Smell Statistics 2026: How Humans Smell, by the Numbers

A 2014 study in Science estimated that humans can discriminate at least 1 trillion olfactory stimuli¹, a figure two 2015 papers dispute^2,3. Smell loss is common with age: in a US smell test of adults aged 40 and over, 12,4% had olfactory dysfunction, rising to 39,4% of those aged 80 and over⁴.

In short

• The “1 trillion odours” estimate (Bushdid et al., 2014) is widely repeated, but two peer-reviewed critiques from 2015 say it is not supported^1,2,3.
• In the US NHANES smell test, 12,4% of adults aged 40 and over had olfactory dysfunction, and 3,2% were anosmic or severely hyposmic⁴.
• Sweden’s Skövde study found olfactory dysfunction in 19,1% of adults, including 5,8% with anosmia⁵.
• Smell loss rises steeply with age: 4,2% at 40 to 49, 12,7% at 60 to 69 and 39,4% at 80 and over in NHANES⁴.

Key smell statistics at a glance

How many smells can humans tell apart?

The best-known number in smell science is also the most contested. In 2014 Bushdid and colleagues wrote that “we calculated that humans can discriminate at least 1 trillion olfactory stimuli”, adding that smell “far outperforms the other senses in the number of physically different stimuli it can discriminate”¹.

The same abstract explains why the study mattered. It says the lay and scientific literature typically claimed that humans can discriminate 10.000 odours, but that “this number has never been empirically validated”. For comparison, it notes that humans can discriminate several million different colours and almost half a million different tones¹.

The two 2015 critiques

Two peer-reviewed papers in eLife challenged the calculation the following year, and they are worth reading alongside the headline figure.

• Markus Meister wrote: “This claim is wrong. I show that the failure lies in the mathematical method used to infer the size of odor space from a limited experimental sample”².
• Richard Gerkin and Jason Castro described “a fragile estimation framework capable of producing nearly any result”. They concluded that “the actual claim supported by the calculation is in fact that humans can discriminate at most one trillion olfactory stimuli”, and that “there is no evidence for the original claim”³.

The fair reading, then, is that the old figure of 10.000 odours was never tested, the trillion figure is a published estimate, and the size of human odour discrimination is still open. We have not found a replacement number that the field accepts.

Receptors, genes and the “weak nose” myth

Two genome studies count the human olfactory receptor gene family slightly differently. Niimura and Nei found “388 potentially functional genes that have intact ORFs and 414 apparent pseudogenes”, and called it “the largest known multigene family in the human genome”¹¹. Malnic, Godfrey and Buck found “339 intact OR genes and 297 OR pseudogenes… on 21 human chromosomes”¹². In plain terms that is roughly 340 to 390 functional receptor genes, commonly rounded to about 400, though 400 is not an exact count.

The idea that humans have a poor sense of smell is also under review. In a 2017 Science article, John McGann traced it to a 19th-century anatomist’s hypothesis. He noted that “the human olfactory bulb is actually quite large in absolute terms and contains a similar number of neurons to that of other mammals”, and that “we are more sensitive than rodents and dogs for some odors, we are capable of tracking odor trails”¹³.

How common is smell loss?

Studies use different tests, age ranges and definitions, so their headline rates differ. Anosmia means a complete loss of smell and hyposmia a reduced one. The sections below keep each study separate rather than blending them into one average.

United States

The NHANES survey in 2012 gave a smell test to adults aged 40 and over. Hoffman and colleagues found olfactory dysfunction in 12,4% of them, equal to 13,3 million adults (55% male, 45% female). This included 3,2% who were anosmic or severely hyposmic, equal to 3,4 million people (74% male, 26% female)⁴.

Smell loss also affects safety. Among adults aged 70 and over in NHANES, 20,3% misidentified the warning odour of smoke and 31,3% misidentified natural gas⁴.

