Smell Training and the Recovery of Lost Smell

Table of Contents

1. What Smell Training Is
2. How Smell Training Is Thought to Work
3. How Smell Loss Is Measured
4. The First Studies
5. What the Pooled Analyses Found
6. Smell Training After COVID-19
7. Other Groups It Has Been Studied In
8. What Changed the Results
9. Parosmia and Smell Training
10. Flavour and Smell From Inside the Mouth
11. What Is Not Known
12. Safety and Side Effects
13. Key Research Papers
14. Connections
15. Featured Videos

What Smell Training Is

The routine comes from a 2009 study led by Thomas Hummel at the smell and taste clinic in Dresden. Patients exposed themselves twice a day, for 12 weeks, to four intense odours. Each odour was a single pure chemical chosen to stand for a familiar smell:

• Phenyl ethyl alcohol — smells of rose.
• Eucalyptol — smells of eucalyptus.
• Citronellal — smells of lemon.
• Eugenol — smells of cloves.

These four became the classic training kit. The same four chemicals were used in later studies from Greece and Germany described below, including the study in people with Parkinson's disease, and the same four odours were used in a study in healthy children.

Researchers stress that the method is not occasional smelling. A 2015 study described it as “the structured sniffing of a defined set of odors, twice daily, for a period of 4 months or longer.” The details vary from study to study. In a 2023 American trial in people with smell loss after COVID-19, participants sniffed four essential oils for 15 seconds each, with a 30-second rest between odours, for 3 months. Some used the classic set (rose, lemon, eucalyptus and clove); others picked 4 scents of their own from a choice of 24, and some were also shown a picture of each scent while they sniffed it.

Attention seems to be part of the method. In a 2023 study in 100 healthy volunteers who trained four times a day, people who trained while watching short videos that matched each odour, or who also had to count the odours one day a week, improved in smell tests after 3 months, especially in odour threshold and odour discrimination. The authors concluded that combining senses and paying attention to the odours appeared to help compared with training alone and with no training. That study was in people with normal smell, so it shows how attention affects training, not how well training repairs a damaged sense.

How Smell Training Is Thought to Work

Why sniffing the same four odours would help a damaged nose is not settled. There are two main ideas, and the research so far supports both without proving either.

Idea 1: help for the regrowing nerve cells in the nose. The smell-sensing nerve cells in the lining high in the nose are renewed over a lifetime (the smell animation shows where they sit). One hypothesis is that regular, strong stimulation encourages new cells to mature and connect properly after damage from a virus or injury. This is a hypothesis; the human studies on this page measured smell and brain scans, not the regrowth of nerve cells in the nose.

Idea 2: changes in the brain. Brain scans give some support to this second idea:

• A 2018 study scanned 30 people whose smell loss followed an upper respiratory tract infection and 31 people with normal smell. Before training, the patients had less grey matter in the limbic system (the brain's emotion and memory circuits) and the thalamus. After 12 weeks of training, grey-matter volume in these two regions had increased. The primary smell areas of the brain showed no change. The authors suggested the gain may reflect paying more attention to odours and linking them with memories.
• A 2014 study using functional brain scans found that training changed how brain networks connect with one another in people with long-term smell loss. The authors described this as evidence that training can bring about reorganisation in the brain.
• A 2025 pooled analysis of COVID-19 smell loss found improvement in every part of smell testing, and the authors read this as a sign that both “top-down” (brain) and “bottom-up” (nose) processes take part in recovery.

A study in healthy people gives one more clue. Among 335 people with normal smell, researchers found 77 who could not smell at least one particular odour (called specific anosmia). After 2 months of training with those odours, they became more sensitive to them, while their sensitivity to odours they could already smell did not change. But the gain faded: 19 months after training ended, 9 of the 10 people followed up could no longer smell androstenone again. The authors concluded that the smell system adjusts to new input, but in healthy people these changes did not appear to be permanent.

How Smell Loss Is Measured

Every result on this page depends on a smell test, so it helps to know what the numbers mean. Two tests dominate the research.

Sniffin' Sticks and the TDI score

The Sniffin' Sticks battery, widely used in Europe, has three parts. Each part produces its own score, and the three are added together into one total, the threshold, discrimination and identification (TDI) score:

• Threshold — the weakest strength of an odour a person can detect.
• Discrimination — telling one odour apart from others.
• Identification — naming an odour from a list of choices.

