Smell and Health

This blog includes information on smell and taste, including the basics but also the latest news, e.g. Smell - what and how, Role of smellFunction of pleasant smell, Theories of smellSmell and MoodSmell and DepressionSmell of disease, Therapeutic use of smellSmell memory and healthAnosmiaSmell trainingTasteHealthy smell (below), and these topics will be updated regularly. I don't deal with "alternative facts" so I try to give a reference to everything I report. There is also a smell NEWS section which reports on the latest stories in smell.

Healthy smell
What is a "healthy" smell? Can smells influence health? Bad smells signify a possible hazard (see Functions of Smell) so some diseases smell bad, but not all, for example the breath of diabetics  smells sweet. What is a “good” smell and why are there good smells? Perfume is just an adornment, good food smells do not please when you are full, and flowers smell for insect pollinators not for us humans, besides many flowers smell bad (attracting flies to pollinate).
The smell after rain - geosmin
To answer the question “why have good smell”? Smell can be therapeutic and curative herbs smell pleasant (see Therapeutic use of smell). Lemon aroma in a doctor's waiting room resulted in fewer reported health symptoms. Lemon is associated with cleanliness and freshness and this aroma altered people's psychological state so they felt better and complained of fewer symptoms. Lemon essential oil has been shown to have anti-depressant properties in both animal and human studies. How does this work? Many studies have been carried out to determine the mechanisms involved (see Therapeutic use of smell).

Night flowering cactus. Flower opens at around 2am, smells beautiful and dies at dawn.
Smell as an indicator of health
Body odour has a number of functions (see "Smell and mate selection"), one of which is to tell us when to wash. We notice a build up of our body odour, some more quickly than others (!) and we find the smell unpleasant - is this another potential hazard? Well, we need to remove dead skin and grime to prevent a build up of fungi and bacteria that could lead to disease and infection. The bacteria are one of the factors in the production of bad smell.Can the smell of your breath reveal information about your health and disease? Exciting new research is beginning to show that it can. Dogs can sniff out cancer and the NHS is thinking of using them to do so - yes really, dogs are cheaper than doctors.
Image from

A fascinating and alarming study by Pinto in Martha McClintock's lab shows that olfactory loss is a predictor of death. In a population of 3000 from 57-85 years old, 39% of those with anosmia (loss of smell) were dead in 5 years compared to 10% of those with a normal sense of smell. This was the greatest predictor of death other than acute liver damage. Why is this? Well, it's not clear but the smell receptor cells are very exposed to the environment and pollution and also to viruses. Viruses can enter the body and get to the brain via the olfactory receptor cells, for example see the viral theory of Parkinson's Disease. Also, these cells are constantly renewing themselves so if they stop doing so this is bad news for the rest of the body. Basically the take home story is that the sense of smell is linked in unknown ways to factor(s) influencing our mortality. We should pay more attention to our sense of smell.

Smell of disease

Plague doctor
There is anecdotal and scientific evidence to suggest that different diseases have particular smells. Most of us recognise that our smell changes when we are ill. In earlier times doctors placed much more importance on a patient's smell - in particular the smell of their stools.
Typhus - sickly odour, another report says "smell of mice"
Smallpox - sweet, pungent rotten stench
Diphtheria - peculiar, penetrating "mousy" smell
Rheumatism - acid smell
Diabetes - sweet, or fermenting fruit (acetone)
Plague - sweet smell (?)
Schizophrenia - some people with the disease have a characteristic smell. The chemical exuded in the sweat of these patients causing the distinct odor is trans-3-methyl-2-hexenoic acid (MHA). It has a goat-like odour. These patients cannot detect the MHA as well as the study group without schizophrenia (full paper here).
Asthma - exhaled nitric oxide levels are much higher in people with asthma. Nitric oxide (NO) is a sharp sweet-smelling gas at room temperature.  After two decades of careful development, a handheld detection device was approved by the FDA some years ago. It is now widely used by doctors to help make a diagnosis.
Parkinson's disease - In a recent news story, Joy Milne is reported as being able to sniff out people with Parkinson's disease. This was followed up by a lab in Edinburgh who tested her ability and found her to be remarkably accurate. Scientists believe that changes in the skin of people with early Parkinson's produces a particular odour linked to the condition. They hope to find the molecular signature responsible for the odour and then develop a simple test such as wiping a person's forehead with a swab. The charity Parkinson's UK is now funding researchers at Manchester, Edinburgh and London to study about 200 people with and without Parkinson's (BBC News Oct, 2015).