The US National Institute on Deafness and Other Communication Disorders (NIDCD) reports a separate, self-reported figure, not a test result. Its page, last updated in September 2024, says nearly 1 in 4 Americans (23%) over 40 report some alteration in their sense of smell, rising to 32% of those over 80. About 1 in 15 (6,5%) say they sometimes experience phantom odour perception¹⁴.

Europe

In the Skövde population study in Sweden, Brämerson and colleagues tested 1.387 adults. The overall prevalence of olfactory dysfunction was 19,1%, made up of 13,3% with hyposmia and 5,8% with anosmia. The study linked impairment to “aging, male gender, and nasal polyps”⁵.

People rarely notice their own loss

In Beaver Dam, Wisconsin, Murphy and colleagues studied adults aged 53 to 97. The mean prevalence of impaired olfaction was 24,5%, yet “self-reported olfactory impairment was low (9,5%)”¹⁵. Self-reported figures such as NIDCD’s may therefore understate or overstate what a test would find; the two measure different things.

Smell loss by age

Age is the strongest pattern in the data. The table below sets three US studies side by side. They used different tests and age bands, so we show them as separate rows and have not merged them.

Age band	Adults with olfactory dysfunction
40 to 49	4,2%
60 to 69	12,7%
80 and over	39,4%

Doty’s 1984 study of 1.955 people reached the same conclusion with a different test: “More than half of those 65 to 80 years old evidenced major olfactory impairment. After 80 years, more than three-quarters evidenced major impairment”¹⁶. The same study found that “peak performance occurred in the third through fifth decades and declined markedly after the seventh”, which means roughly ages 20 to 49 at the top, with a steep fall after 70¹⁶.

COVID-19 and smell: loss, recovery, parosmia

How often smell was lost

Pooled estimates vary widely between studies. One meta-analysis estimated the random prevalence of olfactory dysfunction at 43,0%, with a three times higher prevalence of chemosensory dysfunction in Caucasians (54,8%) than in Asians (17,7%)¹⁷. A BMJ Medicine review cites a separate meta-analysis of 83 studies and 27.492 patients that put the figure at 47,9%¹⁸.

How fast smell returned

A BMJ analysis by Tan and colleagues pooled 18 studies and 3.699 patients. At 30, 60, 90 and 180 days, respectively, 74,1%, 85,8%, 90,0% and 95,7% of patients recovered their sense of smell⁶. The authors estimate that persistent self-reported smell and taste dysfunction could develop in 5,6% and 4,4% of patients, and say “these could be underestimates”. Women were less likely to recover their sense of smell (odds ratio 0.52)⁶.

For long COVID, a review of sensory changes more than 12 weeks after infection found anosmia in 12,2% of patients and hyposmia in 29,9%. The authors rate the evidence as low quality under the GRADE grading system¹⁹.

Parosmia

Parosmia is a distorted sense of smell. The BMJ Medicine review reports that “more than one in three” people with post-COVID smell loss subsequently report parosmia, and that individual studies put it at 28 to 43%. Before COVID-19, a Swedish population survey found a prevalence of self-identified parosmia of 3,9%¹⁸.

Smell, memory and emotion

Willander and Larsson (2006) asked 93 older adults to recall memories triggered by different cues. They found that “most odor-cued memories were located to the first decade of life (<10 years), whereas memories associated with verbal and visual cues peaked in early adulthood (11-20 years)”. Odour-cued memories were also “associated with stronger feelings of being brought back in time”⁸.

A small brain-imaging study points the same way. Herz and colleagues (2004) scanned only 5 women, so treat it as an early signal and not a settled figure. They saw “significantly greater activation in the amygdala and hippocampal regions during recall to the personally significant odor than any other cue”²⁰.

Smell, attraction and perfume

The best-known attraction experiment is Wedekind’s 1995 T-shirt study. Women “scored male body odours as more pleasant when they differed from the men in their MHC [immune genes] than when they were more similar”, and the difference “was reversed when the women… were taking oral contraceptives”²¹.

