What Happens in the Brain When a Scent Feels Strong?

When we smell something and call it strong, we are reporting our own experience. In contrast, the quantity of odor molecules in the air reaching the nose is a concentration that instruments can measure. The two are related, but they do not always correspond one to one.

In 2026, Frans Nordén and colleagues investigated perceived odor intensity and electrical activity in the brain’s olfactory system. They wanted to know whether regions processing nasal information at an early stage correspond to molecular concentration or to a person’s perceived intensity, and how those regions exchange information.

Separating molecular concentration, perceived intensity, and brain activity

The experiment used three odorants, n-butanol, undecanal, and 5-nonanone, each at low and high concentrations. These compounds belong to different chemical classes, an alcohol, an aldehyde, and a ketone, to avoid relying on a single particular odor.

A device delivered each odor to the nose for two seconds, and participants rated how intense it felt on a scale from 0 to 10 after each presentation. Intensity here means the person’s experience on that trial, not a property of the molecule itself or whether the odor is good or bad. Pleasantness and unpleasantness were rated separately and taken into account in the analyses along with breathing.

The other measurement was electrical activity in groups of neurons. Recordings came from 64 scalp electrodes and four electrodes placed just above the eyebrows. A larger amplitude of brain activity and a stronger perceived odor refer to different measurements. The study examined how they correspond.

Following 46 people’s responses within a single breath

Fifty-three people participated. Five whose intensity ratings barely varied and two with too few usable EEG trials were excluded, leaving 46 for analysis. Their mean age was 30, and 27 were women. Before participation, a brief odor-identification test confirmed a minimum level of olfactory function.

Each person received seven conditions: three odors at two concentrations each, plus odorless air. Each condition was presented 20 times in random order. Of the 140 planned trials, an average of 123 per person remained after exclusions for recording artifacts and other reasons. Odors were delivered in time with the natural start of inhalation.

The regions of interest were the olfactory bulb, which receives signals from receptors in the nose, and the piriform cortex, which processes odor information further along. Both are parts of the brain. Rather than inserting electrodes into the brain, the researchers combined head MRI images with electrical recordings from the scalp to estimate activity in each region.

Brain rhythms corresponded closely to perceived intensity

The researchers analyzed electrical activity as rhythms at different speeds. Relatively fast rhythms are called the gamma band, and slower ones the beta band. Neither refers to vibrations of odor molecules; both describe temporal patterns of electrical activity in groups of neurons.

In both the olfactory bulb and piriform cortex, the amplitudes of these activities were significantly associated with participants’ odor-intensity ratings. For example, beta-band associations appeared approximately 0.6–1.6 seconds after odor arrival in the bulb and approximately 0.85–2 seconds afterward in the piriform cortex.

By contrast, associations with low versus high concentration did not meet the statistical criteria in the same main analysis. Because liquid dilution alone does not accurately represent the airborne concentration reaching the nose, the team also analyzed air at the device’s outlet. The main results remained the same when these measured concentrations were used.

This does not mean the brain does not process concentration. Other analyses found early differences in signal patterns according to concentration, and some beta-band activity was also associated with concentration. The finding is narrower: the amplitudes of the rhythms and coordination between regions studied here reflected personal perception more consistently than concentration.

Information did not simply travel onward from the olfactory bulb

When the researchers examined coordination between the two regions, gamma-band activity around 0.2 seconds after odor arrival contained clues distinguishing whether the odor felt weak or strong on that trial. This classification split each person’s ratings at the midpoint of their distribution, rather than separating low and high concentrations.

Around one second after odor onset, beta-band coordination also contained information distinguishing perceived intensity, with an overall classification accuracy of 74% in this time window. This is a classification result within this experiment, using ratings divided into two groups. It does not mean that EEG can now reveal an exact odor-intensity score for any person.

The researchers also examined how well past activity in one region predicted later activity in the other. The results supported a view in which information travels from the olfactory bulb to the piriform cortex in fast gamma-band activity and from the piriform cortex back to the bulb in slower beta-band activity.

Around one second, the relationship between the timing of the slower rhythm and the amplitude of faster activity also varied with perceived intensity. Brief bursts of electrical activity showed corresponding patterns in the same time window. The authors interpret these converging analyses as evidence of back-and-forth exchange between the bulb and piriform cortex.

Is perceived intensity being updated within the brain?

In the authors’ interpretation, the piriform cortex processes the information initially sent from the bulb and returns the result, adjusting the representation of odor intensity. Rather than information simply traveling one way within a breath, perception may take shape through rapid exchanges.

This mechanism might help reduce the difference between expected odor intensity and actual input, maintaining a coherent sensation even when environmental concentrations change. This is an explanation proposed by the authors, not an experiment that manipulated expectations themselves to test it.

The method used to estimate direction likewise examines statistical relationships between recorded signals. The researchers did not stimulate or suppress one part of the brain to test its effects on another. The authors themselves do not claim to have demonstrated strict biological causation.

Because participants rated intensity on every trial, attention and expectations associated with making that judgment may also have contributed to the activity. The authors call for future testing with passive smelling or delayed responses. Whether results from three individual odorants also apply to complex perfumes is another question requiring investigation.

The research received support from the Knut and Alice Wallenberg Foundation, the European Research Council, the Swedish Research Council, and other sources. The authors declared no competing interests.

Summary

By examining both the concentration of molecules reaching the nose and perceived odor intensity, this study found clear relationships between participants’ experience and electrical rhythms in the olfactory bulb and piriform cortex. An important point is that information reflecting perception was already present at an early processing stage such as the bulb.

The findings also suggested that this information may not simply travel onward from the bulb, but may be updated by signals returning from the piriform cortex. The momentary feeling that an odor is strong may depend not only on the entry point that receives molecules, but also on continuing exchanges between brain regions within a single breath.

References

Nordén, F., Zanettin, I., Lundqvist, M., Arshamian, A., & Lundström, J. N. (2026). Olfactory bulb-cortex oscillations encode perceived odor intensity rather than concentration. PLOS Biology, 24(5), e3003810. DOI: 10.1371/journal.pbio.3003810. Full original article. © 2026 Nordén et al. The original article is published under CC BY 4.0.