With the vast training styles available in the world, it has become a common saying that the only one concept any two dog trainers can agree on is what the third one is doing wrong.
Even so, regardless of the training techniques used by today’s trainers, most have some degree of success. Even in the last twenty years the concepts of police dog training have seen a vast variety of techniques used. Some have gained popularity while others have lost their luster. In some cases, the techniques seem to cycle like many popular fads.
Training, in its simplest form, is the application of pain or pleasure to a given behavior in order to obtain the desired results. Over the years however, we have had a tendency to lean towards the use of compulsion, pain compliance and negative reinforcement, applying the human thought process to the application of canine training techniques.
One of the greatest failures of law enforcement trainers is our resistance to ideas from someone outside the law enforcement circle who just may have some techniques that are better than ours.
For example, why do so many trainers continue to use compulsive-based methods of training when motivational techniques seem to provide so much more stability and better long-term success in a finished animal? My entire foundation of training, when I started training dogs, was fundamentally compulsive, however over the years I have watched dogs trained in different sports and for different job-related applications and have learned that there are always better concepts to training if I am willing to be open minded and learn about them.
One example of a training style that I enjoy watching is that of agility dog competitors. If you have ever had an opportunity to watch these dogs compete, then you know the talent that is involved in getting these dogs to perform. Much of what they do is similar to the K9 agility courses police dogs work on, with much more added to it. The dogs run up and down ramps, learn to jump at specific locations off the ramps, weave in and out of poles, run through tunnels and jump over various obstacles. Points are assessed based on the best speed, with time added for any mistakes during the run. At what point during this program would compulsive training be applicable?
Another example of motivational or operant conditioning training techniques comes from the methods used by ocean mammal trainers. Virtually everyone at some point in his or her life has seen the performances of trained dolphins or whales at an aquarium attraction. Consider for a moment how those trainers develop an animal such as a killer whale to the point where the animal has a desire to perform specific actions on request. I think I can pretty safely say that none of them use compulsive methods of training, yet their performances are flawless.
The training styles mentioned above are dependent upon motivational training techniques. These techniques have been created as a direct result of the study of the animal behavior of the species on which they are working.
This is not to say that compulsive methods of training do not work, or are inapplicable in dog training. On the contrary, many compulsive techniques are used with varying degrees of success. Frequently a compulsive technique will solve a training issue very definitively and as such it still has its place, even in my own training regimens. However, the degree to which I use these methods, and the confidence I place in them have changed dramatically over the years. Compulsive training techniques are based on pain induced compliance. From a human thought process, this is effective and obtains the desired result. At least, this is what many trainers espouse.
However, if we look at the same animal and try to obtain the same desired result using motivational methods, experience teaches us that the results obtained are longer lasting with much less recidivism. The dog performs the desired actions with a positive attitude, and actually enjoys the process.
Why is there so much difference in success rates between the two methods? There are a number of reasons, but the most prominent explanation is simply that motivational techniques require the trainer to understand and apply concepts of animal behavior. In other words he or she understands how the canine mind functions and understands how to communicate with and shape the animals behavior. A good dog trainer understands fundamental canine behavior and is able to create desired results by communicating with the animal using techniques the animal is able to understand, rather than trying to get the dog to conform to human behavior.
Compulsive Training Techniques
Compulsive techniques can work to a degree in dog training and in fact, most of your training can be completed using compulsive methods. However the entire process is done using pain compliance or pain avoidance and the levels of compliance can be varied depending on each individual animal.
Take for example a simple heeling exercise and three levels of dogs. For this example we are going to assume that each dog we are working with has had no previous training. The first dog we work with is a dog that for the most part we would consider a “soft” dog, or a dog that is handler sensitive. We put a pinch collar or a choke chain on the dog and start to work a heeling exercise. As we proceed to move forward and the dog has its first correction, the likely reaction you will get is probably going to be one of confusion, maybe even of protest. A really weak dog may even go into a panic. Our compulsive method must adjust to a level where we don’t end up ruining the dog or damaging him to a point where it creates a social problem with him. On some very soft dogs, this method of training would never be appropriate.
