Clicker training is a method of animal training that relies on
positive reinforcement and
reward instead of punishment and negative reinforcement. It requires a slightly different way of
thinking, but yields excellent results. It is particularly nice because it is behavior-based, not language-based, and you can train more than one
behavior at once easily.
It is for the purpose of teaching new behaviors. You will not need to carry a clicker around every time you go out and click every few seconds for the dog's entire life. The clicker means "we are learning something new, pay attention". It also means "good dog!"
It was originally developed for working with dolphins as a way to encourage the animals with no way to be in physical contact with them, but the method has spread to nearly all species of kept animals. It is one of the current favorite training methods for dogs among many owners, trainers, and vets.
The tools are fairly simple. You will need:
- a dog (or other animal)
- a clicker ($1.50 at Petsmart. Buy several. They're easy to lose.)
- treats
A note on treats: because the reward ratio is so high, this method can easily lead to overfeeding of treats. There are several solutions to this. You can break standard treats into pieces and give a piece for each successful action. You can also give normal kibble as a reward, and a special treat if the dog accomplishes a major breakthrough. Or you can remove an equal amount of kibble from the dog's dish with each training session. I took several types of soft treats and crushed them into powder. My Shetland Sheepdog puppy Java gets a pinch of the mix each time, and a whole treat for special successes.
For the FIRST behavior you try to teach, work on ONLY that behavior. This is to get you as the trainer used to the system as much as it is to get the dog used to it. After you are confident in your methods, clicking can be used for many behaviors at once.
Timing is THE most important thing about clicker training. Click WHILE he is performing the action. You want him to have committed to it so he continues it. If you click too soon or too late, he may misunderstand the cue. If a dog is not learning his task it is probably because your clicker timing is off.
Phase 1: introducing the clicker.
As you go about your random chores (or sit on the couch and watch TV) randomly click the clicker. Then throw a treat to the dog. Pretty soon he should start to look at you when he hears a click. At this stage, only reward him if he looks when you click. Eventually he should start looking at the clicker before you click, if he sees you have it or it is sitting around. At this point he understands the clicker and you can move on.
Phase 2: rewards what he already does as he does it.
Pick one behavior you see in your dog. It can be important ones like "sit" or "come" or cute ones like raising a paw. Just make sure it is a common one for him. For this example we'll use "sit". As you're just doing whatever, every time you see him park his butt on the floor, click and give him a treat.
Phase 3: introduce the verbal cue
Pretty soon he'll start sitting a whole lot in hopes of getting a goodie. At this point you introduce the verbal cue. As you see him begin to sit, say "sit!" and click as he continues. Give him a treat, of course. If you miss a cue, go ahead and click anyways. Start phasing out the treats. Click every time and verbally praise him, but treat every second or third or fourth time.
Phase 4: performing on cue
Once your dog is good at sitting and being clicked while hearing the verbal cue, test him. See if he will "sit" AT the command. Start giving the cue BEFORE he begins to sit. Only reward him when he sits within 2 seconds of your commands. His own "sit"s don't count any longer. At first, treat with every sit. As he masters it further and sits at about 70-80% accuracy, phase out the treat every time. If he fails to sit on cue more than half the time, go back to phase 3 for a while.
Phase 5: fading out
Once the behavior is established and he responds at least 80% of the time, phase out the clicker. Still pet him and praise him every time he succeeds, but don't click all the time. Soon the behavior will be established and won't need reinforcing.
Phase 6: building and reinforcing
Once a behavior is established enough in one setting you no longer need to click, go back to phase 4 with new settings. A dog that is happy to "sit!" in your living room most likely will not "sit" outside when he meets other dogs or people or when riding in your car or at the pet store. Issue the command and click and treat if he responds. If he fails to respond the majority of the time, go back to phase 3, and click and cue if he performs the behavior on his own.
* Off to a Good Start--A manual for raising your new puppy, Mary Thompson
this is an excellent book on overall puppy care. There is a large section on training, not much on clicker training
* The Dog Whisperer, Paul Owens
I did not care for this book. It's very mystical new-agey. The clicker training part is pretty good, the rest is nonsense.
* my own experiences
* a large assortment of web pages