Does Knowing an Odor’s Identity Improve Discrimination?

An odor can seem familiar while its name remains out of reach. As soon as someone tells us what it is, we may think, yes, that is it. But does knowing the name also make it easier to distinguish fine differences between odors?

In 2023, Sarah Cormiea and Jason Fischer used mixtures of brown sugar and black pepper to compare a group told the ingredients’ identities with a group that knew them only as A and B. They asked whether name-related changes in impressions extend to the precision of discriminating mixture ratios.

Changing impressions versus distinguishing differences

Earlier research has reported that names or descriptions can change ratings of the same odor’s pleasantness or similarity to other odors. In these researchers’ previous experiments, brown sugar and black pepper were rated as less similar when their identities were known than when they were unknown.

If knowing the names makes the two odors perceptually more distinct, small differences in their proportions in a mixture might also become easier to detect. Rather than simply asking how similar they smelled, the researchers tested this idea using a mixture-discrimination task.

The current participants did not also rate name-related changes in similarity. Evidence for that change in impressions came from earlier studies with other participants.

Judging nine mixtures over 108 trials

Ninety-four fluent English speakers aged 18–35 participated. Two were excluded because their responses were too variable to estimate a reference point reliably for analysis, leaving 92 participants, 46 in each group.

The samples were mixtures at nine proportions, ranging from pure brown sugar to pure black pepper. The reference quantities of the two foods were adjusted so their perceived intensities were approximately matched when smelled separately. The mixture proportions were fractions of these reference quantities.

Samples were placed in identical opaque containers with matched weights. Participants smelled one at a time and answered whether it was more like A or B. Separate pure A and B reference samples could be smelled again whenever needed.

One group was told in advance that A was brown sugar and B was black pepper; the other was not told the identities. They repeatedly evaluated the nine samples in varying order, with breaks, making 108 judgments in total.

The perceptual halfway point varies between people

Even when the two materials are present in equal proportions, people do not necessarily perceive them as equally strong. Some may find the black pepper more prominent, while others perceive more brown sugar.

The researchers therefore estimated each person’s mixture ratio at which responses were equally likely to favor A or B. This was that individual’s subjective midpoint. The physical midpoint of the mixture range was not treated as a common correct boundary for everyone.

They then examined how sharply responses switched from A to B when the proportion changed slightly around that midpoint. A steeper switch was taken to indicate more precise discrimination of small differences.

For group comparisons, the curves were also aligned by these midpoints before combining them. Otherwise, differences between people’s perceptual centers could make discrimination appear imprecise even when individual discrimination was sharp.

No advantage was found for knowing the names

Comparing discrimination precision estimated for each individual revealed no difference between the labeled and unlabeled groups. An analysis of group-level curves after midpoint alignment likewise found almost identical precision.

In other words, this task with two food odors did not show that being told their names made fine differences in mixture proportions easier to discriminate.

The researchers also examined changes with practice. Overall, precision improved slightly as trials progressed, but there was no clear advantage for the labeled group in the rate of improvement.

However, finding no difference does not prove that words have exactly zero effect for every odor and every task. This study concerned one pair of food odors, judged over a limited period mainly by young English speakers.

Do discrimination and evaluation use different information?

The authors propose that finely distinguishing odors and evaluating or describing their impressions may draw on different information or processing stages in the brain.

For example, distinguishing subtle differences may benefit from information closer to the odors’ chemical characteristics, whereas judging what an odor resembles may use information enriched by knowledge and memory. In that case, names could change impressions without changing fine discrimination precision.

This explanation is a hypothesis based on the current behavioral experiment and earlier brain research. The researchers did not measure brain activity here to locate the processing stages. They identify studying how neural information changes before and after labeling as a future research direction.

Why use real foods, and what are the limits?

The researchers chose actual foods rather than single fragrance molecules to ensure that odors and names corresponded correctly to everyday objects. This differs from labeling a molecule that merely smells somewhat like black pepper as black pepper.

Food odors, however, consist of multiple components, and nasal irritation could potentially provide a cue. The authors argue that irritation alone is unlikely to explain performance, noting factors including clear judgments at the pure-odor endpoints. They did not separately remove irritation to test that explanation.

Nor can this mixture-ratio task establish how words affect memory, pleasantness, attention, or long-term training. The result does not mean that communicating fragrance in words is pointless. It helps distinguish functions that words change from functions for which change was not detected.

The research received ethical approval from Johns Hopkins University and participants gave consent. The authors declared no competing interests.

Conclusion

Telling participants that the odors were brown sugar and black pepper provided no clear advantage in mixture discrimination precision or improvement with practice in this experiment with 92 people. The results were similar after accounting for individual differences in subjective midpoints.

Together with earlier evidence of changed impressions, the findings suggest that evaluating an odor and distinguishing its subtle differences may not change in the same way. The authors view this distinction as a clue to where words and knowledge enter the processing of smell.

References

Cormiea, S., & Fischer, J. (2023). Odor discrimination is immune to the effects of verbal labels. Scientific Reports, 13, 1742. DOI: 10.1038/s41598-023-28134-w. Full original paper. © The Author(s) 2023. The original paper is available under CC BY 4.0.