Public Release: 

New studies of smell uncover effects of aging on working memory

How odors are detected and how pheromones may activate nerve cells in the human brain

Society for Neuroscience

In new studies of smell, scientists find that normal aging impairs brain processes involved in olfactory working memory; that olfaction can detect more than one odor at a time; and that 'attraction chemicals' known as pheromones can activate neurons in the main olfactory system, which may explain how humans respond to these substances.

In a new study of humans, scientists find that performance on odor memory tasks was better in the young than in the old. "Interestingly, there was a larger effect of age on identification than on working memory, and both working memory tasks challenged the older adults more than the visual identification task. In fact, performance on visual identification was the same in the young and older adults," says Claire Murphy, PhD, of San Diego State University.

"The results suggest that aging impairs the brain's frontal lobe processes involved in odor working memory as well as visual working memory," says Paul Gilbert, PhD, a consultant to the study. However, says Rose Calhoun, the study's primary author, "it appears that semantic memory for odors is more vulnerable to aging than visual semantic memory, or working memory for either type of stimulus."

Working memory is the type of memory used to keep a telephone number in mind before dialing. Semantic memory is the type of memory used to recall the meaning of a word or what an object is usually associated with.

In the new study, 16 young and 16 older healthy adults were given a working memory task which required them to remember the order in which six items were presented. The items were odors or symbols, depending on the test condition. The semantic memory task required them to identify a series of items, either odors or pictures.

The two types of memory-working and semantic-appear to rely on different regions of the brain, says Murphy. Experiments with visual and auditory memory tasks suggest that prefrontal cortex is essential for working memory, while semantic memory relies heavily on structures in the medial temporal lobe. There is some evidence from studies of brain imaging that the prefrontal areas that are active during visual and auditory working memory tasks are also active during odor working memory tasks.

Very little is known about the effects of aging on odor working memory. However, other types of odor memory tasks-remembering the name of an odor, for example, are very much affected by the aging process. Given that aging affects different areas of the brain and different functions in different ways and at different rates; Murphy and her colleagues wished to know whether odor working memory and odor semantic memory were affected to the same degree by normal aging. They also wanted to know whether odor memory was more vulnerable to the effects of age than visual memory.

In another report, by measuring the speed of smell, researchers have now found that unlike humans, rats can tell two very similar odors apart with just one sniff. And because it's not the 'nose that knows,' but rather the brain, such studies of how animals can rapidly and accurately discriminate odors are revealing vital new information about how the human brain processes information, guides behavior, and even enables us to be consciously aware of our own (though less smelly) world, and our own selves.

"We are trying to understand how systems of neurons participate in the creation of perception, awareness, and behavior," says Cold Spring Harbor Laboratory neuroscientist Zachary Mainen, PhD, who led the new study. By exploring the neural mechanisms by which rodents use odors to guide their behavior, Mainen and his colleagues hope to uncover basic principles of brain function that will apply in many settings, including how our own brains work. But to get there, they needed to start out by measuring seemingly strange things such as how many sniffs a rat takes per second. The answer, according to the new study: about eight sniffs per second.

The "eight sniffs per second" measurement has helped resolve a hotly debated issue in neuroscience. Researchers have previously suggested that the brain requires extra processing time to distinguish among the millions of different chemical signals that can be picked up by the nose. The new study, which appears in the November issue of Nature Neuroscience, overturns this conventional wisdom that smell is a slow sense.

"We found that a rat gets a complete sense of an odor with each sniff. So the animal can reassess what it's smelling quite rapidly, and alter its behavior accordingly. Therefore, compared with other forms of sensory perception, smell is a fast sense, not a slow one," says Mainen.

"Humans are far more attuned to the visual world, but the computations our brains carry out are probably not all that different than in rodents. The neural mechanisms that enable rodents to identify an odor in a single sniff are probably similar to those that help us take in an entire visual scene in a single glance," Mainen adds.

In the study, ten rats were trained to sample an odor and respond with one of two choices (move left or right) according to the identity of the odor. They were rewarded for each correct choice with a drop of water. The scientists measured the odor sampling time, that is, the time between the onset of the odor and the rat's choice. The rats performed 200 to 300 discriminations per day for several weeks, leading to a data set of about 34,000 trials. The odors used were simple organic chemicals such as fatty acids (components of sweaty socks and...some cheeses).

