You Recovered From COVID-19. Now Your Coffee Smells Like Sewage.

The way back from smell loss is its own strange experience.

Illustration of the parts of a face, separated into cubes
Jooeun Bae

Ruby Martinez was eating a banana when she noticed the nothingness. She chewed but tasted no sweetness. She sniffed but got none of the fruit’s redolent musk. “I started freaking out,” she says. She smelled a bottle of perfume. Nothing. She ate a pickle. Still nothing.

That was in June. Since then, her senses of smell and taste have started to come back—but intermittently and in strange ways. There were the two weeks in the summer when all she could smell was phantom smoke. The odor was so strong that she woke up one morning startled, convinced that something in her house was on fire. Sometime later, she was able to smell her boyfriend’s cologne again—but instead of the familiar scent she had always loved, it was a sickening chemical odor. There’s also the hand soap at work, which used to smell generically fruity to her but now smells exactly, and eerily, like Burger King Whoppers. Martinez used to love Whoppers, but she can’t stand the smell of the soap. Her co-workers find her predicament weird and frankly a little funny. “I’m like, ‘I know! What the heck?’” she told me. “Why does it smell like that? Why can’t it be something good?”

Despite these bizarre, lingering sensations, Martinez’s taste and smell have continued to recover. Last month, she went back to her favorite restaurant, Saltgrass, to eat her favorite food, steak, and “it was so good,” she said. Her sense of smell has further improved since then, and she can’t wait to return for another steak.

Martinez, who is 23 and lives outside Houston, credits the gradual improvement to dedicated “smell training,” a once-obscure protocol that has gained new prominence with COVID-19-associated smell loss. (Martinez doesn’t know whether she had COVID-19, but she was sick right before her smell disappeared.) Many years before the emergence of the novel coronavirus, a German doctor helped develop and standardize smell training for patients who had lost the sense, typically because of head trauma or viral infections. It involves sniffing four essential oils for 20 seconds every day over several months; some proponents of smell training recommend that patients recall memories associated with each scent—remembering lemon pie while smelling lemon oil, for instance. While supposed “cures” for smell loss, such as eating a charred orange or poking your forehead while getting flicked in the back of the head, have gone viral on TikTok, smell training is the only scientifically proven intervention for this kind of smell loss.

Nasal congestion and inflammation—as with common colds—often cause some loss of smell, but what happened to Martinez, and to many COVID-19 patients, is markedly different. Their noses are no longer congested nor inflamed, but they still can’t smell a thing. The exact cause is unknown.

In the past year, COVID-19 has drawn much more attention to smell loss, also known as anosmia, as well as to the strange ways smell is regained. Some patients go through a period of phantosmia, in which they experience phantom smells, or parosmia, in which they experience distorted smells—like Martinez’s smoke and Whoppers. These odors are quite foul, for reasons that are poorly understood; people find it extremely distressing to drink coffee that smells of sewage or to come out of the shower smelling like trash. “It’s worse than not being able to smell,” says Pamela Dalton, a cognitive psychologist at the Monell Chemical Senses Center. In most cases, the conditions are temporary. But the process of relearning to smell is as mysterious to us as how we lose it.

In the 1980s, a scientist at Monell named Charles Wysocki was studying androstenone, a pheromone found in the saliva of male pigs. (Commercially, it’s known as Boarmate.) To Wysocki, at first, the androstenone was odorless. But over time, he started noticing that he could pick out the androstenone bottles in his lab by their distinctly musky scent. Could other people learn to smell it too, he wondered? He found 20 volunteers who also could not initially detect androstenone, and after six weeks of sniffing the pheromone for three minutes, three times a day, half of them became sensitive to the molecule. This finding is in some ways intuitive; the olfactory senses of perfumers and sommeliers, after all, become more attuned to perfumes and wine with time.

But it was not until 2009 that Thomas Hummel, an ear, nose, and throat doctor at the University of Dresden, tested a standardized smell-training protocol in the lab. He and his colleagues asked 40 people with anosmia to sniff four essential oils—rose, lemon, eucalyptus, and clove—twice a day for 10 seconds each. After 12 weeks, the volunteers who’d adhered to the smell training regained some of their smell; those in a control group did not. But not everyone who did the smell training improved, and those who did improve didn’t necessarily fully recover. “It’s not a wonder drug,” Hummel told me, “but it helps to increase function faster.” Smell training is less like a cure than like physical therapy for the nose.

Since then, Hummel and his colleagues have tested variations of the smell-training protocol: using more complex scents, using a new set of scents every two months, and adding a picture of an object with the appropriate scent. None has much improved on the original.

Many people who encounter smell training do so through a charity in the U.K. called AbScent. AbScent’s founder, Chrissi Kelly, lost her sense of smell for the first time after a viral infection in 2012. When her doctor suggested smell training, she found little useful information: The studies that existed had been written for other scientists, not patients. So Kelly began writing down her own tips, such as putting drops of the essential oils in a small jar so the scent blooms like that of wine in a wine glass. Kelly says almost any substance with a scent can theoretically be used for training—calamine lotion, hand soap. But people find that beginning with a standardized guide is helpful, so the protocol on AbScent’s website suggests using the four scents from Hummel’s study. “There’s so much anxiety associated with smell loss,” Kelly told me. “Getting it right and doing it the right way is really, really important to people.”

