Joel E. White is a Research Assistant Professor of Neuroscience at Tufts-New England Medical Center. White attended the University of Kansas where he obtained his BA in Biology in 1983. After completing his Ph.D. at Florida State University in 1989, White did his Post-doctoral work at the New England Medical Center.

White works to understand the mechanics of olfaction, the sense of smell. He is currently investigating how the brain detects odors at low concentrations. His findings will be applied to the development of technology designed to detect weapons and explosives.

Kauer and White's response:
This, of course, is a very timely question. The ability of our device to detect a substance requires that the substance release molecules into the air, which are drawn in during a sniff. It seems unlikely that Anthrax spores release any volatile compounds - bacteria spores are essentially inert. Unfortunately, our sensors are not likely to respond directly to the airborne spores. However, it may be possible that chemicals used in the processing of Anthrax may be detectable. This latter possibility would make tuning the device for this application quite a bit safer!

Another possibility is that during metabolism, living Anthrax may release molecules into the air which could be detected. If so, it may be possible for the device to detect the early stages of Anthrax infection, perhaps through sniffing a person's breath. We have given some thought to using our device for other, similar medical diagnoses. There are a number of disease states that lead to changes in the composition of breath and axillary odors.

In terms of "other new threats," it may be that our device could be used to detect chemical warfare agents. A number of these agents (such as Sarin) are volatile, so may be detectable by our device. The goal would be to detect reliably any such agents at concentrations well below a lethal dose.

Kauer and White's response:
The behavior of seizure alert dogs is certainly fascinating, but there seems to be very little in the scientific literature regarding how, exactly, these animals are able to detect an impending seizure. Although odor detection may be involved, dogs may also use other cues. Anecdotal reports indicate that it is difficult to train effective seizure alert dogs. Also, those dogs that are successful in detecting seizures are typically quite attached to their owners. In other words, they are very familiar with their owner's normal and pre-seizure behaviors as well as their odors and may be more in tune to small changes. These observations suggest that the cues the animals are using may be quite subtle. It is certainly possible that a seizure alert dog uses some of its other senses (eyes, ears, and touch) to detect very small changes in their owner's behavior prior to a seizure. In this case, of course, our device would be ineffectual. If, instead, the cues are indeed odors, it may be possible for our device to detect them.

This is an interesting area that needs more research to determine how the dogs are doing what they do.

Kauer and White's response:
Much of our study of the sense of smell (the "olfactory system") has focused on tiger salamanders. While dogs and salamanders are different in a large number of ways, their olfactory systems are actually similar in some important respects. Both animals have sensory cells in their noses that respond to chemicals in the air. These sensory cells communicate with an area of the brain that is remarkably similar in both animals (indeed, this brain area is similar in all vertebrates). The similarities among animals means that studying one animal should provide insight into all animals. There are a number of advantages to studying the olfactory system of the salamander: the animals (and their olfactory systems) are smaller and easier to handle and the anatomy of their olfactory systems are much simpler than in dogs. These advantages have lead to a large scientific literature (from our lab and many others) focusing on the olfactory system of the tiger salamander.

Although there is relatively little information on the physiology of the dog's olfactory system (in other words, "how it works"), the studies of dogs and other mammals that have been published indicate that there are indeed remarkable similarities to what has been seen in the tiger salamander. We conclude from this that the details of the mechanisms that we learn from the salamander are likely to be similar in dogs and other mammals.

From a practical standpoint, there are some great advantages of dogs over other animals (including salamanders!) for odor tasks like landmine detection. Dogs take instructions readily from humans, they learn very quickly, and they are quite adaptable to different situations and environments. In addition, the large size of their olfactory systems (both in the nose and in the brain) is likely to be at least part of the reason they can detect the small amount of odor that comes from a buried landmine.

Kauer and White's response:
Additional funding would certainly be helpful in the development of our device. In addition to device development, more research is necessary on how odors are related to landmines that are actually in the field, how they change over time, and how dogs detect those odors. These details are needed for us to understand better how to find real, live landmines in the field. The events of 9/11 have led to a general increase in interest in explosives detection. There are currently a number of new funding opportunities that we are exploring.

It is unfortunate, but from reports that we have seen, Afghanistan and Cambodia are among the most heavily mined countries in the world.

Kauer and White's response:
As Mark Twain said, "The art of prophecy is very difficult -- especially with respect to the future." We can only guess at a timeframe. If improvements to the device proceed at a pace similar to that of the past year, we may have something ready for extensive field testing in a couple of years or less. We are conservative in our estimates and hope that the final version could be mass-produced easily and relatively cheaply. We are focusing on a hand-held or carried version, but it is certainly possible (and perhaps preferable!) to mount it on a robot for actual mine detection. There are other research groups with robotic expertise that we would work with for that.

Kauer and White's response:
A little bit of both, actually. After a good many years of studying the sense of smell in animals (funded by the National Institutes of Health), we became interested in implementing our ideas about olfactory function in artificial systems, first in computer simulations (funded by the Office of Naval Research), then in hardware devices. We pursued the landmine project (funded by the Defense Advanced Research Projects Agency, or DARPA) because of the need to focus our efforts on a specific odor detection problem. We were also intrigued by the problem from a scientific viewpoint (how could one solve a difficult odor detection task?), as well as from the standpoint of possibly making a positive contribution to a terrible humanitarian problem.