The science of scent: how smells affect your body and mind

Scents genuinely change how you feel, but mostly through the brain, not through "detoxing" or "boosting immunity." The strongest evidence is for modest, short-term effects on anxiety, stress, sleep quality and nausea. Most other claims rest on lab dishes, animal studies or small, poorly blinded trials.

That gap matters. Aromatherapy is a multi-billion-euro market, and marketing often presents a test-tube result as a health benefit. This guide separates the two.

How to read the evidence grades

• Strong: several randomised controlled trials (RCTs) or a systematic review with consistent results in humans.
• Moderate: some human RCTs, but small, mixed or hard to blind.
• Weak: mostly laboratory or animal studies, or a few very small human studies.
• None: no credible human evidence, or evidence against.

One caveat applies to almost every aromatherapy study: blinding is close to impossible. People know when they are smelling lavender rather than a placebo, and expectation alone can lower stress. That does not make the effect fake, but it means the oil may not be doing all the work.

How a smell reaches your body

A scent acts on you through three routes, and the first one does most of the work.

• The nose-to-brain shortcut. Odour molecules bind receptors in the nose. The signal goes to the olfactory bulb, which connects directly to the piriform cortex and the amygdala without first passing through the thalamus, the relay the other senses use (Kadohisa, 2013). The amygdala handles emotion, and nearby links to the hippocampus tie smells to long-term memory. This is why a smell can change your mood or bring back a memory almost at once. Evidence: strong (basic neuroscience).
• Absorption through the lungs. Some aroma compounds, such as linalool from lavender or 1,8-cineole from eucalyptus and rosemary, pass into the blood in small amounts when inhaled. Whether those amounts are enough to act like a drug in everyday use is uncertain. Evidence: weak to moderate.
• Absorption through the skin. In massage, diluted oils penetrate the skin slowly. In practice the massage itself (touch, warmth, rest) probably explains more of the benefit than the oil. Evidence: weak for the oil's own contribution.

A fourth factor is learning. Your brain links a scent to the context in which you first met it. Lavender relaxes many Europeans partly because they associate it with clean linen and calm. The same smell can be neutral or unpleasant for someone with different memories.

Nervous system and mood: where scent works best

This is the area with the most and the best evidence. The effects are real but modest, and usually short-lived.

Anxiety: moderate when inhaled, strong for oral lavender capsules. A 2019 review counted 65 randomised lavender trials, and 54 reported at least one significant benefit for anxiety. The authors flagged the low average quality of the studies as the main limitation (Donelli et al., 2019). The best data are not for sniffing lavender but for Silexan, a standardised oral lavender-oil capsule licensed as a medicine in Germany. In a 539-patient double-blind trial in generalised anxiety disorder, both the 80 mg and the 160 mg dose beat placebo over 10 weeks, with side-effect rates similar to placebo (Kasper et al., 2014). An earlier trial found it comparable to the benzodiazepine lorazepam over 6 weeks (Woelk & Schläfke, 2010). Two cautions: these trials were sponsored by the manufacturer, and a capsule is a medicine, not aromatherapy.

Sleep: moderate. A 2015 meta-analysis of 12 studies found that aromatherapy improved sleep quality, with inhalation outperforming massage (Hwang & Shin, 2015). A larger review of 31 trials also found a benefit, but rated the risk of bias in most trials as unclear and detected publication bias, meaning negative studies probably went unpublished (Taipei Medical University review). In menopausal women, a 2025 pooled analysis found no significant change in overall sleep scores (Tangkeeratichai et al., 2025).

Mood and alertness: moderate. The most rigorous single study comes from Ohio State University. Fifty-six healthy adults smelled lemon, lavender or water under blinded conditions. Lemon reliably lifted mood, while lavender did no better than water (Kiecolt-Glaser et al., 2008). Peppermint tends to raise alertness, and ylang-ylang tends to increase calmness while lowering alertness (Moss et al., 2008).

