News Release

Quick -- what’s that smell?

Time needed to identify odors reveals much about olfaction

Peer-Reviewed Publication

Monell Chemical Senses Center

Researchers at the Monell Chemical Senses Center have found that taking as little as a hundred milliseconds longer to smell an odor results in more accurate identification of that odor. This seemingly simple observation has important implications regarding how olfactory information is processed by the brain. The findings appear in the August issue of Neuron.

By demonstrating a clear relationship between odor sampling time and accurate odor identification, the Monell researchers solved a controversy centering on whether the brain processes olfactory information in a similar manner to how it processes visual and auditory stimuli.

"Previous published work suggested that olfaction was different from vision and audition in lacking this fundamental property," notes senior author Alan Gelperin, PhD, a computational neuroscientist. "We now can use accumulated information about these other sensory systems to help us understand olfaction."

Exactly how the many thousands of different odorants are detected and identified remains a mystery. The human nose probably contains several hundred different types of olfactory receptors, while animals with a highly developed sense of smell - such as dog, rat, or cat - may have over a thousand different receptor types. It is thought that perception of any one odorant probably involves simultaneous stimulation of several different receptors and that an olfactory code enables identification of specific odorants by the brain. Previous experience and motivational state also interact with odorant information to influence processing and identification. It still is not known how the brain deals with all this information to let us perceive odors.

Using an approach that has provided insight into information processing by the visual and auditory systems, the Monell researchers developed a new behavioral paradigm using trained mice to ask whether longer exposure to an odor would result in more accurate identification of that odor. The results indicated that the mice needed extra time to accurately identify more complex odors.

"The well-trained mouse needs almost half a second to solve a difficult olfactory discrimination task," says lead author Dmitry Rinberg, PhD. "This time window is very important as we seek to design experiments and develop models that explain what the brain is doing in the extra time it takes to identify complex odors."

Rinberg, a physicist and computational neuroscientist, comments, "The development of color television was based on extensive studies of visual sensory processing. Modern MP3 players are built based on a deep knowledge about properties of our hearing capabilities. Similarly, increased knowledge of olfactory processing has the obvious potential to open many doors, perhaps including development of electronic olfactory systems that would have capabilities such as identification of odors for medical diagnosis or detection of land mines."

###

Also contributing to the work was Alexei Koulakov, PhD, from Cold Spring Harbor Laboratory.

The Monell Chemical Senses Center is an independent nonprofit basic research institute based in Philadelphia, Pennsylvania. For 35 years, Monell has been the nation's leading research center focused on understanding the senses of smell, taste and chemical irritation: how they function and affect lives from before birth through old age. Using a multidisciplinary approach, scientists collaborate in the areas of: sensation and perception, neuroscience and molecular biology, environmental and occupational health, nutrition and appetite, health and well being, and chemical ecology and communication. For more information about Monell, please visit www.monell.org.

CITATION: Rinberg, D., Koulakov, A., Gelperin A. "Speed accuracy tradeoff in olfaction". Neuron, 2006, 51, 351-358.

FUNDING: Army Research Office, The Whitehall Foundation, and the Monell Chemical Senses Center.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.