Stephan G. Kohlmann
In the last issue, I discussed a few basic principles of how dogs and people can start to communicate by using the same “language”. When training our dogs, it is critical to use a set of commands (signals) and stimuli to make the dog associate a certain signal with a reward. Rewards can be either positive (food, praise) or negative (pain). Positive rewards are usually to reward the dog for a desirable action or behavior, while negative stimuli are applied mot efficiently when we want to discourage a behavior of the dog that may be highly rewarding for the dog, but is undesirable to us (chasing deer, eating garbage, etc.). All this is called conditioning, because after many repetitions the dog starts to associate the signal with the following reward (or negative reinforcement) and consequently engages only in those activities and actions that bring him the highest reward possible. Dogs, similar to us humans are driven to maximize the benefits they gain from their activity. Naturally, our definition of “good doggie”-behavior deviates from those inherited traits and behaviors that allow a dog to find food and a mate in nature. Thus, when we want to reduce the rewards a dog receives from a certain behavior (e.g. chasing ) we must apply a stimulus (pain) to override the reward experienced by the dog. In other words, we compete with the “kick” our dogs are getting out of being “uncivilized”.
Torn between two feelings…
To be effective in this game of competing stimuli, we must apply the stimulus in the right moment and at the right level. Too early or too late can severely affect our training progress, because the dog will try to relate the experience of the negative (or positive) reinforcement with whatever cue he receive at the moment the stimulus was applied. In other words, the dog will think about “what caused this sharp pain?” and if he concludes wrongly because we didn’t superimpose the reinforcement over the exact stimulus we must compete with, the dog will not “learn”. For example, if Fido takes off after a deer in the woods, and all my whistling and shouting doesn’t stop him it is best to forget about this very quickly. If Fido returns and is greeted by an irate master, strap in hand to receive his “negative reinforcement”, he won’t be able to relate the pain to the fun he had chasing Bambi. He’ll get hand shy and will avoid coming home, but that’s all he’ll learn. On the other side, if I slap Fido hard for not backing off the neighbors cat, and Fido actually didn’t see the cat at all, but was looking at the neighbors kid throwing a ball, he’ll become apprehensive the next time he sees a ball, but won’t have learned a thing about cats. Timing is everything!
Good timing is good training
Having established that “good timing is good training”, we should keep in mind that no stimulus is better than one in the wrong moment. If you can’t apply the stimulus in the right moment where it actually overrides the rewards of the undesired behavior, it is better not to even dwell on that situation. Your chance has passed, and rather than making things worse, wait for the next chance and then be prepared. It is also important to realize that in most cases we want the dog to heed a command (signal) rather than just avoid a certain behavior. Thus, it is important to give the signal only when we have the means to apply a correction at the right time and in the right intensity. Old drill sergeants know not to give a command they can’t or are not willing to enforce. All it takes is one or two repetitions of such situations, and the recruits lose their respect of the grizzled veteran. Similarly, Fido will know you can’t reach him, and consequently won’t listen. The result: a dog that is obedient under certain conditions (i.e. when on a leash, or close to the handler) but he is not obedient unconditionally. These dogs will never make reliable hunting dogs, since they will always challenge the handler. A dog that constantly says “try me” isn’t trained, it is ruined and an unmistakable sign of an incapable or unconsistent trainer.
Predictability
It is easy to see that the two most serious problems a trainer may have is (a) to determine the right moment, and (b) to apply the stimulus in the right moment at the right level. Problem (a) will resolve itself by spending time with your dog and actually learning to predict its next move. Here is a simple test of how well you can read your dog’s body language:
1. Can you predict with high certainty from the body posture of your dog when he is about 10-20 seconds from relieving himself?
2. Can you tell what scent (bird, mammal, people, etc) your dog has in its nose, without actually seeing the animal?
If you answered “yes” to both questions, you are well on your way to have overcome problem (a). You are in tune with your dog, and you read him like a book. Incidentally, the same applies to your dog, he too must be able to predict your reactions and read your mood from your body language. Most dogs are better in this than we are, but to avoid time consuming confusion, we as trainers must be entirely predictable. Only if I react in an identical manner every time my dog behaves a certain way, I can hope that he will start associating the right clue with the desired behavior.
