Can a Name Change How the Same Odor Smells?

The same fragrance might seem a little different when labeled mint rather than eucalyptus. Even when the substance entering the nose does not change, what we think it is can shape the experience.

A study published by Okumura and colleagues in 2024 asked whether this linguistic influence also appears at a relatively early stage of odor processing in the brain. It focused on the piriform cortex, part of the primary olfactory cortex. Do activity patterns there change when a name changes the perceived smell? The researchers also examined relationships with regions involved in meaning and memory.

Giving the same odor two plausible names

Four odorants were used: menthol and 1,8-cineole for the minty category, and citral and limonene for the citrus category.

For example, both menthol and cineole were paired with the names mint and eucalyptus. Likewise, two names were crossed with the two citrus odorants. The names had been selected in advance so that neither would seem markedly incongruent with the odor. This was not a comparison of appealing and unpleasant names.

The design allowed comparisons in which only the name changed while the substance stayed the same, and comparisons in which only the substance changed while the name stayed the same. It therefore separated changes in verbal information from changes in input to the nose.

Measuring fine-grained brain activity patterns in 25 people

Of 37 participants, brain images from 25 were analyzed after exclusions for too few trials with detected odors, head movement during scanning, or inability to complete the scan. They were native Japanese-speaking adults with an average age of about 23.

The researchers used an MRI scanner with a high magnetic field of 7 tesla. It measured changes in blood oxygenation associated with neural activity in small, 1-millimeter-sided units. This does not record individual neurons directly, but allows fine-grained investigation of differences within the small piriform cortex.

Participants smelled odors delivered by a device while viewing names on a screen. The same combinations were repeated, and breathing and head movement were also recorded and incorporated into the analyses. After scanning, 20 of these participants rated how different pairs of presentations smelled.

Different names made the same substance seem different

In subjective ratings, the same substance with different names was judged to smell more different than the same substance with the same name. Pairs involving different substances also seemed more different when their names differed than when the names matched.

Thus, the differences participants experienced reflected not just the substances but also the words. However, these were self-reported judgments of odor difference. They were not a test showing that adding names improved accurate discrimination of substances.

Name-dependent differences also appeared in the piriform cortex

The brain-image analysis examined not only average activity strength but also how activity was distributed across small units. In effect, the researchers used patterns of light and shade in the brain to classify which name accompanied the same substance.

When models were trained on pooled data from multiple people and tested on someone left out of training, piriform regions distinguishing the names were found for menthol and cineole. The result did not extend to all four odorants: classification was not statistically clear for the two citrus odorants.

Furthermore, models trained and tested within each individual’s data found no significant regions for any odorant. The results do not mean that a name can be read from anyone’s brain in a single measurement. The authors suggest that pooling participants may make subtle differences easier to detect, or that some features of the activity are shared across people.

The failure to classify citrus labels does not prove that names have no effect on the brain for those odors. Effect size, measurement variability, and trial numbers all affect detectability.

Separate areas for words and substances?

When substances changed under the same name, a region distinguishing citral from limonene was found when both were labeled lemon. Again, this was only one of the four pairs examined.

Its location differed from the regions that classified names. However, with so few successful comparisons, this does not establish separate word-processing and substance-processing areas within the piriform cortex. Both conditions involved an experience combining words and odors.

Connections with regions involved in meaning and memory

The researchers also asked whether activity in the piriform regions showing label differences varied together with activity elsewhere in the brain. This statistical relationship is called functional connectivity.

They found relationships with the temporal poles and parts of the frontal lobe involved in meaning, and with the hippocampus involved in memory. The authors suggest that semantic and memory information may contribute to odor representations in the piriform cortex, helping produce name-dependent changes in perception.

However, the region that classified substances also showed relationships with regions involved in meaning and memory. Direct comparisons found no significant differences in connectivity strength. It is therefore not established that only name-sensitive regions have especially strong connections with meaning and memory regions.

Nor does covariation reveal which way information flowed or which region caused a change. The study did not manipulate a brain region to test the mechanism, and the proposed form of cooperation needs further investigation.

Conclusion

Even for the same odorant, different accompanying names increased perceived odor differences. For two of the four substances, analyses pooling multiple participants also revealed name-dependent activity patterns in the piriform cortex, part of the primary olfactory cortex.

The authors emphasize that brain representations of odors may change flexibly not only through accumulated long-term experience but also in response to words supplied on a particular trial. Smelling is not determined solely by information from the nose, but takes shape in relation to meaning and memory. Fine-grained activity in the primary olfactory cortex provides a way to investigate that process.

References

Okumura, T., Kida, I., Yokoi, A., Nakai, T., Nishimoto, S., Touhara, K., & Okamoto, M. (2024). Semantic context-dependent neural representations of odors in the human piriform cortex revealed by 7T MRI. Human Brain Mapping, 45(6), e26681. DOI: 10.1002/hbm.26681 / Full original paper. Original paper license: CC BY-NC-ND 4.0.