The nose may know, but the brain has a hard time making sense of smells.
Scientists have identified two brain regions that act as an interface between odor and language, helping guide word choices that describe what the nose smells. These regions receive quite crude olfactory information very early in the brain’s smell processing pathways, which may explain why people have such a hard time identifying odors.
“It seems that smell is integrated at a very early stage,” says cognitive psychologist Jonas Olofsson, who led the new study, published November 5 in the Journal of Neuroscience. Unlike visual and auditory information, which travel through much more circuitry before reaching the brain’s language network, smell data seems to come in rough and unedited.
Previous studies in Alzheimer’s patients and other people who can’t name odors had implicated the now pinpointed brain regions. And it has been known for decades that most people are pretty bad at naming smells or their components.
“Smell is very holistic. We’re not very good at isolating multiple objects the way we are with visual objects,” says Olofsson, of Stockholm University. “We make these interesting, obvious mistakes with smell.”
While people can often tell two smells apart, recognizing an individual odor is difficult and often depends on context. “In one context a smell might be identified as a nice cheese, in another it might seem like the person standing next to us needs to wash their clothes,” Olofsson says.
To get at what the brain is doing when someone tries to name a particular smell, the research team put electrodes on the heads of 15 people who then sniffed a particular odor, such as peanut butter or gasoline. The participants were then shown a word that either matched the smell, was related to the smell (the word “chocolate” after smelling peanut butter) or unrelated to the smell (the word “chocolate” after smelling gasoline). When the word was unrelated to the smell, there was a spike in electrical activity toward the back part of the participants’ brains. Using the same setup, the researchers then asked participants to view a picture instead of a smell along with the same word sets. For unrelated words, the spike in electrical activity didn’t come from the same area, suggesting that a different part of the brain was at work when processing an odor cue versus processing a visual one. Participants were also slower and made more mistakes when cued with a smell compared with their response to a picture, suggesting that the odor-language interface is relatively inefficient at making sense of smells.
A separate group of people then underwent similar experiments in an fMRI machine, which allowed the researchers to pinpoint the two areas of the brain that seemed to be deciding what each smell was. These regions, the anterior temporal lobe and orbitofrontal cortex, are both closely connected to the olfactory cortex, the brain’s primary headquarters for smell information. The findings suggest that these regions are talking to brain areas directly involved in language.
The study is “a very good contribution,” says neuropsychologist Johan Lundström of the Monell Chemical Senses Center in Philadelphia. Sensory information typically travels through a major switchboard of the brain, the thalamus, before it’s relayed to the cortex. It’s intriguing that smell information doesn’t seem to do that, Lundström says. By establishing the role of these brain regions in labeling smells and showing that it differs from visual processing, the researchers “have demonstrated something that we’ve been talking about for years,” he says.
The new work raises several interesting questions, including whether it’s possible to train the brain to be better at identifying smells. Research has hinted that these brain regions can be enlarged in people who use their nose for a living, such as experienced perfumers. Olofsson is now conducting experiments to determine whether such experts are born with a great sense of smell or if it can be improved with training.