Can Positive Reinforcement Work to Stop Annoying Behaviours?
People often complain that positive reinforcement doesn't work for their dog's situation. They will say that their dog finds other things more reinforcing than what they have to offer. The reality is that many people who say this try offering a treat once or twice once they have already put their dog in a situation where they are emotionally aroused and incapable of learning or they expect complex behaviours before they are willing to reinforce. The dog is set up to fail and the entire learning process is poisoned. This photo is an example of how positive reinforcement can compete even with other internally reinforcing behaviours if it is applied generously and consistently: Cruz can be seen walking in line behind my husband, with two other dogs behind him and two up front (Rosie right at the back). Now Cruz is an intact male, and as wonderful as he is, he does rather enjoy "sharing the love" and if given the opportunity will happily hump certain other dogs. At the start of our walk we could see he had a gleam in his eye for a couple of the dogs that joined us yesterday, so we kept him on lead at first (so he couldn't rehearse the behaviour) and literally reinforced every single contact or interaction with the other dogs that wasn't humping: if a dog walked past and he ignored them, "Good boy Cruzie" and treat , if he sniffed a dog, "Good boy Cruzie" and treat, if they ran up to him, ran past him, walked shoulder to shoulder, sniffed a smell together, same thing - verbal praise (serves as a marker and secondary reinforcer) and treat. The result is that instead of rehearsing inappropriate or annoying behaviour, Cruz gets to rehearse lots of appropriate and polite behaviour and be reinforced for that. After a while, as you can see here, he can walk off lead with the others and even have them run and push past him without him taking the opportunity to hump. In fact, he will often look at them and check in with us for a reward. In my book, that's pretty awesome - instead of a dog that has learned to fear punishment around other dogs, Cruz has learned to anticipate good stuff from us. Calmness around dogs he fancies is something that I will ALWAYS reinforce heavily - I really don't even think along lines of "will I always have to have treats?" The reality is that humping will always be internally reinforcing for Cruz, so if I want to compete with that, I have to maintain a fantastic reinforcement history for not humping. Positive reinforcement does work - it just requires a mind-shift. It requires us to be proactive rather than reactive. Instead of responding only when the problem behaviour occurs, we need to learn to actively and consistently reinforce the absence of the problem behaviour i.e. we need to reinforce the choice the dog makes not to do whatever it is we find inappropriate - because that is the choice we want them to keep making. It's a simple concept, but one we as human beings struggle to grasp!