Wednesday, August 24, 2011

What Your Dog Trainer Will Not Tell You


I know a man who took his leash-aggressive dog to a "really good bird dog trainer" who told him the dog could be made less leash reactive (i.e. dog aggressive when on a leash) if it was trained with an e-collar.

Did that work? Nope. In fact, that's probably never going to happen.

I explained to my friend what he needed to do, but I could see his interest in finding a solution fading with every word I uttered. You see, it was very clear to him I was talking about a process and he wanted an event -- a common disconnect in the world of dogs and dog training.

I explained the process of making a dog less leash reactive. To begin with he should not feed his dog for a full 24-hours or more in order to increase the dog's food motivation. Then he should take his dog to a large park and, because he was having a hard time controlling his large dog, he should firmly tie the dog's much-shortened leash to a bench. Then, every time a person or another dog approached from the far end of the park, and the dog took notice, he should click his tongue and direct the dog's attention to his face, and then treat when the dog looked at him and not the other person or the dog. Once the dog got the idea that it should look to the owner for a treat and that if it was calm it would get a treat, he should then move to a new position in the park.  Now, with the dog tied to another bench where people and dogs might happen to approach a little more closely, he should repeat the exercise making sure that the dog is always well motivated by hunger and is only treated when it looks to the owner and remains calm.

When the dog seems to have that routine in hand, the owner should walk the dog around the edge of the park but still keep it pretty far from people and other dogs (at least initially), making sure the dog looks to him and sits every time a dog rotates into view within a certain distance. Again, click, praise, and treat every time the dog look to the owner, remains calm, and sits.  If the dog does not sit, walk away from the foreign dog in question, and do not treat and do not praise.  Ignore the dog and remain totally calm.

I stressed this routine had to be done every time, and this was how all the dog's food and other rewards had to be delivered for the next month. It was not going to be an overnight miracle, but if he was consistent and the dog remained hungry (feed a little less!) the dog would learn that the way it got food was when it saw a dog or a person, if it looked at the owner and sat, it would get a food reward (which would later be reduced to a pat and praise sometimes, and a "jackpot" food reward at other times).

Of course my friend was not really looking for this kind of instruction. He was not looking for a process that involved this much work and time. He was looking for a "trick" -- a five-second miracle that did not actually involve spending time and energy on the dog.

And isn't that the problem so often?

People want a dog that's totally calm and obedient right out of the box.

But dog's don't work like that, do they?

People want an animal that will not bark, will not dig up the garden, will not bite the neighbor, and which they can leave at home for 20 hours a day without too much thought.

People say they want a dog, but they need a cat, and they deserve a goldfish.

The simple truth is that dogs are a tyranny, and they are not really suited to the modern world of long hours at the office, and a run to the gym before a quick dash to the Whole Foods for aged balsamic vinegar.

If you live that kind of life, what time does that leave for the dog? It leaves the dog with maybe 10 minutes between the start of Law and Order and the shower. But that's not enough time! That's not enough exercise! That's not enough instruction of any kind.

And so people go to dog trainers, and trainers are put in a frustrating position because what every trainer agrees on is that the dog in front of them deserves time, exercise, consistency, and opportunities to achieve awards for success.

But do people with problem dogs really want to give their dogs time and exercise?

Too often, the answer is NO.

Whatever they say when asked the question is irrelevant; in their day-to-day actions they provide the only answer that counts.

And so the dog "trainer" is left unable to say what he or she really needs to say: "You are lazy and undisciplined, and that's one of the reasons your dog is crazy and undisciplined."

A dog trainer cannot say this, of course, or they will lose their clients and never get another. 

Instead they have to suggest they have "secret knowledge" or a special or "new" system or "philosophy" of dog training.  But the secrets are not really very secret, are they?  And is any part of dog training really new?

The real core secret is that the owner has to be willing to exercise his or her dog, and to put in the tim, and engage in very consistent and well-timed communication. The vast majority of this communication should be earned rewards with perhaps a few mild aversives to bust off self-reinforcing unwanted behavior. 

But can a trainer make a physically lazy and undisciplined person physically industrious and disciplined?

Can a trainer change a person's priorities or increase the number of hours in a day?

Not generally.

And so most trainers show their clients how to do a few basic obedience tricks, and perhaps they show a few rudiments of operant conditioning. That's all well and good, but you can learn the basics of that from a book and a video, and you will only get good with practice and focus -- the very things most folks are so very bad at to begin with.
.

5 comments:

Anonymous said...

Nothing to say but thanks, Patrick!

Kim Helstein
Dogs In Training

Jenn said...

Dogs on leash.

Many reactive dogs are fearful - even if they look like pure aggression, they are acting out of fear. What these dogs need is confidence building - which your long and slow method produces. Confidence that nothing bad is going to happen with that other dog. Confidence that the owner has everything under control. You are building trust here.

When you pop them one with a shock collar, you are provided aversion training. Now when they see another dog, instead of running at them, they will want to run away. Not the desired outcome. And because the dog can't predict what will happen (will the shock come this time?) there is no trust.

There are no shortcuts.

Excellent write up.

HurricaneDeck said...

Amen Patrick!

Along with my "feed your dog a cooked frog" shirt from you, and my sparkly "Alpha Bitch" tee that I adore, I also have my "Praise the Dog, Slap the Handler" shirt that gets plenty of chuckles from those "in the know."

Sue said...

Beautifully said, Patrick. And thank you for advocating the trusty combination of progressive desensitization (to the stimulus) with classical conditioning (changing the dog's emotional reaction toward the stimulus)--also words that we don't usually tell the average dog owner. Not because it's 'secret' but because explaining the science is often counterproductive. It's more effective to all but the truly curious if it looks like magic. :) A slight variation on this method that many of us are using now is to focus on the classical conditioning part at the outset by clicking and treating when the dog LOOKS AT the stimulus, not at the owner. Much simpler than asking for 3 things (look at me, sit and be calm)--the calm will happen, and the sit won't be necessary, because the dog will look at the owner for a treat as soon as he hears the click. Provided the other dog is far enough away, this will very quickly become a fun game for the dog--look at the dog for a treat. You are changing the emotional state from "Oh no, a dog, scary! owner will be upset, I will get choked, I must make it go away!" ----> "Oh boy! A dog! Good things will happen!" As long as you keep the dog far enough away at the beginning that they will be sub-threshold and can attend and eat, And then you can progress closer and closer pretty rapidly. It takes planning and consistency, as you say. And because 'the look' happens in an instant, that's where the clicker is really useful for marking that instant. But the results do look magical and can improve rapidly.

Practice outside the fenced play area of a vet /boarding clinic or dog park--you can't control which dogs run to the fence or how they will react, but you can quickly add or remove distance.

joyce kesling, CDBC said...

Nice post to Sue! Very well stated Patrick, there are countless ways to train dogs in differing contexts, but it's really helpful to know why the dog is acting the way they are, all the training in the world will not necessarily change an owner who has a dog beyond her capabilities. The method you suggested Pat is the same as some of the trainers we know well used 10 years ago. This trains a dog who has issues, possibly with the owners lack of leadership, because you are telling the dog it's not their job to protect you, you've got things under control. The dog trusts you to keep them safe. Ceding this leadership to a human is not always easy nor attainable depending on what the individual dog was bred to do, it's very hard to change instinctive behavior,especially when it is so aligned with one's biology. Think "Misbehavior of Animals" you can't use one method to work and/or train all dogs. BTW, I use positive reinforcement and ignoring what I don't like, it's so freakin easy, but that's basic training, it has nothing to do with how a dog feels in their personal space mentally and/or physically! 8-)