Part One – Detection Canine’s False Response/Nonproductive Response

The basics of detector dogs.This article will bring a fresh look to many decades of canine training myths, mysteries and lies. In recent years I have seen an increase in articles and training seminars revolving around “problem solving” on canines with false responses/nonproductive response (NPR) issues. Much of the information contained in these articles or presented in the seminars is downright false and or very misleading. This misinformation is then utilized by trainers to solve dog training problems that do not exist or that cannot be rectified with the prescribed method of training, many times involving the incorrect application of punishment. Furthermore all of this misinformation is “discoverable” to defense attorneys and much of it is going to be seen again, next time used against law enforcement professionals in court leading to lost cases and more bad case law (more on this later).

Because of the amount of this misinformation, I felt it was time to help clarify and distribute new ideas and opinions on this subject.  The information written here is bound to be controversial at first, but in the end there will be many converts to these new ways of understanding, describing and solving these training and handling issues. I am generally speaking about all detector dogs but most specifically to explosive and narcotic dog teams.  Through the course of this series we will discuss terminology, look at some basic detection principles and dissect some training myths.

I first have to start out with a single detection (trained odor present in the search area) exercise broken down into articulable stages.  We start by assuming the canine has all the intangible’s we test for (health, search, hunt, retrieve, desire, intent etc.) As trainers, the reason’s we MUST have a clear grasp of the different behavior series is for all areas of detection training; basic training, maintenance and problem solving.  Without the in-depth knowledge and understanding of these stages, training in any area becomes more of guessing game (haphazard) than science (articulable and repeatable).

Now in training and deployed on our descriptive “search” break-down of “stages” we have-

  • The “Alert” or “Change of Behavior” (COB) the canines initial recognition of the trained/target odor.
  • Next is the “Interest” or “Tracing to Source”
  • Once at “source” or as close as he can get we have the “Trained Final Response” (TFR).

I like to call it “AIR” in training courses “Alert, Interest, Response.”  If you imagine a dog encountering the trained odor from 50 yards away, the stages easily unfold before you!  The dog makes initial contact with the trained odor, “ALERT” target acquired.  The dog changes direction and with great concentration pursues the trained odor to source “INTEREST,” once at source we see an enthusiastic sit “TFR” awaiting a well-deserved reward!  I usually use more behavioral descriptions to describe these stages but the article is about false responses/NPR’s not basic training.

Next, I would like to start out with a simple true/false statement to get you thinking outside your normal training box and perhaps to plant some new seeds of information.

Every time a detector dog is deployed to search we are just simply asking the dog a question, “Is there trained odor present in the search area or not?”

There are four possible answers that the trained canine is going to provide.  Two of these answers are “true” and correct and two of these answers are “false.”

  1. The dog responds “positive” and there is trained odor in the search area (correct).
  2. The dog does not respond and there is not trained odor in the search area (correct).
  3. The dog responds “positive” that there is trained odor in the search area but there is not trained odor in the search area (false).
  4. The dog does not respond and there is trained odor in the search area (false).

Please ponder these outcomes for little while. The answers to many of the issues we will be discussing in this article are directly related to these four answers and the behavioral training and learning that took place for the canine team to get where it is.

Demystifying detector dog techniques.I realize this is an over simplification of a canines learned/trained behaviors, search variables, odor recognition/discrimination capabilities and most important the proper application of scientific principles of learning.  By starting here we begin our journey and begin to examine how we get (or have already gotten) into these training issues, mysteries and myths. Whether or not you use and understand the basic scientific principles of learning and behavior, you are using them every time you train.  Dr. Robert Bailey, a world-renowned animal behaviorist, trainer and scientist often says “You don’t get what you want you get what you reinforce!”  This means that somewhere in the training program the trainer and/or the handler has rewarded the dog for behaviors that were incorrect and caused confusion as to the basic task at hand.  Which for us is to answer our question correctly- “Is there trained odor present in the search area or not?”  Also remember, “You can’t see what you’re not looking for!”  So during training if the trainer/handler is not constantly aware of what is being taught, how it’s being taught and most importantly how it’s being received by the canine student there will be errors that will surface sooner or later in the form of some other problem, issue or begin false responses/NPR.

Though I will mention certain scientific principles and laws of learning throughout the article with brief explanations, I would be remiss if we did not discuss some of the very basic scientific principles (terminology) that apply directly to our present day training.  I will cover these very briefly and these are in no way, shape or form covered adequately but will suffice for the fundamental understanding of this article.

Pavlov’s dogs- Almost every dog trainer knows the story of Ivan Pavlov and his salivating dogs and the discovery of classical conditioning.  All beginning detector dog training I have ever seen begins with classical conditioning.  We take something that means nothing to the dog and pair it with something that means everything to the dog. So we pair the odor of C4 with a tennis ball or we pair food with the odor of marijuana. The process of this pairing is classical conditioning. As training progresses we quickly roll into a combination of classical and operant conditioning. And throughout training will be using combinations of both classical and operant conditioning to get to our final trained detector dog. Unfortunately there are many bumps along the way, and almost all of them will have something to do with the unknown or misapplication of these scientific principles.