Now let’s consider using the same technique on a good quality, levelheaded dog that is well suited for police work. He has passed all the basic requirements for testing and is now going into his first obedience session. Applying the same techniques as we applied on the “soft” dog, this dog is likely going to adjust quickly and perform very well, understanding and complying with the corrections given. Having said this, even a well-adjusted dog will have some negative reactions to some of the corrections and will at times show a small amount of “softness” to the handler if the handler is required to perform a strong correction. The dog will be somewhat handler-sensitive at least for a moment subsequent to the correction, until the exercise continues to a point where the correction is forgotten. At this point the dog starts to come around and enjoy the exercise again, particularly if the appropriate verbal praise is being applied during the exercise. We then work the exercise to the point where the dog is provided an opportunity to repeat the previous mistake so that another correction can be administered. Again, through a pain stimulus, the dog complies and the exercise continues.
It is important here to note one basic fundamental concept regardless of the training style you use; the dog performs to satisfy himself. If a stimulus is strong enough to overcome the present activity the dog is performing, the canine instinct is to yield to the stronger stimulus. An example here would be in order. If during this heeling exercise a rabbit were to suddenly dart out in front of your dog, his instinct to pursue prey would likely be significant enough to cause him to break from the exercise. A soft dog would likely be so sensitive to being on line that he would not even consider breaking. Our more level-headed dog will likely attempt to break, but with a good sharp correction he will likely comply in order to prevent the correction again. A very strong willed dog with a high pain threshold will very likely ignore your attempts to correct him and may continue to try to break away, even though the level of correction, even in a pinch collar, is very high. Thus stated, the stimulus for what you want your dog to do must place a greater desire in him to do what you want him to do rather than for him to break away to do what comes instinctively.
Now let’s take our heeling exercise and apply it to a dog that is extremely “Alpha”, a dog that would be considered a “hard” dog. These dogs frequently have such intensity that they can tolerate many corrections designed to enforce pain compliance. Many agencies prefer this type of dog as it shows all the strong confidence traits “right out of the box”. Dogs that are at the higher end of this spectrum may have a tendency to challenge the handler if the correction is being administrated physically via a leash and collar. They are also more apt to challenge a handler who corrects them improperly. It is not unusual for a dog that is resistant or capable of tolerating high pain compliance methods to strike out and attack the handler during a training routine. Such a reaction can create a potentially dangerous precedent if the handler does not immediately overcome and defeat the aggressive assault by the dog. Should the dog win this situation in a decisive manner, there is potential for this undesirable behavior to continue and progressively worsen. This dog soon becomes a dangerous animal to attempt to control or work with. Well-meaning handlers/trainers can even create this scenario in a dog that would not normally exhibit this type of behavior if they are consistently heavy-handed or if they correct heavily during inappropriate moments when the dog does not understand the reason for the correction. It is important that if the dog begins to exhibit behavior of an aggressive nature towards the handler, or inappropriate, unsocial behavior, whoever is working the dog should seriously consider their training regimen, and even reconsider whether the dog is an appropriate candidate.
Now take those same three basic types of dogs and put them into a positive training routine using motivation as the main method of training. In all three cases we are going to work these dogs in an enclosed environment, without any collars or other equipment on the dog.
We start each dog the same way, using his or her favorite toy, even food. It doesn’t matter what is used to do the training, as long as whatever is used is their strongest motivator. For my training, I usually prefer a ball such as a Kong. I can hold the Kong in any position I like in order to move the dog into the appropriate position. When the dog sits appropriately I immediately fire him the ball and give him verbal praise as well. The exercise continues as we begin the heeling exercise, and when the dog is in the appropriate position the ball again is thrown directly to the dog as a reward and small playtime is taken.
Regardless of what type of dog we are working, there is no negative involved. From the soft dog to the hard dog, the complete exercises are entirely motivational. Because there is never any negative reinforcement, there is no conflict between handler and dog. All three dogs are enjoying the routines, and even the hard dog is learning that the reward comes if and when the handler desires to give it to him. There is no conflict; therefore any opportunity for a harmful interaction between dog and handler has been removed.
