We walk into a large dairy product manufacturing building we will be hit with many scents. Our mind blends the scents into a few combined aromas. These combined aromas are distinguishable but all have predominantly a milky scent and blend into one. A dog however can smell all the components.

We cannot ask the dog how he is able to discriminate scents but what we can do is make hypothesis and then try to eliminate the hypothesis we created. Though this philosophical method for discovery is neither scientific nor empirical we have out there some good reading and ideas about the scent capabilities of dogs.

For my brainstorming about how dogs use their sense of smell I used the study of a polar bear and a seal’s breathing hole.

Bears are part of the canine family and unlike other bears, the polar bear does not hibernate. Also unlike other bears, polar bears get most of their food by hunting and they rely heavily on their scenting capabilities. I would guess natural selection helped them adapt their minds for this because they need to find food with the least amount of energy expenditure. This is important to a bear who needs plenty of body fat to maintain warmth. If anything I felt polar bears were a good field source that is “outside my dogbox” yet directly relates to my dog studies.

The Polar Bear and His Nose

Polar Bears can smell a seal's breathing hole up to a ¾ a mile away and can smell dead whales and seals from 20 miles away. This fact made me consider the frozen snow and air moisture. Knowing cool preserves the life of the scent and moisture is a delivery system for scents. I then thought about the miles and miles of frozen ice and one seal breathing hole and how this could attract the polar bear. The bear was a good distance away and unaware of a new seal breathing hole and yet he caught the scent and headed straight for it after a few air scent behaviors.

My answer was: The breathing-hole places particles of seal scent that is not part of the ice scent each time a seal would surface in to the air. This scent is mixed with the billions and billions of air molecules and other dominating environmental scents such as the relatively uniform scent of the ice. The fact he did this was not as important to me as was how he did it.

I guessed the bear located the new scent because it was a new scent to the area peaking his interest and he then discriminated the scent favorite food.

Like this bear, I believe dogs smell (for a lack of a better explanation) in a mental dimensions.

They can separate scents and organize them first by scent detraction of their environment. Detraction is defined in behavior as a lowering of attention or concentration without shifting the point of focus. They know the “scent constants” in the environment by sampling the air or detail scenting the location. They then focus their mental energy on the unique intrusions. Some of these will be learned scents and they respond accordingly to the conditioning of their experiences. Some will be new and it is at this time reinforcement or a lack of reinforcement needs to take place.

Barrowing a quote from Sir Arthur Conan Doyle and then applying Occam’s Razor to it: “When you have eliminated the impossible, whatever remains, however improbable, must be the truth. The take the simplest explanation for some phenomenon as it is more likely to be accurate.

I think dogs do not segment or blend scents but use detraction skills and then discriminate the pool of scents that remain individually to identify what they are looking for.

My conclusion: The use the 100% combined components of my scent target is more logical for training from beginning to end.

If I am right this places my use of anything other then the exact combination of scents I want to use for discrimination training into question. My goal is to try to make the association of scent = reward as easy as possible. Consistency being a foundation of training that is gospel.

If I teach a scent that is more closely related to the entire scent spectrum of that particular and unavoidable combination of scents based on purity age and condition. I am refining the “knowledge” of accepted combinations before they are detracted in the dogs mind. These scents would later need to be learned anyway.

To proof the dog from false positives or missing the find he needs good exposure to the acceptable combinations as a whole. A good discerning K9 should be convinced these combined scents mean he needs to have a closer examination of the non-detracted scents.

The argument of using only the COMMON scent in of the whole combination of scents that are found on the street is forceful at first glance because you are signaling out the scent despite the impurities.

The fact is the dog can handle the detraction of multiple scents and then discriminate because he is made to do it. He can decipher the “winning” scent by association fairly easy.

This thinking kind of goes along with the fact I never teach my dog to “stay.” I rationalize that when he is sold to sit stand or down he is staying. Still many people will teach their dog to stay and that is fine. <img src="/ubbthreads/images/graemlins/wink.gif" alt="" />

Dennis, I'm with you I think. As it relates to not teaching the stay command. I do believe that platz means platz and sitz means sitz. Yet I'm guilty of teaching the stay command to my handlers. May be a good topic of discussion tonight at training.

