Reinforcement, What Exactly is That?
My dog, Huntington sits when cued, I click and deliver a treat to him. He takes it with some hesitation, slowly chews some of it and lets some of it fall on the ground. He was reinforced, right?
Well, maybe or maybe not. If it had been a piece of chicken or bacon, he would have taken the offered treat without hesitation and no crumbs would have been left unattended on the ground.
In both cases, I may feel like Huntington was reinforced, after all, he took the treats. The question isn’t about what I think though, it’s about what Huntington thinks! He doesn’t speak in human language, so I can’t ask him if he thought he was reinforced. So how do I know if he was reinforced?
The answer lies in future behavior. Does he repeat the behavior? Does he repeat the behavior as quickly or is he taking much longer to start to sit? Is the behavior deteriorating? If the behavior is repeating, i.e. Huntington sits when cued, those treats were reinforcing. If Huntington’s cued sit behavior is starting to deteriorate, taking longer to start to sit or sometimes not sitting at all when cued, those treats probably weren’t reinforcing.
A lot of dogs bark at the mailman. He walks up to the house, puts mail in our mailbox or mail slot on our door and then walks away to the next house. That’s his job and he may or may not give any thought to the dog in the house barking and snarling. To the dog in the house, it could be a scenario where this man walks up to my house, I bark to scare him off as he is doing something to the house or shooting all this paper into the door slot. Look, he moves away from the house as I tell him to leave in my “protection” voice! Job well done, I protected the house! Then tomorrow comes and the mailman is just doing his job, but to the dog it could be “oh my gosh, here he comes again! I must need to get out my scarier voice to make him leave”. The barking and snarling get more animated, and the mailman, having delivered the mail, walks away. The dog pats himself on the back for a job well done. He, once again, protected the house from that man!
In the mailman scenario, the dog will only get louder, snarl more and perhaps work himself into a frenzy as days go on. Why? Because the mailman leaves the house each time. To the mailman, he is just going to the next house. To the dog, he scared the potential intruder away! The mailman leaving is the reinforcement for all the barking and snarling. It is highly likely, that the dog will continue to bark and snarl when the mailman comes to the house each day.
When dealing with behaviors we would rather our dogs not do, it is important to discover what is reinforcing the behavior. If your dog is repeating the behavior it must have some sort of reinforcement. The laws of behavior tell us that organisms behave to either get something they want or to get away from something they don’t want. Those “somethings” are consequences. If the consequence is reinforcing, the behavior is likely to increase, if the consequence is not reinforcing, the behavior decreases.
Sometimes recognizing what is reinforcing behavior may be obvious, a lot of times it takes some investigative work. Why does your dog bark in training class? Perhaps he just likes to hear himself, perhaps your hand is in the treat bag prior to the behavior being performed, perhaps you haven’t marked/clicked the behavior, but rather just handed out treats, perhaps you are giving attention to your dog that you don’t even realize you have been giving. Sometimes, to the dog, the behavior itself is just fun and therefore self-rewarding. Once we figure out what is reinforcing the behavior, we can begin to change the behavior.
We change the behavior by training or cueing a different behavior that will be able to gain the same reinforcement as the behavior we are trying to change. If we don’t like the barking in training class, we can cue a down on the mat where your dog can enjoy a chew toy or food puzzle. This can quiet a barking dog who wants to do something in class when we are listening to the instructor. By giving your dog the chance to earn the reinforcement he is seeking in a behavior that we find appealing, he won’t need to perform the unwanted behavior.
The other side of this coin is that the unwanted behavior’s reinforcement must cease. Again, because you have investigated and discovered what the reinforcement for this unwanted behavior is, you can now manage the environment so that reinforcement doesn’t occur when the barking in class happens. Without that reinforcement, the behavior should begin to decrease and eventually stop.
One last thing to keep in mind. Most behavior we humans don’t care for is normal dog behavior. Dogs don’t think up behaviors to irritate us. When your dog performs a behavior you find unappealing, find something else for your dog to do instead. Saying “no”, “stop” “quiet” doesn’t give the dog any information about what you would rather they do. Train a few alternate behaviors such as nose touch to your palm, sit, down and spins. Cue these behaviors in situations where your dog is likely to behave in a manner that you find unappealing and reinforce the performance of the alternate behaviors. You should see an increase in the alternate behaviors and a decrease in the unappealing behaviors.
It takes knowing what the reinforcement is, patience and consistency to change behavior. With a click and a treat, you can be on your way to behavior you want to see!
Jan Gould, KPA CTP
KPA, CTP: Karen Pryor Academy, Certified Training Partner
Knowledge, training and teaching assessed.