Parosmia
The nose does not go silent. It goes wrong. Coffee becomes burning refuse, and the past stops arriving.
Most people understand losing your sense of smell as a silence. Parosmia is not that. It is a distortion, in which familiar things arrive as something else and usually as something foul, and it typically appears after a person has begun to recover. The current explanation is a wiring error: olfactory neurons regenerate after viral damage and a proportion connect to the wrong targets, so the brain receives a correctly transmitted signal from the wrong address. In 2022 a research group identified the specific molecules most often responsible. This essay covers the mechanism, those molecules, what it costs the people who have it, and what is actually known to help.
Everything this house has ever written rests on an assumption we have never examined in public: that a smell will arrive intact. That the jasmine outside a window in 1998 is still recoverable, that the scent of a grandmother's kitchen is stored somewhere and can be reached, that fragrance carries memory the way a road carries a car.
For a significant number of people, that road has been rebuilt wrong. They can smell. The channel is open. What comes down it is not what was sent.
A woman picks up her infant and recoils, because the child now smells of sewage and she knows, intellectually and completely, that the child does not smell of sewage. A man makes coffee every morning for eleven months and every morning it smells of burning refuse. Somebody stops being able to sit at the family table. Somebody else cannot be in the same bed as a partner whose skin has become, through no fault of anyone, unbearable.
This is parosmia, and until recently almost nobody outside ENT clinics had heard of it.
Four different things, routinely confused.
The words matter here, because people are frequently told they have one thing when they have another, and because the outlook and the treatment differ.
No smell at all. The most widely known and the most straightforward to describe. Food becomes texture and temperature; the sense of taste largely goes with it, because most of what we call flavour is retronasal smell.
Smell is present but weak. Extremely common, frequently unnoticed, and increases with age. Many people have it mildly and have never thought about it.
A stimulus is present and is perceived as something else, nearly always unpleasant. The trigger is real and external. This is the subject of this essay.
Perceiving an odour that is not there at all. Frequently smoke, burning or chemical. It can accompany parosmia and can also occur alone, including in grief, where people report smelling somebody who has died.
The signal is fine. The address is wrong.
Olfactory sensory neurons are unusual. They sit in the roof of the nasal cavity, exposed to the outside world, and they are among the very few neurons in the human body that routinely regenerate throughout life. That regenerative ability is the reason smell often comes back after viral damage. It is also the reason it can come back wrong.
Each neuron carries one receptor type, and its axon has to find its way to a specific target in the olfactory bulb called a glomerulus. All the neurons carrying a given receptor converge on the same one or two glomeruli. That arrangement produces an ordered map, and the brain reads position on the map as identity. When a virus destroys a large population of these neurons and the tissue regrows, the new axons have to find their targets again, through an environment that is still inflamed. A proportion of them arrive at the wrong glomerulus.
Nothing is broken after that point. The nose detects the molecule correctly, the neuron fires correctly, the signal travels correctly, and it is delivered to a location the brain has spent your entire life interpreting as something else. The system is working. The map has been redrawn.
Nothing is broken. The nose detects correctly, the neuron fires correctly, and the message is delivered to the wrong door.
Fifteen compounds do most of the damage.
For a long time the accounts were purely anecdotal, and the same foods kept appearing: coffee, meat, onion, garlic, chocolate, egg, toothpaste. In 2022 a group led by Jane Parker at the University of Reading did the obvious and difficult thing. They put food extracts through a gas chromatograph and had people with parosmia smell the separated components one at a time, so that each molecule could be judged alone.
Twenty-nine participants. More than thirty molecules identified as distorted. And a shortlist of fifteen compounds that came up again and again. Furanthiols, disulfides and pyrazines accounted for nine of the ten most frequently reported. They are all small, potent, heterocyclic molecules carrying an exposed sulfur or nitrogen, and they are all things that ordinarily smell of roasting, cooking and browning.
Notice what these molecules have in common in ordinary life. They are the smell of the Maillard reaction: coffee roasting, onions browning, meat searing, bread crusting, chocolate conching. They are, almost exactly, the chemistry of cooked food and therefore of hospitality, family and home.
Of all the possible sets of molecules that could have been affected, the one that was affected is the one that runs through every kitchen in the world.
2-furanmethanethiol.
The single most frequently reported trigger in the Reading study was 2-furanmethanethiol, also called furfuryl mercaptan. In an unaffected nose it is one of the defining smells of roasted coffee, and it is detectable at an exceptionally low threshold, which is to say your nose is extraordinarily good at finding it.
