University of São Paulo: Experiment shows how different odors can influence the visual perception of emotions


Smells influence the human ability to visually perceive and correctly judge other people’s emotions – even if the person affected by the odor is unaware of its presence. As a result of the master ‘s research by Matheus Henrique Ferreira , currently a doctoral student at the Institute of Psychology (IP) at USP, an article with detailed measurements of this effect was published in the journal Plos One .

“If I’m subjected to a pleasant smell, my perception of pleasant emotion improves,” says Mirella Gualtieri , PhD in Neuroscience and Behavior and professor of Experimental Psychology at IP. “The same happens with unpleasant odors, which improve the discernment of fear and disgust”, explains the scientist, Ferreira’s advisor.

The research team started from the premise that the smell stimulus has the peculiar characteristic of almost always being connected to a pleasantness judgment. “We can be faced with many visual scenes and we won’t necessarily classify them as something we like or don’t like to see, but often the only thing an individual can describe about a certain smell is whether it’s good or bad”, says Mirella, who develops applied research in sensory psychology.

Using this premise, the group detailed their experimental design with the aim of evaluating how being subjected to an environment where there is a pleasant or unpleasant smell can affect the way a person evaluates the emotions they see in others. Mirella points out that this is not an unprecedented experiment: “Evaluating emotional expressions on people’s faces is a very old thing; What is interesting about us, and few studies have this, is that we do not use expressions of very strong emotions. People in this area usually work with features that are not very common on a daily basis. Joy, sadness, anger are portrayed in an almost stereotyped or caricatured way and this is not how we convey emotions in everyday life”.

In the experiment, images were used that portrayed a gradual of emotions. Cartoon faces, which expressed, for example, extreme joy or sadness (rated as 100% gradient), were blended with a neutral face. Thus, gradations from 10% to 10% of a given emotional content were created (see figure A). The group then tracked how people rated or judged that emotion. Volunteers looked at the face in the drawing and said whether it expressed joy, sadness, anger, disgust or fear.

“We observed how much intensity of this expression would be the minimum for the person to begin to correct the emotion that was present there – we know that it does not need 100%, but we wanted to know what the minimum value was – and we saw that it was usually around 20% to 30% of the total content of that emotion”, says Mirella.

Having determined the intensity threshold of an emotion that people needed to discriminate against it, the speed (reaction time) with which they made this judgment was evaluated. Finally, it was observed how all this could be modified with the presence of bad or good smells.

“Our contribution is to show how this effect between sensory modalities happens. We have our five senses, but for us to adapt to the environment, communicate and be able to live, these senses must be interacting. What we show in this article is an example of how this can happen”, says Mirella. “The presence of a smell — and I don’t even necessarily need to be aware that it’s there — is going to affect my visual processing and the way I attribute emotions to the visual stimulus.”

Another differential of the experiment is that the qualification of a smell as good or bad was determined by each participant, instead of adopting predefined conventions. “Many works use a categorical classification, that is, people will necessarily think that the smell of strawberry is good and the smell of foot odor is bad. There are these ready-made labels, but we know from experience that, especially with odor, this is very complicated, it doesn’t always work. What was guiding our work is what the person judged, whether it was pleasant or unpleasant for them, and that changed our analysis process a lot compared to when we used a label, assuming that the smell was always bad. This choice changed our results in a very important way,

The participants – 35 people, 20 women and 15 men – were unaware that the experiment was about smell. They were only informed that their speed would be measured in detecting which emotions certain facial expressions indicated. “They didn’t know it had a smell. They would sit in front of the screen and, in the foam of the headset they used, we would put a very minimal amount of some substance [butyric acid, with an odor of rancid butter; isoamyl acetate, strong banana-like odor; or lemongrass] . The participant did the entire experimental session, identified the emotions, we saw the hit rate and reaction times.”

Only when that part was complete did the team explain that the goal was to see if having smells reaching the nose simultaneously with the emotion judgment affected discernment. Then the participants pointed on a scale, by means of a dial , to show how much they liked or disliked the smell.

“Although previous studies have highlighted the role of the hedonic valence of smells in the emotional processing of visual stimuli, there are several other factors that are possibly involved. This study demonstrates that there is an important bilateral effect involved between olfactory and visual stimuli. It was possible to verify that smells influence the identification of facial expressions, as well as they influence the emotional reaction to odor as well”, comments Patricia Renovato Tobo , scientific manager of Natura Inovação e Tecnologia de Produtos and co-author of the article.

The research was conducted within the scope of the Center for Applied Research in Welfare and Human Behavior (CPBEC) maintained by the Fundação de Amparo à Pesquisa do Estado de São Paulo (Fapesp) and by Natura at the Institute of Psychology at USP between 2016 and 2021. The work also had the participation of Carla Barrichello, manager of Welfare Sciences at Natura.

Leave A Reply

Your email address will not be published.