Pain: weak to moderate, as an add-on only. A meta-analysis of 12 studies found that aromatherapy reduced reported pain, most clearly after surgery and in childbirth and gynaecological pain. The effect on chronic pain was small and not clearly significant (Lakhan et al., 2016). It may help alongside pain relief prescribed by a doctor, not instead of it.

Heart, blood pressure and stress hormones

Scents can nudge heart rate and blood pressure down briefly, mostly as a side effect of relaxing. They are not a treatment for high blood pressure. Evidence: weak to moderate.

• A meta-analysis of 11 trials found that essential-oil inhalation and massage lowered systolic pressure by about 4.7 mmHg and diastolic pressure by about 2.4 mmHg on average (Lee et al., 2020). That is a small change, and it was measured over short sessions.
• In heart patients, a 2021 review graded the evidence as high for a drop in heart rate but low for blood pressure (Turan Kavradim et al., 2021).
• In the blinded Ohio State study, neither lemon nor lavender changed cortisol, heart rate or blood pressure (Kiecolt-Glaser et al., 2008).

Dose matters, and more is not better. In a study of 100 spa workers in Taiwan, diffused bergamot oil lowered heart rate and blood pressure slightly during the first hour. After about 75 minutes of exposure the trend reversed and both began to climb (Chuang et al., reported by MDedge). The practical lesson: diffuse in short sessions of 30 to 60 minutes, not all day.

Immune system: the most overstated claim

"Boosts immunity" is the most common aromatherapy claim and the least supported one. Evidence that inhaled scents change human immune function: weak.

What the lab shows. Many essential oils (tea tree, thyme, oregano, clove, eucalyptus) kill bacteria and fungi in a petri dish. That is real chemistry, but it happens at concentrations you cannot safely reach by breathing near a diffuser or rubbing diluted oil on your skin. A dish result is not a health benefit.

What the best human study shows. In the blinded Ohio State trial, neither lemon nor lavender reliably changed the inflammatory markers IL-6 and IL-10, and recovery of the skin barrier did not improve. One immune skin test actually responded more strongly after water than after either oil (Kiecolt-Glaser et al., 2008).

The forest-bathing exception: interesting but early. The Japanese researcher Qing Li found that a three-day forest trip raised the activity and number of natural killer (NK) cells, a type of immune cell, in a small group of office workers (Li et al., 2007). He attributes part of the effect to phytoncides, the aromatic compounds trees release, such as pinene. The caveats are large: the group was about a dozen people, there was no blinding, and the trip also included walking, rest, sleep and time away from work. Forest bathing plausibly lowers stress. The scent-specific immune claim is unproven.

Where a real immune link exists. Chronic stress suppresses immunity. If a scent helps you relax and sleep, it may support immune health indirectly. That is a reasonable expectation; fighting colds by diffusing oils is not. SEDINA's guide to massage and the immune system draws the same line for massage.

Memory, focus and the sense of smell itself

Rosemary and memory: weak to moderate. In a study of 144 adults, rosemary aroma improved memory accuracy but not speed, while lavender slowed performance (Moss et al., 2003). A later study of 20 people found that higher blood levels of 1,8-cineole, a rosemary compound absorbed by breathing, tracked with better scores on mental arithmetic (Moss & Oliver, 2012). That suggests a drug-like effect and not just mood. The studies are small and mostly from one lab, so read this as promising, not proven.

Peppermint and alertness: weak to moderate. Peppermint aroma raises self-rated alertness in several small studies. It is a reasonable pick-me-up for a dull afternoon, not a cognitive enhancer.

Smell training: moderate to strong, and the most underrated use of scent. If you lost your sense of smell after a viral infection, including COVID-19, sniffing a set of strong scents (classically rose, lemon, eucalyptus and clove) twice a day for several months helps recovery. A 2020 systematic review of 16 studies found clinically meaningful improvement (Kattar et al., 2020), and a 2024 review covering 1,596 post-COVID patients reached the same conclusion (Treder-Rochna et al., 2024). Ear, nose and throat specialists now recommend it routinely. If your sense of smell has not returned, that specialist is the person to see.

