If only I could be as clear, unambiguous, and have excellent timing like Ranger when I train. He really is a master and I learn a lot from him.
Recently we went for a walk with a friend and her dog. She has a young cattle dog typical of the breed. His favorite game is chase and herd and he barks at the head of the animal he's trying to turn, in this case Ranger. Ranger loved the off leash running in the woods and playing herding games with his pal but Ranger does not love having someone bark in his face. Because he doesn't like having someone bark in his face Ranger set about training his pal. Running, chasing, cutting each other off, and even stare downs were fine and Ranger would happily engage in herding dog play but bark in his face and Ranger immediately stopped engaging with his friend. He wouldn't run, he wouldn't chase, he wouldn't even look at his playmate. He would simply freeze and pretend his friend wasn't there. As soon as the barking would stop Ranger would re-engage. They had maybe half an hour of off leash time together. In the beginning there was a lot of barking by the end there was hardly any.
Let me recap. Ranger knew exactly what behavior he wanted to end--barking in his face. Ranger rewarded any play behavior that wasn't barking in his face and Ranger never rewarded the behavior he didn't like. His playmate got reinforced for all play behavior except barking. The barking was nearly extinguished in half an hour of play. What Ranger did not do was use positive punishment to end the undesired behavior.
I need to take a brief digression here to talk about punishment and reward as defined by behaviorists. Behaviorists assign things to four categories, positive reward, negative reward, positive punishment and negative punishment. Please, remember I am not a trained behaviorist but as I understand these categories, positive reward is adding something to increase the chance of the behavior being repeated--Ranger playing with his pal for any behavior that wasn't barking was giving his friend a positive reward, play, for playing that didn't involve barking. Negative reward is removing something to increase the likelihood of the behavior being repeated. When Finna glances away from one of her triggers and I immediately take her further away from that trigger I am giving her a negative reward--something that she found scary is not as close anymore. Positive punishment, adds something to decrease the likelihood of the behavior being repeated--if Ranger had growled at his playmate to get him not to bark that would have been positive punishment. Negative punishment removes something to decrease the likelihood that a behavior will be repeated. Ranger removed his attention and engagement from his friend in response to barking. The calculation for his friend became simple and went something like this, "Bark and Ranger won't play with me, don't bark and Ranger will play with me. I like to play with Ranger so I won't bark."
Ranger is easily twice the size of his cattle dog pal so physical intimidation would be pretty easy for him but Ranger didn't use positive punishment. He used positive reward and negative punishment. His friend was given clear feedback and could make his own choices about how he was going to play with Ranger. If barking was the only way the pal could play he could have kept barking but he wouldn't have had a playmate. Rather than shutting his friend down with a growl or snap every time he barked Ranger provided a clear choice don't bark and we can play or bark and we can't play.
I don't consciously use positive punishment in training Ranger and Finna although I'm aware that sometimes they may interpret thing that I do, such as hugs, as positive punishment. I do my best to be clear and consistent but after watching Ranger training his friend I'm even more aware than usual that my timing is lousy and my consistency is slipshod. I'll just have to keep practicing and keep watching Ranger for training tips.