First I wanted to apologize for being silent on the blog for so long. As I mentioned in my last entry, we've been making a lot of changes and adding some new features (check out our newsletter for one) and these things take time. We've also been working on several outside projects including Paws in the Park and the upcoming Open House at Columbia Canine Sports Center this weekend. Hopefully, once again, I can get back to the blog and share some relevant information about training, behavior, and the like.
There has been a lot of discussion in the forums I participate in on disclosure of training techniques. I would be apprehensive of anyone who is not willing to share their methods. Usually these equate to some method of aversive training, aversive being hurting your dog to get him to do what you want. This is not always the case, but see my blog on choosing a trainer.
So, with that in mind, here is my training method. I'll try to stay to the point and not add too much back information here, but save it for another time.
It is important to understand that a dog, or any animal, learns more for being reinforced for the behavior you want than punished for the behavior you don't. If you reinforce the desired behavior, the dog, person, chinchilla, or whatever will try harder each successive time. Using punishment, the dog will only do enough to avoid being punished.
For most of my training I use what is called "marker training," also known as "clicker training." I like to use a clicker because it is a distinct sound that carries over a lot of other noises. In class I teach the use of the word "yessssss," drawing out the "s" to make it distinct from just saying "yes" in normal conversation. You can use whatever you like. It works like this: I start with charging the marker. You simple click and give the dog a treat, click and give the dog a treat, click and give the dog a treat. This teaches the dog that every time he hears the click, good things happen.
I'll use a sit for this example, just for clarity's sake. You can wait for the dog to sit on his own, or you can lure the dog into position. As soon as the dog's bottom hits the floor, click and treat. Do this a few times. Then add the word "sit." By this time he should know the behavior you're looking for and will sit quickly on his own. If not, lure him into position, click and treat.
Usually you only want to spend about five minutes on training at a time to begin with. You can go up to about 15 minutes, but I don't recommend any more than that in one session.
Most trainings I've talked to or read say if the dog does the behavior five to seven times in a row without fail, you can move to the next step. In this case, the next step would be duration. Ask for the sit. When the dog sits, delay the click and treat for about a second. Once he can do this, move up to two seconds, then three, then five, etc. Be sure to go at a slow enough pace that your dog doesn't get frustrated, but fast enough to keep his interest.
Suppose in the first session you get a three second sit. Next session, take a step or two back and just go for a one or two second sit. This should go pretty fast and you should be able to move past it and past the three second sit quickly.
One thing detractors often say about this method is they don't want their dog working for treats. This is a total misunderstanding of the method. Just like people work for a paycheck or do hobbies for enjoyment, we want the dog to work because he wants to. But, for basic obedience in particular, the point is to have the dog offer the behavior in real life situations when you may not have a treat handy or the dog is not in "training mode."
So, a word about fading treats. Once the dog is sitting consistently, ask for a sit, click and treat. Ask for a sit, say "good boy" and give lots of praise. ask for a sit, click and treat. Ask for a sit, "good boy." Next session, ask for a sit, click and treat. Ask for a sit, "good boy." Ask for a sit, "good boy." Ask for a sit, click and treat. You want to put the click and treat on an intermittent schedule so the behavior is only occasionally reinforced that way. Because the sit has a history of being reinforced with a treat, his emotional response will be the same whether the treat is delivered or not. (You should never click without giving a treat or it will start to lose its effectiveness. This is why I like to use "good boy.) For "proofing," you may want to reinforce the sit with a click and treat from time to time. Because I think a sit is so important, I start each training session with a sit, click and treat to get my dogs into the training mindset, then I end the session with a sit so we always end on a good note but also to proof the behavior.
I won't get into problem behaviors here because this is about my training techniques, not behavior modification. If you have questions about that though leave them here or feel free to e-mail me.
Two final thoughts. First, I try to never prompt (touch or force) the dog into position. Pushing on the hips to get a dog to sit, for example, can cause or aggravate hip problems and although they make take a little longer to figure it out, once they know you want them to sit, they've got it, and a lot better than if they wait for you to put them in position. Second, I never punish a dog for the wrong behavior, but reinforce the right behavior. If the dog is doing something he shouldn't, I don't tell him what I don't want, but what I do want. I simply can't understand why someone would want a dog if they felt it was going to be a constant struggle to maintain some misconception of dominance. I want a companion and hope the same for you.
If you want to learn more about my techniques, here's an excellent list of some of my favorite books to get you started:
Family Friendly Dog Training by Patricia McConnell and Aimee Moore
The Puppy Primer by Patricia McConnell and Brenda Skidmore
Clicker Training for Dogs by Karen Pryor
The Thinking Dog: Crossover to Clicker Training by Gail Tamases Fisher
No comments:
Post a Comment