In their experiments, researchers led by Edmund T. Rolls of the University of Oxford presented subjects with a cheddar cheese odorant and showed them labels that read either "cheddar cheese" or "body odor." They found that the subjects rated the odor significantly more pleasant when it was labeled "cheddar cheese" than "body odor."
They then scanned the subjects' brains using functional magnetic resonance imaging (fMRI) during the presentation of labels and odors to explore which brain regions were activated. They also analyzed brain activity when the subjects were presented with clean air labeled either "cheddar cheese" or "body odor." The widely used analytical technique of fMRI uses harmless magnetic fields and radio waves to measure blood flow in regions of the brain, which reflects brain activity.
The researchers found that labeling the odor "cheddar cheese" produced an activation in a specific part of the brain region that processes olfactory information. Clean air labeled as "cheddar cheese" activated the same area, but to a lesser extent. The "body odor" label, however, did not produce activation in this area, either with the cheddar cheese odor or clean air.
The researchers also used correctly labeled pleasant ("flowers") and unpleasant ("burned plastic") odors as reference odors to test the subjects' responses to such odors and to identify areas of the brain activated by either pleasant or unpleasant odors. Also, they tested whether a change in the amount of "sniffing" in response to an odor label might influence the results, finding no effect.
They emphasized that they used word labels so that the cognitive input would be "high level and semantic," as opposed to a picture, which could have been a lower-level association in the brain.
"The results thus show that cognitive inputs can be very important in influencing subjective responses, including affective responses to olfactory stimuli, and show that some of the brain areas activated by odors . . . show an effect of this high-level cognitive influence," wrote de Araujo and Rolls. They also noted that their inclusion of clean air in the study showed that semantic labels influenced judgment, even when the absence of odor was paired with the labels.
Whether the words caused subjects to imagine a smell or simply affected their brain's processing of odors, "the important new point being made in this paper is that high-level cognitive inputs, such as the sight of a word, can influence the activity in brain regions that are activated by olfactory stimuli," they wrote.
The researchers include Ivan E. de Araujo1,2, Edmund T. Rolls1,2, Maria Inés Velazco3, Christian Margot3 & Isabelle Cayeux3. This research was supported by grants from the Medical Research Council.
1 Department of Experimental Psychology University of Oxford South Parks Road Oxford OX1 3UD United Kingdom
2 Centre for Functional Magnetic Resonance Imaging of the Brain John Radcliffe Hospital Oxford OX3 9DU United Kingdom
3 Firmenich SA, 1 Route des Jeunes, CH-1211 Geneva 8, Switzerland
de Araujo, I.E., Rolls, E.T., Velazco, M.I., Margot, C., and Cayeux, I. (2005). Cognitive Modulation of Olfactory Processing. Neuron 46, 671-679. www.neuron.org