Imagine trying to solve a puzzle using one of the most personal and elusive human senses: smell. Unlike colours or sounds, odours don’t come with obvious labels, and people often experience the same scent in strikingly different ways. Yet in everyday life – from cooking to detecting danger – we regularly talk about smells and rely on others’ noses alongside our own. A new study, led by Mi3 researcher Mustafa Yavuz, asks whether this everyday intuition has scientific backing: can people actually make better decisions about smells by working together?
To find out, researchers tested volunteers using standardised clinical smell tests. Participants were asked to discriminate between similar odours and to identify specific smells from multiple-choice options. Everyone completed these tasks alone first, giving researchers a clear measure of each person’s individual smelling ability. Several weeks later, the same participants returned, this time paired up. Some pairs were carefully matched so both people had similar levels of olfactory skill, while others were deliberately mismatched. Each person still made an initial decision privately, but then the pair discussed the smell and had to agree on a single joint answer.
The results reveal something striking: when two people worked together, they often outperformed the average of the individuals. In other words, under the right conditions, two noses really were better than one. This wasn’t true for pairs with very different abilities, where collaboration added little benefit. The pattern closely mirrors what scientists have previously found for vision and other senses: collaboration works best when partners are roughly equally skilled.
Crucially, this improvement wasn’t just a matter of taking turns or voting. The pairs’ joint decisions were better than what you would expect if they were simply guessing whenever they disagreed. Something about talking – even with limited and imperfect language – was helping people combine their information more effectively.
One reason this worked is that people turned out to be more insightful about their own sense of smell than often assumed. Participants’ confidence ratings reliably tracked how accurate they actually were. This meant that during discussions, people weren’t only exchanging guesses like “I think it’s coffee,” but also signals such as “I’m pretty sure” or “I’m unsure,” which helped pairs resolve disagreements.
Interestingly, the most successful conversations were not the longest ones. When pairs talked a lot, especially during difficult identification tasks, they were more likely to be wrong. Short, focused exchanges worked better than lengthy debates. In fact, the more talkative person was less likely to have their answer chosen as the final decision, suggesting that confidence calibration – not persuasion or verbosity – was doing the real work.
Together, these findings challenge the idea that smell is too subjective or too hard to talk about to benefit from collaboration. Even without a rich vocabulary for odours, people can align their judgments, share uncertainty, and improve performance through interaction. The study suggests that olfaction, much like vision, can support collective intelligence when the social conditions are right.
Beyond the lab, this has intriguing implications. From wine tasting and perfumery to food quality control and clinical testing, smell judgments are usually treated as solitary acts. This research hints that carefully matched pairs – or groups – might actually reach better conclusions together. More broadly, it reminds us that human communication doesn’t rely on perfect descriptions. Sometimes, simply knowing how sure someone else feels is enough to sharpen our own perceptions.
Check out the original article and Mustafa Yavuz’s profile.
Reference:
Yavuz, M., Sayahpour, S., Bahrami, B., Deroy, O. (2025). Human Pairs Show Collective Benefit in Olfactory Perception Despite Individual Differences and Verbal Limits. iScience, Volume 0, Issue 0, 114535. https://doi.org/10.1016/j.isci.2025.114535