One person may describe a fragrance as energizing, while another calls it nostalgic or says it reminds them of prayer. Feelings evoked by odors cannot be reduced to liking and disliking.
A 2011 study developed scales for reporting odor-related feelings in Liverpool, United Kingdom, and Singapore, then compared them with a scale previously developed in Geneva, Switzerland. It identified clusters of feelings shared across the three locations, alongside differences that seemed to reflect local life.
- Which words should measure the feelings associated with odors?
- Starting with 480 terms and narrowing them down in three stages
- Four clusters of feelings shared across the three locations
- Nostalgia, incense, and appetite: differences that reflect everyday life
- Local scales had an advantage, but shared terms gave a different result
- Differences between scales are not national character traits
- Summary
Which words should measure the feelings associated with odors?
The study centered on developing a measurement tool for investigating how odors affect people’s feelings. A few dimensions, such as pleasant versus unpleasant or excited versus calm, might not adequately describe the experience of smelling.
Earlier research had developed GEOS, a self-report scale for odor-related emotions, among French speakers in Geneva. The researchers followed the same procedure in Liverpool and Singapore to develop locally appropriate scales, LEOS and SEOS.
They then examined whether the organization of terms was shared across locations and whether locally developed scales better explained local responses. Participants smelled actual odors and described their own experiences in words.
Starting with 480 terms and narrowing them down in three stages
In the first stage, participants rated whether 480 emotion-related terms appropriately described feelings they had experienced when smelling odors in the past. There were 148 participants in Liverpool and 105 in Singapore. Removing poorly understood, less relevant, and redundant terms reduced the lists to 79 and 81 terms, respectively.
In the next stage, different participants smelled odors and used these terms to rate their feelings. Forty-one participants in Liverpool and 38 in Singapore each smelled 14 odors drawn from a set of 24.
The researchers statistically grouped terms that tended to receive high ratings together. For example, similar patterns for energetic and refreshed could indicate a shared dimension of feeling. These analyses and agreement among responses produced final lists of 37 terms for Liverpool and 36 for Singapore.
Finally, the structure was tested with members of the public recruited at museums and science centers: 351 people in Liverpool and 211 in Singapore. The complete sets contained 56 and 59 odors, respectively, but each person smelled only seven or eight. Familiar local odors of pandan, soy sauce, and coconut were added in Singapore.
The Geneva comparison data came from earlier research involving 220, 37, and 210 participants in its respective stages. These were not longitudinal studies following the same individuals throughout: selection and validation proceeded with different groups.
Four clusters of feelings shared across the three locations
In Liverpool and Singapore, models with seven dimensions fitted responses better than models based only on pleasantness or on pleasantness and arousal. The earlier Geneva scale had six dimensions.
Four dimensions appeared across all three locations: Disgust, Happiness/Well-being, Sensuality/Desire, and Energy. Although their constituent terms were not identical, the broader clusters shared common features.
Specifically, citrus odors such as grapefruit and mandarin, along with peppermint and eucalyptus, tended to receive high Energy ratings. Sweet odors, particularly strawberry, were associated with Happiness/Well-being. Civet and dirty-sock-like odors stood out on the Disgust dimension.
A shared Disgust dimension did not mean identical reactions to every odor. Durian received high Disgust ratings in the two European samples, but the same pattern did not appear in Singapore. Shared frameworks for organizing feelings must be distinguished from reactions to individual odors.
Nostalgia, incense, and appetite: differences that reflect everyday life
Geneva and Liverpool shared a Soothing/Peacefulness dimension. In Liverpool, however, it also included terms such as dreamy and sleepy, so dimensions with similar names could differ in content.
Liverpool additionally showed separate Nostalgia and Hunger/Thirst clusters. Nostalgia was associated with floral odors and confectionery-like odors such as anise and strawberry. The researchers suggested links with childhood memories, but did not establish the content of individual memories.
Singapore showed dimensions of Intellectual Stimulation, Spirituality, and Negative Feelings. Incense and woody odors were strongly associated with Spirituality. Many participants interpreted woody fragrance materials as incense, and some mentioned temples, prayer, or meditation.
Geneva had a Sensory Pleasure cluster incorporating nostalgia, enjoyment, and salivation. Because it partly overlapped with Liverpool’s Nostalgia dimension, it is important to examine the actual terms within a dimension rather than simply its local name.
Local scales had an advantage, but shared terms gave a different result
An important test applied each scale to data from other locations. When the comparison retained a broad range of locally used terms, the structure developed in each location explained that location’s responses better than structures from elsewhere. This supported the value of locally appropriate vocabulary and organization.
However, when the comparison was restricted to the 21 terms shared across all three locations, the Geneva-style model fitted best in every location. This comparison may also have been influenced by adding Geneva terms in the other locations for comparison purposes.
Together, these results suggest that a short shared scale may enable cross-regional comparisons, while local scales can help capture finer, locally specific experiences. They do not establish either that one scale is sufficient everywhere or that cross-regional comparisons are impossible.
Differences between scales are not national character traits
The research sampled one city in each country, included more women than men, and had a Singapore sample that was 79% ethnically Chinese. Geneva used French, whereas the other two locations used English, and terms may differ in their range of meanings. Familiarity with and liking for the odors were also not matched across locations.
Furthermore, the absence of a separate dimension does not mean that people in a location do not experience that feeling. It may be incorporated into another dimension or may not have stood out with the particular odors and vocabulary selected.
Even so, the suggestion that the questions used to evaluate odors influence what researchers can discover has practical value. Providing words suited to someone’s experience, such as nostalgia, energy, or calm, rather than asking only whether they like an odor, can capture its reception in greater detail.
Summary
Odor-related feelings included shared clusters of Disgust, Happiness/Well-being, Sensuality/Desire, and Energy across the three locations. At the same time, differences involving nostalgia and spirituality associated with incense reflected local experience.
The study developed two self-report scales and showed both the advantages of local adaptation and the possibility of comparisons using shared vocabulary. Understanding feelings about odors requires attention not only to the odors themselves but also to the words used to describe them.
The study received funding from Firmenich and the NCCR Affective Sciences, supported by the Swiss National Science Foundation. The coauthors included Firmenich employees.
References
Ferdenzi C, et al. (2011). Affective Dimensions of Odor Perception: A Comparison Between Swiss, British, and Singaporean Populations. Emotion, 11, 1168–1181. Original paper