A 2019 study pooled test results from 9,139 healthy people aged 5 to 96 to set normal values. It placed the line for reduced smell (hyposmia) at a TDI score below 30.75. Scores changed strongly with age: people aged 20 to 30 did best, while children under 10 and adults over 71 scored only half as well. Women scored higher than men on average.

How large a change matters in daily life was worked out in a 2006 study of 83 people with anosmia or hyposmia who were tested twice. More than 60% of them said their smell had improved when the TDI score rose by 5.5 points. Since then, a rise of 5.5 points has been the usual benchmark for a “clinically important” improvement, and several studies below report how many people reached it.

The University of Pennsylvania Smell Identification Test

The University of Pennsylvania Smell Identification Test (UPSIT), widely used in the United States, is a booklet of “scratch and sniff” odours. Its 1984 development paper describes work with more than 1,600 people across five experiments. Retested 6 months apart, scores were highly consistent (a correlation of 0.918), the test could pick out people who were pretending to have no sense of smell, and it could be taken without supervision, even by post. Scores fell with age, with the steepest decline between the sixth and tenth decades of life, and women did better than men at every age. In the 2023 American COVID-19 trial, smell loss was defined as an UPSIT score below 34 for men and below 35 for women.

Some studies also use patients' own ratings, questionnaires about quality of life, or tests of smell from inside the mouth (see Flavour and Smell From Inside the Mouth).

The First Studies

Dresden, 2009. In the first study, 40 patients with smell loss trained and 16 did not. People with disease of the nose and sinuses were excluded. After 12 weeks, the trained group's Sniffin' Sticks scores and their thresholds for the four training odours had improved; the untrained group's smell was unchanged. The study was prospective but not randomised, so the two groups were not assigned by chance.

Thessaloniki, 2013. A Greek study followed 119 patients with smell loss after infection or after head injury. Training groups (49 after infection and 23 after injury) trained twice daily for 16 weeks with the classic four odours and were compared with untrained groups of the same causes (32 and 15). Again, sinus and nasal disease, neurological disease and smell loss of unknown cause were excluded. Both training groups scored significantly higher than their controls. An improvement was measured in 67.8% of the post-infection patients and 33.2% of the post-injury patients, mostly in identification and then discrimination, and patients' own ratings matched the test results.

Germany, 2014: the first large randomised trial. Twelve university centres took part in a randomised, single-blind crossover trial in people whose smell loss after infection had lasted less than 24 months. For 18 weeks, 70 people trained with high concentrations of four odours and 74 with low concentrations; then the groups switched for a further 18 weeks. After the first 18 weeks:

• Smell improved in 18 of 70 people (26%) training with strong odours, against 11 of 74 (15%) training with weak odours.
• In people whose smell loss had lasted less than 12 months, smell improved in 15 of 24 (63%) with strong odours against 6 of 31 (19%) with weak odours.

The authors called training a safe procedure, found that stronger odours helped, and concluded that training appears particularly useful when started within 12 months of the smell loss.

What the Pooled Analyses Found

Pooled analyses (meta-analyses) combine many studies to get a steadier estimate. Several have now looked at smell training.

• 2017, 13 studies. Training had a significant positive effect on every smell ability tested, with large effects on identification, discrimination and the TDI total, and a small-to-moderate effect on threshold. The pattern differed by the cause of the smell loss, and longer training was linked to larger gains in identification and the TDI total.
• 2020, an evidence-based review with recommendations. Thirty-six studies with data on 2,183 people with smell loss after a virus were included. The review rated smell training as “a recommendation” for this group, rated short-term steroids, taken by mouth or applied in the nose, as “an option” in selected patients after careful consideration of the risks of oral steroids, and found too few high-quality studies to make firm recommendations on the other medicines tested.
• 2021, smell loss after a virus. Of 16 studies included, 15 used Sniffin' Sticks, and all 16 reported clinically important improvement (a rise of more than 5.5 TDI points) after training. Four studies with control groups could be pooled: people who trained had 2.77 times the odds (95% confidence interval 1.67 to 4.58) of a clinically important improvement compared with controls. The authors noted that training methods varied from study to study.
• 2024, 36 studies. The largest pooled analysis so far combined 45 before-and-after effect sizes. The overall effect on smell function was moderate to large (a standardised effect size, Hedges' g, of 0.755; 95% confidence interval 0.572 to 0.937), and results varied widely between studies. The length of training, the age of the patients and whether smell was completely lost all changed the size of the effect.

