How Does Clicker Training Work?
Thursday 19 January 2012
It seems amazing, that with a simple little device horses can be taught to perform tricks of all sorts. With nothing more than a little click as a reward, horses bow, nod, and answer numeric problems. But how is it done?
To understand clicker training, it is important to understand the theory of replacement. When training a horse, you often start out with one aid, and gradually replace it with another. For example, when you teach a horse to walk forward, you might start out by tapping it on the quarters with a whip. The horse responds to the tap by moving away from it. Then, you add the verbal cue “walk on”. As the horse connects the words to the tap of the whip, you can remove the whip and have the horse move forward from the verbal cue alone. Once you get on the horse, the process is repeated, adding a cue from the leg to the verbal cue, and eventually taking away the voice altogether.
Clicker training works the same way. Most trainers start out with a treat reward. They ask for some kind of trick, and reward the horse for success with a treat. A bow is taught by offering a treat between the front legs. The horse reaches for the treat, gradually progressing until he reaches far enough to bow.
Now trainers add the clicker. When giving the treat, the trainer also clicks the clicker. The horse begins to associate the treat reward with the click of the clicker. After a while, the trainer stops offering treats every time, gradually replacing them entirely with the sound of the click.
Once the horse understands that the click is the reward, he can be persuaded to learn new trick with nothing more than a clicker as a reward.
The entire process is simple, and any number of alternate rewards could be exchanged for a clicker. Either way, the horse learns to understand that the reward is no longer a physical treat, but something that is apparently unrelated.