Researchers finds brain processes certain aromas as real tastes, shaping a person’s eating behaviors.
Researchers at Karolinska Institutet in Sweden have found that the brain interprets certain smells as if they were actual tastes, providing new insight into how people perceive flavor. Their findings, published in Nature Communications, help explain why drinks with added aromas, such as flavored waters, can seem sweet even when they contain no sugar. Flavor is more than just taste on the tongue. It comes from a mix of taste and smell, with aromas from food traveling through the mouth into the nose in a process called retronasal odor. Until now, scientists thought the brain kept taste and smell signals apart until they reached higher brain regions. The new study, however, shows that certain smells trigger taste and that these signals merge much earlier, in the insula, a part of the brain recognized as the taste cortex.
According to lead researcher Putu Agus Khorisantono from the Department of Clinical Neuroscience, specific aromas trigger taste when activated in the brain in the same way as actual tastes. That finding explains why a smell alone can trigger taste when the brain is experiencing flavors such as sparkling water or smells food that seems to be sweet before it even touches the tongue.
The study tested 25 healthy adults who were taught to recognize both sweet and savory flavors by pairing tastes with matching aromas. Participants then underwent brain scans using functional magnetic resonance imaging (fMRI). During the scans, they were exposed either to pure aromas without taste or to tastes without smell. A computer algorithm trained to recognize brain activity patterns for sweet and savory tastes was then applied to see if the same patterns appeared with aromas alone.

The results showed a striking overlap. Smells linked to sweetness or savoriness lit up the same brain regions as the tastes themselves and produced similar activity patterns. The overlap was strongest in areas of the taste cortex that handle the integration of different sensory inputs. Senior researcher Janina Seubert, also from Karolinska Institutet, explained that this confirms the brain does not handle taste and smell separately but builds a shared representation of flavor directly in the taste cortex.
This mechanism could have a direct influence on eating behavior. If aromas alone can trigger the perception of flavor, they may play a stronger role than previously understood in shaping food preferences and cravings. This may also help explain why certain foods are more tempting than others and why aromas can encourage overeating.
The team now wants to expand the research to external smells, called orthonasal odors. The goal is to see whether the same process takes place when people encounter scents from the environment rather than from food inside the mouth. As Khorisantono put it, one question is whether a person walking from the cheese section of a grocery store to the pastry aisle experiences a shift in brain activity patterns, from savory to sweet, even before eating anything. If proven, this could have wide implications for how food choices are influenced by the surrounding environment.
The study was carried out with researchers in Turkey and received support from the European Research Council and the Swedish Research Council. While more work is needed to fully understand the effect, the findings highlight the brain’s ability to merge senses in ways that shape behavior, sometimes without conscious awareness.
By showing that the brain treats certain smells as tastes, the research helps explain everyday experiences, from why flavored drinks seem sweet without sugar to why certain aromas can instantly trigger cravings. It also suggests that flavor perception is deeply tied to brain activity patterns formed in the taste cortex itself, not just a combination of separate inputs that come together later.
Sources:
Study reveals how the brain interprets certain aromas as taste
Tastes and retronasal odours evoke a shared flavour-specific neural code in the human insula


Join the conversation!