Monday, September 24, 2007

My Training Philosophy

This post is a carryover from my Amazon blog, so if you’ve read it before, well, I’d apologize, but I’ve expanded on some previous thoughts, and pared down others. Besides, I thought this would be the perfect post to inaugurate my new home. Enjoy!

My Training Philosophy
 Sigmund Freud with Two of His Chow-Chow Pups
I’m a Neo-Freudian dog trainer.

Seriously. As I see it, all canine behavior is either an expression of unconscious drives or  a way of sublimating the energy of those drives into “obedience” behaviors, something the dog is more than happy to do, as long as it results in a strong feeling of social cohesion.

Following this line of reasoning, positive reinforcements only work—that is they only have a long-term, lasting effect, under any and all conditions—when they give the dog an outlet for his drive-energy as part of a group dynamic.  So even though the names Pavlov and Skinner are more likely to ring a bell when we think of animal training, I think Freud’s view—that behavior and misbehavior run a tightrope between the urges for self-gratification and the need to maintain social harmony—fits our canine companions to a tee. In fact, the entire philosophy behind operant conditioning is essentially just a clinical and experimental outgrowth of Freud’s “pleasure principle,” the idea that animals tend to seek pleasure and avoid painful or aversive experiences. That’s true (except in rare cases). But underneath that principle there is still the dog’s natural, vibrant energy, energy that needs to either be expressed naturally or channeled into socially-acceptable behaviors; it's the Id vs. the Ego.

As a dog interacts with the environment, he’s stimulated by things that on the most basic level (i.e, emotional energy) stir feelings of either fear or desire. Loud noises or tall, imposing people might create fear, while seeing a pizza crust on the sidewalk or a squirrel in the park might stimulate a desire. These two basic feelings always create some form of visceral, palpable tension, which you’ll be able to detect in the dog's behavior. Tension can be a good thing too, as long as the dog knows what to do with it. If not, it's not good for him or us (or that poor squirrel).

I think we also need to teach the dog that his excess energy should always flow toward us or, better yet, into us, that we’re the place for that energy, those emotions to be safely grounded. Remember, dogs need to be part of a social dynamic in order to feel whole. When we become part of the dog’s flow of energy, just as the pizza crust and the squirrel are, and if he also sees us as the place where his energy can safely and easily go to ground, everything is resolved and the dog will gladly do anything we ask of him, and will in fact want us to give him instructions on what to do with his energy. If he has no safe way to release this energy, then the tension builds up, creating more and more stress.

Tension and stress will always need some kind of outlet, just as electrical energy does; it needs to “flow” in some direction. Clearly it’s best for the dog if his flow of energy doesn’t make him constantly run away from loud noises or try to kill every squirrel he sees. That’s why it’s our role to teach the dog how to reduce this inner tension in ways that don’t create problems for the dog or for us.

Over time behaviors which are successful at reducing tension, without disturbing the social dynamic, are preferred by the dog and become learned. If you want to apply the behaviorist paradigm to this it means that learning happens, or is reinforced, through the pleasurable feeling of releasing internal tension. If you want to apply the dominance paradigm, you could say that when a dog knows who’s “in charge” of his energy, he’s much less likely to have any internal tension—that’s all taken care of. (Though we now know the alpha concept is a myth, all myths have their basis in some truth.) Two other things to remember are that the behaviorist paradigm works best when the kind of reward/reinforcement used is in some way relevant to the source of the dog’s internal tension. And the dominance paradigm tends to work best when the trainer’s need to be “in charge” doesn’t manifest in punishing or scaring the dog too much. It's also important to realize that all living things experience tension as part of being alive. So when I refer to internal tension I don't mean to imply that the dog is a nervous wreck, just that even the vaguest feeling of dissatisfaction drives the dog to look for an outlet, whether that's through chasing a squirrel, picking up garbage, or bringing you a toy. (The last is obviously the best choice!)

There’s no question that both dominance training and positive reinforcement methods create well-trained dogs a certain percentage of the time. If trainers in both camps weren’t getting some positive results, they’d surely start looking for a new way of doing things. I think the other part of the equation is that, generally speaking, for 85% of most dog training outcomes, the credit goes to the dog, while only 15% goes to the trainer or her methodology. However, it’s been my experience (and I’ve studied and used both dominance and behaviorist techniques) that it’s only through understanding where and how a dog’s internal tension comes from, and how best to use it, activate it, and help the dog reduce it, that you get results that work much closer 100% of the time with all the dogs you train. So, to me at least, it’s always better to at least try to understand how a dog’s emotional energy influences his behavior.