Problem (b) is more challenging, because it is largely a techical probleem. Our arms are too short to reach Fido everywhere and under all situations, and if we have to rely on running Fido down or waiting til we have control again, we violate the principle of “Good timing is good training”. In other words, we’re not training, we just intermittently beat our dog, and in most cases he won’t have a clue why.
Whips and chains
Over the years, dog trainers have come up with an amazing collection of gadgets and tools to extend our reach and to apply a stimulus from afar. Since our dogs work often quite far from the handler, we need a long-range connection with Fido that will not break down. Once Fido is trained, that connection is called obedience. But before we reach that stage, we must apply a variety of tools. Most importantly, there is the leash and the check cord. Both are indispensible training aid when it comes to maintaining control over a a young dog. No matter what we teach our dogs, we should always start out on a leash or check cord. Only when Fido has mastered the subject without needing reprimands or stimuli while on the leash can we work at the next level of “freedom”. If your dog can’t walk at heel on-leash, you shouldn’t try to teach anything at all (other than heeling!).
Few training aids are an aid at all, most are either to cumbersome to use, too ineffective, or too inhumane. When using force we must keep in mind that only enough discomfort is inflicted to override the rewards of disobedience. If we continually apply force too strong, the dog will become confused and will panic. Pain and fear locks up brains in humans and in dogs! Although we have a magnitude of gray matter more between our ears, we can’t think straight if we hurt somewhere, so how’s a dog supposed to learn when we apply pain galore? The key to fast, happy and successful training is in gentle application of discomfort, not causing bodily harm.
In the old days, trainers used throw chains, sling shots, and ultimately their shotgun. They even designed special leather skirts to protect the dogs vital areas while they peppered the dogs rump with bird shot. Aside from the possible negative effects of using a gun on our dogs (gun shyness!) and the public outcry that would result should anyone watch such practices, the gun is also not very practical, since it can only be used “safely” at a distance from ca 40-60 yards, and it delivers one incredibly harsh level of intensity only. Harsh treatment of a dog beyond the necessary minimum level is a sign of a stupid and insensitive trainer! Pain closes minds and slows training.
The electric collar
A tool often abused and seldomly used right is the electic collar. If the trainer has little self control or no appreciation of the canine learning, the use of electric collars can amount to incomprehensible animal abuse. But if we keep in mind that application of stimuli in the right time at the right level is critical for successful training, we also must concede that no other tool allows that level of accuracy and flexibility in the application of discomfort. The emphasis is on “discomfort”, not pain! The modern shock collar has at least several levels of instantly modifiable intensity — just enough to override different rewards for different undesired behaviors. If you are interested in buying a shock collar, don’t even look at a model where you can’t instantly adjust the level of electricity without changing something on the collar (i.e. plugs). Next in importance is the availability of continuous stimulation. Since we want to use as little pain as possible, and want to reward the dog for his obedience by “absence of discomfort” or a “black and white” situation, we need a continuous mode of stimulation. Dogs learn best when “good” and “bad” follow each other closely. They are very bad in distinguishing among “shades of grey”! For example, if I want to reinforce the “here” command in a free ranging dog (which, of course, has learned following the whistle signal first on the long leash), I set my collar to a low level, where the dog won’t cry or whine. Next I give the signal, and if Fido decides to ignore it, I give him a gentle tickle with the collar. I keep this gentle stimulation up until Fido gets “fed up” with this tickling and turns to look at me. Immediately I let go of the button and praise the dog for turning. Only a few situation like that and Fido will turn around quickly to “beat” that mosquito on his neck. Obviously, if Fido saw a kitty and wanted to give chase, I wouldn’t accomplish much with a gentle tickle, so up goes the “juice” until I match my correction with Fido’s level of excitement — naturally only after I gave a command! All I need to do is reduce the fun of chasing enough that Fido would rather do something else, — which is comply with my command. The key to that is the continuous stimulation, which can be applied as long as Fido’s undesirable behavior continues. Once he stops, the discomfort goes away immediately. Good timing is good training!