As you can see, in all three personality types, the dogs are going to finish on a positive note, enjoying the interaction with the handler. The soft dog is going to enjoy the exercise with little or no negative reaction. The moderate dog is going to perform just as well without any negative effects of handler softness that may be created by physical corrections, and the hard dog is also going to finish the exercise with a strong positive outcome. In all three cases, the dogs’ attitudes towards the training experience are going to be positive ones.
A major advantage to operant conditioning techniques is that the behaviors can be taught without ever putting a collar on the dog. The handler learns to interact and work with the dog in a manner that the dog understands. The dog has a desire to complete the exercise as directed.
Your dog will never perform any task incorrectly, or do anything wrong if never given the opportunity to make a mistake. This is where operant conditioning techniques shine. Compulsive techniques are based on correcting the dog for erroneous behavior, utilizing pain compliance only. However, if the dog gets some degree of satisfaction before the correction comes, he has learned that he can succeed through inappropriate behavior. Operant conditioning techniques provide the constant desire for the dog to perform correctly in order to receive a given reward, never allowing him to experience any level of satisfaction from an erroneous behavior.
Dogs trained using motivational techniques are easy to spot when watching them work and interact with their handlers. These are dogs that can be very intense, strong working dogs, however they are quite social and interact well with their handlers. In short, they enjoy the interactive relationship with their handlers. The end result at the end of a full course of training is a team that works well together and a dog that is stress-free and is obviously enjoying the work.
This motivational training style is also known as “operant conditioning”. In short, you are conditioning the animal in a manner where it will respond in a specific way to a stimulus or command provided. Once successfully performed, the dog receives a reward for the correct behavior. Thus you are shaping the behavior of the dog, conditioning him to respond to specific stimuli.
The use of electrical stimulation via remote collars works using the same concepts. The dog learns that he can turn off the stimulation when he is in the proper position or performing in an appropriate manner. If you apply the principles of learning concepts using a remote collar in the previous scenarios, you will find that you will have results similar to that of using positive motivation, even though you are in fact using negative reinforcement to obtain the desired results. In fact, you may even find that the responses are quicker, as the dog knows that at certain points if he does not perform an established behavior fast enough, then the electrical stimulation will occur. Many trainers have a tendency to push their dogs to the limit using this concept, making the dog crisper and quicker to react to direction given. The end result is a dog that looks sharp and performs very well. Two major advantages in the use of remote collars are that the dog does not relate the correction to the handler/trainer, as occurs with on lead corrections, and that the stimulation can be done at increasing distances from the trainer.
There are negative sides to the use of electrical stimulation. If you watch carefully you will find that dogs trained using remote collars will have a tendency to anticipate the commands. While there is the benefit of instant correction, dogs trained using this method can have a tendency to break and commit to a particular behavior in an attempt to beat the negative reinforcement. (i.e.) when setting up to do a bite work exercise you may see the dog literally vibrating in anticipation of the command, his stress level will be up and sometimes the dog will break and start the exercise before receiving the command to do so.
Many dogs trained in this manner will constantly be “on the edge” due to the stress and intensity that comes with the use of electrical stimulation. While the dog performs as directed, the relationship with the handler, and the desire to accomplish the task at hand is achieved by an entirely different type of motivation than that of a dog trained using primarily operant conditioning techniques. This is a dog that is working to avoid the stimulation. His entire life is based on pain avoidance. This system is very effective, with the exception of situations where a dog becomes resistant to the stimulation, at which point the trainer must re-evaluate his options and the workability of the dog.
It should also be noted that not every dog is capable of coping with or understanding what is desired of him with the concepts of electrical stimulation. While applicable to dogs properly selected for law enforcement purposes, the use of remote collars would not be applicable to all dogs.
Looking objectively at all three methods of training, it becomes very clear that the best all around method of training is one that shapes the dog’s behavior by using techniques which use the dog’s natural instincts and does so in a positive light.
The training techniques I now endorse are those that use motivational, operant conditioning techniques. Wherever possible in my training regimens, the use of positive reinforcement is applied with minimal compulsion or negative reinforcements applied.
- Concept is simple
- Easy to administer, handler only required to learn to correct inappropriate behavior.
