User:Lindsay Ridgeway/The Case for Dog Training in the High School Curriculum

= THE CASE FOR DOG TRAINING IN THE HIGH SCHOOL CURRICULUM =

Abstract
Because dog training is at heart the study and practice of behavioral science, training dogs can materially enhance the trainer's understanding, not only of dogs, but of how humans interact at every level, from daily life to global affairs. So fundamental and sweeping is the field of behavioral science that the world would be transformed in a single generation if every high school student were required, by a mandatory semester or more of dog training, to gain that kind of practical experience and share in that body of knowledge.

Introduction
Dog training is one of America's niche avocations, with only a few thousand enthusiasts. Many are intuitive trainers, who learn from experience and from other trainers what seems to work and what doesn't. Others are serious students of behavioral science who strive to be highly disciplined in its practical applications to dog training. For both groups, what they learn about training dogs can enhance their understanding of parts of their lives that have nothing to do with dogs.

I'm a dog trainer. Though it is only a hobby for me, after I became interested in the subject in 2004, I soon began to see the world in a new way. I would see someone, possibly a stranger in a public place, having a conflict with a child or a friend, and I'd see the parallel of that problem in dog training, which in turn would suggest insights I might not have had in the past. Or I would hear a news report and wonder that no one involved in shaping such important events seemed to be aware of principles that are commonplace in the dog trainer's world.

Now, after four years training dogs, I've come to feel that a great deal of harm occurs unnecessarily in this world because so few people understand some of the principles that for many dog trainers have become second nature. And I've thought that if every high school student took a semester or two, or even more, of dog training as one of his required courses, along with math and English and current events, the world would change dramatically for the better in only a few short years, as all those students went on to take their places in life and apply, day in and day out, the knowledge of behavior that had become woven into their very fabric.

I don't attempt to solve the practicalities of such a program in this paper. I freely stipulate that the barriers would be enormous, and impossible to overcome without serious commitment. Who's going to pay for the dog food and the vaccinations? Who gets sued if someone gets bitten? Where do the dogs live, both during and after the school year?

But practicalities are not the first issue. The earlier questions are, what would be the benefits of such a program, and would those benefits be so immense as to make overcoming those barriers a worthwhile endeavor?

In this paper, I try to outline some of the dog training principles with which I've become familiar, and to show how those principles might apply to everyday living as well as the shaping of world events. To me, the soaring scope of those applications makes a compelling case for every student in America to learn those lessons, as practical experience in training a dog can teach them far better than any book.

Behavior and Outcome
As you'll see in later sections, a dog trainer juggles many principles. But the holy grail of dog training is the principle that a dog's behavior is largely shaped by outcome. The name of this principle is operant conditioning, and it's generally represented as four quadrants: two outcomes (based on pleasant or unpleasant stimuli) that result in behavior increasing, and two outcomes (again, based on pleasant or unpleasant stimuli) that result in behavior decreasing. Here are names and descriptions for each:


 * Positive reinforcement (+R or R+) A pleasant outcome increases behavior.  The dog sits on cue and receives a treat.  The next time she hears "sit", she's more likely to sit again.
 * Negative reinforcement (-R or R-) Escape or avoidance from an unpleasant outcome increases behavior.  Initially, the dog ignores "sit" and is whacked with a stick until the dog sits (escape).  The next time she hears "sit", she's more likely to sit before she gets whacked (avoidance) and she is not whacked, which she likes better than being whacked.  Thereafter, she's even more likely to sit when she hears the cue.  Both the escape and avoidance outcomes shaped her behavior.
 * Positive punishment (+P or P+) An unpleasant outcome decreases behavior.  The dog tries to pick up a burning ember and feels pain.  Thereafter, the dog is less likely to repeat the behavior.
 * Negative punishment (-P or P-) An outcome of losing something pleasant, or of losing the opportunity to obtain something pleasant, decreases behavior.  The dog barks as someone arrives and is put in her crate, depriving her of freedom and removing the opportunity to play with the new arrival.  If she perceives the loss of freedom as undesirable, she is less likely to bark the next time someone arrives.  If she perceives the loss of opportunity to play with someone as undesirable, she is also less likely to bark the next time.  For some dogs, both outcomes may be undesirable and the effect is doubled.