That last property is the problem. Sensitivity is not a volume control that parosmia turns down; it is a gain that stays high. A molecule you could previously detect in vanishingly small amounts is still detected in vanishingly small amounts, and now arrives labelled as something rotten. The compounds that hurt most are the ones the nose was best at.
This is why the advice to simply avoid triggers fails in practice. A person can avoid coffee. They cannot avoid a city that smells of coffee, a colleague who has just had one, or the residue in a cup across a room.
It is not a quirky symptom.
It is worth being plain about this, because parosmia is often reported in a tone of mild novelty, and the people living with it consistently say that the tone is part of the injury.
Eating becomes difficult and sometimes frightening, and weight loss is common. Social life contracts, because most of it happens around food. Intimacy is affected in ways people find humiliating to describe. And there is a documented relationship between olfactory dysfunction and mood: people with olfactory disorders score worse on standard depression measures, the effect scales with severity, and improvement in olfactory scores has been associated with a reduction in depression severity. The relationship runs in both directions and it is not small.
There is also a specific loss that this journal is obliged to name. Smell is the sense most directly wired to memory, and a person with parosmia does not merely lose pleasant sensations. They lose access. The smell that used to return them to a childhood house now returns something else, and the door does not open. People describe this as a bereavement, and they are not being dramatic. It is one.
Smell training, and patience nobody wants to be asked for.
The intervention with the best evidence behind it is unglamorous and free. Olfactory training, developed by Thomas Hummel and colleagues, asks a person to smell four strong, familiar odours for around ten seconds each, twice a day, every day, for months. The classical set is rose, lemon, clove and eucalyptus, chosen because they sit in four different categories of odour rather than because of anything special about those plants.
The mechanism is thought to be the obvious one: repeated, deliberate stimulation supports the regenerating system in re-establishing a usable map, and gives the brain consistent material to recalibrate against. Studies in people with parosmia specifically have found improvement not only in identification but in hedonic perception, which is the technical way of saying that things stop being so revolting.
The compounds that hurt the most are the ones the nose was best at finding.
What this asks of an industry that sells pleasure.
Fragrance is built on an assumption of a shared nose. Every brief, every panel test, every review culture, every note pyramid assumes that when we say rose we are both receiving roughly the same thing. Parosmia is the largest-scale demonstration in living memory that this assumption is a convenience.
The industry has been almost silent on it, and the reason is not sinister so much as commercial: there is no product here. You cannot sell a fragrance to somebody for whom fragrance has become a hazard. So an enormous population acquired a condition that sits precisely in this industry's area of expertise, and the industry mostly looked away, while the useful work was done by ENT researchers, a food chemistry group at Reading, and a charity.
There are things a house can actually do, and they are small. Say plainly that a distorted response to a perfume is a known medical phenomenon and not a matter of taste. Do not treat every negative reaction as a failure of the wearer's sophistication. Understand that for some readers a description of a warm gourmand base is a description of something frightening. And remember that the vocabulary we all use, notes and accords and facets, quietly assumes a functioning map.
The premise, examined.
This house is named Harum Oleh Memori, fragrance through memory. We have written more than forty essays on the proposition that a smell can return you to a place, and we have never once written about what happens when it cannot.
Parosmia is the strongest test of that premise we know of, and we do not think it breaks it. What it does is clarify what the premise was actually claiming. The memory is not in the molecule. It never was. A molecule is a shape moving through air; it carries no meaning of its own, and it means what it means only because of where it lands and what that place has been taught. Parosmia demonstrates that with unusual cruelty by leaving the molecule intact and moving the place.
Which means the sentence we have built a house on is more literally true than we understood. Fragrance does not contain memory. Fragrance is a key, and the lock is a person, and the lock can change.
For the people reading this whose locks changed, we do not have anything to sell you and we are not going to pretend otherwise. What we can do is write it down properly, because one of the things people with this condition report most often is that nobody around them understood it was real.
This essay is journalism about published research, not medical advice. Anyone experiencing distorted or lost smell should speak to a clinician.
- The trigger molecules, the gas chromatography olfactometry method and the fifteen compounds: Parker et al., Insights into the molecular triggers of parosmia based on gas chromatography olfactometry, Communications Medicine 2022; University of Reading; Chemistry World.
- Aberrant reinnervation, the miswiring account and the timing of onset: Post-viral olfactory loss and parosmia; Kim et al., Long-term olfactory loss post-COVID-19, 2024.
- Olfactory training and its effect in parosmia specifically: Effects of olfactory training on patients with parosmia; Smell training, overview of the Hummel protocol.
- Support and training kits: AbScent.
- Related reading in this journal: Nobody Can Predict a Smell, on the receptor map, and The Proust Machine, on the route from smell to memory.