- Basic principles can be taught in a short period of time
- Concept based on pain compliance
- Any administered correction increases stress in the dog
- Dog will have a tendency to test the waters at times, recidivism is common (i.e.: constant bite work correction)
- Dog knows the handler is the origin of the correction
- Handler sensitivity or conflict with the handler occurs
- Requires a mature dog or a dog with confidence, not good for starting younger dogs
- Concept based on canine instincts and behavior (approach is from the dogs point of view)
- Can be used on any level of dog, including younger candidates without any negative effect
- Once learned, recidivism is rare
- Minimal upkeep on learned concepts
- Learning is a constant positive
- Entirely stress-free for the dog
- Dog consistently enjoys the working environment
- Working relationship with the handler is always positive
- Process requires many repetitions and can be time consuming
- Remote capability
- Maximum control of the dog at all times
- Concept includes operant conditioning ideals based on canine behavior
- Use of collar takes the handler out of the picture as to where the corrections originate
- Training time is less than using strictly positive reinforcement
This is a concept bsed on pain avoidance, often abused or misused as a corrective device rather than as a training tool. As a result, the following applies:
- Increases stress in the animal
- Animal has a tendency to be “on edge”
- Dependency on the collar to overcome training inadequacies
- Requires a mature dog or a dog with confidence, not good for starting younger dogs
- Requires basics be taught using other techniques before the collar concepts can be applied.
Genetic and Environmental Development
It is important to understand that there are basically two types of behavior inherent in any species, genetic (animal is born with built-in survival instincts) and environmental (learned behaviors through life experiences).
This is important to us as dog trainers in that we must understand that we cannot control what behaviors are genetically created in the dog. We know that in order to obtain quality dogs capable of meeting basic courage, stress, stability and hunting requirements for law enforcement work, we must have strong genetics to start with. For example, we cannot create courage or hunting instinct in the animal. We can however manipulate and build on these behaviors if they are there. The only way to control the genetic qualities of the stock we work with is through a selective breeding process using animals that exhibit the strongest genetic traits we require for police work.
We do however have control over environmentally created or learned behaviors. Just because we have a genetically sound dog that tests well in the initial selection process doesn’t mean the dog will work out.
We can make or break the dog by how well we develop him through the various stages of his life. Negative experiences can bring on learned behavior that will have long-term effects to the point where the dog will never overcome that experience. While initially he may have had the genetics to deal with most situations, he may have become environmentally unsound through improper upbringing, abuse or accidental incidents that caused him insecurity, pain or injury. These are developmental behaviors that we refer to as environmental. We have control over the animal’s environment and his learned behavior is a direct result of his lifelong experiences within his environment.
Therefore, in order for us to obtain the best possible candidates for law enforcement applications, we must be determined to develop the most genetically sound stock, and raise that stock in a manner that produces a challenging, yet positive learning environment, constantly building on the various basic behavioral building blocks given us by his genetic imprint.
Neural Basis of Learning
The following principles may seem somewhat complicated and irrelevant, however if you understand how the brain functions and how an animal learns, you will have a superior understanding of how to communicate and work with your dog.
For us to obtain maximum results from our training it is important for us to understand how the dog thinks and how he learns from a canine perspective rather than from a human perspective. We must realize that the dog cannot understand higher reasoning concepts. Therefore, in order to be good handlers and/or trainers, it is imperative for us to understand how the dogs mind works, how he thinks and learns. The concepts are similar in all creatures with a developed brain, each species having its own limitations. In other words, we will first learn the concept of learning, then determine at what level the canine species is capable of learning. We then apply motivational principles that shape the dogs behavior in a way that gives the animal the desire to perform a specific action based on a specific situation and/or command given by the trainer.
Learning concepts hold true for all creatures with a developed brain. Behavior is a response to a given stimulus, producing a subsequent instinctive reaction (inherently genetic in the animal when it is born) or a reaction based on the learned behavior (environmentally-learned behavior) of the animal.
Lesser species rely more on instinct than learned behavior and higher species such as human beings have the ability to rely on both instinct as well as learned behavior involving cognitive reasoning.
