Nosy Intelligence
The immune intelligence
your nose has been running
all along.
On the MHC, why your olfactory preferences change across your cycle, what copulins are doing, and what happens to all of it on the pill.
When we say there was "chemistry" between two people, we are — without knowing it — describing a biological reality. What we call chemistry is, in part, a real chemical event: volatile organic compounds in the air, processed by olfactory receptors, routed to the limbic system, converted into the subjective experience we name attraction. The metaphor was always literal.
This post is the second of two. The first was about what hormonal contraception does that the prescribing conversation omitted — including a mechanism I named but didn't fully explain: that combined oral contraceptives reverse the olfactory-immune preference system that naturally cycling women run continuously. This post is what that system is, how it works across your cycle, and what it means when it's altered.
If you haven't read the first post, that's fine — this one stands alone. But the connection is worth knowing: the biology here is one of the specific omissions from the consultation described there. Not a side effect. A documented reversal of an ancient and precisely calibrated immune intelligence system, running in your nose, that nobody mentioned.
The MHC: the immune system's recognition library
The major histocompatibility complex — MHC, or in humans specifically, the HLA complex — is the most genetically polymorphic region of the human genome. More variation between individuals exists here than anywhere else in the genome. This is not random. It is a conserved feature of vertebrate immune architecture, maintained across hundreds of millions of years of evolution because diversity at this locus is protective at the population level.
The MHC encodes proteins that sit on the surface of cells and present fragments of pathogens — viral peptides, bacterial components — to T cells. Think of them as the immune system's reference library: each HLA allele enables the immune system to recognise a particular range of pathogen fragments. The more diverse your HLA alleles, the wider the pathogen range your immune system can detect.
When two individuals with different MHC profiles have children, those children inherit HLA alleles from both parents — a wider immune recognition range than either parent alone. This is basic immunogenetics. And it is the reason that the drive to select MHC-dissimilar partners is not merely romantic preference — it is population-level immune defence encoded as attraction.¹
Your nose is not evaluating aesthetics when it responds differently to different people. It is running an immune compatibility algorithm on their volatiles, comparing their MHC signature to yours, and generating an output in the language your brain has available: interest or absence of interest, draw or neutrality, the thing we call chemistry.
The Wedekind experiment — and what came after it
In 1995, Claus Wedekind and colleagues at the University of Bern published what became the most cited study in the biology of attraction. Women rated the pleasantness of body odour samples from T-shirts worn by men with varying MHC profiles. Naturally cycling women rated the odour of MHC-dissimilar men as more pleasant, more attractive, and — notably — more reminiscent of their current or former partners.¹
The study has been treated, in popular science writing, as a curiosity: a party fact about sweaty t-shirts and immune genes. What is less discussed is what the Wedekind experiment was measuring. The finding was not that women like the smell of genetically different men. The finding was that the olfactory system is functioning as an immune compatibility sensor — running a comparative assay between the subject's own MHC profile and the candidate's, and producing a preference signal based on the degree of dissimilarity. The preference was for the candidates whose immune profile would extend, rather than overlap, the woman's own.
The field that followed expanded considerably on this. Subsequent research confirmed the MHC-odour preference link across multiple populations, documented its cycle-phase dependency, and — crucially — documented what happens to it when the hormonal state is pharmacologically altered.
Your olfactory sensitivity changes across your cycle
Olfactory acuity is not fixed. It varies measurably across the menstrual cycle, peaking in the follicular phase and reaching its maximum in the pre-ovulatory window. Detection thresholds — the minimum concentration of a scent required to be perceptible — decrease significantly toward ovulation. You are, in the days preceding ovulation, measurably more sensitive to smell than at any other point in your cycle.³
This heightened sensitivity is not uniformly distributed across odour types. Sensitivity to androgen-related compounds — specifically androstenone and androstenol, volatile steroids found in male axillary secretions — peaks most sharply in the pre-ovulatory window. The same compounds rated as neutral or mildly aversive in the luteal phase are rated as significantly more attractive around ovulation. Your olfactory preference system is not providing consistent information across the cycle — it is a phase-dependent instrument that sharpens exactly when it matters reproductively and relaxes when it doesn't.
The MHC-based preference follows the same pattern. The preference for MHC-dissimilar odours is strongest in the follicular and pre-ovulatory phases, and attenuates in the luteal phase — when conception is not possible and the reproductive signal is not being actively generated.⁵
Click any day on the heatmap above to read the olfactory profile for that cycle day.
Copulins: the signal you're sending
The chemical communication is not one-directional. You are not only processing information from others — you are emitting it, in concentrations that vary with your cycle and that have documented biological effects in those around you.
Vaginal secretions change in chemical composition across the menstrual cycle. In the pre-ovulatory window, concentrations of short-chain aliphatic acids — a class of compounds termed copulins — reach their peak. These compounds are not passive by-products of the cervical environment. Research by Grammer and colleagues documented measurable effects of copulin exposure on male testosterone levels and on the ratings men assign to female attractiveness.⁴ Preti and colleagues documented that male axillary compounds, conversely, affect LH pulsatility and mood in women recipients.⁵ The system is bidirectional — a chemical conversation, running at concentrations below conscious detection, between two endocrine systems.
Combined oral contraceptives suppress the ovulatory hormonal cycle that drives copulin variation. Women on the pill show reduced cyclic variation in copulin concentrations — the signal flattens. The chemical conversation that the natural cycle conducts is diminished. This is one of several specific ways in which pharmacological HPO suppression changes not just internal hormonal architecture but the chemical information being exchanged with the external world.