Of course, since I don’t try to intimidate Fido, I don’t have to holler and cuss. I simply keep my finger on the button and my eye on Fido, ready to let go of the stimulation as soon as he complies with my earlier signal. And I have the praise on the tip of my tongue! Only a few occasions, and Fido will obey my commands willingly amd happy, because he has learned to be in control of the collar and the stimulation. He is confident and willing, because he knows exactly how to turn the thing off — by doing what we want him to do. Now we speak the same language, our goals are the same. Incommunicado? No more!
As an example of how easy, relaxed and efficient this training is, compared to the old ways of whips and chains, let me compare my experience using both methods. When training the “Halt” command, I used to work with the dog for weeks at heel on a leash, whip behind my back to teach them to collapse instantly on whistle blast. The whip helped to speed up the “deflation” of the dog. Weeks and weeks passed, and my dogs grew weary and tired of this unhappy routine. They complied because they feared the pain and because the “halt” became a reflex they had no longer mental control over. They learned it, allright, but it was a part of the program I always dreaded, because it was such an unhappy time. A few years back I bought my first electic collar, partly to correct an extremely hardmouthed dog from crunching ducks in the water. More advanced models (of collars and dogs!) followed, and my training has never been the same. Now it takes me about a week to teach my dogs to lie down in the desired position (head between both front paws), which is accomplished by using the leash and gentle and firm pushing down. Once Fido has learned that he isn’t supposed to lift his head once commanded “Halt”, I simply turn the collar upside down on his neck, so that the electrodes are sitting on the top of the neck, rather than on the throat. A couple more days on the leash, now with the collar doing the “gentle pushing” and me helping out should he not assume the correct position, and we’re ready to go solo. Now I just give the “Halt” command, and apply the stimulus to the back of the dog’s neck until his head is firmly pressed to the ground. In that moment I praise and make him feel good, all the time ready to turn on the lights again should his head come up only an inch. The combination of my praise and the collar pushing Fido down creates the necassary contrast between discomfort and “feel good” of the correct Halt position, which the dog starts to like. All is friendly and calm, no hollering, dogs squealing or trying to escape. Within a week, I have my pup running full throttle across the field, and upon whistle blast throwing himself into the dirt like someone shot his legs out from underneath him. There he is, flat as a pancake in a big cloud of dust from the deep skidmarks he left in the soil, an the best of it all is: Fido is wagging his tail! This is happy, confident learning and at the same time humane and effective training. I challenge all “traditionalists” to show me a similar obedience with traditional training by punishment. I think it can’t be done, or only with incredible patience, time and effort. Needless to say, I’m rather training a dog in a friendly and happy learning atmosphere than having to beat the unfortunate creature into submission. Again, the secret is the good timing of applying (and letting off!) the stimulation in right moment, and the right level of stimulation that matches my dog’s level of excitement of desire to follow alternative rewards. Such a training program creates trust, confidence and pride in both handler and dogs.
A word of caution at last. If you start using the collar, it is absolutely indipensible to keep the collar on the dog at all times until your dog is trained. Dogs are very good at figuring out that the collar does the “biting”. Collar-smart dogs are very hard to cure with the collar, and usually require going back to old, less humane methods. If you use the collar you are “married” to it until ol’ Fido is fully trained. My young dogs go on the collar at about 8 months and I take it off the day they pass the VGP. Then I’ll never use it again! To prevent any degree of collar-smartness I make it a rule that none of my young dogs ever leaves the yard without a collar on their neck, whether it is actually turned on doesn’t matter so much. Of course, I’d rather have the collar on, even if I don’t plan on using it, should the situation require it (good timing….). Now when I walk to the kennel, collar in hand, my youngsters are jumping out of their skin — they know we’re going afield. They even push each other trying to get their neck first into the collar! With the advent of the modern electric collar, dog training has finally reached a level of humane interaction and communcation between man and dog. Now it is up to us not to let our best friends down again, by being “incommunicado”.
Comments