Although by no means all of a dog's behavior is determined by those principles, dog trainers constantly see the principles accurately describing and predicting how their dogs will respond to a training procedure. The result of all those observations is that the principles become deeply ingrained in the trainer's view of dog behavior, and are soon seen as predominant forces in human behavior as well. He then becomes mystified by the countless obvious missteps he sees taken by others who haven't formed the habit of analyzing the effect of outcome on behavior. A student who had learned to habitually look for correlations of behavior and outcome in high school could avoid those missteps.

Management
By management, dog trainers mean using tools that prevents unwanted behavior, but that do not use classical or operant conditioning to modify the dog's behavior. An example would be a fence.

It turns out that management is an excellent training tool, because while management tools do not allow the behavior, and therefore no outcome is available for the dog to learn from as in the OC model, habits are nonetheless formed and learning does occur. An example is using a slip cord to prevent a dog in field competition from breaking toward a mark (a thrown retrieval object) before the handler sends the dog. If the handler has placed a slip cord through the dog's collar and the dog tries to break, the handler can hold on tight and the dog won't succeed in breaking. Over time, the dog stops even trying to break, because she learns she won't succeed. The habit becomes so strong that, even in competition, where no slip cord is allowed, the dog usually still doesn't break because the habit of not attempting a break has been formed.

But trainers learn something else, too. If you try to depend entirely on management, the chances are good that your attempt to manage will occasionally fail. Someday, someone will leave the gate open; someday, the collar will break; someday, no one will be watching the dog when Grandma walks in the front door.

Therefore, the trainer doesn't depend entirely on management if he can help it. He makes the confined area attractive to the dog; he trains the dog not to pull on the leash, and to sit or lie down or come back on cue; he teaches the dog not to jump on people even if no one is there to stop the dog or call her off.

One more point. Although the dog gets in the habit of not attempting a behavior, that habit may well be temporary. In the movie Jurassic Park, the T-Rex discovers within hours that the fence is no longer electrified. As a general rule all organisms continue to test boundaries, and if the management tool is no longer in place, after some interval, a behavior previously suppressed spontaneously begins to recur.

Competing Influences
One of the central facts of life when you're training a dog is that you can't ask the dog why she exhibited a particular behavior. You often wish you could, but on the other hand, you don't waste time with incorrect explanations for behavior. After all, you can ask a man why he exhibited a particular behavior, but does he always, or even often, know the real reason? Twin studies suggest that human behavior is far more pre-programmed than we sense when we think we're making choices.

If a high school student learned to train dogs, he'd learn that some of the influences on a dog's behavior in a particular context are:


 * Instinct
 * Classical conditioning
 * Operant conditioning
 * Establishing operations

It's also possible that dogs reason to a certain extent, and some trainers believe that quite strongly. Another widely held belief is that, in some cases, dogs consciously evaluate outcomes and choose the one they prefer.

Without going into detail about each of those principles, it's clear that having such a template for understanding behavior would put a person at an advantage in everyday events life and would put a leader at an advantage in shaping world events.

Of further interest is the fact that those influences on behavior are sometimes, in fact usually, in conflict. The dog's natural instincts tell it to do one thing, but life experiences tell it to do another. Or, when the dog is hungry, one behavior occurs, whereas when the dog has just eaten, she responds differently to identical stimuli. In addition, life experiences themselves are contradictory in their influence.

For example, a dog with an undesirable tendency to be aggressive toward other dogs is being walked on a leash, and as another dog comes within sight, the first dog begins to show aggression. The trainer responds by giving the dog a treat. From an operant conditioning point of view, the trainer has reinforced the dog's aggressive behavior, and that behavior should become more probable in the future. But from a classical conditioning point of view, the dog has learned that seeing another dog predicts receiving a treat, which suggests that the dog will be happier the next time she sees another dog and will be less likely to act aggressively. Which of those two forces will predominate?

The answer happens to be that, in most cases, classical conditioning trumps operant conditioning, and the procedure mentioned above will tend to reduce aggression, though it's probably not be the most efficient way to go about it.

But more important than the answer is the fact that dog trainers habitually think in these terms, and use that thinking to plan their own behavior and also to adjust their behavior when their observations conflict with what they expected. A student who had trained dogs in high school would grow into an adult who habitually monitored his behavior and then modified it according to objective evaluations of how his plans were working out.

