Last month I explained the different operant conditioning quadrants. This month I'll dive deeper into positive reinforcement and punishment and explain why I choose to work the way I do.
With positive reinforcement, you add something of value to the learner so that a behavior is more likely to repeat. For example, when I call my dog and he comes, I reinforce that behavior by giving him a high value treat. Next time he's more likely to come when called. Because Ralphie has a long reinforcement history for coming when called, he's got a solid recall. Also he's a very good boy.
In addition to being the most humane way to train, positive reinforcement has several other selling points.
Most of our dogs live completely controlled lives: we decide when they go for walks, when, what, and how much they eat, and where they are allowed and not allowed to be. In most public spaces, there are laws requiring us to tether our dogs to ourselves with leashes, which prevents them from moving in a way that is more natural to them (meandering, running ahead, doubling back, changing directions, stopping to sniff, sniff, sniff). We often curb natural dog behavior like barking at people passing our house or growling when they are uncomfortable (don't punish your dog for growling--it's an important communication!). The very least we can do for our dogs is give them directions on how to live in our human-centric world. With positive reinforcement, we do that by teaching our dogs what to do and rewarding them when they do it, making it easier for them to understand our expectations.
Because the learner decides what is reinforcing, when we train with positive reinforcement, we take into account our dog’s preferences. My dog finds all food reinforcing—sometimes I train him using his kibble and he is an avid participant. But if I offered him pets and attention when he comes when called, he might stop responding to my request. So when we determine what the dog in front of us finds reinforcing, we think about who they are and what they like rather than simply expecting them to do things because we told them to. That kind of consideration can build your dog's confidence and strengthen your bond with him.
Positive reinforcement training can also help create happy associations to the world because you are reinforcing behavior with something the dog enjoys. Over time, your dog will learn that in many different circumstances (as many as you introduce her to during training) good things happen with her person. That’s why I use easy pattern games when I’m working with dogs who react strongly to their environment and things in it whether from fear or frustration or another emotion. The games provide a rapid rate of reinforcement (lots of treats in a short time period) that in turn can create happier associations.
Finally, positive reinforcement, as the American Veterinary Society of Animal Behavior points out, has been shown in multiple studies to be more effective than punishment in training. I can't imagine using any other methods with my dog or my clients' dogs.
Comments