Can Repeated Exposure Make You Like a Smell You Dislike?

Repeated encounters can gradually make us like something more. Does this mere exposure effect also work for smells we dislike? In a 2015 study, lilac and shampoo fragrances became more pleasant over six sniffing sessions held on separate days. In contrast, there was no clear increase for two unpleasant odors or for strawberry, which was especially pleasant from the outset.

Smelling each odor the same number of times did not produce a uniform change in liking. This experiment examined those differences in relation to initial likes and dislikes.

Six encounters on separate days, from pleasant scents to unpleasant odors

The participants were 40 people with an average age of about 22, including 10 men. All reported having a normal sense of smell. They were presented with six odors: cheese-like isovaleric acid, a fecal odor, leather, lilac, shampoo, and strawberry. A pen containing no odorant was also included.

The stimuli were presented in pens containing liquid, with participants taking one sniff about 1 centimeter below the nose. At each session, the seven pens were presented in random order, with 30–45 seconds between odors. Participants rated each stimulus for pleasantness, familiarity, and intensity on continuous scales.

They completed these ratings six times, with at least one day between sessions. The median interval was three days, ranging from 1 to 19 days, and data collection for the whole study lasted five weeks. The design aimed to examine changes from encountering an odor again on another day, rather than adaptation from smelling it continuously for a long time.

Greater familiarity did not mean liking everything more

In the initial ratings, cheese and the fecal odor were unpleasant; leather and the blank pen were neutral; and lilac, shampoo, and strawberry were on the pleasant side. Initial perceived intensities were not perfectly matched either, with the blank pen rated as particularly weak.

Comparing the first and sixth sessions, familiarity ratings increased overall. However, pleasantness increased significantly only for lilac, shampoo, and the blank pen. Leather showed only a trend toward an increase, falling short of the conventional 5% significance threshold.

Cheese, the fecal odor, and strawberry showed no clear increase in pleasantness. Becoming more familiar and becoming more liked did not necessarily progress together.

Changes were larger for stimuli that were not initially at either extreme

The researchers compared each stimulus’s initial average pleasantness with its subsequent increase in pleasantness. They found an inverted-U relationship: changes were larger for stimuli initially rated closer to neutral, and smaller for those at the unpleasant or very pleasant end. This relationship remained when the blank pen was excluded.

The authors concluded that the mere exposure effect is more likely to appear for neutral to mildly pleasant odors, and is smaller when strong likes or dislikes already exist. However, this inverted-U pattern was based on the means for six or seven stimuli. It did not establish that every fragrance must follow the same curve for every individual.

It is also interesting that the blank pen showed the largest change. This pen was smelled and rated at every session just like the other stimuli; it was not an unexposed control. Its ratings therefore need to be read as potentially including growing familiarity with the pen or the task itself.

Did intensity or knowing an odor’s name matter?

Intensity ratings were higher in the sixth session than in the first for the blank pen and cheese. However, there was no clear correlation across stimuli between changes in intensity and changes in pleasantness. The authors therefore considered it unlikely that the observed changes in pleasantness simply resulted from odors becoming weaker.

In the final session, participants also completed an odor-naming task. Correct responses averaged 38% in free identification, rising to 88% when participants chose among four options. In free identification, strawberry and shampoo were relatively easy to identify, whereas cheese was difficult. Correct answers included not only exact names but also appropriate categories, such as flowers or cosmetics.

Across stimuli, more pleasant odors were also associated with easier recognition of their names. The authors suggested that stimuli whose meaning is already understood and whose pleasantness is already high may leave less room for new learning through repetition. However, the study did not compare groups that were or were not taught the names, so whether recognition causally determines changes in pleasantness remains unresolved.

Why unpleasant odors resist change remains an open question

The authors discussed whether it might have survival value for wariness of unpleasant odors not to disappear through familiarity alone. They also considered the processing-fluency explanation: repetition may make a stimulus easier to process and therefore more likable. These are hypotheses to explain the findings; the experiment did not measure danger-avoidance behavior or brain processing.

Everyone repeatedly encountered all the stimuli, and there was no unexposed comparison group. Only six odors were studied. It would therefore be premature to conclude that disliked smells can never change over a lifetime. The appropriate conclusion is narrower: under these conditions, smelling the unpleasant odors six times did not produce a clear increase in pleasantness.

Summary

After six encounters on separate days, lilac, shampoo, and the blank pen were rated as more pleasant than at the first session. Leather showed only a trend toward an increase, while cheese, the fecal odor, and strawberry showed no clear change.

Increasing familiarity alone cannot fully explain changes in liking. The study’s important finding is that odors initially rated from neutral to mildly pleasant were more likely to become better liked with repetition.

The research was supported by an affective-sciences research center funded by the Swiss National Science Foundation and by a research grant from Firmenich. Firmenich also supplied the odorants, and one author was affiliated with the company at the time of the research. The paper declares no conflicts of interest.

References

Delplanque S, Coppin G, Bloesch L, Cayeux I, Sander D. The mere exposure effect depends on an odor’s initial pleasantness. Frontiers in Psychology. 2015;6:920. doi:10.3389/fpsyg.2015.00920