Digestion, nausea and breathing

The strongest physical effects of aromatic plants come when their compounds are swallowed in measured doses, not smelled. That is medicine, not aromatherapy, but it explains where some of the reputation comes from.

Peppermint oil for irritable bowel syndrome: strong, as enteric-coated capsules. A meta-analysis of 12 randomised trials with 835 patients found that peppermint oil capsules improved abdominal pain and overall symptoms compared with placebo (Alammar et al., 2019). Menthol relaxes gut muscle. Sniffing peppermint will not do this.

Nausea after surgery: weak for essential oils. A Cochrane review found that sniffing isopropyl (rubbing) alcohol gave fast, short relief and reduced the need for anti-nausea drugs. For peppermint and other oils, the reviewers found no reliable evidence (Hines et al., Cochrane 2018). A 2024 meta-analysis was more positive about ginger, lavender and peppermint, but called the evidence preliminary (PMC, 2024). The hospital alcohol swab outperformed the boutique oil.

Eucalyptus and breathing: moderate as capsules, weak as vapour. In a double-blind trial of 242 patients with chronic obstructive pulmonary disease (COPD), 200 mg of 1,8-cineole (eucalyptol) three times a day for six winter months reduced flare-ups and breathlessness alongside usual treatment (Worth et al., 2009). The trial was industry-linked and needs replication. Inhaled eucalyptus or menthol makes the nose feel clearer by stimulating cold receptors, but it does not measurably widen the airways.

The scents, one by one

Lavender and peppermint have the best evidence. Most others rest on small studies and tradition. The table covers the oils most used in aromatherapy plus scents people meet every day, sorted from the strongest evidence to the weakest.

The studies linked earlier in this guide back the lavender, peppermint, eucalyptus, lemon and rosemary grades. The remaining grades are our reading of the broader research literature; treat them as estimates.

Scent in massage and spa treatments

In spas and treatments, scent is one ingredient among several, and usually not the main one. The touch, warmth, quiet and time off do most of the work; scent adds pleasure and a learned cue to relax. The unseen details that define a massage covers how oil, scent, textiles, sound and light fit together in a session.

• Aromatherapy massage. The most common spa use, typically lavender, sweet orange, frankincense or chamomile at 1–3% in a carrier oil. In cancer care, aromatherapy massage gives short-term relief from anxiety, but reviews have not shown that it beats plain massage. Evidence: moderate for the massage, weak for the oil's added value. Types of massage explains where it sits among the other options.
• Sauna and steam rooms. Eucalyptus, pine, birch or menthol are poured on the stones or added to the steam. They create a feeling of easier breathing and a strong "sauna" association. The heat itself drives the cardiovascular and relaxation effects. Evidence: weak for the scent's own contribution.
• Waiting rooms and clinics. Diffused orange or lavender in dental waiting rooms lowered anxiety and improved mood in Austrian studies (Lehrner et al., 2000 and 2005). Some hospitals use inhaled lavender or peppermint for patients who are anxious before procedures. Evidence: moderate.
• Bathing and balneotherapy. Oils added to bath water are common in German and Central European spa traditions. Oils float instead of dissolving, so undiluted droplets can irritate skin. Mix them first with a carrier oil, milk or bath salts.
• Forest bathing (shinrin-yoku). Slow walks in woodland, now offered by spa resorts and promoted through designated forest-therapy sites in Japan. Lower stress and blood pressure are reasonably well documented; the immune claims are not. Evidence: moderate for stress, weak for immunity.
• Scented yoga, meditation and sleep rituals. Using the same scent each time trains the brain to link it with calm, a form of conditioning. This works with any scent you like; the brand does not matter.

Before an aromatherapy massage, three questions are worth asking: which oils are used, at what dilution, and whether an unscented oil is available. A professional venue answers all three without fuss. One that promises to detox you or strengthen your defences with a blend is telling you something about its rigour, not about your body.

