When it is necessary to sniff out a trail--be it to a bomb, a drug stash or even to a pie sitting on a windowsill--the task is normally better left to our canine companions. And while our four-legged friends still hold the crown for tracking scents, a new study published in this week's issue of Nature Neuroscience proves that humans may not be bloodhounds, but they can follow a scent--and they get better with training. In addition, the paper provides evidence that mammals make comparisons between what is sensed by each of their nostrils to synthesize information on scent much the same way that they use both ears to hear.

In a series of experiments designed to uncover the mechanisms involved in tracking odors, neuroscientist Noam Sobel and his team at the University of California, Berkeley, worked with human subjects because, as Sobel's graduate student, Jess Porter points out, they could "conduct manipulations such as blocking one of their nostrils, and & make sure that they understand the task we want them to do." Further, she says, "we wanted to test whether the dual-nostril configuration that humans and other mammals have is utilized in the scent-tracking task."

In the first experiment, 32 subjects wearing blindfolds, earplugs and thick gloves attempted to track a scent of chocolate essential oil through a grassy field. Twenty-one of the participants successfully completed the 10-meter-long course, consisting of two straightaways and a 45-degree turn. This proved that humans, though inefficient at doing so, could follow the path of a scent. Next, the researchers trained four other subjects on the same task, having them track a scent nine times over a two-week period. The team found that the trained odor detectors were able to stay closer to the trail mapped out by the scent, without zigzagging wildly around it, and that they finished the course twice as fast in their last attempt as they did the first time around. The researchers note that as the participants began tracking faster, they also began sniffing faster, presumably to gather information more quickly.

The humans, however, still sniffed much more slowly than dogs, which may partially account for canines' greater efficiency at scent tracking. Gordon Shepherd, a neurobiologist at the Yale University School of Medicine, says that despite their relatively sluggish speed, the fact that subjects improved with training is noteworthy. "I think that shows the effect of our distinctively different behavior in actually using this sense," he says. "The dog [has] been doing this its whole life, and humans [were] just asked to plunge in the first time they've ever done it."

A second question addressed by the Berkeley study was how the human nose processes scents. Since the late 1960s scientists hypothesized that our closely spaced nostrils sample essentially the same air--unlike our ears, which, due to the space between them, collect different inputs and compare both signals to glean auditory information. Sobel's findings indicate otherwise. In collaboration with a group at Penn State University, and using a technique called particle image velocimetry, the researchers took rapid snapshots of airstreams while subjects inhaled them. "The results showed us that the two nostrils draw air from separate regions of space," Porter explains. More specifically, the nostrils draw samples from areas whose centers are 3.5 centimeters apart. "Thus," he says, "in principle, they could carry different information about odor concentration in two regions." This finding was further supported when they returned to the grassy field and had subjects follow scents with one nostril blocked or while sampling only the air present at the center of the nose. In either case, tracking was significantly less successful.

"This is the first evidence I've seen that humans might have access to that kind of information--that there's some spatial separation between the air that's coming into the two nostrils and that could potentially give them some information about which way to go to follow the plume," says Neil Vickers, a neurophysiologist at the University of Utah who works with olfaction in moths, which are known to make bilateral comparisons about the air around them. Porter notes that while the Berkeley study does show a benefit to the "dual-nostril configuration," the exact utility is not yet clear. "For instance, we were not simultaneously recording brain activity, or visualizing the concentration of odorant entering each nostril--so we don't have a complete picture of exactly what information the brain has access to, and how it is processing that information."