Beyond that, the dog trainer also learns in due time that sometimes, the dog's instincts will reawaken a behavior pattern that cancels out the effects of training. The dog isn't behaving irrationally, she's doing what she was born to do in that context. Instead of being provoked by the situation, the trainer recognizes the nature of the challenge and simply rolls up his sleeves and renews his training of the desired response. In the same way, we want new parents to recognize that the natural behavior of their children may sometimes undo their parental influence, and our leaders to recognize that cultural forces may sometimes skewer their influence as politicians. The high school student who had faced such situations with the dog he trained in high school would be better prepared by that experience to deal with an analogous situation in the real world than an adult who had no idea what was thwarting his well-intended and well-planned influence.

Recognizing the Quadrant

 * Intent is irrelevant, behavior is only way to determine R vs. P: Is behavior increasing or decreasing?
 * Is outcome based on pleasant or unpleasant stimuli?
 * Superstitious behavior
 * Unintended shaping, such as viciousness (not giving into growling) and danger (emergency recall)

Reward-Based Training

 * Coercion and side effects
 * Coercion and trainer's feelings about it: -R is highly +R for trainer, and is generally more persistent than +R; but for some, such as those who see themselves in spiritual bond with dog, the end does not justify the means
 * Coercion and loss of temper
 * Training without coercion: the 2-quadrant dog
 * Tug
 * Premack
 * Self-reinforcement and discovery training

Markers

 * Bridges (secondary/conditioned reinforcers), NRMs, and warnings
 * Anything that predicts +R becomes +R, including subsequent cues
 * Clickers
 * Delayed reinforcement reinforces something else, and intended effect is lost

Training Procedures

 * Motivation vs. learning: motivation is temporary.
 * Motivation vs. drive: drive is inborn.
 * Reinforcement vs. luring: the reward is produced after the behavior is complete.
 * Establishing operations, such as hunger, promote motivation and may promote learning.
 * Shaping complex behaviors with successive approximations.
 * Errorless learning (cues and prompts).
 * In shaping, trainer and dog train one another to get what they want.
 * Videotaping ("greatest invention since the secondary reinforcer").
 * "If the training method you're using isn't working, change your behavior." Generalized to, "If what you're doing isn't working . . ."
 * Intermittent reinforcement produces persistence, but may also produce variability
 * Train without distractions, then introduce distractions and generalize for location
 * The no-glance rule to avoid associating frustration with trained behavior, which can reverse the effect of the training
 * Extinction includes extinction bursts and spontaneous recurrence
 * Short sessions, high rate of success and reinforcement (compare to typical education practices)
 * Building duration

Conclusion
Beyond the principles discussed in this paper are even more that always come up when a group of people train dogs: special considerations for the shy dog, the aggressive dog, the performance dog (the author's primary area of interest), and training several dogs simultaneously (the perhaps unfortunate natural trajectory of this hobby). To train dogs is to delve into a rich and many-faceted sea of ideas and practices. And much of what we learn about dog behavior carries over to the behavior of all organisms, especially other mammals, and even more especially mammals who readily form companionships with humans, such as other humans.

Does this mean that all dog trainers become skillful in dealing with their fellow men? Hardly. A personal observation by the author is that people often seem drawn to dog training specifically because of social discomfort. The author may well be a case in point: "The more I learn about people, the more I like my dogs," as the bumper sticker says.

Yet most dog trainers do have well thought out beliefs about child rearing, friendship, marriage, and politics, beliefs that are deeply informed by their dog training experience and, in some cases, the behavior science they also come to study. Dog training may not be the cure for social awkwardness, but a kid who's made a mistake training his dog, and seen with his own eyes the nature and consequences of that mistake, is that much less likely to make the same mistake raising a child when he's older. And a kid who's been rewarded with success in training a dog with gentle clarity is that much more likely to bring that lesson to his interaction with his fellow humans as well.

How would the world change if all the reflexive, unthinking, and negligent actions toward fellow humans that we see in our daily lives and our daily news media were replaced by the conduct of people who had studied firsthand the effects of their actions on the canine species, both in theory and in living practice, not once but countless times, and not only personally but in an entire community of others facing identical problems and testing the identical palette of solutions, and had discussed what they'd learned day in and day out in an academic milieu for months or years on end? Can there be any doubt that a new age would dawn?