The OCP reversal: what the research actually says
In 2008, S.C. Roberts and colleagues published a study examining MHC-correlated odour preferences specifically in women using oral contraceptives.² The finding was this: women on the pill preferred the smell of MHC-similar individuals. The opposite of what naturally cycling women preferred.
The leading mechanistic hypothesis is that the hormonal state the combined oral contraceptive induces — which resembles, broadly, the hormonal profile of early pregnancy — reorients the preference algorithm from immune diversity toward immune familiarity. During pregnancy, proximity to genetically similar individuals signals family — safety, resource-sharing, protection. For a pregnant woman, MHC similarity codes for kinship. The preference reversal is not irrational. It is adaptive in its original context. The problem is that a non-pregnant woman is running it.
The practical consequence is documented and specific. Women who meet a partner while on the pill, form a long-term relationship, and then stop the pill — to try to conceive, or for any other reason — may find that their chemical response to the same person has shifted. Not the person. Not the relationship. The information system generating the response.
This has been in the published literature since 2008. The Roberts et al. paper was published in the Proceedings of the Royal Society — not a fringe journal, not a contested finding. It has not been incorporated into prescribing conversations. Women who experience this shift have largely had no language for it, and no mechanism to understand it as anything other than a change in their feelings.
The consent conversation this post belongs to
The MHC preference reversal is one specific mechanism in a larger pattern of undisclosed information about hormonal contraception. The previous post in this series documented SHBG, synthetic progestin receptor profiles, HPA axis effects, and the depression association. This post adds the olfactory-immune dimension — not because it is the most pharmacologically significant, but because it is perhaps the most intimate in its consequences.
Attraction — what we experience as the subjective pull toward one person and not another — is not purely psychological. Part of it is a chemosensory output of a continuously running immune compatibility assessment. When a pharmacological intervention reverses the preference direction of that assessment, and when women are not told this is happening, the implications extend beyond side effects into the domain of decision-making about relationships, partners, and life trajectories.
Women have the right to know that the chemical information they are processing about potential partners — and the preferences that information generates — may be running on a different orientation while they are on the pill than when they are not. That is not an anti-pill argument. It is an informed consent argument. The science has been available. The conversation has not been had.
What to do with this
- If you met your partner while on the pill and are considering stopping, give yourself time. The preference shift — if it occurs — is a recalibration of sensory information, not a verdict on your relationship. Attraction is more complex than a single chemical signal. Many relationships formed on the pill remain fully intact after discontinuation. But if something feels different chemically after stopping, you are not imagining it, and you have a mechanism to understand it.
- If you have never been on hormonal contraception, your MHC preferences have been running on natural cycle information throughout. This gives you a data point that women who have been on the pill since early adolescence — before the preference system was fully established and expressed in its natural state — may never have had access to. This is worth noting, not as a judgement either way, but as information about what you have and haven't been working with.
- Pay attention to olfactory response as information, not just aesthetics. The smell of a person is not random preference. It is a complex of volatile compounds that encode genetic, immune, and endocrine information. Finding someone's natural scent unpleasant is not a superficial response. Finding it compelling is not coincidence. The olfactory system is giving you a read that the visual system cannot.
- Understand that your cycle phase changes what you smell and how you interpret it. The same person's scent may be rated differently at different cycle phases — not because the scent changed, but because your detection threshold and preference weighting changed. This is worth knowing when you are making evaluations — particularly pre-ovulatory, when sensitivity and MHC preference strength are both at their maximum and the signal is the most distinct it will be.
The word we use is "chemistry." We use it as a shorthand for something that feels ineffable — an attraction that can't be reasoned through, a compatibility that exists at a level below language.
What the research shows is that it is not a metaphor. It is a specific biological process. A set of volatile compounds, processed by receptors that evolved for exactly this purpose, routed through the olfactory bulb to the limbic system, translated by a brain that has been calibrating this signal since puberty — comparing immune profiles, weighting dissimilarity, generating the subjective output we call draw.
And that process — running continuously, sharpening at ovulation, softening in the luteal phase, calibrated to your cycle and to your immune identity — changes when you take a pill that tells your hormonal system you are pregnant. It reverses. It begins preferring what it previously wasn't looking for.
Nobody told you this was happening. It was happening anyway.
These two posts together — the consultation that wasn't had, and the immune intelligence it failed to mention — are an attempt to return some of that information to the people it describes. What you do with it is yours.
– Nina- Wedekind, C., Seebeck, T., Bettens, F., & Paepke, A.J. (1995). MHC-dependent mate preferences in humans. Proceedings of the Royal Society B, 260(1359), 245–249. doi:10.1098/rspb.1995.0087
- Roberts, S.C., Gosling, L.M., Carter, V., & Petrie, M. (2008). MHC-correlated odour preferences in humans and the use of oral contraceptives. Proceedings of the Royal Society B, 275(1649), 2715–2722. doi:10.1098/rspb.2008.0825
- Gangestad, S.W., & Thornhill, R. (2008). Human oestrus. Proceedings of the Royal Society B, 275(1638), 991–1000. doi:10.1098/rspb.2007.1425
- Grammer, K., Fink, B., & Neave, N. (2005). Human pheromones and sexual attraction. European Journal of Obstetrics & Gynecology and Reproductive Biology, 118(2), 135–142. doi:10.1016/j.ejogrb.2004.08.010
- Preti, G., Wysocki, C.J., Barnhart, K.T., Sondheimer, S.J., & Leyden, J.J. (2003). Male axillary extracts contain pheromone-like activity that affects pulsatile secretion of luteinizing hormone and mood in women recipients. Biology of Reproduction, 68(6), 2107–2113. doi:10.1095/biolreprod.102.008136