Does Masking an Unpleasant Odor with Lemon Make It Disappear?

How far does covering an unpleasant odor with a lemon-like fragrance change the impression it creates? A 2019 study compared the smell of citral alone with a mixture containing citral and a small amount of caproic acid.

In a task asking whether two stimuli were the same or different, discrimination of the different stimuli did not significantly exceed chance. Yet on a scale from goat-like to lemon-like, the mixture was rated as less lemon-like, and brain activity also differed. Different questions and measurements appear to reveal different aspects of odor masking.

Examining odor masking and the effects of training

The study had two aims. One was to examine whether perception and brain processing differed between a mixture containing a masked unpleasant component and the masking fragrance alone. The other was to test whether discrimination improved when people were told whether their answers were correct and offered the possibility of a reward.

Thirty people were recruited. After one was excluded because of equipment problems, the analysis included 29 healthy participants aged 20–36. Citral provided the lemon-like fragrance, while caproic acid, described as goat-like, was used to mimic unpleasant sweat odor. In a preliminary evaluation, caproic acid alone was rated as less pleasant than citral.

Smelling two odors in succession and receiving feedback

The main comparison was between odors generated from a solution of 80% citral in propylene glycol and a mixture of 80% citral with 4% caproic acid. These percentages describe the liquids placed in the apparatus, not the airborne concentrations reaching the nose. Citral emitted by the apparatus was measured at approximately 2 ppm in the two main conditions.

Inside an MRI scanner, participants smelled two odors in succession, each lasting two seconds and separated by at least eight seconds. There were 25 pairs of identical stimuli, 25 pairs of the main different stimuli, and five pairs containing a more easily distinguished mixture with less citral.

Participants pressed a button to indicate whether the odors were the same or different and immediately received feedback on whether they were correct. They were told that correct answers added points, incorrect answers subtracted points, and a higher score increased their chance of receiving a monetary reward. Pleasantness, intensity, and odor quality were also rated on scales.

Lemon-like quality differed even without clear differences in pleasantness or intensity

Citral alone and the main mixture containing caproic acid did not differ significantly in pleasantness or intensity. This supports the authors’ interpretation that the unpleasant odor was masked, but does not prove that the conditions were completely identical in every respect.

Indeed, on a scale with 0 representing goat-like and 100 representing lemon-like, citral alone averaged 71.49 and the mixture 65.52. The mixture remained on the lemon-like side of the scale but was rated as less lemon-like than citral alone, a statistically significant difference.

By contrast, in the sequential smelling task, accuracy in correctly identifying the main different-stimulus pairs as different did not significantly exceed the 50% chance level. Accuracy was above chance for identical-stimulus pairs and for the more easily distinguished comparison pairs.

Twenty-one of the 29 participants also showed a tendency to answer “same.” It is therefore inappropriate to interpret accuracy alone as showing that nobody perceived any difference at all. Rating quality on a continuous scale captured a distinction that the two-choice task did not.

Some brain regions were more active for citral alone

In the fMRI analysis, citral alone produced greater activity than the mixture in the left putamen and anterior insula. This difference was found in an analysis corrected for the effects of making comparisons across many locations. No region met that criterion in the reverse comparison, testing for greater activity with the mixture.

The authors suggested that odor quality, fit with the concept of lemon, or rewarding value might be involved. However, this activity difference alone cannot establish a person’s unknown true preference or show that the brain identified the unpleasant component separately. What it demonstrated was a difference in processing the two stimuli within the same task.

The first odor also elicited relatively greater activity in areas including the orbitofrontal cortex and hippocampus, while the second elicited greater activity in areas including the anterior insula. The researchers suggested that this could reflect stages of the task, remembering the first stimulus and comparing the next one with it. Brain responses depended not only on odor type but also on the evaluation procedure.

Feedback did not produce a confirmed improvement within this experiment

Feedback indicating a correct answer produced greater activity than feedback indicating an incorrect answer in regions including the reward-related ventral striatum. However, comparing the first and second halves of the experiment showed no significant improvement in discrimination performance. There was also no increase in odor-related brain activity over time at the specified statistical threshold.

Thus, the hypothesis that reward-associated feedback could improve discrimination of masked odors was not supported by this procedure. The study tested one pair of components at particular concentrations; whether other combinations or longer training would give the same results remains an open question.

Summary

The mixture in which citral masked caproic acid showed no clear difference from citral alone in pleasantness or intensity, and discrimination of different stimuli in the two-choice task did not significantly exceed chance. Nevertheless, differences remained: the mixture was less lemon-like, and activity in the left putamen and anterior insula was greater for citral alone.

No improvement from reward-associated feedback was confirmed. The authors showed that even when masking makes unpleasantness less noticeable, odor quality and processing do not necessarily become identical. This was not a test of removing an odor component, but a study using several methods to examine how mixed odors are perceived.

The study received funding from Symrise’s Scent & Care Division and support from a research facility in Aachen. One coauthor was employed by Symrise at the time of the research. The paper states that the funder was not involved in study design, data collection, analysis or interpretation, writing, or the decision to submit for publication.

Reference

Rodriguez-Raecke R, Loos HM, Sijben R, et al. A Masked Aversive Odor Cannot Be Discriminated From the Masking Odor but Can Be Identified Through Odor Quality Ratings and Neural Activation Patterns. Frontiers in Neuroscience. 2019;13:1219. Original research article