The principle goes like this:
- You can increase the frequency of any low probability behavior by making it contingent upon a high probability behavior.
The principle supports rewarding the dog with what they most wanted in the first place; it tells us to use the environment to reward the dog. For example, if your dog loves to swim in the lake, do your obedience exercises near the lake and send the dog for a swim as his reward. Or, if you are working on recalls, send your dog into the lake, call him out again and then send him back in as his reward for coming.
Positive training can too easily descend into stuffing your dog with food. It is good to remember that there are lots of things out there that are equally, or more, rewarding than food. Garrett publishes a list of possible reinforcers and suggests that you mark them in order of your dog's preference.