back to Barry's Place
Things to remember with clicker training
> > When the dog is in the next room making a racket, wait outside and wait for him to calm down. After he is quiet for 10 seconds, you can walk in, click and treat for being quiet. Before too long, he should get the idea that quiet, calm behavior is rewarding to him!
> Actually, I think I should have said to click before you go into the room. If you walk in and then click, you may be rewarding jumping or mouthing or whatever other behaviors the dog offers when he is excited. If you click first, then walk in and treat, you should be marking the calm, quiet behavior.
One of the most important pieces of clicker knowledge is to know when NOT to use a clicker.
The exercise of leaving the dog for a while, and then entering only when it is quiet is fine, but I would not use a clicker or a food reward here at all for several reasons.
| 1. |
The clicker tells the dog. "Good job, you are done." I do not want a dog that I confine to a crate or X-pen or room to think in terms of its solitude STARTING and ENDING. Learning to spend time alone in calmness is best accomplished by convincing the dog that spending time alone in calmness is an ordinary everyday thing, not a grand accomplishment. Thus, we leave the dog casually without any big good- bye, and we enter the room when we return casually as though nothing difficult has been demanded.
(This is the same principle that is used with a dog that is afraid of thunder. When the thunder is booming loudly, the best thing you can do is go about your daily routine and prentend that you don't even hear it.) This is also a behavior we want to be seamless, not well defined. |
| 2. |
It is important to try to have the dog calm and quiet when you enter the room, and as you approach to let it out. Again, we do not want the dog jumping for joy, thinking, "Thank God, I am being liberated!" The click itself cause the dog to feel excitement, and the reward again suggests that "You just did a tough thing very well." Not the impression I want to make! Also, no reward is necessary because my presence is itself rewarding, as is the liberation of the dog. How many rewards do we need to give it? |
| 3. |
The use of the click as a Reward Marker is also misplaced here. The most imprortant rule in CT is to know exactly what you are clicking for before you click! So, the dog has been alone for 3 hours. During that time it fussed for the first ten minutes, and then spent a fairly calm afternoon alone. That is one behavior we would like to see, but did we click and treat the dog for that? No! It is in fact impossible to C & T a dog for being alone because as soon as you enter and C & T, the dog is no longer alone! So, now, after three hours of good behavior, we enter and C & T at a moment when the dog has become excited and is about to be released. So what are we reinforcing? We are reinforcing getting excited and being released!
(BTW, clicking before entering the room is not practical because unless you have the dumbest dog in the world, it will know you are home when you pull in the driveway! Also, clicking right before you enter reinforces what? It reinforces the dog recognizing that you are about to enter the room.) |
| 4. |
We are also dealing with a long duration activity here. In long duration activities, the click is often confusing and self-defeating. For example, when I teach my dog to do a ten minute Sit Stay, the basic rules of CT would suggest that I should C & T at the end of the behavior--at the end of the ten minute Stay. But I do not want to reinforce the end of the Stay--the Release! I want to reinforce the Staying as it is happening, not when it is done.
Therefore, I would not use a clicker to mark the end of the Stay. I would simply use a release command. I would also not reward at the end of the Stay. Instead, I would continuously give the dog a treat every minute or so during the ten minutes it is Staying. The treat without the click becomes what is called a "Keep Going Signal." It says to the dog, "You are doing the right thing, but we are not done yet," whereas a click would be saying, "You are doing the right thing, and we are now done!" A KGS is a type of BRIDGING STIMULUS. So by rewrding during the stay, and not rewarding at the end of the stay, I am teaching the dog that good things happen when you stay, but there is nothing particularly special about ending your stay, so let's not get all excited about that!" Unfortunately, this KGS is not workable in the situation we have been discussing because you can not give a KGS to tell a dog to keep relaxing while it is alone without causing it to not be alone anymore. |
| 5. |
Finally, if properly encouraged and developed, having a quiet time alone can become self-rewarding for many dogs.
So this situation is an interesting one because first, it requires a careful evaluation of why you are clicking and what you are reinforcing. Second, it points out that C & T is not always the desireable or most effective thing to do. And finally, the situation is a bit Zen-like since you cannot celebrate being calm, and you cannot be with someone to tell them they are doing a good job of being alone!
I meditate regularly, and this situation reminds me of a saying that meditators use to jokingly refer to the fact that if you are sitting there feeling happy that you are meditating so well, then your mind must be wandering and you are not meditating!
The saying is: "Here I am, wasn't I?" |
|