A 2009 study by Craig Roberts and colleagues looked at fragrance directly. There was “no difference between groups in mean attractiveness ratings of men’s photographs”, but “the same women judged men using the active spray as more attractive in video-clips”. The active spray also changed how the men saw themselves, raising “psychometric self-confidence and self-perceived attractiveness”. One detail matters: the fragrance spray was combined with antimicrobial agents, so the result cannot be credited to fragrance alone²².

Are women better smellers?

On average, slightly. A 2019 meta-analysis by Sorokowski and colleagues, covering more than 13.000 people for identification, found that “women generally outperform men… in every aspect of olfaction”. The authors add: “However, the effect sizes were weak and ranged between g = 0.08 and g = 0.30”²³.

Smell loss, by contrast, hits men harder. Men had 1.92 times the prevalence of smell impairment in Beaver Dam¹⁵, and they make up 74% of severe cases in NHANES⁴.

Smell as a health warning sign

Several studies link smell loss in older adults to later illness. These are associations, not proof that losing smell causes the outcome.

• Mortality. Pinto and colleagues followed 3.005 adults aged 57 to 85. “Mortality for anosmic older adults was four times that of normosmic individuals”, with “over three times the odds of death (OR, 3.37)”, making smell “one of the strongest predictors of 5-year mortality”⁷.
• Dementia. “Older adults with olfactory dysfunction had more than twice the odds of having developed dementia 5 years later (odds ratio = 2.13)”²⁴.
• Parkinson’s disease. Smell loss “is present in approximately 90% of early-stage PD cases and can precede the onset of motor symptoms by years”²⁵.
• Smell training. A 13-study meta-analysis found “large effects of training on identification, discrimination and TDI-score and small-to-moderate effect in the case of threshold”²⁶.

Fragrance ingredients and allergens in Europe

The IFRA Transparency List for 2025 records “3.691 fragrance and functional ingredients”: 3.312 fragrance ingredients and 379 functional ingredients, of which 1.021 are Natural Complex Substances⁹. IFRA republishes the list about every four years²⁷.

EU rules on labelled allergens are changing. Commission Regulation (EU) 2023/1545 notes that “currently, 24 fragrance allergens… are to be mentioned in the list of ingredients” and identifies 56 additional ones, which makes 80 in total. The labelling threshold is “1 % in leave-on products and 0,01 % in rinse-off products”¹⁰. According to the European Commission, since 31 July 2026 cosmetic products that do not comply may no longer be placed on the EU market; products placed on the market before that date may continue to be made available until 31 July 2028²⁸.

Fragrance allergy itself is not rare. The EDEN study patch-tested 3.119 people in five European countries. It estimated the prevalence of fragrance contact allergy conservatively at 1,9%, with rates in women about twice those in men. The single most common reaction was to fragrance mix I, at 2,6%²⁹.

What this means for perfume lovers

The studies above were not about buying perfume, so the points below are our reasoning from the facts, not findings.

• Memory. Odour-cued memories cluster in the first decade of life⁸. Reasoning: that may be why a scent from childhood can feel so vivid, and why a fragrance tied to a place or person is hard to replace.
• Attraction. The Craig Roberts result came from a spray that also contained antimicrobial agents²², and Wedekind’s study found that women’s preferences depended on how different a man’s immune genes were from their own, and reversed on oral contraceptives²¹. Reasoning: there is no single scent that suits everyone, and a perfume is better chosen for how you like to smell than for a promised effect on others.
• Age. Smell loss becomes much more common after 60⁴, and many people do not notice their own loss¹⁵. Reasoning: someone who feels a perfume has “gone quiet” may be noticing a change in their own nose as much as in the bottle.
• Trying before committing. Differences between the sexes are small on average²³. Reasoning: sensitivity varies from person to person, so the dependable way to learn how a fragrance suits you is to wear it, which is why many people start with a small vial before buying a full bottle. Scento sells 1.000+ fragrances in 2 ml, 5 ml and 8 ml vials, plus a 30 ml bottle decant on most perfumes.

For market-side figures on what people buy and spend, see our perfume industry statistics for 2026 and the Scento research hub.