Electronic noses are being developed to identify infections and lung diseases (eg TB) and dogs can be trained to detect cancer; they have been approved for use in an NHS trial. Whereas it is more obvious why bacterial infections smell it is less clear why and how other diseases such as cancer smell. What it is precisely that dogs are detecting is unknown at present. We all have a distinct and unique normal body odour and this clearly changes when we are sick.

In a recent study Professor Matts Olsson at the Karolinska Institute gave human volunteers something that activated their immune system and then collected their sweat. Their smell was rated as more unpleasant than controls. It is not known which chemicals are given off to account for this change but we communicate our health state by changing our smell - diesease smells.

There is a smell of disease. Is there therefore a smell of health or is it just the absence of "bad" aroma?

Function of pleasant smell
Smell is difficult to describe in words and on that topic here is an interesting WIRED magazine article. An ethnic tribe in Thailand, the Maniq, have a rich vocabulary for smells. Asifa Majid, a psycholinguist at Radboud University in the Netherlands has been studying the languages of the Jahai and Maniq tribes (read full paper). In the Maniq culture herbs are sought after for their health-giving properties.
A lot of medicinal herbs have a pleasant fragrance and are healing or disease-preventing and are hunted in the forests by their smell. Also, in Maniq culture pleasant smells are associated with cosmetic products, cleanliness and hygeine and they wear such pleasant "cosmetic" smells in necklaces.  Here is a positive function of pleasant smell that confers adaptive advantage. Furthermore it points to the origins of our fascination (and obsession) with perfume. Early cultures collected pleasant smelling herbs, plants because they were medicinal, curative, or represented cleanliness and hygeine.  Pleasant smell equates with health.
A Maniq child in Satun province, Thailand

Smell memory influences health via immune system
The phenomenon in which a certain smell evokes a specific memory is known as the Proust phenomenon. Odour-evoked autobiographic memories are more emotional than those elicited by other sensory stimuli. Rachel Herz has been pre-eminent in this research area and Herz and colleagues characterized odour-evoked autobiographic memories as being more emotional than those elicited by other sensory stimuli. Functional magnetic resonance imaging (fMRI) was used to demonstrate that the amygdala is more strongly activated when autobiographic memories are evoked by olfactory cues than by other sensory cues, such as visual ones. The amygdala is critical for the expression of emotion, and heightened activity of the amygdala represents an intense emotional experience. Odour-evoked autobiographic memory is accompanied by positive emotions has remarkable effects on various psychological and physiological activities, including the secretion of cytokines, which are immune-signalling molecules that modulate inflammation. Participants in a study by Matsunaga and colleagues experienced positive emotions and autobiographic memories when nostalgic odours were presented to them. The levels of peripheral proinflammatory cytokines (e.g. interleukin-2 (IL-2), tumor necrosis factor-α (TNF-α) and interferon-γ (IFN-γ)) were significantly. Brain imaging indicated that the medial orbitofrontal cortex (mOFC) and precuneus/posterior cingulate cortex (PCC) were significantly activated during experiences of odour-evoked autobiographic memory correlating with a reduction in inflammatory mediators.

Inflammation is linked to many diseases including depression so if odour-evoked memories can reduce inflammation this would be beneficial. Treatments for depression targeting the immune system are being explored.


If you have lost your sense of smell then check my Anosmia website and you should definitely go to the excellent website 5th Sense dealing with this problem and the super helpful Smell Training website, of which more below.

Smell Training

There is accumulating evidence that you can enhance your sense of smell by smell training that is, sniffing a number of distinctive aromas regularly (once or twice a day). This is particularly relevant to the recovery of the sense of smell following temporary viral anosmia/hyposmia. Visit the Smell Training website for much helpful and interesting information and follow @smelltraining on twitter.

Olfactory bulb volume correlates with olfactory function (2) and volume loss occurs in some pathologies such as depression, schizophrenia, epilepsy, MS, Alzheimer's disease and posttraumatic or postviral olfactory loss.

Repeated exposure to odours leads to improvement in olfactory function and has been shown to improve recovery of patients suffering from postviral olfactory loss (1, 3) and to prevent olfactory deterioration in older people (5).

