Reinforcement Contingency refers to the direct relationship between the properties (approximations) of a behavior and the reinforcement that follows.
Reinforcement contingency allows trainers to work all aspects of a trained behavior. Reinforcing the entirety of a behavior (from start to end) is very important, but maintaining a behavior and preventing break down is equally important. Behaviors are maintained by reinforcing all aspects (aka steps or approximations) of the specific behavior.
For example, when sending Hope on bows (jumps), we as trainers can work on the different approximations like her run (agility to get moving up out of the water), the height of her bows, number of bows, her body position when reentering the water, etc. By reinforcing specific approximations, this communicates to Hope specifics of the behavior like the speed, height, or body positioning. In the end she will be able to put all the parts together to create a complete picture, thus shaping and maintaining the behavior “Bow” as a whole from start to finish. This way, when asked for bows, she will be able to consistently give a desired speed for her run, height to her bows, reentry to the water, etc.
Not only does reinforcement of specific aspects of a behavior strengthen each part of the behavior, it also keeps training variable and unpredictable, which can strengthen the behavior as a whole and keep the animal engaged.