One caution applies to the larger figures. Most of the 2024 effect sizes compare the same people before and after training, a design that cannot separate the effect of training from recovery that would have happened anyway. The comparisons against untrained or differently trained control groups, such as the 2021 pooled odds and the randomised trials in the next section, are the better guide to how much training itself adds.

Smell Training After COVID-19

The COVID-19 pandemic brought a large wave of smell loss and, with it, a new set of randomised training trials and pooled analyses. The main Anosmia page covers how often COVID-19 smell loss recovers on its own.

The 2023 American trial

This randomised, single-blinded trial enrolled 275 adults (mean age 41; 86% women) whose smell loss had begun within 2 weeks of COVID-19 and had lasted 3 months or more. Groups trained with the classic scents or with 4 scents of their own choosing, with or without pictures, and a separate group did not train. The main measure, the change in UPSIT score, did not differ between the training styles. In exploring results, 18 of the people (53%) who trained with pictures and self-chosen scents reached a clinically important UPSIT improvement, against 5 (24%) in the untrained group, a difference of 29% (95% confidence interval 4% to 54%). For patients' own impression of improvement, the figures were 12 (35%) against 4 (19%), a difference that was not statistically clear (95% confidence interval −7% to 39%). The authors described the findings as not definitive.

The pooled analyses

• 2025, long-lasting COVID-19 smell loss. Nine studies were included in which people with smell loss lasting more than 16 weeks trained for more than 8 weeks. Among 179 patients in 7 studies tested with Sniffin' Sticks, the TDI total and each of threshold, discrimination and identification rose significantly after training. Two studies tested 63 patients with the UPSIT, and the pooled identification results also improved.
• 2025, eight randomised trials and add-on treatments. Training improved smell scores compared with controls. Training combined with palmitoylethanolamide and luteolin (PEA-luteolin) gave the largest added gain (mean difference 4.62 points, 95% confidence interval 2.17 to 7.06), followed by ethylenediaminetetraacetic acid (EDTA) (2.33 points, 0.58 to 4.08). Corticosteroids showed a borderline benefit (1.34 points, 0.01 to 2.67), and alpha-lipoic acid showed no significant effect. Combination treatments were linked to higher recovery rates (risk ratio 1.65, 95% confidence interval 1.13 to 2.42). The authors called for larger, standardised trials.
• 2026, ten randomised trials, 628 patients. Compared with control groups, training was linked to better objective smell scores (standardised mean difference 0.30, 95% confidence interval 0.08 to 0.51) and better quality-of-life scores (standardised mean difference −0.40, 95% confidence interval −0.65 to −0.15), with low variation between studies. The authors urged caution because training methods differed between trials. This comparison against controls gives a smaller figure than the before-and-after pooling in the previous section.

Training in the first weeks

Whether starting any treatment early prevents lasting smell loss is a separate question. A 2022 Cochrane living review of people with COVID-19 smell loss of less than 4 weeks stated that it was not known whether early intervention with any treatment, such as medication or olfactory training, promotes recovery. Its five included trials (691 participants) tested steroid sprays, steroid drops, salty (hypertonic) saline and zinc sulphate; none tested training, and the review concluded that there is very limited evidence on the benefits and harms of early treatments.

Other Groups It Has Been Studied In

After a head injury

In the 2013 Greek study, improvement was measured in 33.2% of trained post-injury patients, against 67.8% of trained post-infection patients. A 2021 pooled analysis of 601 participants from 8 earlier studies found that people with smell loss after a head injury had lower odds of a meaningful improvement after training than people whose loss followed an infection (adjusted odds ratio 0.29).

No known cause

In the same 601-participant analysis, smell loss of unknown cause (idiopathic) carried the lowest odds of a meaningful improvement compared with post-infection loss (adjusted odds ratio 0.18).

Parkinson's disease

Smell loss is common in Parkinson's disease, and the study's authors noted that Parkinson's medicines had not been found to restore it. In a 2013 controlled, non-blinded study, 70 people with Parkinson's disease and smell loss were split into a training group (35) and a group that did not train (35). After 12 weeks with the classic four odours, the trained group's Sniffin' Sticks scores and their thresholds for the training odours had risen significantly; the untrained group's smell was unchanged.

