Does the Effect of Detergent Fragrance Depend on Where It Is Used?

How closely does liking a detergent fragrance in a laboratory relate to experiencing it while doing laundry and actually moving one’s hands? When fragrance is used in a product, an important question is whether evaluating the scent in isolation also tells us something useful about the situation in which it will be used.

A study published in 2020 by Jasper H. B. de Groot and colleagues created a virtual reality (VR) laundry scene and compared it with smelling the same odors alongside a still image. The researchers wanted to know whether odor pleasantness or associations with laundry motivated washing behavior, and whether the surrounding context altered that effect.

Entering a laundry scene versus looking at a picture

The participants were 90 healthy, nonsmoking female university students. Forty-five used VR goggles and headphones to experience a washing machine spinning in front of them and sounding an end-of-cycle beep. When they opened its door with a controller, a lightly stained T-shirt fell into a laundry basket, and an odor reached their noses at that moment.

The remaining 45 viewed a still image of a stained T-shirt and basket on a computer screen. Odor onset and the 15-second presentation duration were matched, while rich cues such as sound, interaction, and movement were reduced. In both groups, a device controlled odor delivery to the nose.

Each participant experienced three conditions: a laundry odor evoking detergent or soap, vanillin evoking sweet vanilla, and air without added odor. Presentation was arranged to counterbalance order effects. Vanillin was included to test the possibility that any pleasant odor would work similarly.

Measuring movements on real fabric and inhalation through the nose

After viewing the VR scene or still image and smelling the odor, participants removed a stain from linen fixed to a real desk. The pencil-made stain was about 2 cm in diameter. They dipped cloth wrapped around a finger into cleaning liquid and rubbed until they decided they were finished. The washing task did not take place only inside VR.

A small accelerometer held in the hand and video recordings yielded measures of the magnitude of hand movements per unit time, time to finish, and the frequency of repeated movements. The researchers used these as behavioral indicators of motivation to clean. The equipment did not directly measure motivation; the approach inferred it from vigor and speed.

A thin tube placed in the nose also recorded air-pressure changes to examine how the first three inhalations changed relative to before odor presentation. The main cleaning-task analyses used data from 87 people after missing recordings were excluded; sample sizes for breathing and other analyses varied with usable-record criteria. The 90 recruited participants were not necessarily included in every analysis.

Behavior changed when laundry odor was combined with VR

In the VR group, hand movements per unit time were greater and participants finished stain removal sooner after laundry odor than after odorless air. Direct comparisons between laundry odor and vanillin on these same two measures did not show clear differences.

For hand-movement frequency, the comparison differed: laundry odor produced a higher frequency than vanillin, but its difference from odorless air was not clear. The finding therefore does not show that laundry odor outperformed both controls on every measure.

Across the study, a planned comparison of laundry odor with the average of the two control conditions also found differences in all three cleaning measures in VR. The still-image group did not show the same combination of improvements, and magnitude and frequency of hand movements sometimes showed an opposite pattern. Statistical tests likewise indicated that odor effects depended on whether the context was VR or a still image.

Nasal inhalation also reflected the combination with context. Smelling laundry odor in VR increased the measure of inhaled volume compared with air alone, and inhalation-onset speed was higher than with either air or vanillin. Similar differences were absent with still images. The way participants used their noses to take in odors thus changed along with context.

Evaluations of cleaning effectiveness changed too

Participants were told that they would use different cleaning liquids for the three stain-removal tasks. In fact, all were the same mixture of water and one drop of fragrance-free detergent. The odors were not mixed into the liquids, but presented during the preceding VR or still-image scene.

Nevertheless, the VR group was more likely to choose the liquid used after laundry odor as the best stain remover. The paper reported a selection rate of 53.1%, exceeding the 33.3% expected from chance selection among three alternatives. This was not evidence of a superior cleaning-liquid formulation: it showed that evaluations of experiences with the same liquid were influenced by the preceding odor and scene.

Participants themselves decided when to stop cleaning, and the study did not demonstrate improved cleaning performance by measuring the amount of stain remaining. Its distinctive feature was examining both hand movements and the feeling that cleaning had worked well.

Connections with context that pleasantness alone cannot explain

Pilot experiments aimed to select laundry odor and vanillin with similar pleasantness and intensity, but in the main experiment the laundry odor was rated more pleasant and stronger. The principal differences involving odor and context remained after the researchers statistically adjusted for these differences.

Vanillin also did not consistently encourage cleaning more than air. The authors interpreted this as indicating that associations with laundry and cleaning mattered more than pleasantness alone. This explanation combines control conditions and additional analyses; it does not come from an experiment in which pleasantness was perfectly matched.

The authors further argue that combining laundry-associated odor with matching images, sounds, and interactions may make behavior consistent with past laundry experience easier to evoke. Rather than automatically responding identically to the same odor every time, the current situation connects odor with behavior. This mechanism was not directly tested through brain measurements or similar methods, but is an important hypothesis explaining why results differed between VR and still images.

Including real-use contexts in fragrance evaluation

A major implication concerns methods for studying product fragrances. Asking people to smell an odor alone in a quiet booth permits strict control, but may not fully capture how it works in daily life. Home testing, meanwhile, is realistic but makes surrounding conditions difficult to control.

The authors propose combining VR with odor-delivery equipment to create an experience closer to product use while retaining experimental control. Recording movements and sniffing, rather than only asking whether an odor is liked, provides another perspective on users’ responses.

However, this VR experience bundled visual depth, sound, movement, and interaction together, so which elements were necessary is unknown. Participants were limited to female university students; whether the same results occur in men, people with different laundry habits or cultural backgrounds, or actual homes remains to be investigated. The study did not establish that VR testing invariably predicts real-life responses.

The research was supported by Unilever Research & Development. Several coauthors were affiliated with the company and contributed to study design and consideration of the findings.

Summary

Laundry odor altered washing hand movements, sniffing, and evaluations of cleaning liquid when combined with an immersive laundry setting. Presenting the same odor alongside a still image did not produce the same overall pattern of findings.

The authors concluded from this difference that understanding how a product fragrance works requires examining what it evokes and the context in which it is used. The study indicates that controlled VR experiments may help measure experiences during use alongside fragrance pleasantness.

References

de Groot, J. H. B., Beetsma, D. J. V., van Aerts, T. J. A., le Berre, E., Gallagher, D., Shaw, E., Aarts, H., & Smeets, M. A. M. (2020). From sterile labs to rich VR: Immersive multisensory context critical for odors to induce motivated cleaning behavior. Behavior Research Methods, 52, 1657–1670. DOI: 10.3758/s13428-019-01341-y · Full original article.

Original article: © The Author(s) 2020. CC BY 4.0.