Thursday, February 13, 2014
How to use positive reinforcement training to stop an undesirable behavior.
Positive reinforcement training is a wonderful way of getting animals to do desired behaviors. But what if you want to stop an undesirable behavior?
Can positive reinforcement and operant conditioning be used to stop an undesirable behavior? It can, as long as the trainer has a good answer to the following question: What would you like to have the animal do instead of the undesirable behavior?
Let’s say that you’d like to stop a dog from constantly barking at everything. Instead of trying to teach the dog not to bark, an easier solution would be to train the dog using positive reinforcement to bark when asked. Why teach an already barking dog to bark on cue? Well the dog will soon associate the command to bark with a reward. It will notice that when it barks without the command it doesn’t get a reward. It soon learns to only bark when given the signal to bark. You can then gradually ask for the bark less and less until it is extinguished as a behavior. We answered the question of what we want the animal to do instead by saying that we only want the dog to bark when asked. Then we don’t ask for it again. Easy!
We used this type of training on one of our otters that had a habit of continuously rolling around in the water during a training session. This actually hindered the training process because the otter’s full attention would not be on the trainer during a roll. So the otter was trained to roll only on cue, and then eventually not asked to roll again during a session. The undesired behavior was soon extinguished during subsequent training sessions.
There are other ways to stop an undesired behavior. You could figure out what stimulus triggers the behavior. You can then ask for a behavior during that stimulus that is incompatible with the undesired behavior, but more rewarding. You also want to make the right behaviors easier to do. Let’s say that you want to stop a dog from rushing to the door whenever someone arrives. The stimulus of the people arriving triggers the undesired behavior. You can train the dog to go to a station, say a pad on the living room floor, when asked—an easy behavior for the dog to do. You then ask the animal to go to that pad whenever someone arrives at the door. Going to the pad in the living room is incompatible with the behavior of rushing the door, but more rewarding. The dog soon associates that when someone arrives at the house it should go to that pad instead of the door. Problem solved.
There is a lot more to positive reinforcement training than just the idea of rewarding a desired behavior. Perhaps in future blogs we’ll go deeper into some of the finer details of training.
Have Something to Say? Leave a Comment!
All blogs and comments represent the views of the individual authors and not necessarily those of the Aquarium.