A recent study from Hummel's lab put smell training to the test (4). Normal subjects were selected for smell training - 4 months, one nostril (other nostril blocked by finger pressure), 10s, 4 odorants: phenyl ethyl alcohol (rose-like), eucalyptol (eucalyptus), citronellal (lemon) and eugenol (cloves) - based on the 4 primary odours of the Henning prism (Henning, 1916). Both OBs increased in size 11.3% in trained side, 13.1% in untrained side.
Olfactory input therefore increased OB volume. The suggestion is that both bottom-up and top-down mechanisms must be involved since both untrained and trained OBs increased in size.
They found no correlation between OB volume and olfactory function and no effects of gender or age were observed. Furthermore, training did not improve olfactory function, in fact, PEA thresholds worsened. However, it has been previously shown that training improves function in those with olfactory loss and specific anosmias. As they admit the study was hampered by a lack of control group, self-report of compliance and failure to record an increase olfactory function.
  1. Damm, M. et al. (2014) doi:10.1002/lary.24340.
  2. Hummel, T. et al. (2013) doi:10.1016/j.neuroscience.2013.01.044.
  3. Hummel, T. et al. (2009) doi:10.1002 /lary.20101.
  4. Negoias, al. (2016)  doi 10.1007/s11682-016-9567-9
  5. Schriever, V.A. et al. (2014). doi:10.1111/jgs.12669.

Improving your sense of smell

If you want to improve your sense of smell then sniff things a lot. Wear perfume, experiment with them, mix them. Learn a smell vocabulary.  The Maniq tribe (see "Function of pleasant smell" above) have many "smell" words which are abstract and don't just describe the smelly object. Perfumers use terms like "green", "woody", "chypre" (Chypre is French for Cyprus and the concept of the chypre smell was coined by Francois Coty in 1917 for a perfume - fresh citrus accord, woody-oakmoss base and patchouli). Smell words get stored in another brain region. Storing information about a smell in two brain regions, one concerned with smell and the other concerned with language, improves recall and recognition. Get pregnant! Mum's-to-be often have an increased sense of smell but it's very unpredictable. If you are a mouse, have sex, this causes the release of prolactin (it does in humans too) and, in mice, prolactin causes neurogenesis of olfactory cells. Humans? Don't know! But what I do know is that women of child bearing age can increase their sensitivity by regular training. A colleague, Pam Dalton, had a group of such women test their thresholds for odorants every day. They improved their sensitivity by several orders of magnitude. Men couldn't do this. It is thought to be due to the fact that women can recruit more brain processing power to such a smell task than men. Women have the capacity to do this because in evolution they have had to be responsible for early childcare and this means choosing non-toxic food for themselves and their offspring. Other possibilities for enhancing a normal sense of smell were suggested by Wired magazine.

Role of Smell

What is smell and what does the sense of smell do for us?

Hazard warning | Memory prompt | Chemical signalling | Antidepressant | Flavour detector |
Appetite control | Mate selector (HLA detector)
also: Smell and the pill

Hazard warning
Smell is one of the chemical senses, the other being taste (see "Taste" page). They are so called because they sense chemicals, and smells are, of course, chemicals. With these senses we sample our environment for information. We are continuously testing the quality of the air we breathe (this will alert us to potential dangers, e.g. smoke) as well as using this sense to inform us of other relevant information, such as the presence of food or another individual. Many dangerous and toxic compounds have a sulphur atom in them (e.g. bacterial decomposition) and we are very sensitive to sulphur (think of rotten eggs or skunks smell, both contain sulphur). This sends a signal to the amgydala, we interpret it as a "bad" smell and this alerts us to possible danger. Often we don't even have to think about it, our response is sub-conscious. We are then primed to act if necessary. The chemicals detected by our sensory systems need to have certain properties. For instance, odour molecules must be small enough to be volatile (less than 300-400 relative molecular mass) so that they can vapourise, reach the nose and then dissolve in the mucus. This tells us that smell, unlike taste, can signal over long distances (an early warning device). We appear to have an innate ability to detect bad, aversive smells. One-day old babies give facial expressions that indicate rejection when given fish or rotten egg odour.
Body odour has a number of functions (see "Smell and mate selection"), one of which is to tell us when to wash. We notice a build up of our body odour, some more quickly than others (!) and we find the smell unpleasant - is this another potential hazard? Well, we need to remove dead skin and grime to prevent a build up of fungi and bacteria that could lead to disease and infection. The bacteria are one of the factors in the production of bad smell.

