Dogs rely heavily on their sense of smell for survival, communication, and work, yet the neural mechanisms underlying odor discrimination have remained poorly understood. Prichard et al. (2019) used awake fMRI in 18 trained dogs to examine how the canine brain processes single odors compared to odor mixtures.
In the experiment, one odor (A) was paired with a food reward, another odor (B) had no association, and a mixture of both odors (A+B) was also presented. Neural activity was measured across brain regions including the caudate nucleus, amygdala, and olfactory bulbs.
Findings showed that the amygdala rapidly distinguished between reward and non-reward odors, with the mixture (A+B) eliciting responses most similar to the non-reward odor (B). Importantly, machine learning models revealed that the amygdala, piriform cortex, and posterior cingulate cortex were the most critical for decoding odors. Dogs did not perceive the mixture as simply a combination of A and B but rather treated it as a novel configuration.
This research demonstrates that odor perception in dogs is not purely elemental but involves unique neural coding of mixtures. Such findings advance our understanding of canine olfaction, with implications for working dogs in detection roles and insights into how dogs interpret the complex scent world around them.
Source: Prichard, A., Chhibber, R., King, J., Athanassiades, K., Spivak, M., & Berns, G. (2019). Decoding Odor Mixtures in the Dog Brain: An Awake fMRI Study. bioRxiv. Published September 1, 2019.