Anosmia is associated with depression, loss of appetite, and diminished quality of life, but it’s also an invisible and underappreciated phenomenon. So Kelly started a Facebook group where people with anosmia could connect. When COVID-19 pandemic began, membership in the group took off; she could almost track the global spread of COVID-19 based on the locations of new members. Kelly eventually created another group just for COVID-19 patients, which now has 25,000 members. Some COVID-19 patients also lose their sense of taste, as Ruby Martinez did; others lose their ability to sense mintiness or spiciness, which are chemical sensations distinct from pure smell or taste. But smell loss seems to be the most common sensory disruption associated with COVID-19.

Kelly’s smell got quite good after smell training but in April, she got COVID-19, and lost her sense of smell for the second time. She’s now going through the process again. Unfortunately, she’s still experiencing terrible parosmia. “I’m of course very sad for the second loss of my sense of smell,” she said. “But I’m not giving up. I don’t think that it’s implausible that it will once again come back.”

Research on how this coronavirus attacks the olfactory system is beginning to emerge. In July, a team led by Sandeep Robert Datta, a neurobiologist at Harvard, published a paper suggesting that the virus does not directly infect smell-receptor neurons in the nose. Instead, the virus is probably infecting support cells in the olfactory system, which normally help replenish the supply of smell-receptor neurons. COVID-19 infection might interrupt this replenishment, leading to sudden but usually temporary smell loss.

Infection of the support cells may also help explain the parosmia that people find especially distressing. When the smell-receptor neurons do finally regenerate, they have to find their way up the nasal cavity, through tiny holes in the base of the skull, and finally to the right structures in the olfactory bulb. “That is an error-prone process,” Datta told me. Humans have only 400 distinct smell receptors, but can distinguish potentially 1 trillion different odors. That’s because a single molecule can bind to multiple smell receptors, and one recognizable scent can be made up of hundreds of different molecules that together activate a unique combination of receptors. If some receptors are missing or miswired, the brain might get a scrambled signal that results in parosmia. The miswiring normally gets sorted out after several weeks or months.

“But to be honest, I think there are many things about parosmia we simply don’t understand yet that suggest that more complicated things are going on,” Datta said. For one, why are the scents associated with parosmia usually so foul? The exact smell is “impossible to describe,” Kelly said, but people almost always grasp for words that evoke disgust: sewage, rotten, putrid. In parosmia, Hummel hypothesizes that the brain is interpreting unfamiliar, scrambled signals from miswired olfactory neurons as danger. “Why should something you’ve never smelled before be pleasant?” he says. In our evolutionary history, smells like smoke or rot have often acted as warnings.

Strangely enough, though, babies do not seem to have a natural aversion to bad smells. They certainly react to bad tastes or irritants, but they don’t turn their heads away from foul odors. “When they’re very young, their diapers do not disgust them,” Dalton, the psychologist at Monell Chemical Senses Center, notes. The aversion to certain smells seems to be learned over a lifetime, but once we’ve learned it, the reaction is very strong indeed.

The connections among smell, emotion, and memory are apparent in our neurobiology, too. Smell-receptor neurons converge on the olfactory bulb, which then sends signals to be processed by the olfactory cortex. But the olfactory bulb is also connected to the amygdala, which is involved in emotions and memory, and the hippocampus, which is also important in memory. “When you smell something, it drives activity in all these areas, and all of these areas talk to each other,” Datta said. “The way your olfactory cortex understands odors depends at least in part on your memories of the odors.” How a particular scent is represented in the brain remains an unsolved mystery. Neurobiologists are able to quantify visual or audio cues, but mapping the relationship between smells is a lot more complicated. Certain smells are definitely related to one another—lemon and lime seem to activate overlapping sets of neurons in the brain, for instance—but scale that to billions of potentially smellable molecules, and mapping them quickly gets impossible.

What’s clear is that the experience of smell is not simply what your nose detects but what your brain interprets. It is top-down as well as bottom-up. What you’re actively paying attention to—or not—will determine how you smell. We all get used to the smell of our own house. But we can also learn, as sommeliers do and the pig-hormone sniffers did, to pick out specific scents.

This is why smell training is not just passive sniffing. “It’s a process that really needs to take their concentration,” says Sunthosh Kumar Sivam, Ruby Martinez’s ENT doctor at Baylor College of Medicine. After Sivam suggested smell training, Martinez’s boyfriend took her straight to Whole Foods to pick out four essential oils. They followed her doctor’s guidelines for rose and eucalyptus, but swapped lime for lemon and cinnamon for clove. When Martinez does her smell training twice a day, she thinks of her mom’s arroz con leche as she sniffs the cinnamon. She remembers how its scent used to fill the house. Fortunately, she can smell and taste it again.