There’s another piece, in that dogs who are trained either by conditioning-based techniques or through dominance aren’t necessarily taught how to be emotionally open and flexible. That generally only comes through understanding the underlying flow of energy in the dog’s instinctive and emotional systems. This is why you often see dogs in households run either by dominance trainers or +R trainers, who aren’t always able to get along together peacefully without some control mechanisms in place. Dogs who are trained using the concept of tension and release, are always flexible and able to get along with any dog they have to spend any amount of time with.

I see each dog as a discrete energy exchange system, composed of progressively smaller energy exchange systems—like organs, cells, genes, etc.—inside its body. Dogs are also part of a larger energy exchange system of network consciousness, which for lack of a better word we might call “the environment.” Dogs receive energy from their internal systems in the form of heat, instinctive drives, etc., and from the environment in the forms of food, water, oxygen, sensory input (light, sound, and smells each carry forms of energy), and from the network through energy exchanges with other dogs, people, etc. They enjoy their interactions with the larger energy exchange system most when the kinds of inputs and outputs they engage in are successful at either reducing their internal tension, or successful at building up that tension in a way that when it finally is released (through behavior) there’s a more pleasurable reduction of that tension as an end result. (I think this explains some of the really strange behavioral problems we see in dogs who seem to be almost masochistic in the ways that they keep doing harmful things to themselves—because, as the old punchline goes, “It feels so good when I stop.”)

Another thing to remember is that the dog is generally happiest when he feels aligned harmonically with the members of his social group, so oftentimes he’ll choose behaviors that don’t seem to be beneficial to him individually but are beneficial to his group dynamic. (This is where the old canard “dogs want to please their owners” comes from.)

One of the reasons I think tension-and-release is so important in dog training is that dogs and wolves are, in a manner of speaking, two manifestations of the same species, one wild, one domesticated. And, to a certain degree, they share the same predatory instincts, though dogs don’t have the “complete predatory package” wolves do. That’s because humans have bred dogs to use certain aspects of their prey drive, that is certain specific predatory motor patterns like the search, the eye-stalk, the chase, the grab bite, and the kill bite, for specific purposes: bloodhounds are all search, border collies are mostly eye-stalk with a little chase thrown in, retrievers shouldn’t have a strong kill bite, etc. So some of these motor patterns inherited from the wolf have been amplified in some breeds, some have been diminished.

Most pet dogs don’t really need to use these instincts to “make a living”—they get their food in a bowl—yet that instinctive, predatory energy is still active, creating behavioral problems if it’s not given a satisfying outlet, which I think should come through playing games or doing some sort of work related to the prey drive. In fact, the prey drive is the raison d’etre for the pack instinct—wolves who live near garbage dumps don’t form packs, coyotes form packs but only when they need to hunt large prey—so dogs get most of their natural sociability from their genetic history as social predators. However, their social history with humans has also given them a remarkable ability to read our conscious and unconscious signaling, giving many dog owners the impression that, “My dog seems to know what I’m thinking.” And that’s true, though I’d put it this way: “Your dog knows what you’re feeling.”

In wolves, one of the ways the predatory instincts function is through a natural mechanism of increasing internal tension whenever the pack hasn’t hunted for a while. This seems to be a necessary motivator, an incentive to get the wolves to leave the safety of their den (or home turf), and go out to hunt and kill animals that could very easily kill or severely injure them individually. In other words, there seems to be a kind of balance scale between the “survival” instincts of the individual pack members, on the one hand, and the needs of the prey drive of the pack as a whole, on the other. As this internal tension between fear and desire builds the pack seeks a release, and finds it in the form of hunting.

Captive wolves and village dogs don’t have this release mechanism, so they express their internal tension through what I think are commonly misconstrued as dominant and submissive, or hierarchical behaviors. In other words, their natural aggressive energy, which is designed to be directed only toward prey animals, or toward other wolf packs who might invade their turf, becomes directed to members of their own social group instead. (I say “social group” because without the need or opportunity to hunt large prey captive wolves and village dogs don’t form real packs.)

This same build up of tension (or unused predatory energy) happens in pet dogs unless they get a chance to use that energy in some way that’s acceptable to the social dynamic they’re a part of. (If they choose unacceptable behaviors and are punished for them, the internal tension can often worsen, creating even more “behavioral problems” or withdrawal from all household comity). I think most behavioral problems are caused by this increase in the dog’s underlying nervous or emotional tension; the dog simply doesn’t know what else to do with his energy, how to use it safely. I think that’s why techniques like desensitization—which are actually designed to reduce the dog’s energetic output (behavioral responses) rather than targeting his stress and giving it a satisfying release—don’t work very often to solve behavioral issues in the long run (though they can be effective for a few weeks or months). It's also why things like playing tug, fetch, and push-of-war almost always work (unless there’s an underlying physiological or neurological condition causing the “misbehavior,” and play therapy sometimes works even then).

