Dressage riders who use positive reinforcement have asked me: ‘I can’t click and reward my horse during a dressage test. So how can I practice a test and still use clicker training?’
Or they say: ‘I don’t want my horse to stop in the middle of the test because he expects a treat’ or ‘He stops because he is used to a click and treat every few minutes’ or ‘If I don’t click and treat often he stops and gives up trying’.
One possible solution to prevent this is to use ‘back chaining’.
The rider has to memorize the test. If you are a visual learner you can use a dressage-test-white-board.
If you are a practical learner you can memorize the test by walking it yourself. Make a little arena on your lawn or in your living room with letters you’ve printed out and walk the test several times until you know it by heart.
Once you know what to do you want to practice with your horse. The expression ‘chaining’ in positive reinforcement training refers to splitting the behaviour into smaller steps and train every step separately. Each step is one link of the chain.
After you practiced each link separately, you can start pairing two links together before clicking and reinforcing. If that goes well add another link of the chain before that. This is how you make a behaviour ‘chain’.
In ‘back chaining’ you also start training every exercise (link of the behaviour chain) separately. It doesn’t matter in what order. Once the horse knows all the separate steps you can start ‘back chaining’. Start to reinforce the last exercise in your chain of exercises.
Almost every dressage test ends with ‘A: Down center line, X: Halt, salute, leave the arena in free walk’.
In back chaining you start with this last exercise (free walk and exit the arena). Train the free walk consciously: click and reinforce right after leaving the arena. You can’t click and reinforce during the test, so you have to do it after the test.
Then you add one exercise before the last one (X: Halt, salute) leave the arena in free walk, click and reinforce these two links. Then add a third link before ‘X: Halt, salute’ and so on.
The power of back chaining is that your horse will anticipate and he will learn what to expect. The last part of your chain becomes very predictable and easy because it is always the same. It only becomes longer because the trainer adds exercises ahead.
In this way your horse doesn’t expect a treat during the test, but he will know at the end will be a tasty reward waiting.
The chain can also become a reward in itself: you have reinforced the last link so many times it has a really positive and strong association with something pleasurable in the horses’ brain.
If you are too predictable in your use of your bridge signal and or too predictable in the rewards you offer and the reward schedule you are using, back chaining, can backfire on you. You get the opposite result of what you want: a horse that performs worse instead of doing the best he can.
Keep in mind that you need to vary your reward schedule and your reinforcers in order to keep your horse motivated. Don’t be afraid to experiment with back chaining.
As always: start small, reward big.