1 of 14

A Pilot Study: The Effects of Odor Congruency on Visual Perception

Xiong, T., Elliott, L., Greenberg, A.

MCW & Marquette University Department of Biomedical Engineering

Sensory Neuroscience, Attention, & Perception (SNAP) Laboratory

2 of 14

Sensory Neuroscience, Attention, & Perception (SNAP) Laboratory

MCW & Marquette University
Neurobiology of attention and perception in the visual, auditory, & olfactory domains
Psychophysical and neuroimaging methods (fMRI, TMS, DSI, & Computational Modeling)

3 of 14

Background

Odor & Perception

Survival Instinct

Congruency

Odor (Olfactory)

Picture (Perception)

COVID-19

Smell Test

Quick and Proper Diagnosis

This Photo by Unknown Author is licensed under CC BY-SA

So, olfaction plays an important role in perception. It is also a survival instinct as it aids in identifying toxins or expired foods. It can affect our behavior, especially when it comes to odor and visual congruency. What we smell can somewhat affect what we see. Research has also shown that mild to severe smell dysfunction and complete smell loss are potential indicators of COVID-19 disease; because of this, olfactory testing could be used to aid in early diagnosis. Although COVID-19 smell dysfunction is an area of ongoing investigation, the possible cognitive effects of this dysfunction are still unknown. Our research aimed to measure the effect of odor context on participants’ ability to discriminate between visual images of edible and nonedible items. We plan to extend this research towards COVID-19 patients in the future.

4 of 14

Method

Healthy Participants

Testable.org

UPSIT

3 Sections in Experiment

8-item Test

Picture Identification Task

Congruency Task

5 of 14

8-Item Test

Familiarity (1-9)
Intensity (1-9)
Edibility (1-9)
Pleasantness (-4-4)
Odor Identification

6 of 14

8-Item Test: Odor Identification Results

Figure 1. 8-Item Test Accuracy of Identifying Odors

Participant	Score	Level
1	5	Microsomia
2	7	Normal
3	5	Microsomia
4	7	Normal
5	7	Normal

Table 1. 8-Item Test Scores

7 of 14

8-Item Test: Ratings

Figure 2. Odor Ratings for Familiarity, Pleasantness, Intensity, and Edibility

Familiarity

Pleasantness

Intensity

Edibility

Figure 3. Intensity Ratings and Odor Identification Accuracy

Here are the data for the ratings of each odor within the 4 categories of familiarity, pleasantness, intensity, and edibility in figure 2. What I want to highlight here is the similarity between intensity ratings and accuracy scores on the previous slide. And I combined them on figure 3 to the right. Each odor is represented by a dot. On the x axis we plotted the average intensity ratings for each odor, and on the y axis it depicts the average accuracy for an odor for our group of participants. Notice that all intensity values are at a 5 or higher as there is a gap on the left side of figure 3. All participants thought that the odors had a strong intensity and were intense enough to smell. Although it was marginally significant, we predict that correlation will increase with more data and participants. There is a modest correlation with accuracy and intensity as R squared is about 0.24. The more intense the smell of an odor, the accuracy of identifying the odor will be higher, too.

8 of 14

Picture Identification Task

Correct: Great job! You correctly identified the picture of menthol.
Incorrect: Sorry, this is a picture of menthol.

9 of 14

Congruency Task

Microencapsulated odorant patch

Held with left hand
“z” scratch using pencil

Response

Standard QWERTY keyboard
“J” key (edible, right index finger)
“K” key (inedible, right middle finger)

Time

10-second break between every trial
1-minute break between each block
Total time approximately 30-minutes

J

K

Moving onto the congruency task, participants are directed to scratch a “z” pattern into a microencapsulated odorant patch using a pencil and then sniff the odor before the image is shown and then press the spacebar as an indicator that they could smell the scent and to move onto see the image while continuing to smell the odor. This is so the experiment does not move on until the participant has been able to smell the odor and can continue by clicking the spacebar and to ensure that they continue to smell the odorant patch while the image is being presented on the screen. Participants are instructed to respond by pressing the “J” key using their right index finger on a standard QWERTY keyboard if the image is edible and the “K” key with their right middle finger if the image is not edible.

The card that has the microencapsulated patches will be held with their left hand while the right hand is used for responding with the keyboard. After eight practice trials, participants complete 128 trials that are divided into four randomized blocks of 32 trials. A ten-second break is provided between each trial to allow for clearance of all odorants from the nasal cavity. And then a longer one-minute break is provided between each block to avoid fatigue. So, the total duration of each session (including the 8-item test, image identification, and congruency task) is approximately 1 hour. Again, the main point of this task is for participants to smell and odor and then view an image and then determine whether the image is edible or not edible.

10 of 14

Congruency Task Conditions

Identical Congruent

8 trials

i.e., chocolate odor and image
i.e., rubber tire odor and image

Categorical Congruent

8 trials

i.e., chocolate odor and strawberry image
i.e., rubber tire odor and motor oil image

Incongruent

16 trials

i.e., chocolate odor and rubber tire image

Total

4 blocks of 32 trials each
128 trials in total
Randomized

Each block of 32 contains three distinct conditions: identical congruent, categorical congruent, and incongruent trials. There are 8 trials of identical congruent and this is where the odor and image are the same. So, you can have chocolate odor and an image of chocolate; both are edible. You can also have rubber tire odor and an image of a rubber tire; this set is not edible but still identical congruent. The second condition is 8 trials of categorical congruent. Both the odor and image are the same edibility, but they are not an exact match. An example of this would be chocolate odor and an image of a strawberry. Although both are not the same, they are still edible. You can also have rubber tire odor and an image of motor oil. Both are not edible and not an exact match, so they are still categorically the same. Lastly, the third condition is 16 incongruent trials. This is where the odor and image are different when it comes to edibility. An example of this is a chocolate odor that is followed by an image of a rubber tire. And again, all the conditions are randomized within the 4 blocks of 32 trials. (Read last box, too)

11 of 14

Congruency Task Results

Figure 4. Inverse Efficiency between Congruent and Incongruent Trials *p = 0.03

*

Figure 5. Accuracy within Identical and Categorical Congruent Trials *p = 0.03

*

12 of 14

Conclusion

Healthy control group

Compare to patients who have COVID-19

COVID-19 Patient Performance
Design of Task

Because participants are healthy, these data could be used as a potential control group to compare to patients who have smell loss due to COVID-19. With the papers that we read during our olfaction group meetings, lots of the research didn’t have a baseline or control to compare to as COVID-19 smell dysfunction is still an area of ongoing research.

We also speculate that COVID-positive patients with smell loss will be unaffected by our odor-picture congruency manipulation; whereas COVID patients without smell loss will yield results similar to normal control subjects. This means that those with regular or normal smell function will be impacted by the congruency task as they can smell. This task is designed to be sensitive to long-term effects of cognitive decline in the olfactory domain which may continue after odor perception dysfunction has subsided.

13 of 14

Acknowledgements

Dr. Adam S. Greenberg (PI)
Lauren Elliott (Lab Manager)
Olfaction Group (Dr. Leslie Cameron, Carthage)
SNAP Lab members (Graduate and Undergraduate Students)
Kelly Agone and Tony Perez (SPUR)

14 of 14

Questions

Xiong, T., Elliott, L., Greenberg, A.

MCW & Marquette University Department of Biomedical Engineering

Sensory Neuroscience, Attention, & Perception (SNAP) Laboratory