How Scent Works: Smell, the Brain and the Placebo Question

The question. When a scent seems to calm or wake someone up, is the cause the chemistry of the oil, the brain's response to smell, or the person's expectation? The research we read suggests the answer is probably a mix, and that expectation is hard to separate out. This page describes research only. It is not advice to use any oil.

How smell reaches the brain

A review of the olfactory (smell) system describes odor molecules touching the first smell nerve cells in the nose. From there, signals pass through the olfactory bulb and onward to deeper brain areas, including the amygdala and hippocampus, which are also involved in emotion. [1] The same review says odors can change mood, thinking and behavior and describes brain structures shared by smell and emotion. [1]

That is a description of anatomy. It shows why a link between smell and feeling is plausible. It does not show that any particular essential oil treats a condition.

What research shows about aroma and mood

• Reviewed experiments. A systematic review of experiments on smell and mood, body measures and behavior analyzed 18 studies that met strict criteria. It found credible evidence that odors can affect mood, physiology and behavior, and judged a psychological explanation to fit the data better than a pharmacological one. [2]
• Lemon versus lavender in healthy adults. In a randomized trial of 56 healthy people, lemon oil raised positive mood compared with water and lavender, whether or not people had been told what to expect. Odors did not reliably change cortisol, heart rate, blood pressure or pain ratings. [3] This is a small trial in healthy volunteers, so it cannot show what happens in a clinical setting.
• Rosemary and lavender and thinking tasks. In a trial of 144 healthy adults, participants were not told the true aim until testing ended. Lavender was linked to poorer working memory and slower reaction times, and rosemary to better overall memory quality but slower memory speed, compared with no odor. [4]

Evidence label for all three: preliminary human research in healthy volunteers, not evidence of clinical benefit.

The placebo and expectation question

A placebo effect happens when a person improves, or feels they do, because they expect to. Several trials we read tested this for scents.

These studies do not show that essential oils have no effect. They show that expectation can produce changes on its own, and that a scent trial has to account for it.

Why blinding scent trials is hard

In a drug trial, a sugar pill looks like the real pill. A smell is different: people can usually tell a lavender scent from no scent, and often from a placebo scent. The trial authors we read point to this problem:

• One group wrote that failure to use adequate placebos, proper blinding, objective measures, or screening of prior beliefs means many earlier findings could reflect expectancy bias. [6]
• A Cochrane review of aromatherapy in dementia only included trials that compared aromatherapy with placebo aromatherapy or usual care. Many of its trials were small, and it found no convincing evidence of benefit. [8]
• An overview of systematic reviews called most of the reviews it found poor in methodological quality. [9]

Many trials also rely on self-reported anxiety, pain or sleep. If a person knows which scent they are smelling, those ratings can be shaped by expectation.