The difficulty of the discrimination was varied by mixing pairs of very similar odor molecules. The researchers used a functional brain imaging technique called "intrinsic optical imaging" to measure the neural responses of the rats' olfactory system to these odors. They also recorded the number of sniffs during odor sampling by implanting a small temperature probe in the rats' noses to measure the warming and cooling of air during breathing.

Median odor sampling times for 4 similar and 2 dissimilar pairs of chemical stimuli ranged from 250 to 300 milliseconds. Accuracy on these problems was over 90 percent in all cases. When more difficult mixture stimuli were used, accuracy dropped for the most difficult mixtures but odor sampling times remained quite constant, increasing only 35 milliseconds from the easiest to the most difficult discriminations. The results were insensitive to odor concentration and other parameters of the task.

When the data was divided by odor sampling time, the scientists found that the rat's discrimination accuracy reached a maximum after about 200 milliseconds regardless of the difficulty. That is, a choice made in 200 milliseconds was about as likely to be correct as a choice made after 300 milliseconds or even 500 milliseconds. When odor sampling time was measured in terms of sniffs, it could be seen that maximum accuracy was achieved in just a single sniff.

Mainen and his colleagues are currently recording electrical signals from neurons in the brains of rats as they perform the odor discrimination task. In this way, the researchers hope to learn more about information processing in the olfactory system, and to explore the neural basis of perception, decision-making, and other aspects of behavior.

Another new study uncovers the first direct evidence that known pheromones can activate neurons in the main olfactory system, which may explain human responses to pheromones.

Pheromones are chemicals created and exuded by the bodies of individual animals, and carry messages that convey gender, individual identity, and even social status to other members of the same species. Pheromone processing has generally been linked to a specialized system called the accessory olfactory system, which is absent in humans. However, the main olfactory system, which is used for all kinds of smelling, and is present in humans, now appears to be activated by certain pheromones. Bodily secretions, such as urine, have long been known to play a crucial role in the reproductive and sexual behaviors of many mammals, including rodents, ungulates and carnivores.

By recording the electrical activity of individual nerve cells in the olfactory bulb of mice (the first olfactory processing station in the brain) scientists have uncovered nerve cells whose patterns of activity distinguish between male or female urine.

"By combining these recordings with a chemical separation technique called gas chromatography, we have identified specific components present in urine that animals may use to differentiate gender based on chemical cues," says Lawrence Katz, PhD, of Duke University Medical Center. "Using this approach, we've uncovered the first direct evidence that known pheromones can activate neurons in the main olfactory system, which may explain human responses to pheromones."

After recording from hundreds of individual neurons, Katz and his colleague, graduate student Dayu Lin, found that neurons responsive to urine are found in a restricted region of the olfactory bulb. Neurons frequently showed robust activation to the urine of either male or female animals, but not both. Urine contains a complex mixture of over 100 odorous molecules. When the researchers compared the gas chromatography profiles of the different urines, they found significant differences between males and females, both in the presence of certain components or in their abundance.

To determine the odorants that produced the distinctive responses, the scientists used gas chromatography to separate the urine into its component odorants. They then directed the effluent of the chromatograph (containing the individual odorants) to the nose of the mouse, while recording the activity of the urine-selective cell. Remarkably, only one or two of the hundred or so compounds present was responsible for activating the neuron. Thus, the urine responses represent an exquisite level of selectivity.

The scientists were especially excited to find that some of the urine-specific responses were to chemicals that had previously been identified as pheromones by behavioral assays. This provides the first direct evidence that pheromones can be detected by the main olfactory system. In addition, they have uncovered a number of other molecules which give sex-specific responses, and these are currently being identified by mass spectroscopy.

In the course of evolution humans have lost much of the molecular machinery usually associated with pheromone detection, a loss that coincided with the emergence of visual capabilities. "Our finding that pheromones can be detected by the main olfactory system suggests that we may still be able to detect and respond to such signals," says Katz. "Recent human brain imaging studies show that certain odorants can activate brain areas in humans that pheromones activate in other mammals. Perhaps this is why perfumes - which contain pheromone-like substances - still smell sexy."


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.