When someone smells a perfume they like, can heart rate and the skin’s electrical response alone reveal that preference? In a 2015 study, physiological responses differed across a diverse set of odors spanning a wide range of pleasantness. When only perfumes were compared, however, heart rate and skin conductance did not show clear differences, even when participants’ preferences did.
Small differences remained in facial muscle activity even for perfumes. The study was not about developing a machine that could read preferences. It examined which emotional measures were sensitive to differences across different ranges of odors.
Comparing diverse odors with perfumes alone
The researchers compared data from two separately recruited groups. One smelled a wide variety of odors, including food-like, floral, and animal-like examples. The other smelled nine commercially available perfumes of the time, including Chanel N°5, CK One, and J’adore. For the diverse-odor group, the researchers used previously unreported data from the presentation phase of an earlier experiment.
All participants were students at the University of Geneva. The recruitment description reports 18 participants in the diverse-odor group and 21 in the perfume group. However, these numbers are inconsistent with the later electromyography exclusions and analysis sample sizes. The text also reports 32 diverse stimuli, whereas the table lists 48 entries. Because the exact correspondence between participant and stimulus counts is uncertain in the report, we do not reconstruct it by guesswork here, instead focusing on the two-group design and the reported results.
The stimuli were placed in pen-shaped containers and presented approximately 1 cm below the nose for two seconds. Presentation order was randomized. Participants breathed in time with on-screen instructions to standardize inhalation and rated pleasantness, intensity, and familiarity. Electrocardiography, finger skin conductance, respiration, and facial electromyography were recorded at the same time.
Preferred odors were defined individually
Because preferences vary between people, the researchers did not define a single set of pleasant and unpleasant odors for everyone. Instead, they selected each person’s two most pleasant and two least pleasant stimuli from their own ratings and compared the corresponding responses.
In both groups, participants’ pleasantness ratings clearly separated the highest- and lowest-rated stimuli. However, the pleasantness range was wider in the diverse-odor group, particularly because its lowest-rated odors were more unpleasant than the lowest-rated perfumes. The pleasant stimuli were also more familiar, while the unpleasant stimuli tended to smell stronger.
This comparison was based on what each individual liked. It did not show that a particular perfume was preferred by everyone or that physiological responses could predict the ranking of the nine products.
Heart rate and skin conductance differed across the broad odor range
In the diverse-odor group, the unpleasant stimuli produced larger skin conductance responses than the pleasant ones. Skin conductance measures changes in how easily electricity passes through the skin as sweat-gland activity changes.
When moderately pleasant stimuli were included as well, the relationship was U-shaped: responses were larger at both the very pleasant and very unpleasant ends and smaller in the middle. In other words, skin conductance response size did not increase linearly with liking. The authors discuss the possibility that it captures arousal or stimulation arising from both pleasant and unpleasant experiences.
For heart rate, the researchers examined the deceleration phase five to eight seconds after inhalation. Across the diverse odors, heart rate slowed more for pleasant stimuli and less for unpleasant ones.
In the perfume group, by contrast, neither heart rate nor skin conductance showed significant pleasantness-related differences or clear correlations. Even when participants distinguished their preferences among perfumes, these two measures had difficulty capturing those differences.
Small differences remained in facial muscles even for perfumes
The corrugator muscle, involved in frowning, was more active for unpleasant stimuli in the diverse-odor group. The perfume group also showed greater activity for the unpleasant stimuli, but the response was smaller than in the diverse-odor group.
For the zygomaticus major muscle, involved in raising the corners of the mouth, the analysis combining both groups showed greater activity for pleasant stimuli. Changes over time were larger for the diverse odors. It would therefore be inaccurate to conclude that perfumes produced no differences at all in bodily responses. Facial muscle activity gave different results from heart rate and skin conductance.
Respiratory amplitude did not differ clearly between pleasant and unpleasant stimuli. The researchers therefore considered differences in inhalation alone unlikely to explain the physiological response differences. However, overall thoracic respiratory amplitude did differ between groups, so not all physiological conditions were the same.
Verbal preferences and physiological responses capture different aspects
The authors describe emotion as having multiple aspects, including what a person feels and reports, how the body responds, and what appears in facial expression. Their interpretation is that subtle differences in felt preference do not necessarily produce corresponding differences in heart rate or sweating.
A very unpleasant odor and a pleasant fragrance may involve quite different preparation for actions such as withdrawing or directing attention. Subtle preferences between perfumes, on the other hand, might be distinguished verbally while bodily responses remain similar. This is the authors’ theoretical explanation; the study did not measure approach, avoidance, or purchasing behavior.
The groups also comprised different people and differed in age, sex composition, stimulus range, and physiological recording equipment. This design therefore cannot fully distinguish whether the differences disappeared because the stimuli were perfumes as a product category or because their pleasantness range was narrower. Together with the paper’s inconsistent participant counts, these limitations make further testing desirable for precise estimates of the effects.
Summary
Across diverse odors with large differences in pleasantness, unpleasant stimuli produced larger skin conductance responses, while pleasant stimuli produced greater heart-rate deceleration. Among perfumes, these two measures did not show clear differences despite participants’ different preferences, although relatively small differences remained in facial muscle activity.
The authors conclude that odor-evoked emotion should not be represented by a single measurement: participants’ ratings need to be considered alongside several bodily indicators. When exploring subtle perfume preferences, people’s own words remain important information.
The research received support from the Swiss National Science Foundation’s research center for affective sciences and from Firmenich. Two authors were affiliated with Firmenich at the time, and the paper acknowledges the company’s research division. The paper declares no commercial or financial relationships that could be construed as a conflict of interest.
Reference
Pichon, A. M., Coppin, G., Cayeux, I., Porcherot, C., Sander, D., & Delplanque, S. (2015) Sensitivity of Physiological Emotional Measures to Odors Depends on the Product and the Pleasantness Ranges Used. Frontiers in Psychology, 6, 1821. Original research paper
