It has been a law of mammalian behavior since Pavlov and Lorenz and other behavioral scientists, studied, experimented and proven that if there is a pay off for any given behavior, the likelihood of that behavior being repeated is greatly increased. In addition, if there is no payoff for a given behavior, the likelihood of that behavior of being repeated is greatly reduced.
The trouble is....why some of you and many trainers/owners don't buy this is because you are NOT seeing that payoff. It can slip past us all if we're not looking for it or thinking. If a dog keeps doing something that you don't like, look, think, try to see where is he being rewarded...how is he being rewarded.
Think of your own behavior. What do you engage in repeatedly that you don't get any benefit from what so ever? But, before you answer that, think long and hard and make sure you're not missing something. What do you engage in often that you do get something good from?
One more little example of something my Dobe has learned fairly recently: He was running through my vegetable garden and making a mess. I didn't scream at him or scruff him, yank him or tell him "NO." He didn't know any better. And his payoff was that it was fun to run through there. I taught him an incompatible behavior and that was "out." How can he be in the garden if he is out of the garden? I showed him to stay out on the grass by walking close to him and coaxing him out.... and gave the word, "out." Everytime he would step into the garden, I said, "Out" and when he stepped out, I'd reward profusely. Even when he wasn't in the garden, but maybe walking along side it, I'd say, "Gooooood Out." And reward. He doesn't go in the garden at all anymore.