Older adults

• In a 2021 randomised study of 68 people aged 50 to 88, assigned to training with single-chemical odours, training with odour mixtures, or no training, the authors reported beneficial effects of the single-chemical training on cognitive assessment, symptoms of cognitive decline and smell sensitivity.
• In a 2025 randomised study of 128 people aged 50 to 85, training with either pleasant or unpleasant odours improved verbal fluency (a measure of thinking speed with words); physical exercise alone also improved it. Depression scores improved with training only in the people who had marked depression symptoms at the start.
• In a 2026 study of 71 people aged 52 to 86, 3 months of training improved total smell function. How physically active the participants were made little or no difference to the smell, thinking or mood results.

Mild cognitive impairment

A 2022 randomised, blinded study gave 38 people with mild cognitive impairment (an early stage that can precede Alzheimer's disease) either smell training or a control training twice a day for 4 months. Training had no significant effect on smell or on thinking skills. Brain scans showed a stronger response to odours in the frontal lobe, but no change in grey-matter volume.

Children

In a 2015 controlled study of healthy children, 12 weeks of training with the four classic odours improved their thresholds for all four odours and their identification scores compared with children who did not train. In a 2025 study of healthy children aged 6 to 9, 12 weeks of training with the four odours, compared with an odourless placebo, produced no change in working memory.

What Changed the Results

Across the studies, several factors were linked to larger or smaller gains.

• How long the smell had been lost. In the 2014 German trial, 63% of people with smell loss of less than 12 months improved with strong odours, against 26% of the whole strong-odour group.
• How strong the odours were. In the same trial, high concentrations did better than low ones.
• How long people trained. The 2017 and 2024 pooled analyses both found training duration changed the size of the effect.
• Changing the odours. A 2015 study of 85 people with smell loss after infection (mean age 45.6) compared three groups over 36 weeks: classic training with the same four odours throughout, “modified” training with three different sets of four odours used one after another (switching after week 12 and week 24), and no training. Both training groups beat the untrained group in discrimination and identification, and switching odours did better than keeping the same four.
• The number of odours. A 2024 pilot study in 40 patients compared 4 odours with 7 (adding stewed apple, balm and mint) for 3 months. Results were similar, and the authors noted the small sample size.
• Cause and starting level. In the 601-participant analysis, smell loss after infection did better than loss after head injury or of unknown cause, and people with a higher starting score had lower odds of a meaningful improvement (adjusted odds ratio 0.93). The 2024 pooled analysis also found that age and complete smell loss changed the effect.
• Pictures and choice. In the 2023 American trial, choosing one's own scents and seeing pictures did not change the main measure, but the group that did both reached a clinically important improvement more often than the untrained group (53% against 24%).
• Add-on treatments. The 2025 pooled analysis of COVID-19 trials found the largest gains when training was combined with certain add-on treatments (see Smell Training After COVID-19). A trial combining training with a steroid nasal rinse is described on the Axel and Buck page.

Parosmia and Smell Training

Parosmia is distorted smell: real odours come across as different, and usually unpleasant, smells. It often appears while smell is coming back after an infection. The main Anosmia page describes its triggers and how people cope with it. Two questions come up with training: whether having parosmia changes the chance of recovery, and whether training helps the distortion itself.

• Parosmia and recovery. A 2021 study looked back at 153 people with smell loss after infection who trained at German clinics between 2008 and 2018. People who had parosmia at the first visit were more likely to reach a meaningful improvement in discrimination and identification. A lower starting score also predicted improvement, and older age predicted improvement in threshold. As a look-back study without a comparison group, it shows an association, not a cause.
• Training for COVID-19 parosmia. A 2022 study looked back at 75 people with parosmia after COVID-19 (mean age 33, range 16 to 60). One group used modified training with 12 odours in three sets over 36 weeks; the other did not train. Both groups improved at 3, 6 and 9 months, the training group more so, and extending training from 6 to 9 months was linked to fewer parosmia complaints and better discrimination scores.
• Parosmia ratings and smell scores. In the 2024 pilot comparing 4 and 7 odours, the clinically important 5.5-point rise in TDI score went together with changes in patients' ratings of their parosmia (correlation 0.62).