The neural basis of learning is a process of cell interaction. Every time we learn something, we activate billions of brain cells that interact with one another. Whenever we complete the same exercise repetitively, generally speaking, the same cells in the brain are used to communicate what action needs to take place. A simple example would be in order here. When you are first learning to drive a car as a young person you were taught the concept of signaling your intentions before turning or making a lane change. In the beginning it was a conscious thought process, however as time progressed and your driving experience continued, you found yourself signaling without thinking.
The same process occurs in firearms training. We train using repetitive actions in order to learn how to do things quickly and without thinking. The purpose behind this is so that we will react instantly to a threat without having to go through a conscious thought process. Virtually anyone reading this book will have heard the term “muscle memory” as it refers to firearms training.
In reality, it is not our muscles that has retained that memory, but a series of brain cells that have been exercised through conscious repetition so many times that the action becomes autonomic.
How does this occur? Simply stated, cells interact with each other using a chemical to close the connection and communicate, referred to as a chemical synapse. Think of each connection as a “bridge” between two cells. This “bridge” is called an axon and is coated with a myelin chemical sheath. Each time those two cells communicate to perform a function a chemical is deposited on that “bridge”, continually building up the amount of chemical between the communicating cells. Thus, the frequency in which those cells are used increases the amount of chemical that is bridging that synapse. The thicker the chemical connection, the less conscious thought is required to accomplish the task, thus creating an automatic response without conscious thought. Each task we learn requires billions of these connections to repeatedly occur before the resulting response becomes automatic and without conscious thought.
In scientific terms, a chemical synapse coats the myelin sheath, which over time builds a coating on the axon so responses are autonomic and without conscious thought.
If we stop training for an extended period of time, our skills become substandard because the chemical synapse on the applicable axons begins to break down. Without constant updated training, we lose our skills, as the chemical coatings are not being renewed. However, skills that are well ingrained where the myelin sheath is extensively coated may take longer to lose and may be easier to recover with limited repetition.
This learning process continues through our lifetime and our skill levels are entirely dependent upon the amount of time and repetitions we are willing to perform in order to achieve those levels.
Applying the Learning Concepts to Training
Understanding how the brain functions for the learning process to develop makes it much easier to understand how the dog learns. Take away the human ability of cognitive reasoning and apply the neural learning concepts as stated to the limitations of the canine species and it helps us to comprehend not only our training limitations but also our training potential.
Consider the following rules when preparing your individual training goals:
Patience is paramount above all else.
- Timing is everything. Any corrections given, as well as praise and reward must be perfectly timed, appropriate, and sufficient to deal with the circumstances.
- Consistency is another primary ingredient to success. Do not change style or technique just because your first attempt fails. A dog learns best from consistent technique.
- If compulsion or electrical stimulation is used to teach a concept, it is vitally important that the dog be verbally praised or otherwise immediately rewarded for proper behavior.
- If a correction is given for inappropriate behavior, it must be sufficient to suppress that behavior once and for all. To do only a minimum correction and continue having to correct the dog for the same inappropriate behavior is in and of itself abusive. Better a single definitively unforgiving correction given only once, than a less forgiving correction given numerous times, or on a continual basis.
- You must work on the level of the dogs learning aptitude. The dog must understand what is expected of him. Do not expect him to understand human reasoning.
- Before doing any training exercise, ensure you have defined the outcome you desire, then design the exercise to ensure that outcome is reached. The exercise must be run in a manner that ensures success each and every time.
- Never allow the dog to make a mistake. If a mistake occurs and the dog wins his own objective, he has learned that he can achieve results using his own means.
- Repetition is fundamental to proper learning development.
- Keep exercises brief and upbeat. Exercises of repetition teach the dog proper behavior. Exercises that are short and repetitive will teach the skill.
- Extended exercises should only be attempted once repetitions have taught the skills flawlessly. Longer duration exercises are designed to build stamina.
- Use positive motivation whenever and wherever possible.
- Keep the use of compulsion or the use of electrical stimulation to a minimum.
- Use compulsion and/or electrical stimulation very judiciously followed with abundant praise when the task is completed properly.
- Break training profiles down into small stages; teach each stage slowly, gaining fluidity with the routine as the dog becomes familiar with each stage.
- Once each stage is flawless, put them together to complete that profile from start to finish.
- Combine the profiles to complete dynamic training exercises by training at speed. (Live scenario training)