Memory prompt

But, is our olfactory system doing more than just giving us warnings? Yes, of course. Smell is the backdrop to our every waking moment, it is part of our consciousness and, even in sleep (see Smell and dreams) the sense of smell keeps working. It also serves a recognition function. We all have our own unique smell (some more pleasant than others! - but that's another story, see Smelly you on the "Smell - What & How" page) and can recognise and be recognised by our smell. Smell evokes memory and, to a greater extent than the other senses, evokes the mood and emotion associated with the memory. The brain regions dealing with smell (piriform cortex) and memory (hippocampus) are close together and there is a two way highway of connections between them (see interesting work by Matsunaga's lab on smell memory, the immune system and health). The amygdala is also part of this collection of brain centres, called the limbic system, and sets the the appropriate emotional tone in response to memories.

Chemical signalling

Dogs can distinguish between the smell of T-shirts worn by non-identical twins (they couldn't tell the difference between identical twins - presumably because they smell identical!). Children can distinguish between the smell of their siblings and other children of the same age. Babies recognise their own mothers' smell and mothers recognise their own babies' smell. Emotion can be communicated by smell. Dogs and horses are very sensitive to the smell of fear in humans and recently it has been demonstrated that humans can detect fear. Dr Mujica Parodi used fMRI to show that "fear sweat" (from skydivers) activates the amygdala in other people whereas gym sweat does not. Research has shown that a panel of women can discriminate between armpit swabs taken from people watching "happy" and "sad" films. Men were less good at this. Mum's learn the smell of their own babies and babies learn the smell of their own mothers. The emotions of others, for example fear, contentment, sexuality, may therefore be experienced and communicated by smell. Memory is often associated with smell. Smell and memory are intimately linked - although this phenomenon is not well understood.

We certainly underestimate the importance of smell to our well-being - ask an anosmic (someone who has lost some or all of their sense of smell). There are suggestions that smell can influence mood, memory, emotions, mate choice,

the immune system and the endocrine system (hormones). In fact the sense of smell could be said to be at the mind-body interface working in the background to modulate our response to the outside world.


A proportion of those who lose their sense of smell become depressed. So you could say that smell acts as an antidepressant. This works both ways. More detail on smell and depression can be found in my page on "Smell and Mood" and I look into new, non-drug treatments for depression in "Therapeutic use of smell".

Smell and flavour

Flavour is SMELL + TASTE, but mainly smell. Without smell food tastes really dull, so does drink. When we are hungry our appetite increases and our sense of smell becomes more acute, we smell things more intensely and this eases off once we've eaten. Those that loose their sense of smell often have difficulty in controlling their weight either eating too much to compensate or not eating at all because they have lost the pleasure in food.

Smell and mate selection

I have always believed that the prime function of smell is mate selection and that this has driven (human) evolution. We like the smell of some people and don't like the smell of others. Our sense of smell is subconsciously identifying the immune system of others. It is our MHC genes (called HLA in humans), part of of our immune system, that confer our own unique body odour and we like the smell of those with very different HLA genes and do NOT like the smell of those with similar HLA genes. The attraction to the smell of those with different HLA alleles encourages outbreeding, increases disease resistance and prevents incest. It is not so much that we actively sniif out our perfect partner more that we are unlikely to end up with someone whose smell we can't stand. The more scientific details about this are given below.

HLA and mate selection - non random pairing

One of the major functions of our sense of smell is as an immune system detector. A specific set of genes determines our immune system (Human Leucocyte Antigen, or HLA, genes). We are all different (except identical twins) and this huge range of differences (called “polymorphism”) results from our innate ability to choose partners with different HLA genes to our own. This phenomenon is called “disassortative mating preference” (Penn and Potts, 1999) where unlike genotypes are attracted to one another.  Evidence for this has been found in mice (Yamazaki et al., 1976; 1978; Beauchamp et al., 1988), humans (Ober et al., 1997; Jacob et al., 2002; Chaix, Cao and Donnelly, 2008), birds (Freeman-Gallant et al., 2003), reptiles (Olsson et al., 2003) and fish (Landry et al., 2001). This mating preference acts to increase disease resistance (Hughes and Nei, 1992) and prevents kin mating (incest avoidance; Weisfeld et al., J. Exp. Child Psychol. 85:279-295, 2003). Mate choice appears to be mediated by smell and this immune-dependent odour is referred to as “odourtype”. There are four hypotheses as to how our immune system can influence odourtype (reviewed by Penn, 2002): (1) HLA molecules may provide the odours themselves, (2) HLA molecules may act as carriers for volatile odorants (the carrier hypothesis), (3) metabolites of HLA-bound peptides may provide the source of volatile odours (the peptide hypothesis), and (4) HLA genes may influence the odour indirectly by shaping an individual’s personal microflora composition (the microflora hypothesis), i.e. the bacteria responsible for your smelly armpits.

