When you describe a perfume as “refreshing,” what impressions does that include? It might feel cool or light, or bring a blue landscape to mind. Several impressions can overlap within a single word. If music is playing while you smell that perfume, which of those impressions might change?
A study published in 2024 created music to match the freshness a Brazilian perfume brand wanted to convey and tested it with actual perfumes. Some combinations strengthened associations with blue and coldness, while associations with the word “fresh” weakened. The research examines the relationship between music and scent one impression at a time.
Turning the brand’s desired “freshness” into sound
The study began with a specific request from a perfume brand: to communicate through sound a blue-centered world evoking exploration of cold, humid nature. The researchers separated these images into attributes such as coldness, lightness, humidity, and color, and searched for suitable musical features.
In the first experiment, 794 Brazilian adults listened to one of eight sounds and reported impressions of temperature and weight, associated scents, matching colors, and other responses. They did not smell perfume at this stage. The experiment examined which sensations sound alone brought to mind.
Candidates featuring piano or woodwinds, slow tempos, and smoothly connected playing were associated with coldness and calmness relative to contrasting sounds. Blue and green were frequently selected, and fresh scents were also evoked. However, the experiment compared sounds combining several features; it did not isolate the effects of individual instruments or tempo.
Based on those results, two original 30-second compositions were made using the two sounds closest to the brand’s wishes as foundations. The paper calls them C1 and C2. A subsequent experiment with 90 participants examined associations between the compositions and words representing freshness or the brand’s image. C2 showed positive associations with both the average of freshness-related attributes and the average of brand-image attributes.
Measuring perfume–word associations through response speed
The second and third experiments used a task in which participants pressed keys according to instructions on a screen. For example, a sentence such as “this fragrance is cold” would appear briefly, followed by “YES” or “NO.” Participants pressed the indicated key as quickly as possible. For the same attribute, they were instructed to respond with both YES and NO.
This differs from a questionnaire in which participants choose YES or NO as their own answer. The researchers used the time taken to press the instructed key to calculate an index of association between perfume and words. The paper calls this method an IAT. The result concerns differences in responses to words, not a direct measure of how much participants liked the perfume or whether their mood improved.
The third experiment, which used perfume, involved 200 Brazilian men aged 29–49. They were regular perfume users, and 80% used the target brand. The stimuli were two finished perfumes in the same brand’s fougère family. Product names were not disclosed, and fragrance concentration was reported as 9–15%.
One offered a contemporary freshness combining aquatic notes with mint, rosemary, and other elements. The other conveyed an after-bath freshness through citrus, lavender, neroli, and other notes. Here, we call them Fragrance 1 and Fragrance 2.
Participants first smelled one of the two fragrances on a blotter without music and completed the task. They then tried both fragrances in sequence while listening to either C1 or C2. Fragrance order during the music phase was randomized, but everyone completed the no-music phase first. Repetition and order effects therefore remain relevant when interpreting differences between music and no-music conditions.
A stronger blue impression does not mean everything becomes fresher
With C2, Fragrance 1 showed stronger associations with “cold” and “blue” than without music. Fragrance 2 showed stronger associations with “blue” and “light.” The authors emphasize that both fragrances, initially negatively associated with blue, showed positive associations when accompanied by C2.
The same results also contain changes in the opposite direction. With C2, associations with “fresh” weakened for both fragrances. Fragrance 1 retained a positive association, but Fragrance 2 shifted to a negative one. Fragrance 2’s association with “cold” also weakened. Associations with cleanliness and a watery impression, both important to the brand, weakened for both perfumes as well.
In other words, blue, coldness, lightness, and freshness do not all move in the same direction with music. Even when the paper groups them as freshness-related attributes, individual words capture different impressions. Coldness increasing for Fragrance 1 but decreasing for Fragrance 2 also suggests that the perfume’s own characteristics and its combination with sound matter.
Changes also differed between perfumes with C1, the other composition. The association with “light” weakened for Fragrance 1 and strengthened for Fragrance 2. Impressions from hearing a composition alone cannot simply be added to a perfume.
How should we think about music that communicates fragrance?
The authors propose breaking the intended fragrance impression into specific words, finding sounds that match those impressions, and testing their combination with the actual product. It is an approach to expressing a brand’s world through music that combines composers’ intuitions with checks of consumers’ responses.
The authors also note that verbal evaluations of sounds and associations measured through reaction times do not always agree. Across the first two sound-only experiments, coldness and lightness showed similar tendencies, but humidity differed. Although the comparisons involve different sound stimuli and participants, they highlight the importance of asking which impression was measured and how.
The paper also discusses possible uses when customers sample perfumes in shops or encounter products online without being able to smell them directly. However, the experiment did not demonstrate increased purchase intention or sales. It found changes in perfume–word associations within combinations of a particular consumer group, two perfumes, and two compositions.
The study was funded by cosmetics company Grupo Boticário, and three coauthors were employees of the company. It was a collaboration arising from a brand’s practical question. The experimental data are not public, and C1 and C2 are also unavailable publicly for copyright reasons.
Conclusion
Adding music may change some of the impressions a perfume creates. An interesting feature of this study is that it used finished perfumes and examined, one by one, the blue, cold, and light impressions the brand wanted to communicate.
Yet a stronger blue impression did not coincide with stronger freshness. Even the same composition produced different directions of change depending on the perfume. Communicating fragrance through sound may therefore require more than selecting a single “refreshing composition”: it calls for considering exactly which impression of that perfume should be conveyed. The authors’ proposal is a step toward exploring those combinations through both intuition and experiment.
References
Rodríguez, B., Cantu, M. A. F., Reyes, L. H., Pereira, V. J. D. A. R., Santos, L. C. Z., & Reinoso-Carvalho, F. (2024). Sound of freshness: Crafting multisensory experience in perfumery. Food Quality and Preference, 119, 105228.
Original article (DOI) / Author-shared full text (ResearchGate)
