Male vs. Female Brain Differences & How They Arise From Genes & Hormones | Dr. Nirao Shah

The spine of this stretch of the conversation is a single claim, stated early and returned to repeatedly: chromosomal sex in mammals runs through one gene. Huberman sets up the textbook frame of 23 chromosome pairs, with autosomes identical between the sexes and a sex-chromosome pair that is XX in females and XY in males, and Shah locates the decisive element not in the Y chromosome as a whole but in a gene sitting on it, SRY, described as the sex determining region on the Y. That gene, in Shah's account, dictates whether the embryo will have testes; if testes form, they make testosterone, which masculinizes the genitalia, the brain and the rest of the body.

The mechanistic detail matters for everything that follows. SRY is a transcription factor, meaning it is transcribed into RNA and translated into a protein that regulates the expression of other genes, switching on or silencing suites of them in the bipotential gonad, the structure that can go either way. Shah dates gonadal bipotency in humans to the late first or early second trimester, and notes the same logic holds in the mouse, where the gonads stay bipotential until day 12 of a roughly 20-day gestation. Huberman pushes the inference one step further, asking whether the fetal brain is therefore essentially identical between the sexes before SRY acts, and Shah answers, "That's the thinking, yes." That answer is worth flagging as an inference rather than a measurement, and Shah makes the same point about the mouse, where the brain is thought to remain bipotential almost until birth.

SRY alone does not build the body, and the episode is careful about the division of labor. Once SRY makes a testis, the testis secretes at least two hormones: testosterone, and an anti-Müllerian hormone that suppresses differentiation of the uterus, the vaginal tract and the fallopian tubes. Testosterone then masculinizes bipotential external genitalia. Dihydrotestosterone, or DHT, produced from testosterone by the single enzyme 5-alpha-reductase, is the more potent actor on external genitalia because it binds the same androgen receptor at much higher affinity. Huberman compresses the logic into a formula: you need suppression of femaleness plus amplification of maleness, which is why androgens reaching an XX fetus do not by themselves produce a male.

The symmetry the audience might expect is not there, and Shah is explicit about it. Asked whether the absence of SRY is what determines femaleness or whether a separate female-determining gene exists, he answers that no such single gene has been identified in mammals, in mouse or human. What exists instead is a genetically programmed pathway that produces a female body and brain when SRY is absent, with SRY acting to tamp that program down and boost maleness. The strongest supporting cases are the chromosomal anomalies: SRY can translocate onto an autosome, producing XX animals that develop as males, and loss-of-function mutation of SRY produces XY females. Shah declines to convert any of this into an evolutionary story, saying plainly that he does not know what it says about evolution.

irreversible differentiation of a bipotential brain along a male pathway with testosterone

Shah's framework is the classical two-stage model of steroid hormone action. Testosterone, estrogen and progesterone act early in development, embryonically or in utero in humans and perinatally in mice, to produce what is thought to be an irreversible differentiation of the brain along a female or male pathway. After that species-specific critical window closes, the testes and ovaries go quiescent, and the circuits sit waiting. At puberty the hormones return and activate the circuits that were laid down earlier, which is why adult behavior appears only then. The organizing effect builds the hardware; the activating effect runs it.

The mouse work lets Shah extend the window past birth in a way human data cannot. A female mouse given testosterone on the first day of life will show masculinized behavior as an adult even though she has no testes, which is what makes the hormone, rather than the gonad, the operative variable in the experiment. The human version of the same manipulation is read off clinical conditions rather than done deliberately, and the timing difference is large: SRY switches on around the start of the second trimester in humans, whereas the mouse brain is still described as plastic to testosterone almost until birth.

Huberman repeatedly tries to convert this into a version of the nature-nurture question, and Shah resists the strong reading in one specific place. Humans are different, in his account, because of cortical volume: a much larger cortex grants many more degrees of freedom over when, what and where to act, while the basal structures for the relevant behaviors, the hypothalamus and the amygdala, remain highly conserved. The behaviors are still encoded in the brain, but they can be controlled or inhibited at appropriate moments. That is a claim about modulation, not about erasure of the underlying circuitry.

The empirical anchor Shah offers is a 1959 experiment by Charles Phoenix on guinea pigs, chosen because guinea pigs, like humans, undergo sex differentiation prenatally rather than at birth. When pregnant females were given testosterone, the female offspring were born having seen testosterone in utero and, as adults, showed male-typical sexual behavior, including mounting and thrusting, with very little female-typical receptivity. The result was not a matter of activation alone: when the same adult females were given estrogen and progesterone boosts intended to increase female sexual behavior, they still showed little of it and continued to mount like males. Shah reads this as the organizing effect in its clearest form, and Huberman summarizes it as the presence of something and the absence of something in the same animal.

Aggression was not part of that paper, and Shah is explicit that the extension comes from his own mouse work rather than from Phoenix. Giving day-one pups testosterone produces females that become territorial as adults, which is a male-typical trait in mice. The qualification is important: male mice are territorial, while female mice in the laboratory fight less, except when they are mothers nursing a litter. Huberman asks whether maternal aggression is testosterone-mediated, and Shah answers that this is not known. That single exchange marks the boundary of the model, since maternal aggression is a robust behavioral difference that the organizing-hormone account in this conversation does not yet explain.

Congenital adrenal hyperplasia is the episode's cleanest human case of androgen exposure without SRY. Shah corrects a premise in Huberman's question: the mutation is in the baby, not the mother, and it disables an enzyme in the cortisol pathway, so cortisol precursors are shunted into androgens. The resulting XX infant is born with masculinized external genitalia, corrected surgically now that the condition is recognized, and given cortisol because it is essential for survival. Crucially, she keeps her ovaries and remains fertile as a female, since she has no SRY gene.

The degree of virilization is not fixed, and both speakers treat that as a substantive point rather than a rhetorical one. Shah says the outcome depends on exactly when the adrenal androgens kicked in, so the range runs from an enlarged clitoris to a small or normal-sized penis. Huberman's phrase for the intermediate zone, offered as a description rather than a joke, is that there is a continuum, and Shah agrees. This is the one place in the conversation where the binary framing is relaxed for human anatomy on the speakers' own initiative, and it comes from timing and dose rather than from a different mechanism.

The mirror-image case is complete androgen insensitivity syndrome. These individuals are XY with a functional SRY and have testes, but a mutation in the androgen receptor means they cannot respond to testosterone. Externally they are completely feminized, are raised as females, regard themselves as females, and are typically diagnosed at puberty when menstruation fails to start. Shah's summary is that their inability to respond to testosterone is what feminized them, which isolates receptor function from hormone presence as separate causal steps.

The third condition, 5-alpha-reductase deficiency, is the one Shah uses to argue that the brain is not simply waiting for puberty. These children cannot convert testosterone to DHT, so their external genitalia are feminized early and they are raised as girls, only to virilize at puberty when testosterone rises enough to drive penile development on its own. Shah's correction to Huberman's framing is technically central: these individuals had always been making testosterone, just not DHT, so testosterone was available to act on the brain throughout development, while the visible genital change at puberty reflects the later, DHT-independent route. He also notes that many of these individuals switch to living as boys and men at that point, and that the switch is not forced on them.