This odour-based mate selection encourages disassortative pairing and there are penalties if it is subverted (it is thought that oral contraceptive use undermines this process). Couples sharing HLA alleles have a higher rate of spontaneous abortion (Ober, 1995), and significantly lower success at achieving pregnancy following artificial reproductive therapy (reviewed in Ober and van der Ven, 1997) and some studies have suggested that females reject HLA-similar sperm or zygotes (Rulicke et al., 1998). Consanguineous married couples in China have been found to have low fertility and it has been suggested that they fail to initiate pregnancy due to the increased incidence of HLA-sharing (Liu et al., 2005) a phenomenon that has been long reported in the West (Thomas et al, 1985). Ober et al. (1985) concluded that fertility is reduced in couples sharing HLA-DR alleles. HLA-dissimilar mate preference may influence the psychology of sexual attraction. As HLA sharing increased, women’s sexual responsiveness to their partners decreased, their number of extrapair sexual partners increased, and their attraction to men other than their primary partners increased, particularly during the most fertile period of their menstrual cycles (ovulation) (Garver-Apgar et al., 2006).

From the above it can be suggested that it is important to make the appropriate, genetically-compatible choice when selecting a partner. Clearly this is not an active, conscious process in humans. More likely, it exerts a negative influence, viz. couples who find the smell of their partner continuously unpleasant would be less likely to stay together. To confound this biological selection mechanism, use of the oral contraceptive interferes with womens’ ability to select HLA-compatible partners, reversing their preference, and they have been found to prefer the smell of HLA-similar men (Wedekind et al., 1995; Craig Roberts et al., 2008). Is this the end of human evolution? Actually, not a silly question and no-one has looked into it. They should.

Smell and the pill
A study by Salvatore Caruso found women taking a birth control pill were less sensitive to scents, which may negatively affect their sex drives. Researchers tested the ability of 60 women not taking the pill to detect six smells at various points in their menstrual cycles. Sense of smell was shown to be most sensitive around ovulation, a time when women are most fertile. The subjects were then put on birth control pills and their sense of smell was re-tested. After three months of pill use, the increased sensitivity to smell usually seen during ovulation was absent. This contradicts earlier work carried out by Doty and colleagues who demonstrated fluctuations in smell sensitivity during the menstrual cycle that were not affected by contraceptive pill use.
Changes in olfactory perception were seen in a study by Hummel and colleagues. They tested sensitivity to phenylethyl alcohol, androstenone and nicotine at various phases of the menstrual cycle. They found that it was only the pleasantness of androstenone that changed and sensitivity (to all three compounds) was unaffected.  Naturally cycling women in the periovulatory phase were significantly more sensitive to androstenone, androsterone, and musk than women taking oral contraceptives. These findings give support to odour-specific hormonal modulation of olfaction. Renfro and Hoffman tested the sensitivity of women to 6 odours - lemon, peppermint, rose, musk, androstenone and androsterone. They found that naturally cycling women were significantly more sensitive around ovulation to androstenone, androsterone, and musk than women taking oral contraceptives.
Another study has uncovered other relationships between the sense of smell and pill use. Research by Thornhill and colleagues, showed that women taking oral contraceptives had no significant preference for the smell of males with symmetrical features vs. asymmetrical features. Symmetry in features is a measure of attractiveness and genetic fitness. Women not taking the pill preferred the smell of males with symmetrical features during their most fertile time. Pill use can also decrease sex drive in some women. However, it should be said that sex drive and choosing a mate are very complex issues that involve many physical and psychological variables.

So it seems that oral contraceptive use does affect smell and may alter sensitivity to factors that have bearing on partner choice, sex and relationships.

Smell and Health