Getting Results With Advanced Clicker Training
Clicker training is an effective way to teach dogs the basics — things like sit, stay, down and come — but it also is useful in teaching more complex behaviors. This basic training method, a form of classical conditioning, is an effective way to train a dog to associate desired behavior with a reward.
Once a dog understands a desirable reward is to come, the clicker instantly and consistently communicates he has performed the correct behavior. This instant feedback helps create an active learner eager to work for the next reward. Here are three primary ways to teach advanced behaviors using the clicker.
Shaping for Results
Shaping is rewarding a dog for performing successive closer approximations to a desired end behavior. In teaching the down command, for example, you would click and reward the dog for each step that comes closer to lying down. You would first reward the dog for putting his head down, then one shoulder, then two shoulders, then the hind end, until he’s in a full down.
Avoid progressing too quickly by not allowing your dog time to understand the first step before moving to the next. Conversely, lingering on one step too long can result in the dog thinking that step is the desired behavior and cause a delay in progressing to the next steps. The associated command for the final behavior should not be used until the dog is completely proficient in the final behavior.
Target Training by Direction
Target training is teaching a dog to move to an object or location by direction and perform a behavior once there, if desired. The target can be anything, including a target spot such as small plastic disc used to mark the location, a stick, or even a laser or flashlight to indicate the direction in which the dog should move. Click and reward the dog for each progression he makes to the target until he finally reaches it.
Do not incorporate a command until the dog shows a clear understanding that he is being rewarded for going to or touching the target. Target training has diverse applications and can be used for training dogs for competition, assistance work or to perform tricks. Among the possibilities are training for AKC obedience or agility trials. For example, in an obedience trial, the handler must send the dog to a specific area or object to await further instruction. In agility, when a dog reaches a contact obstacle, he must learn to touch it with two feet. When performing a running obstacle course, a dog must learn to touch obstacles with his feet before moving to the next obstacle.
Assistance and police dogs are taught to go to a specific location identified with a flashlight or laser pointer and wait for instruction. Even dogs trained for television and movies must be able to move to a target to ensure they are in the right place for a scene. Target training can even be used to help a dog overcome fear when a target is placed progressively closer to the object of fear until the dog is finally able to calmly approach it.
Combining Simple Behaviors
Chaining is the term used for training that combines two or more behaviors. It involves performing the behaviors in a fixed order followed by a reinforcer, such as food or toy reward, after the last behavior is performed. Any two or more behaviors can be chained, or paired. In fact, pairing simple behaviors helps to accomplish very complex behaviors. One behavior can serve as a cue to the next behavior, and the reward is reserved for the end behavior. Each behavior in the chain is taught separately in any order; however, the dog must be fully fluent in each individual behavior before attempting to perform the chain as a whole.
Two methods are used in chaining: forward and backward. In forward chaining, the behaviors are taught in the sequence in which they will be finally performed. The dog is rewarded at the completion of the first step, and then moves on to the next. This can cause difficulty when the reward is given at the first step if a dog gets stuck or becomes reluctant to move forward to the next step. Some dogs continue to expect to be rewarded for that first behavior without moving on.
Backward chaining reaps the strongest results and is easier for dogs to learn. In this method, the final step is taught first, and the following steps are taught in reverse order. Since the dog is rewarded at the end behavior, he will be more willing to perform preceding behaviors to get to the end reward.
If you are teaching an obedience retrieve, for example, the first step would be to teach the dog to front, which is sitting facing you. Next, have the dog hold the dumbbell while sitting in front of you, and then step back and have the dog bring the dumbbell to you. With the dumbbell on the ground, have the dog pick up the dumbbell and bring it to you.
The next steps will require the help of another person. Have the helper toss the dumbbell a few feet in front of you and allow the dog to bring it to you. Next have the helper hold the dog sideways and toss the dumbbell, allowing the dog to retrieve the dumbbell and bring it to you. The dog already understands he is being rewarded when he brings the dumbbell to you and sits, so he is more likely to fully complete the preceding steps so he can get his reward at the end.
Some dogs may test you or try to skip steps in the chain. If at any point a step is skipped by the dog, restart training the chain from the beginning. The dog must eventually understand that this is one action as a whole and all parts are necessary to obtain the reward.
Remember that patience and positive motivation are key in teaching any behavior. With shaping, target training and chaining, there are limitless possibilities to what you can teach your dog.
Triple Crown Dog Academy near Austin, Texas, offers training programs for trainers and dogs. Visit www.triplecrowndogs.com for information.
This article orginally appeared in Today's Breeder, A Nestlé Purina Publication Dedicated to the Needs of Canine Enthusiasts.
Help Future Generations of Dogs
Participate in canine health research by providing samples or by enrolling in a clinical trial. Samples are needed from healthy dogs and dogs affected by specific diseases.