• HOME
• PRACTICE AREAS
• ATTORNEYS
• ARTICLES
• CONTACT US
• CONTACT US
• CONTACT US

Clicker Training

Animals learn from the consequences of their behavior.  Behaviors that are rewarded are repeated, those that are not, go away.  This is how all animals learn, in nature as well as in our homes.  Clicker training takes advantage of this learning process.  This method was first used to train marine mammals and is now used to train all kinds of wild animals in zoological parks.  It is a method that is also extremely effective in training dogs.  It is a communication system between you and your dog.  It’s positive and fun and the dog actually learns better because he understands what you want him to do.  It is also very mentally stimulating because it actually encourages your dog to think.

We start by establishing the marker signal (a “click” of the clicker) as a conditioned reinforcer.  What this means is that the dog learns that the click sound predicts that a treat is coming.  To accomplish this we simply click and treat, click and treat, click and treat – about 5-6 times.  Once the association has been made, you can use the marker signal to clearly communicate to your dog the exact behavior you liked.  The clicker is a wonderful conditioned reinforcer because it is a clear, unambiguous sound that your dog has never heard before.  It sounds the same no matter who is using it and it carries no emotion, unlike our voice.  In addition, the clicker requires no processing in your dogs brain unlike our spoken word (whenever we speak to our dogs they must process the sounds to determine if they recognize anything, and if so, what it is associated with).  Once you have established your conditioned reinforcer (the click sound) your dog knows that he will be rewarded for the behavior he was doing when he heard the sound. 

Food is most often used as the primary reinforcer (the reward that follows the marker signal) because it is very desirable to most dogs.  Some dogs, however, work just as well for a chance to chase a ball, play tug, or access to the outside world.  It doesn’t really matter what the reward is as long as it is rewarding to the dog (praise and petting is usually not enough!). The reward is the “salary” that the dog “earns” for performing a behavior.  We do not bribe the dog to do something with the reward.  He must work for it.

Timing is very important.  We want the marker signal (“click”) to give specific information to the dog.  For example – putting his rear end on the ground (sit) is the behavior you want to reinforce so you must click at the precise time he puts his rump on the floor.  If the dog is not doing what you want, your timing might be off.  For example: if you are late with your click and it happens when the dog stands up, he has been rewarded for standing, not sitting.  Whenever things are not going well, ask yourself “what am I reinforcing?”   It often helps to think of the clicker as a mini camera.  If you wanted to take a picture of a certain behavior like sit, you would push the shutter button on the camera when the dog is doing what you want a picture of.  The moment you would take the picture is the same time you want to click. 

Clicker training is not a command based training system.  We do not tell the dog what to do at first.  Once he learns that certain behaviors are rewarded and starts to offer them, then we give them names.   To start, we either wait for a particular behavior to happen (capturing the behavior), or we lure the dog into position the first few times (never use a food lure more than 3 times or your dog will become dependent on it).  Eventually he will learn that this motion of putting his rear on the ground or lying down earns treats.  When he is offering the behavior to you and it is executed to your liking, you can then name it “Sit” or “Down”.  This is called “adding the cue”.  When you think your dog is about to do the behavior you are working on – say the cue just before he goes into position.  Repeat this over and over so he learns what the word means.  Eventually, you will only want to reinforce the behavior when you give the cue. 

Training sessions should be short and fun, only 3-5 minutes at a time.  Never get mad, never punish your dog if he doesn’t understand or messes up. Always end a session on a positive note.  If you decide, one more time, and he messes up, don’t end there.  Go back to the last step that he did successfully before you end the session.

At first you want to click and treat every time your dog does the behavior you are working on.  This is important during the shaping process.  When he is reliably and consistently offering the behavior on cue (when you will bet money that he will perform the behavior if you give the cue) you should switch to a variable schedule of reinforcement.  This means that you only reinforce every second time or third time he does the behavior.  Ask the dog to do two or three repetitions of the same behavior before clicking and treating.  Reward each execution of the behavior with verbal praise (“Good boy”), encourage him to move out of position and then repeat the cue.  Introduce variable reinforcement for position changes.  Ask him to make two different changes before you click and treat (sit to down, down to stand, etc).  Mix it up.  This will actually strengthen the response.  He learns that reinforcement will come, but he is no longer exactly sure when, or what quality of performance will produce it.  So he continues to perform in the absence of reinforcement, pushing farther and harder to stimulate you to produce the magic “click”.  Think of yourself as a giant walking slot machine (variable reinforcement is why people get so addicted to slot machines).

Dogs do not generalize well.  This means that just because he performs the behavior on cue for you in the house does not mean he will do it in the park.  You must teach your dog the behavior in a non-distracting environment first and once he has learned the cue you can start adding distractions slowly.  Start in the yard, then in front of the house, then in a strange location (down the street), then at a park, then a park with lots of kids or dogs etc.  Go slowly.  Set your dog up for success!

Remember the clicker is a tool that we use to train a new behavior.  Once the behavior is successfully on cue you no longer need to use the clicker for that behavior (make sure you fade it out using the slot machine method).  If you are adding new distractions or trying to shape a new behavior, get the clicker and treats back out.  You can use this wonderful communication tool for the rest of your dog’s life to teach new behaviors or strengthen old ones.

We also do not want to have to have food with us whenever we are with our dog in order for him to perform a requested behavior.  We must therefore associate verbal praise with the treat during the training phase, so that eventually the praise becomes the reinforcer that the dog will work for.  To do this always say “good boy” as you give him the treat.  It is, however, a good idea to occasionally give a treat to your dog when he performs an “old hat” behavior.  This will keep that behavior strong and reliable.