As it relates to scent discrimination. I know a very successful trainer who trains his detector dogs with all of the odors at once. At least in the imprinting stage the narcotics are in the same box. After all isn't it the dog that has the ability to discriminate. As the training progresses and the narcotic hides are seperated the dog has no problems indicating. Are we not guilty of anthropomorphasizing when we assume that the indication on the odors must me taught one at a time and almost always marijuana first because it has the strongest odor?

BTW you gave me a great idea the other day. We have been sitting on our thumbs waiting for our DEA license. My plan has been to cross train our patrol dogs. In order to stay on the up and up we applied for a license. Very slow process. We have all of the narcotics that we need in evidence. I've been sitting back and waiting. I like the idea of scenting towels/narc bags etc. in order to go ahead and get started.

A researcher, will place more importance on the WHY a dog can do something than a trainer. As a trainer, from my perspective at least, the important point is that we can document the behavior, and demonstrate a dog's ability to discriminate.

DFrost

As it relates to scent discrimination. I know a very successful trainer who trains his detector dogs with all of the odors at once. At least in the imprinting stage the narcotics are in the same box. After all isn't it the dog that has the ability to discriminate. As the training progresses and the narcotic hides are seperated the dog has no problems indicating. Are we not guilty of anthropomorphasizing when we assume that the indication on the odors must me taught one at a time and almost always marijuana first because it has the strongest odor?

The Germans did quite a bit of study on this subject and still today most start with multiple odors initially. I also start them with multiple odors, using a primary reward system. It provides reinforcement of the multiple odors quickly and has the German studies to support it.

Howard,

If you are going to use scent impregnated cotton towels they should be free of dye and 100% COTTON.

Wash them about 5 times in a washing machine (NO SOAP) and dry them in the air on after the last cycle.

You containers need to be as free as possible from odor as well. Use CLEAR containers that seal tight and wash them without soap place them in the open sun and air for at least 5 days. This weathering will make the containers fairly free of any “plastic scent.” Even so, keep a few “proofing containers” and “proof towels” that never come in contact with any drug so you can test your dog on them. I can never remember it being a false positive problem though.

The towels go into the container first then the “in bag” narcotics. Open the top of the sealed bags in away that none of the narcotic comes in contact with the towels. Leave it the bag open and seal it for a few days or even a few weeks.

The amount of scent impregnated into the towel fibers is relative to the amount and quality of the narcotic and the exposure time in the container. You will need to make a judgment call on how much towel to use and based on your supply and training level and goals.

Generally speaking the longer the bag is sealed with the towels the more scent is impregnated up to a scent saturation point. Sorry no time for saturation time windows yet. So I would paper tag my containers with date so I could breakdown the scent amounts more accurately until you get the OK from DEA.

Good luck.

The Germans did quite a bit of study on this subject and still today most start with multiple odors initially. I also start them with multiple odors, using a primary reward system. It provides reinforcement of the multiple odors quickly and has the German studies to support it.

I wasn't aware that there were studies supporting it. That is good to know. I'm amazed at the number of trainers who still train one odor at a time and always start with marijuana because of the belief that it smells the strongest.

It's evident that they still get there but simply from the perspective of saving time it makes sense to start with all of the odors at once.

I have had countless conversions with trainers on this, and most are hell bent on starting one at a time. I have found that after training at least five dogs like this, I am happy with the end result.

Kevin, are there any web pages that have some of these studies published?

It is tough to change a training method that you have done for years that is quite successfull. Why bother changing? Well, a year or so ago I started using scent mixing and have not looked back since. Either way works just fine if done correctly but scent mixing just saves some time. Scent mixing goes against the foundation of the Customs training method which is what I started on. The preaching always was never let the odors come into contact and don't do this or that. Well, low and behold the dogs out did us once again. I welcome change and the challenge it brings to training. The end result is what counts even if some accept different paths to get there.