Mouse experiments have shown that the olfactory system classifies olfactory stimuli based on their chemical and electrochemical properties.
The work of scientists from Harvard Medical School, published in Nature, sheds light on how our brains perceive and structure olfactory information. Researchers first described how the relationship between different odors is encoded in the olfactory cortex.
The sense of smell allows animals to determine the chemical nature of the world. Sensory neurons in the nose detect odoriferous molecules and transmit signals to the olfactory bulb, a special brain structure where the odor is initially processed. The olfactory bulb mainly transfers information to the pyrophoric, or pear-shaped, cortex for more complete processing.
Unlike the perception of light or sound — stimuli that are easily controlled by adjusting individual characteristics, such as frequency and wavelength — it is much more difficult to study smell perception. Often minor chemical changes (for example, several carbon or oxygen atoms) can lead to significant differences in the perception of the olfactory stimulus.
The authors of the work tried to check how the brain identifies related, but still different smells. Selecting odorous substances with a well-studied molecular structure, scientists let them smell the mice while analyzing the neural activity of rodents. Substances were pre-classified using machine learning methods, taking into account thousands of different characteristics, such as the number of atoms, molecular weight, electrochemical properties.
Before the experiment, three sets of odorous substances were formed, differing in their diversity. The first was the most versatile, the second was medium diversified, and the third consisted of very similar molecules in structure, differing only in the length of the carbon chain. Mice were exposed to different sets of odorous molecules.
It turned out that neuronal ideas about the smell in the olfactory cortex reflect the chemical similarity between the stimuli. Associated odors passed through the same nerve cell patterns in both the olfactory bulb and the pyrophoric cortex. This allows animals to categorize odors and possibly associate them with sensory experience.
An additional analysis revealed that a variety of chemical features, such as molecular weight or certain electrochemical properties, were reliably associated with the nature of the neural activity. The information obtained from these signs made it possible to reliably predict the reactions of cortical neurons to smell in one animal based on experiments with another rodent.
Researchers also found that these neural notions of odors were plastic enough. Mice were repeatedly given a mixture of two odors, and over time, the corresponding neural patterns of these aromas in the cerebral cortex began to correlate more strongly. This happened even when the two odors had a different chemical structure.
The discovery allows us to understand why we all have a common system of views on smells and how it turns out that different people experience with olfactory stimuli is very similar. “We all think that the smell of lemon and lime is similar, and we agree that they do not smell like pizza, but until now we did not know how the brain organizes such information,” said the study’s senior author, neuroscientist Sandeep Robert Datta.