Risks and safety

"Natural" does not mean harmless. Essential oils are concentrated chemicals, far more concentrated than the plant they come from.

• Swallowing is the main danger. Eucalyptus and camphor oils can cause seizures in small children; a few millilitres can be enough (Toxicology International, 2015). A Toronto poison centre logged 251 ingestions of these products by children aged one to five in about 16 months (Flaman et al., 2001). Store oils locked away, in child-resistant bottles.
• Skin allergy. A meta-analysis cited by the EDEN Fragrance Study puts contact allergy to the standard fragrance mix at about 3.5% of the European general population (van Amerongen et al., 2021). Old, oxidised tea tree and lavender oils are stronger sensitisers than fresh ones. Always dilute (1–3% for adults) and patch-test on the inner arm.
• Sunburn. Cold-pressed bergamot, lime and lemon oils can make skin burn and stain in sunlight for hours after application.
• Children and hormones. Repeated use of lavender and tea tree products was linked to breast growth in three prepubertal boys, which reversed when use stopped (Henley et al., 2007). The link is debated, but caution with daily use on children is sensible. Keep peppermint and eucalyptus away from the faces of young children.
• Pets. Cats lack liver enzymes that break down many oil compounds. Tea tree, eucalyptus, peppermint, citrus and pine oils are among the most commonly reported problems, and veterinary guidance is not to diffuse near cats (Hill's Pet).
• Indoor air and asthma. Diffusers, candles and incense release volatile organic compounds, and some react with ozone to form fine particles and formaldehyde. People with asthma often find strong scents a trigger. Use short sessions and ventilate.
• Pregnancy, epilepsy and medication. Ask a doctor before using rosemary, sage, camphor or high doses of any oil in pregnancy or with epilepsy. Oral products such as Silexan or peppermint capsules are medicines and can interact with other drugs.

The bottom line

Use scent for what it does well: shifting mood, easing everyday tension and helping you wind down. Do not use it for what it cannot do: fighting infections, "detoxing" or replacing medicine.

• Worth trying: lavender before sleep or before a stressful appointment; lemon or orange for a mood lift; peppermint for alertness; smell training after losing your sense of smell.
• Medicine, to discuss with a doctor: peppermint oil capsules for IBS; oral lavender (Silexan) for anxiety; cineole capsules for COPD as an add-on.
• Enjoy, but do not expect a health effect: immune boosting, antiviral diffusing, frankincense "cell renewal" and most blend-specific claims.
• Use safely: dilute on skin, diffuse in short sessions, keep oils away from children and cats, and stop if you get headaches, rashes or breathing trouble.

The pleasure you get from a scent you like is real and worth having. It does not need to be dressed up as medicine.

Frequently asked questions

Does aromatherapy actually work?
For a few things, modestly. The best-supported effects of inhaled scents are short-term: lower anxiety, a better mood and somewhat better sleep quality. Those effects run mostly through the brain and through expectation, and the trials are hard to blind because people can smell what they were given. Claims about detoxing, fighting infections or strengthening immunity are not supported by human studies.

Does lavender help you sleep?
Probably a little. Meta-analyses of small trials find that inhaled aromatherapy, lavender most often, improves self-rated sleep quality, but most of those trials carry a risk of bias and negative results tend to go unpublished. It is a reasonable, low-risk thing to try as part of a wind-down routine. Insomnia that lasts for weeks belongs to a doctor.

Can essential oils boost your immune system?
There is no good human evidence that they can. Many oils kill bacteria in a laboratory dish, at concentrations you cannot safely reach by diffusing or by rubbing diluted oil on skin. In the best blinded human study, neither lemon nor lavender improved immune or inflammatory markers. If a scent helps you relax and sleep, that may support your health indirectly, and that is the honest limit of the claim.

