Food is laid out in front of you, accompanied by the sounds of dishes and a restaurant. Adding scent to this virtual-reality dining setting seems likely to make it feel more real. But do users notice the scent in the first place?
In 2020, Susan Persky and Alexander P. Dolwick explored this question in a virtual buffet where parents selected a meal for their child. They wanted to know whether people notice scent without being told it is present, and how noticing it relates to the feeling of being there and to food choices.
Adding the smell of fried food to VR meal selection for a child
The participants were 190 parents of children aged 3–7, recruited near Washington, D.C., in the United States. Eligibility criteria included reporting themselves as overweight and having some responsibility for feeding their child. This was not an experiment in which children themselves used VR.
Using a headset and controllers, participants chose a meal for their child from a virtual buffet. French fries were among the options. Meanwhile, in the physical room, two scent diffusers and a fan dispersed a fragrance marketed as the smell of french fries.
The scent was released at the equipment’s highest setting for about 45 seconds at the beginning, while instructions for using the VR environment were given. Research staff reported finding it very strong. However, airborne concentration was not measured to confirm that everyone received the same amount. Participants were not told in advance that scent was included.
The experiment was part of a larger study examining how information about a child’s future obesity risk relates to parents’ food choices. This paper analyzed odor perception in those data. All participants were exposed to scent; this was not a comparison with unscented VR.
Thirty-two of 182 respondents reported smelling food
After the VR session, participants rated whether they had smelled food on a five-point scale. The researchers classified those choosing the affirmative responses, 4 or 5, as having noticed it.
Of the 182 people whose answers were tabulated, 32, or 17.6%, were classified as noticing the smell. The denominator was not all 190 participants. The other 150 did not choose an affirmative answer, but this alone does not establish that their sense of smell was not functioning or that scent had no influence whatsoever.
Of the 32 who noticed the odor, 25, or approximately 78%, correctly identified what it smelled like. A correct answer was not limited to french fries, however. Fifteen gave that answer; seven said fried food, two said chicken, and one said a fast-food restaurant. These were also counted as matching the fragrance’s character.
Even though staff perceived the scent as strong, many VR users did not subsequently report that they had smelled food. This discrepancy was the starting point for the study.
Those who noticed the smell felt more present in VR
The researchers analyzed factors associated with noticing the scent while accounting for age, sex, susceptibility to motion sickness, food choices, and other variables. People who gave higher ratings of presence, the feeling of being in the virtual space, were more likely to report smelling food.
An additional analysis of people who both noticed and correctly identified the smell found an association not only with presence but also with selecting virtual french fries. In contrast, the analysis of noticing the smell alone did not show a statistically clear association with choosing fries.
What remains unknown is which came first. Did scent increase presence, or did people who felt more present pick up a scent that matched the scene? Did the smell make them want fries, or did looking at fries help them notice the smell? These associations cannot distinguish the possibilities. The selected food was virtual, and the amount actually eaten was not measured.
Does attention to visuals and controls push scent into the background?
The authors suggest that participants may have devoted their attention to rich visual input and the actions involved in choosing food, leaving little attention for smell. They also considered that participants might have perceived the odor but treated it as part of the physical room rather than linking it to the VR scene. Both are hypotheses to explain the findings, not explanations tested by varying visual richness or the amount of interaction.
Scent was presented throughout the room, the equipment was unobtrusive, and there was no advance announcement. Telling users that scent is included, or making the scent-producing device visible, might change what they notice. The authors identify advance information about scent as another condition worth investigating.
Other possibilities include becoming accustomed to the odor, effects of the headset on nasal breathing, and differences in participants’ olfactory ability. The study did not include a general smell test. The researchers also considered lingering scent from previous participants: even among the 138 people who were the first participant of the day, the proportion noticing the odor remained low, at 13%.
Summary
Experiencing scented VR does not necessarily mean reporting that you noticed the scent. At this virtual buffet, only around 18% of respondents did so. Reporting the smell was associated with presence, while correctly identifying it as well was also associated with selecting the corresponding food.
The study highlights why a simple design assumption that adding scent will always enrich VR is insufficient. What are users attending to? Can they connect the scent to the scene? Have they been told it is there? The authors suggest that studying these conditions can help identify when scent contributes to the user experience. VR is both a technology for that purpose and a research setting for studying human smell in situations resembling everyday life.
References
Persky, S., & Dolwick, A. P. (2020). Olfactory Perception and Presence in a Virtual Reality Food Environment. Frontiers in Virtual Reality, 1, 571812. DOI: 10.3389/frvir.2020.571812 / Full original article.
Original article license: CC BY 4.0. The original article was written as part of U.S. government work and is not subject to copyright protection in the United States. Copyright may apply outside the United States and in other circumstances.