All dogs are different so the release of tension doesn’t always have to come through play. For some dogs it can come from doing a lot of sniffing on their walks (the search), “killing” their toys (the kill bite), or, since many obedience behaviors are analogues to the predatory motor patterns found in wild wolves, dogs can reduce their internal tension simply through obeying their owner’s commands, as long as the commands are taught with some a payoff that satisfies the dog’s urge to bite.

As you can probably tell by now, in my view, most behaviors are best learned through teaching the dog that obedience successfully reduces his internal tension. One of the most direct and satisfying way to do this is through biting games. Dogs who’ve had their oral impulses punished as puppies, or pups who’ve had their natural inquisitive nature quashed by well-meaning positive trainers who believe “almost everything needs to be taught right away,” will either display too much energy—in the forms of destructive chewing, excessive jumping up, barking, digging, aggression, etc.—or will display too little energy because they’ve shut themselves off from their instinctive release systems in the interest of feeling safe within their social system. This is why some good hard games of tug, fetch, "chase me," keep away, and push-of-war, can be successful “cure-alls,” especially when played outside, because for the first class of dogs—those with too much energy—it’s an incredibly satisfying outlet. For the others, it’s a safe way to bring their natural, normal energy back up to the surface and finally find a way to release it.

I’m actually not against using treats, or prescribing desensitization (on the off-chance that it might work, since it does initially, and its effects can sometimes be longer lasting). I am against punishing and scolding a dog for anything, but I see nothing wrong with letting a dog know when he’s out of bounds. (A simple, “Hey, that’s my toe!” followed by “Good doggie!” seems to work just fine for me.) I’ve also learned that what would normally be perceived as punishment by the dog—as long as it’s done as part of a high-energy game, and as long as the dog is given an immediate release for the shock of energy the stimulus has caused—can actually make the dog that much more interested and invested in the game. It can also make him more willing to work with and for his trainer. In fact, if anything, invigorates and enlivens the dog.

Also, when you use the prey drive in training, and make the exercises more and more difficult (things like the “down while running” and the “stay with a ball throw right past the dog’s nose”), you’re strengthening his emotional flexibility and adaptability, which, again, come directly from the prey drive. That’s because when wolves are hunting large prey they have an incredible capacity to be aware of everything going on around them, to stop on a dime emotionally when needed, to instantly pump up their emotions to the max when needed, to cooperate and be in harmony with one another to an almost telepathic degree (it may actually be telepathic), and to manage and use their own levels of internal tension in ways that are simply incredible. To give your dog even a tiny taste of that does wonders for his morale, not to mention his ability to adapt to almost anything that life throws at him.

So that’s basically it. The only thing to add is that since most dog owners and dog trainers are unaware of the kind of training I prefer, I have, of necessity, framed a good portion of my philosophy in contrast to the two most common and well-known approaches. Since I’ve studiously researched and used both the alpha model and conditioning, and found them both lacking (for me, anyway), I believe I have a unique perspective on all three forms of training.

Lee Charles Kelley


Lee Charles Kelley, said...

Summer's Mommy wrote:

Call it kismet, Lee!!! I just directed a friend to your blog because she asked with interest about training through prey drive, and I was struggling to give her my take on it correctly!! I love your post!!! I hope she reads it and gains good info from it, because to me, it is golden!

I'm currently teaching Summer to jump over a four foot hurdle. She's a real pro now! And as I teach her more new things, and proof the stuff she already knows, she's really gaining in confidence. I can safely say I can get her in a strict Sit/Stay or Down/Stay WHILE there are a ton of dogs milling around her, walk 100 yards away, and get her to recall like her patootie is on fire. She'll do a sharp front/sit, then the Heel and side sit finish. Working on Down while walking with her on leash. Oh, and she generalized the Jump command on her own - I was walking her to prospect park, and there was a rope cordon where I wanted to take a short cut, so I said Jump, and she cleared the barrier with two feet to spare! She COULD have just stepped over it, but I wanted to test her. And many times, I can actually get her to recall to me IN midplay with her favorite pups. You'd be proud of her! She's coming a long way! Also going to start working on her learning how to jump over her leash if I hold it in a U-shape (without her being attached, of course...) and eventually, want her to learn how to jump through a hoop made with my arms. :D

I hope you're doing well.

Lee Charles Kelley, said...

That's great stuff, SM! It's amazing how dogs can "generalize" stuff so quickly when you make it part of a high energy game.