Not every study points the same way. The main Anosmia page describes a Danish one-year follow-up of smell loss in the COVID-19 era in which parosmia was linked to lower recovery rates. That study measured self-reported recovery in a general group of people with smell loss, while the 2021 German study measured test scores in people who were training, so the two do not answer exactly the same question.

Flavour and Smell From Inside the Mouth

Much of what people call the taste of food is smell that reaches the nose from the back of the mouth while chewing and swallowing, called retronasal smell (the Zinc: Loss of Taste and Smell page explains how taste and smell are told apart). Most training studies only test smell through the nostrils.

A 2026 study tested both in 53 patients, mostly after COVID-19, who completed at least 3 months of training with four odours twice a day. Retronasal identification scores rose from 9.40 to 10.85, and the usual nostril identification scores rose from 8.75 to 10.53. Changes in the two were moderately linked (correlation 0.38), and people with a meaningful retronasal gain were more likely to recover nostril smell as well. The study had no untrained comparison group, so it cannot show how much of this gain came from training rather than time.

What Is Not Known

• How much training adds to natural recovery. Many people with smell loss after a virus improve without treatment. Before-and-after results include that natural recovery, and the comparisons against control groups give smaller figures.
• The best routine. The odours, their strength, the number of odours, the time per sniff and the length of training all vary between studies. The 2021 and 2026 pooled analyses both pointed to this variation, and the 2025 and 2026 analyses called for large, standardised randomised trials.
• Blinding. It is hard to give someone a convincing “fake” smell training. Several studies, including the Parkinson's study, were not blinded, and the 2014 German trial used weak odours rather than no odours as its comparison.
• How lasting the gains are. In healthy people, training-induced sensitivity to an odour faded in 9 of 10 people followed up after 19 months. How long gains last in people with smell loss after they stop training is less well studied.
• Very early training. Whether training in the first weeks of COVID-19 smell loss prevents lasting loss was untested in the trials reviewed by Cochrane in 2022.
• Smell loss from sinus disease. The first studies excluded people with nasal and sinus disease, so they say little about smell loss caused by inflammation or polyps (see Sinusitis).
• Effects on thinking and mood. Results in older adults were encouraging, but the randomised study in mild cognitive impairment found no effect on smell or thinking skills.
• The mechanism. Whether training works by helping nerve cells in the nose, by changing the brain, or both, is unresolved.

Safety and Side Effects

Smell training involves sniffing odours, not swallowing them or putting them on the skin. The authors of the 2014 multicentre randomised trial described training as a safe procedure, and none of the study abstracts read for this page reported harm from training itself. Adverse effects are reported less systematically in training studies than in drug trials; in the 2022 Cochrane review of early COVID-19 treatments, for example, the included steroid-spray studies either did not report adverse effects or did not plan in advance how to collect them.

The add-on treatments are a different matter. The 2020 evidence-based review set the use of steroids against the potential risks of oral steroids, and each add-on treatment (corticosteroids, ethylenediaminetetraacetic acid, palmitoylethanolamide with luteolin) carries its own safety questions; decisions about medicines and supplements belong with a clinician.

Training does not restore a warning sense while recovery is under way. During months of training, a person with smell loss still may not detect smoke, leaking gas or spoiled food; the main Anosmia page sets out these hazards.

