Can Odors Serve as Landmarks for Remembering a Route?

Turn this corner, and there is that familiar smell. Just as we learn routes from scenery, could odors mark a turning point? And could those cues remain useful over time?

Mira Schwarz and Kai Hamburger used a virtual maze with pictures or odors as landmarks to examine route decisions and memory for the landmarks themselves. The study was published online in 2023 and appeared in print in 2024. Its central question was how wayfinding memory differs between vision and smell, and how it changes after one month.

Pictures or odors as landmarks at 18 intersections

The 52 participants, aged 19–62, were divided into an odor-landmark group of 25 and a picture-landmark group of 27. All wore a virtual-reality headset and watched a first-person video moving through a maze built in Minecraft. They did not move freely around the maze themselves.

The route had 18 intersections: six right turns, six left turns, and six places to continue straight. Each intersection had a different landmark. In the visual group, a picture appeared for three seconds; in the olfactory group, the experimenter brought an open vial near the nose and presented an odor for five seconds.

The olfactory group also watched the maze video. This was not a study in which people closed their eyes and followed an odor plume or walked toward a scent. The task was to associate the odor presented at each intersection with the direction to take there.

Testing route decisions separately from landmark recognition

After participants first learned the route from the video, they watched the same video again, paused at each intersection. They saw or smelled the landmark and indicated whether the next step was right, left, or straight ahead. Whether their answer was correct or incorrect, the video continued along the correct route.

Next, the 18 landmarks used along the route and 18 unused landmarks were presented in random order, and participants indicated whether each had appeared before. This was a recognition task, judging whether the landmark itself had been seen or smelled previously rather than deciding which way to go.

One month later, both the route-decision and recognition tasks were repeated without showing the initial learning video again. However, landmarks were presented again during route decisions, and the correct route became visible after each answer. The subsequent recognition task therefore did not measure only the memory remaining after a month without any re-exposure.

The odor-landmark group also answered above chance

Route-decision accuracy was approximately 66% immediately after learning and 55% one month later in the picture group. In the odor group, it was about 59% immediately and 53% after one month. At both times, the odor group performed significantly above the approximately 33% expected from guessing among three directions.

The difference between the picture and odor groups was not significant either immediately or after one month. This suggests that odors can serve as cues for route decisions in the experiment. However, the absence of a significant difference does not prove that the two methods are exactly equivalent.

There was also no comparison with a group without landmarks. Participants could have memorized the sequence itself, such as right, left, straight. An additional analysis comparing performance at the beginning and end of the route found no significant difference between groups, but it cannot establish that sequence memorization was not used at all.

Was odor memory more resistant to decline?

In the picture group, route performance declined significantly from the immediate test to the one-month test. In the odor group, the average also fell, but the decline was not significant. The authors interpret this as suggesting that route memory cued by odors may be relatively well preserved.

However, the test directly comparing the decline between the two groups found no significant difference. A significant decline in one group and a nonsignificant decline in the other do not by themselves establish different rates of forgetting. The authors also identify this limitation, and a conclusion that long-term odor memory is superior to picture memory requires further testing.

Possible explanations considered by the authors include the connection between odors and emotions and the possibility that odor memories are less vulnerable to interference from other information. These ideas are consistent with earlier research suggesting distinctive mechanisms for odor memory, but the present study did not test them by measuring brain activity or emotional intensity.

The picture group was better at recognizing the landmarks themselves

Landmark-recognition accuracy was about 97% immediately and 96% at the one-month visit in the picture group, compared with about 76% and 75% in the odor group. Here, the picture group performed significantly better at both times, unlike in the route-decision task.

This accuracy measure included not only correctly identifying landmarks that had been used but also correctly rejecting unused ones. Because participants were asked to respond quickly, differences in the time needed to process pictures and odors may also have affected performance.

Furthermore, landmarks were seen or smelled again during the route task immediately before recognition. High recognition scores after one month therefore cannot be treated as evidence that long-term memory for the landmarks themselves was preserved. The important distinction is that saying a landmark is familiar and saying to turn right at that point were not the same measure.

Thinking beyond vision when considering mental maps

The authors argue that research on spatial memory should include multiple senses, such as smell and sound, rather than focusing primarily on visual landmarks. One important reason is that the advantage for pictures in landmark recognition did not translate into a similar difference in route decisions.

However, this was basic research. Because the vials were brought to the nose by hand, the distance from the nose and odor intensity could not be perfectly standardized. The video also followed the correct route even after an error, so the study did not test whether odors could help someone find their way again after actually getting lost.

Outdoors, wind direction changes, odors mix, and the smell of the same location can vary. Applying these findings to everyday wayfinding or signs would require testing in real environments that include such instability. Memory over longer periods and combinations with other senses also remain questions for future work.

Open-access publication was supported by Projekt DEAL, and the authors declared no conflicts of interest.

Summary

The group using odors as intersection landmarks performed above chance on route decisions both immediately and one month later. There are indications that odors could provide lasting cues, but the direct comparison of changes between groups is insufficient to establish that they are forgotten less readily than pictures.

Nevertheless, the difference between recognizing a landmark and identifying the direction to take from it is important. Understanding how routes are learned requires separating what is remembered as a landmark from the action associated with it. Including odors makes it possible to study human spatial memory as something that cannot be fully captured by vision alone.

References

Schwarz, M., & Hamburger, K. (2024; published online 2023). Memory effects of visual and olfactory landmark information in human wayfinding. Cognitive Processing, 25, 37–51. DOI: 10.1007/s10339-023-01169-7. Full original article. The original article is published under CC BY 4.0.