limited conclusion. Male and female brains contain groups of measurable differences that are consistent enough for a computer to detect biological sex fairly accurately in that study.
But this does not mean that every person has one of two completely separate brain types. Human brains overlap, and an individual can have some features that are more common in males and others that are more common in females.
Most importantly, the study does not show that a person with some sex-atypical brain features has the brain of the opposite sex. The computer was trained to classify biological sex, not gender identity. It can recognize patterns that tend to differ between males and females, but it cannot determine what those patterns mean about a person’s identity.
That is why this evidence works against the “trans brains” thesis. If the brain contains many overlapping, sex-linked features rather than two distinct male and female types, then finding a few features that resemble the opposite sex does not prove that someone has an opposite-sex brain. It shows only that human brains vary, and that some individuals have combinations of features that are less typical for their biological sex.
Slogan, Not Discovery
None of this means gender dysphoria is not real. The distress it causes is genuine, and compassionate care for people navigating it is not in question here. The narrower question is whether neuroscience has identified a “transgender brain” that proves transgender identity is a fixed biological category present from birth.
The current evidence does not support that claim.
Some transgender populations show neurological differences. Some findings appear sex-atypical. Others remain aligned with biological sex. Still others concern networks involved in self-body perception rather than sexually dimorphic regions at all. Sexual orientation, hormone exposure, development and other confounders complicate the picture further.
More fundamentally, an anatomical difference observed in adulthood cannot simply be read backwards as evidence of an inborn identity.
Until longitudinal studies can distinguish cause from consequence, control major confounds, and produce robustly replicated findings, “the transgender brain” remains an activist’s slogan, not an established neuroscientific discovery.