Key Research Papers

1. Hummel T, Rissom K, Reden J, Hähner A, Weidenbecher M, Hüttenbrink KB. Effects of olfactory training in patients with olfactory loss. Laryngoscope. 2009;119(3):496-9. PubMed PMID: 19235739
2. Konstantinidis I, Tsakiropoulou E, Bekiaridou P, Kazantzidou C, Constantinidis J. Use of olfactory training in post-traumatic and postinfectious olfactory dysfunction. Laryngoscope. 2013;123(12):E85-90. PubMed PMID: 24114690
3. Damm M, Pikart LK, Reimann H, Burkert S, Göktas Ö, Haxel B, Frey S, Charalampakis I, Beule A, Renner B, Hummel T, Hüttenbrink KB. Olfactory training is helpful in postinfectious olfactory loss: a randomized, controlled, multicenter study. Laryngoscope. 2014;124(4):826-31. PubMed PMID: 23929687
4. Altundag A, Cayonu M, Kayabasoglu G, Salihoglu M, Tekeli H, Saglam O, Hummel T. Modified olfactory training in patients with postinfectious olfactory loss. Laryngoscope. 2015;125(8):1763-6. PubMed PMID: 26031472
5. Sorokowska A, Drechsler E, Karwowski M, Hummel T. Effects of olfactory training: a meta-analysis. Rhinology. 2017;55(1):17-26. PubMed PMID: 28040824
6. Hura N, Xie DX, Choby GW, Schlosser RJ, Orlov CP, Seal SM, Rowan NR. Treatment of post-viral olfactory dysfunction: an evidence-based review with recommendations. Int Forum Allergy Rhinol. 2020;10(9):1065-1086. PubMed PMID: 32567798
7. Kattar N, Do TM, Unis GD, Migneron MR, Thomas AJ, McCoul ED. Olfactory Training for Postviral Olfactory Dysfunction: Systematic Review and Meta-analysis. Otolaryngol Head Neck Surg. 2021;164(2):244-254. PubMed PMID: 32660334
8. Delgado-Lima AH, Bouhaben J, Delgado-Losada ML. The efficacy of olfactory training in improving olfactory function: a meta-analysis. Eur Arch Otorhinolaryngol. 2024;281(10):5267-5284. PubMed PMID: 38802578
9. Khan AM, Piccirillo J, Kallogjeri D, Piccirillo JF. Efficacy of Combined Visual-Olfactory Training With Patient-Preferred Scents as Treatment for Patients With COVID-19 Resultant Olfactory Loss: A Randomized Clinical Trial. JAMA Otolaryngol Head Neck Surg. 2023;149(2):141-149. PubMed PMID: 36580304
10. Chen CH, Shih CF, Hummel T, Chao YT. Multidimensional benefits of olfactory training for chronic COVID-19-related olfactory dysfunction: a systematic review and meta-analysis. Rhinology. 2025;63(4):431-440. PubMed PMID: 40465208
11. Asseri AA, Aldukain M, Aldukain A, Alzuhairi A. Olfactory Training for Post-COVID-19 Olfactory Dysfunction: A Meta-Analysis of Efficacy and Combination Therapies. J Clin Med. 2025;14(18):6578. PubMed PMID: 41010778
12. Albazee E, Aldawas G, Alzeeby M, Alebrahim AH, Alwael A. Olfactory Training for COVID-19-Related Olfactory Dysfunction: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Turk Arch Otorhinolaryngol. 2026;64(2):94-104. PubMed PMID: 42311058
13. Webster KE, O'Byrne L, MacKeith S, Philpott C, Hopkins C, Burton MJ. Interventions for the prevention of persistent post-COVID-19 olfactory dysfunction. Cochrane Database Syst Rev. 2022;9(9):CD013877. PubMed PMID: 36063364
14. Liu DT, Pellegrino R, Sabha M, Aytug A, Aytug A, Damm M, Poletti SC, Croy I, Hohner A, Oleszkiewicz A, Mandy C, Hummel T. Factors associated with relevant olfactory recovery after olfactory training: a retrospective study including 601 participants. Rhinology. 2021;59(1):91-97. PubMed PMID: 33544097
15. Liu DT, Sabha M, Damm M, Philpott C, Oleszkiewicz A, Hähner A, Hummel T. Parosmia is Associated with Relevant Olfactory Recovery After Olfactory Training. Laryngoscope. 2021;131(3):618-623. PubMed PMID: 33210732
16. Altundag A, Yilmaz E, Kesimli MC. Modified Olfactory Training Is an Effective Treatment Method for COVID-19 Induced Parosmia. Laryngoscope. 2022;132(7):1433-1438. PubMed PMID: 35257391
17. Power Guerra N, Kruschwitz E, Krautwurst D, Hummel T. Four odorants for olfactory training are enough: a pilot study. Eur Arch Otorhinolaryngol. 2024;281(12):6445-6458. PubMed PMID: 39242414
