Clicker Training By Adrienne
Since there are a few of us that have new family members I thought it would be good to discuss training. We are choosing to attend Puppy Kindergarten classes at a local facility. They practice “Clicker” training. This style of training is becoming more well known and accepted as a better practice with pets. Here's the basics of how it works and how to get started.
Dogs learn by association, positive and negative. When you walk towards the fridge they usually follow you and get excited that something might be for them, as well as shying away from you when you get out the nail clippers if they have had a bad experience with them. The other way they learn is visually, such as using a treat or our finger to teach them the “down” command. The basic principal of clicker training is to capture a behavior you want. A clicker is a quicker response than a voice response. You “click” when they have done something that you want, like sit. The clicker is a marker or communication tool. It is a unique sound that gets the dogs attention. Once they have sat, you click and give them a treat.( A SMALL treat Brett. Something the size of a pencil eraser, no matter the size of the dog and only one at a time.)LOL! All you want is for them to get a taste and be happy about the response they are getting. When your dog is conditioned to the clicker, they will understand that it means two things. (1) Whatever they are doing at the time the click is heard is a desirable (and therefore rewardable) behavior and (2) a reward is coming. The reward is the primary reinforcer. The reward can be anything the dog finds desirable. Use a combination of food and toys, food for stationary exercises and a toy for moving exercises.
Until your dog becomes “clicker wise” you can start a training session by “loading” the clicker. Click 5-10 times and follow each click with a desirable treat. Food should never be in sight or in the handler's hand since you do not want the dog to become focused on the food. This would apply to toys as well. You should only click once. If your dog does something exceptionally well or has a significant breakthrough during a training session, use multiple rewards not multiple clicks, like a jackpot. The rule is, “if you click you MUST reward the dog”. If you do not reward after each click, then the power of the clicker is diluted. Also remember not to give away that they are going to get a reward by putting your hand in your pocket or treat pouch as you are clicking. This should be done after you have clicked. The click “ends” the behavior. What you dog does after the click or while waiting for the reward does not matter. Remember not to add extra body language before or as you click. Your dog should be listening for the click, not watching for a signal from you.
Once you have the basics down you can start to “shape behaviors”, teaching them something new that you want them to do. With clicker training the dog has to take responsibility for learning. You do not physically manipulate the dog to teach the desired behavior. Instead, you wait for the dog to offer increments of the behavior and use the clicker to communicate to the dog that such behavior is the desired response. If the dog is not offering the desired behavior, you can begin by “luring” them using a piece of food. Try not to do this too often, the dog may become dependent on the lure. As the dog progresses, you raise your standards so more is expected for them to get the click. You then start to switch from the continuous reinforcement to a variable reinforcement (clicking randomly, usually after 2,3 or more offerings of the correct behavior). When your dog does not know when they are going to receive a click they will put forth more effort.
There are 4 main parts of the the training called “Operant Conditioning”.
Positive Reinforcement: Adding something the dog values (food, toys/play, petting, praise) to increase the probability of the behavior happening.
Negative Reinforcement: Removing something from the dog to increase the probability of the behavior happening.
Positive Punishment: Adding something the dog dislikes (collar jerk) to inhibit behavior.
Negative Punishment: Removing something the dog wants (such as your attention or the opportunity to earn rewards) to decreased the behavior from happening. This is a good method to refocus an inattentive dog.
Always remember to make sure that your dog is physically capable of the behavior you want. Keep your training sessions short so that you keep there attention. Use rewards your dog likes not what you want him to like. Split behavior into small parts for the most success. If something you are trying is not working, don't blame the dog, break the behavior down so they can understand it and be successful. Up the criteria when your dog is 85% successful at there current level. And most importantly, if your dog is confused, frustrated or has made repeated attempts that are “wrong” have them do something they already know and reward them. If they keep trying, are getting it wrong and not getting rewarded, they will stop trying.
Above is a picture of Stella and her daddy Janus before we left CO. Janus had to be put down last December for bone cancer. Shiloh, my mom's Shepard mix who went to training , but still thinks the rules don't apply to her and Cody, my mom's golden who was trained using the clicker method and now competes in agility trails.
posted by Adrienne at 9/07/2007 10:53:00 PM