This is a question I hear at least once a week from clients new to clicker training, and the answer is absolutely not.

In fact, green clicker trainers often accidentally get clicker addiction and they don’t seem to mind, since their dogs are having a great time and being more responsive than ever, but it’s still not good dog training.  What I mean by this is that people start clicking their dogs to teach a behavior and continue to reward the behavior with a click and a treat long after the behavior is learned.   Clicking and treating every single time for the same exact behavior not only dulls your dog’s responsiveness for future cueing, but also dulls the meaning of the clicker.  Think about it: when you use a clicker to shape or capture behaviors, you are using it to identify specific moments in time, and you want your dog to think about what got clicked and build on or repeat it.  If you are clicking a learned behavior it should only be to point out something specific about that behavior that you like–such as your dog performed a tight, tuck-sit, instead of a lazy rock back sit.

In fact, you should get rid of the use of the clicker once your dog is performing the end behavior.  In my pet training classes, where we teach dogs to “sit”and “down” using shaping* or capturing* we focus first on sit, and then the next week move right to down, and in just two weeks’ time the dogs have a reliable sit and down verbal cue.  This is not because my students work harder than other people, it’s because they are taught how to use clicker training effectively.  The process should look like this:

First, get the behavior.  Shape or capture the behavior you are looking for, when the behavior is what you want it to look be (dog goes directly into a down, or dog sits immediately without backing up, etc.) and it is happening predictabley with 80% accuracy, move on.

Second, when you can predict when the behavior will occur and that it will occur with 80% reliability, add a cue.  The cue should happen right before the behavior happens.  For instance, you throw a cookie and your dog gets up and eats it.  You know that your dog will return to you and offer a down for a click, so say the cue “lie down” as your dog is returning to you.  Click when he lies down, and toss the cookie again to “reset” for another repetition.  When you have done this is a variety of locations (living room, front yard, training class, kitchen, garage, back yard, with human sitting, standing, kneeling, etc.) and your dog is still having an 80% accuracy rate, move on.

Lastly, stop using the clicker and move to a random reinforcement schedule (meaning you don’t give your dog a cookie every time).  It is important to get to this stage as quickly as possible to avoid clicker or food dependency.  Since this is the final stage of training, start to use your cue in a lot of situations, and be sure to back it up with reinforcement randomly, but frequently.  It is also vital to vary what the reward is, meaning sometimes your dog gets a cookie, sometimes praise, sometimes a toy, sometimes a chase game, etc. 

Happy training!

*Shaping:  Or shaping by successive approximation, is teaching by marking and rewarding small increments (approximations) of a behavior until the final behavior is achieved.  It’s brilliant, and where the real fun is at in dog training.  Shaping is what makes clicker training so fantastic.

*Capturing:  This is a method of teaching a behavior where the trainer marks and rewards the final behavior as it happens.  It is only practical to use if the desired behavior is already happening to some degree.  Most dogs that enroll in training classes have already learned that sitting earns them things in their life, so capturing can be used quite easily to teach a more reliable sit cue.  Other behaviors capturing can be used for include barking on cue and lying on a bed.

Advertisements