Spring Training: Beginning Clicker
I had been taking our young dog to clicker training classes in the fall, and decided that it was time to try this method out on Mattie and Sam. I’ve been reading about it in horse books and on the web for some time now. It’s a form of classical conditioning–like Pavlov’s dogs, who were conditioned to respond to a bell. When I first heard about it, I felt that this was a method to take the art out of horse training, that it would make the horse’s response too mechanical, but three things have made me more open to trying it: taking Jeter to clicker class and seeing it work on dogs, learning that this method is used to train zoo animals to be handled for routine care, and, well, Mattie and Sam’s annoying habits.
I’ve done a little conditioning with them and find that they do respond to rewards better than other types of training. This seems obvious when I reflect on it–pleasure rewards more than pain. Last spring, Janet Zadina, the Brain Lady, who has branched out from her own neurological research to offer a synthesis of brain research applied to learning, came to Fairbanks to give a workshop. One thing I remember her telling us was that the brain functions better in pleasure than pain–in other words students learn more when they’re having fun and are engaged than when they’re anxious. This has profound implications for all levels of teaching, and it applies to horses as well.
Take Mattie, for example. When she arrived here, she was so anxious about simple things that she would pin her ears back and snake her neck at me when I brought hay. When working with her, I’m always reading her anxiety level and backing off a little when I seen her stressed nearly to the reaction point. Using pressure or pain to communicate with her is not only counterproductive but it’s dangerous. But with a pocketful of beet pellets–the reward is as much the crunchy sound they make as their food value–I can persuade her to do things she is worried about doing. In fact the first thing I taught her was to put her ears up on command, something I didn’t know I could teach her, like retraining an attitude. There are a few things we just haven’t really worked out yet, though, such as standing still at the mounting block. I’m hoping clicker training will help with this.
The first session with Sam went well. Nina was out for a “mentorship” opportunity, so she helped me with all the details. According to the clicker training website, it helps if you start with a target, training the horse not only to associate the clicker with a treat, but with an activity to complete before the treat is delivered. Ultimately this generalizes into a request for any activity that can be paired with the clicker.
That day, a week ago, it had warmed to 20 above; Nina and I stood in the snow, across the gate from Sam, who was haltered and on a lead rope. We showed him the clicker and he drew back, especially once we had clicked it. But we had beet pellets, so we clicked the clicker and offered the beet pellets and soon the click didn’t bother him. Then we introduced the Nancy’s Yogurt container. Nancy suggests right on the container that we should reuse it in some way, so we took her suggestion. We held it up to his nose and, when he had accidentally bumped the container, clicked the clicker and gave him a treat. He was a bit surprised to get a treat for nothing, and tried a number of things to get us to give him another one. He’s been taught to beg by turning his head away at the command, “wait,” so he tried that. He tried bumping us with his nose. We held the yogurt container up so he would bump it again, and, after a few tries, he was clearly bumping it deliberately to get the treat. Then we noticed that he would turn toward my coat pocket when I clicked the clicker–he knew where I kept the beet pellets.
We tried moving the container around and finally putting it on the ground, and he learned to bump it with his nose each time. He learned it so quickly that, as always with Sam, I had the sense that he’d done all this before. We stopped after about ten or fifteen minutes, so not to burn him out on the first attempt. When I tried it again a few days later, he remembered, and I could move the container around on the ground and he’d bump it and look at me. A good start. Now to figure out how to teach him something useful, like not bumping me with his nose while I’m leading or grooming him!
I tried this with Mattie today. She picked it up quickly, too, and would touch the container anywhere I put it. She clearly got that it was an action to get a treat, too, because she would do the behavior I’d already taught her, “ears up,” at the same time she was bumping the container. The clicker training people write about this process as like learning a mutual language that allows humans and animals to communicate. It did seem like Mattie, Sam, and I were doing something akin to Helen Keller’s first experience of connecting finger language with the meaning of water, opening the possibility of communicating her experience to others and vice versa.
More on this as spring training progresses.
Tags: brain research, clicker training, dogs, horses, spring, training horses
Mattie's Pillow 
February 19, 2009 at 12:11 am
This clicker training is interesting. Does it work on children?
February 19, 2009 at 3:55 am
Well, I’m not an expert–but we all make this kind of association. Positive reinforcement–like promising yourself chocolate after a boring meeting, or promising a kid a reward for doing homework–is the best way to teach anyone anything. Still, I think that kids’ minds are more complex than a horse’s.
I’ll add a link to Janet Zadina’s site on this blog.
I’ve just started reading a book, The Beautiful Jim Key, which is about a famous “clever” horse at the turn of the 20th century. I’m looking forward to learning the methods used to train him to spell and make change.