Is an aromatherapy massage better than a plain massage?
More pleasant for many people, but not shown to be more effective. Reviews in cancer care find short-term relief from anxiety after aromatherapy massage without showing that it beats massage with plain oil. The touch, the warmth and the hour of rest do most of the work. Choose the scent because you like it, and tell the therapist beforehand if you have asthma, allergies or sensitive skin.

Are essential oils safe around children and pets?
They need care. Swallowed eucalyptus or camphor oil can cause seizures in small children, so bottles should be stored locked away. Peppermint and eucalyptus should be kept away from the faces of young children. Cats cannot break down many oil compounds, and veterinary guidance is not to diffuse around them. Dilute oils on skin, diffuse in short sessions and ventilate the room.

Sources

• Alammar N. et al. (2019). The impact of peppermint oil on the irritable bowel syndrome: a meta-analysis. BMC Complement Altern Med.
• Donelli D. et al. (2019). Effects of lavender on anxiety: a systematic review and meta-analysis. Phytomedicine.
• van Amerongen C.C.A. et al. (2021). Skin exposure to scented products and fragrance contact allergy in the European general population: the EDEN Fragrance Study. Contact Dermatitis.
• Flaman Z. et al. (2001). Unintentional exposure of young children to camphor and eucalyptus oils. Paediatr Child Health.
• Hines S. et al. (2018). Aromatherapy for treatment of postoperative nausea and vomiting. Cochrane Database Syst Rev.
• Hwang E., Shin S. (2015). The effects of aromatherapy on sleep improvement: a systematic literature review and meta-analysis. J Altern Complement Med.
• Kadohisa M. (2013). Effects of odor on emotion, with implications. Front Syst Neurosci.
• Kasper S. et al. (2014). Lavender oil preparation Silexan is effective in generalized anxiety disorder: a randomized, double-blind comparison to placebo and paroxetine. Int J Neuropsychopharmacol.
• Kattar N. et al. (2020). Olfactory training for postviral olfactory dysfunction: systematic review and meta-analysis. Otolaryngol Head Neck Surg.
• Kiecolt-Glaser J.K. et al. (2008). Olfactory influences on mood and autonomic, endocrine, and immune function. Psychoneuroendocrinology.
• Lakhan S.E. et al. (2016). The effectiveness of aromatherapy in reducing pain: a systematic review and meta-analysis. Pain Res Treat.
• Lee M. et al. (2020). A systemic review and meta-analysis on the antihypertensive effect of aromatherapy essential oils.
• Li Q. et al. (2007). Forest bathing enhances human natural killer activity and expression of anti-cancer proteins. Int J Immunopathol Pharmacol.
• Moss M., Oliver L. (2012). Plasma 1,8-cineole correlates with cognitive performance following exposure to rosemary essential oil aroma. Ther Adv Psychopharmacol.
• Tangkeeratichai C. et al. (2025). Aromatherapy and sleep in menopausal women. Women.
• Treder-Rochna N. et al. (2024). Olfactory training for chronic post-COVID smell disorder. Front Hum Neurosci.
• Turan Kavradim S. et al. (2021). Aromatherapy inhalation in cardiovascular disease. Int J Clin Pract.
• Woelk H., Schläfke S. (2010). A multi-center, double-blind, randomised study of the lavender oil preparation Silexan in comparison to lorazepam for generalized anxiety disorder. Phytomedicine.
• Worth H. et al. (2009). Concomitant therapy with cineole (eucalyptole) reduces exacerbations in COPD. Respir Res.
• Moss M. et al. (2003, 2008), Int J Neurosci; Lehrner J. et al. (2000, 2005), dental waiting-room studies; Henley D.V. et al. (2007), N Engl J Med.

SEDINA is free, takes no commission and sells no placement. Answer three questions and get three verified venues matched to what you need, whether or not you want scent in the room. Every venue has cleared the same gate: at least 30 Google reviews, a rating of 4.2 or better, transparent pricing and no grey-zone signals. SEDINA covers Barcelona (81) · Madrid (96) · Málaga (42) · Valencia (64) · Seville (24).