Training Science Articles:
What is Clicker Training?
Terminology
The Science Behind Learning
Creating Behavior
Reinforcement
Cues
Eliminating Behavior
Stimulus Control
Building Behavior Chains
The ABCs of horse training
Stimulus Stacking
Dealing with Fear

Putting it into Practice, Ethically:
What is an Aversive?
Over-excited about Food?
Enrichment
Protective Contact
Control vs. Communication
Transition Troubles
Making it a Lifestyle
"Un-ridable"
The Importance of Choice


Behavior Chains


Once a behavior is taught well through R+, something with a strong reinforcement history, the behavior becomes so pleasant and happy to the horse it actually becomes reinforcing itself. Being told to do that behavior is not just a good thing, but a reinforcing thing. Targeting or stationing on a mat are the most obvious of these reinforcing behaviors. When we ask a horse to do a behavior that might be difficult or not as fondly enjoyed, we can add a cue to do a simple, well loved behavior to reinforce the last. This is taking advantage of the Premack Principle.

You can use this priciple to build behavior chains. For example when riding you don't want to stop and reinforce every step, or every behavior the horse does. You can ask for "walk on", "turn left", "woah" then click and reinforce. If each behavior has been taught individually, thoroughly and are enjoyed by the horse, each cue for the next behavior will reinforce the previous behavior. You could do "turn left", "woah", "back up" then click and reinforce. Depending on your horse and the reinforcement history and your mission - you can adjust these behaviors to make any chain you'd like. If a behavior starts to slip, not perform up to par, spend some time ending on that behavior or reinforcing it directly.

Some chains though we want to be "automatic". We don't want to have to cue each individual step - we want it to be one fluid behavior. The best way to do this is what's called "back chaining".

This is where you teach the last behavior on the list first - usually the first behavior we teach is the one that will be repeated the most and thus be the most heavily reinforced. So if you teach each step backwards, each new behavior is reinforced by an easier old behavior that's already pleasurable.