How we compiled this

Every figure comes from a published paper, an official health agency page, an industry body or EU law, all read on 12 October 2026. Abstracts were read through Europe PMC, and quotes are verbatim from abstracts or official pages. Where two sources round the same figure differently, we use the underlying paper’s number. This page uses no Scento order data.

• Studies use different tests and age bands, so we show them side by side and do not average or merge them.
• Several studies are small or observational, and we describe them as such.
• We left out figures we could not trace to a primary source.

Sources

1. 1Humans can discriminate more than 1 trillion olfactory stimuli, Bushdid C. et al., Science, 2014.
2. 2On the dimensionality of odor space, Meister M., eLife, 2015.
3. 3The number of olfactory stimuli that humans can discriminate is still unknown, Gerkin R.C. and Castro J.B., eLife, 2015.
4. 4New chemosensory component in NHANES: first-year results for measured olfactory dysfunction, Hoffman H.J. et al., Reviews in Endocrine and Metabolic Disorders, 2016.
5. 5Prevalence of olfactory dysfunction: the Skövde population-based study, Brämerson A. et al., Laryngoscope, 2004.
6. 6Prognosis and persistence of smell and taste dysfunction in patients with covid-19, Tan B.K.J. et al., BMJ, 2022.
7. 7Olfactory dysfunction predicts 5-year mortality in older adults, Pinto J.M. et al., PLOS ONE, 2014.
8. 8Smell your way back to childhood, Willander J. and Larsson M., Psychonomic Bulletin & Review, 2006.
9. 9IFRA launches updated Transparency List, IFRA press release, 2025.
10. 10Commission Regulation (EU) 2023/1545, EUR-Lex, 2023.
11. 11Evolution of olfactory receptor genes in the human genome, Niimura Y. and Nei M., PNAS, 2003.
12. 12The human olfactory receptor gene family, Malnic B., Godfrey P.A. and Buck L.B., PNAS, 2004.
13. 13Poor human olfaction is a 19th-century myth, McGann J.P., Science, 2017.
14. 14Quick Statistics About Taste and Smell, NIDCD/NIH, updated 20 September 2024.
15. 15Prevalence of olfactory impairment in older adults, Murphy C. et al., JAMA, 2002.
16. 16Smell identification ability: changes with age, Doty R.L. et al., Science, 1984.
17. 17Prevalence of Chemosensory Dysfunction in COVID-19 Patients: A Systematic Review and Meta-analysis Reveals Significant Ethnic Differences, von Bartheld C.S. et al., ACS Chemical Neuroscience, 2020.
18. 18Post-viral olfactory loss and parosmia, Liu Z.Y. et al., BMJ Medicine, 2023.
19. 19The prevalence of sensory changes in post-COVID syndrome, Trott M. et al., Frontiers in Medicine, 2022.
20. 20Neuroimaging evidence for the emotional potency of odor-evoked memory, Herz R.S. et al., Neuropsychologia, 2004.
21. 21MHC-dependent mate preferences in humans, Wedekind C. et al., Proceedings of the Royal Society B, 1995.
22. 22Manipulation of body odour alters men’s self-confidence and judgements of their visual attractiveness by women, Craig Roberts S. et al., International Journal of Cosmetic Science, 2009.
23. 23Sex Differences in Human Olfaction: A Meta-Analysis, Sorokowski P. et al., Frontiers in Psychology, 2019.
24. 24Olfactory Dysfunction Predicts Subsequent Dementia in Older U.S. Adults, Adams D.R. et al., Journal of the American Geriatrics Society, 2018.
25. 25Olfactory dysfunction in Parkinson disease, Doty R.L., Nature Reviews Neurology, 2012.
26. 26Effects of olfactory training: a meta-analysis, Sorokowska A. et al., Rhinology, 2017.
27. 27IFRA Releases Updated Transparency List, Perfumer & Flavorist, 2025.
28. 28Fragrance allergens labelling, European Commission, undated page.
29. 29Prevalence of fragrance contact allergy in the general population of five European countries, Diepgen T.L. et al., British Journal of Dermatology, 2015.