Neural Research Shows That the Nose Needs Time To Smell
PASADENA— New research from the California Institute of Technology shows that it literally takes some time to smell the roses.
In the current issue of Nature, Caltech neuroscientists Michael Wehr and Gilles Laurent present work demonstrating that information about odors is contained in the temporal activity patterns of groups of neurons over an interval of time.
"Perfumers sometimes speak of 'top notes' and 'medium notes' in a bouquet," says Laurent, associate professor of biology and computation and neural systems. "These refer to early and late perceptions that unfold over time during long sniffs or successive sniffs. Our new research suggests that the brain is actually representing odors by making a neural melody of its own."
A helpful analogy Laurent offers for the research is the musical notes that make up a tune. A listener can perceive one note in an instant, but must listen for a time before he or she can recognize the tune. Therefore, the specific manner in which the notes follow one another is the very thing that gives a song its individual character.
"It is the order in which specific neurons are activated that appears to contain useful information about the identity of the odor," says Laurent, adding that different odors cause different neural "melodies."
Laurent and Wehr, a graduate student in computation and neural systems, did their research by analyzing the brain waves of locusts. When an odor was wafted by the olfactory organ of the locust, the collective response of neurons in the olfactory brain was such that specificity in the responses arose from considerations of their temporal characteristics. And because olfactory systems are very similar among most animals, the researchers think that these coding principles may be common to most, including humans.
What happens in the brain during the act of smelling is not well understood, but has been known for a long time to involve neural synchronization and oscillations of the EEG. The function of oscillations, which are observed also in all other sensory areas of the brain, remains totally speculative.
"If our hypothesis is right, oscillations are a kind of clock for the temporal codes we observe," Laurent says. In a parallel study by Laurent and Caltech behavioral biology graduate student Katrina MacLeod in the November 8 issue of Science, the authors described a method by which the neurons representing odors can be simply desynchronized, thereby eliminating the clock signal for the temporal codes. This result will now allow the researchers to directly test, in future experiments, whether these temporal codes are essential for odor perception.
The conclusions of the researchers is that animals as primitive as snails and as complex as humans do some mental "data crunching" each time they pick up a smell. This allows the neurons to separate a certain odor from the background, provided a window of time is available.
In a manner of speaking, the research shows that time is of the essence, and vice versa.