18. Haehner A, Tosch C, Wolz M, Klingelhoefer L, Fauser M, Storch A, Reichmann H, Hummel T. Olfactory training in patients with Parkinson's disease. PLoS One. 2013;8(4):e61680. PubMed PMID: 23613901
19. Oleszkiewicz A, Abriat A, Doelz G, Azema E, Hummel T. Beyond olfaction: Beneficial effects of olfactory training extend to aging-related cognitive decline. Behav Neurosci. 2021;135(6):732-740. PubMed PMID: 34110862
20. Sabiniewicz A, Reichert A, Oleszkiewicz A, Hähner A, Hummel T. Effects of physical exercise and olfactory training with pleasant and unpleasant odors on verbal fluency and depression. Chem Senses. 2025;50:bjaf013. PubMed PMID: 40317228
21. Sabiniewicz A, Hernandez AK, Franke C, Haehner A, Hummel T. Olfactory training improves olfactory function in older adults, while physical activity is not particularly effective. Physiol Behav. 2026;311:115318. PubMed PMID: 41887278
22. Chen B, Espin M, Haussmann R, Matthes C, Donix M, Hummel T, Haehner A. The Effect of Olfactory Training on Olfaction, Cognition, and Brain Function in Patients with Mild Cognitive Impairment. J Alzheimers Dis. 2022;85(2):745-754. PubMed PMID: 34864678
23. Mori E, Petters W, Schriever VA, Valder C, Hummel T. Exposure to odours improves olfactory function in healthy children. Rhinology. 2015;53(3):221-6. PubMed PMID: 26363163
24. Pieniak M, Rokosz M, Nawrocka P, Reichert A, Zyzelewicz B, Mahmut MK, Oleszkiewicz A. Null cross-modal effects of olfactory training on visual, auditory or olfactory working memory in 6- to 9-year-old children. Neuropsychol Rehabil. 2025;35(3):524-545. PubMed PMID: 38762780
25. Gellrich J, Han P, Manesse C, Betz A, Junghanns A, Raue C, Schriever VA, Hummel T. Brain volume changes in hyposmic patients before and after olfactory training. Laryngoscope. 2018;128(7):1531-1536. PubMed PMID: 29238983
26. Kollndorfer K, Kowalczyk K, Hoche E, Mueller CA, Pollak M, Trattnig S, Schöpf V. Recovery of olfactory function induces neuroplasticity effects in patients with smell loss. Neural Plast. 2014;2014:140419. PubMed PMID: 25544900
27. Gillmeister A, Pieniak M, Hummel T. Investigating plasticity of the olfactory system through the lens of specific anosmia. Cortex. 2025;192:53-63. PubMed PMID: 40967089
28. Li Z, Anne A, Hummel T. Olfactory training: effects of multisensory integration, attention towards odors and physical activity. Chem Senses. 2023;48:bjad037. PubMed PMID: 37715960
29. Pieniak M, Abou Nasr M, Bayindir S, Hummel T. Retronasal Olfactory Function After Olfactory Training: A Prospective Single-Arm Study. Clin Otolaryngol. 2026;51(6):938-943. PubMed PMID: 42587340
30. Gudziol V, Lötsch J, Hähner A, Zahnert T, Hummel T. Clinical significance of results from olfactory testing. Laryngoscope. 2006;116(10):1858-63. PubMed PMID: 17003712
31. Oleszkiewicz A, Schriever VA, Croy I, Hähner A, Hummel T. Updated Sniffin' Sticks normative data based on an extended sample of 9139 subjects. Eur Arch Otorhinolaryngol. 2019;276(3):719-728. PubMed PMID: 30554358
32. Doty RL, Shaman P, Dann M. Development of the University of Pennsylvania Smell Identification Test: a standardized microencapsulated test of olfactory function. Physiol Behav. 1984;32(3):489-502. PubMed PMID: 6463130

PubMed Topic Searches

1. PubMed: olfactory training randomised trials
2. PubMed: olfactory training after COVID-19, pooled analyses
3. PubMed: modified olfactory training
4. PubMed: olfactory training and parosmia
5. PubMed: olfactory training in older adults

Connections

• Anosmia (Loss of Smell) — the main page: causes, recovery odds and other treatments
• ENT (Ear, Nose and Throat) — the section this page belongs to
• Axel and Buck: How You Smell — the Nobel-winning science of smell receptors, and what helps when smell is lost
• Smell: How Your Nose Reads a Trillion Odours — an animation of the smell pathway
• Zinc Deficiency: Loss of Taste and Smell — when low zinc dulls taste and smell
• Sinusitis — the sinus disease that was excluded from the first training studies
• Parkinson's Disease — where smell loss is common and training has been tested

Featured Videos