The ABCs of training – The basics of operant conditioning revolve around three events the antecedent, the behavior and the consequence. The antecedent or “cue” is twofold for our discussions the primary “cue” for us in training is the presence of the trained odor. Depending on your type of training though it may be the command to begin the search behavior.  The behavior we are looking for is a chain of behaviors, meaning simply that the dog is required to do more than one thing before it can be rewarded. Upon hearing the command to search (search, seek, find it etc.) the dog is expected to begin to search for the trained/target odor, (if it’s available) locate it and then trace it to its source (if no trained/target odor is present, no alert or response should occur). The consequence is that the dog is rewarded. Just as in all training the ABCs evolve from very simple in early training too much more complex as in actual deployments. But just keep in mind the consequence drives behavior. So if a consequence is pleasurable the behavior will increase, if the consequence is un-pleasurable the behavior will decrease.

Training fundamentals- There are numerous factors that go into a well-trained reliable detection canine, but the basic factors of learning, retention; performance and reliability do not change and are based on three principles. These principles are timing, criteria and the rate of reinforcement.  Let’s look at each one more closely.

Timing- The majority of mistakes performed early in training are directly related to timing. All animals including dogs are masters at tying in behavior and consequences. So the presentation of the trained odor and its immediate tie-in with a high value reward (value to the dog) is vital to success early in training. Timing throughout training never diminishes in value. A poorly thrown ball when the dog is getting up or leaving source rewards getting up and leaving source (you get what you reward not what you want). Problems revolving around timing of the reward are often hard to spot because the person delivering the reward is usually consistently late or consistently early. Ask anyone who’s trained a chicken how critical timing is when you’re getting behaviors!

Criteria- What are you looking for? What behavior are you going to pay, ignore or maybe even punish? If your criteria is inconsistent it is all but impossible for the canine to perform consistently.  This is because the canine is confused as to exactly what is needed or being asked of it in order to obtain its reward. You should be able to articulate your criteria or a training session in one simple sentence, if you cannot, you are asking for too much.  A well written training plan always comes in handy. It tracks your progress, your trouble spots and is an excellent reference manual for your next dog.  Remember if you don’t know exactly what you’re looking for how is the dog supposed to figure it out.

Rate of reinforcement- In early training such as the trained odor association training the dog is normally on a 100% reward ratio, simply meaning every correct response is rewarded every time. For most detection dog trainers this does not vary too often throughout the career of the dog. In some instances handler or trainer may use a “praise-off” by physically praising the dog off of the trained odor source without its being given its primary reward.  What normally happens in detection dog “field acclimation” training is that the workload increases exponentially while the rate of reinforcement drops like a stone. This is where we begin to cause confusion and frustration in the dog.  If work ratios and rates of reinforcement are not carefully calculated for every individual dog, each one will react slightly different. Some begin to shut down and are said to be lazy or stupid, some begin to false respond/NPR which is normally called a time and distance issue and others will react in a combination of ways. The root cause is that the rate of reinforcement has dropped too quickly or the amount of the expected workload.  There is an entire branch of science called Behavioral Economics.  In a nutshell, it studies how much behavioral effort an organism will put out to obtain a certain reward or to avoid a certain punisher. Many times when I have seen the mysterious time and distance issue, it had nothing to do with either time or distance! It had to do with increasing the workload and reducing the rate of reinforcement to quickly on that particular dog.  Just remember the example that if your garage was full to the ceiling with debris garbage and goods and you offered your son five dollars to clean it, it would probably never get done. However, if you offered your son $100.00 that might be worth the time, effort and energy to get the job done. Every human and every animal makes up their minds if the consequences are worth the behavior if it’s not they won’t do it (remember- Consequences drive behavior!). This is where many trainers start adding punishment, so rather than simply increasing the rewards and/or increase the workload more slowly we now begin to correct the dog into what looks like extended search to the trainer/handler. This new problem now creates a whole new set of dynamics that I have normally seen when dogs begin to walk odor. The reason they walk odor is that they’re going through the physical mechanics that appeared to be searching behavior that they are not. Dogs perform the behavior that they believe will stop the corrections.

Let’s look at a quick breakdown of detector dog training and some of the scientific principles that apply. It is irrelevant whether or not you agree with the order or steps; the idea is to understand the action with the science because in the end that is how we will solve the training problems.  Most detection dog programs start with a recipe similar to this: select the dog, do some form of trained odor association (classical and operant conditioning), trained final response training (TFR), trained odor search sequence (classical and operant conditioning) and then a gradual progression in difficulty into different environments (classical and operant conditioning).  Upon completion of this basic course there is some type of certification examination and/or “field acclimation” or both before the team is deemed to be street worthy.

In part two we will look deeper into each of these steps, dissect some misunderstood canine behaviors and look at some legal ramifications.

Authored by: Don Blair

Photo credit: The U.S. Army / Foter / CC BY and isafmedia / Foter / CC BY

Tags

 

2 Comments

  1. Edward Meyers says:

    Excellent article. I think a lot of trainers forget, or just don’t know, that training is simply the practical application of classical and operant conditioning theory. The best trainers I have trained with and learned from are the ones who had a thorough understanding of the two. I have also had the misfortune of training with individuals who couldn’t give even a rudimentary summary of either.

  2. Roger Lautt says:

    Thanks for reminding me that Pavlov is always “sitting my my shoulder”.
    Nice explanation of the Classical and Operant, great reminders about consequences driving behavior. Looking forward to your next article.

Leave a Comment