One of the things that I love about blogging is its interactive nature. My last post spawned a great conversation about training methods, the “right” way to train, and the role of punishment in dog training. The thread really made me think about my training philosophy, so much so that I felt the discussion deserved its own post.
Let me start by saying that I think it's impossible to use 100% positive reinforcement. Personally, I probably use approximately 85% positive reinforcement, 10% negative punishment, and 5% positive punishment. (If you're not familiar with these terms, this link does a nice job of discussing them.) This makes me a decidedly lopsided trainer, and while I do use primarily clicks and treats, I'm not afraid to use the occasional “correction.” My “corrections” trend pretty light because of who Maisy is- sensitive and anxious- but also because I believe that it is possible to train without using pain or fear.
Does this mean it's wrong if someone trains differently than me? No, of course not. If a given training method is getting the desired results, if it's fair and consistent, and if it's improving your relationship with your dog, then who am I to say that it's wrong? However, I stand by my statement: I believe you can train a dog without using pain or fear.
This does not mean that I think training should be one-size-fits-all. Frankly, I think that's impossible. Every person, dog and situation is different, and as such, needs a different approach. Further, because the dog defines what's reinforcing and what's punishing, it's impossible to make blanket statements about whether or not a technique is acceptable. For example, some dogs find being sprayed in the face with water incredibly aversive. Maisy happens to love it. If I were to try to stop her from doing something by using a squirt bottle, it probably wouldn't be very effective.
The cool thing about positive training is that there are many different ways to teach the same behavior. If I want my dog to go lie on a mat, I can shape her to do that by clicking small movements in the right direction, I can lure her to do it by tossing a cookie on the mat, or I can capture it by waiting for her to go to the mat herself. More than that, though, I can also manipulate the environment to make it more likely that she'll go to the mat by placing it in the hallway and wait for her to walk down it, or putting it in her favorite napping spot, or putting it in her crate... the possibilities are limited only by your own creativity.
I definitely believe that teaching new behaviors can be done with almost completely positive reinforcement. However, stopping an already existing and unwanted behavior is much more difficult. Personally, I call this type of training “behavior modification” but perhaps that's splitting hairs. Regardless, my approach is to try positive methods first, and if those fail, use punishment in a thoughtful, fair, and consistent manner. The punishment used should be the least invasive and minimally aversive option possible.
I think we can find pain and fear-free punishments that will stop a behavior, however, I will concede that there are situations where we might not have the time to figure out how to do that. Life-and-death situations like a dog trying to eat something poisonous or running towards a busy highway are no time to futz with a clicker. You do what you have to do in that moment, even if it hurts or scares your dog. Of course, I don't think that's training so much as damage control, and once your dog is safe, you teach a stronger "leave it" or a better recall to prevent the situation from occurring again. It should go without saying that I would do that training without pain or fear.
This leads to the question of whether it's ever necessary to use pain or fear in training. Philosophically, I'd say no, but I will grudgingly admit that there are times where using pain or fear in training is the lesser of two evils. For example, if a dog is going to lose his home because he's barking too much while his owner is gone, a bark/shock collar is probably the better option. Do I like this? Of course not. In fact, it makes me really uncomfortable to say it, because I believe there's always a “positive” solution. Unfortunately, people don't always have the time, money or knowledge to find it.
All of which is to say that while I think some training methods are better than others, I recognize that those other methods that not only work but that they might make sense in a given situation. I may not like those methods, I may think they're unnecessary, but I'm trying hard to avoid judging people who use them. I haven't walked in their shoes, and I don't know what they're up against. I will speak out against abuse when I see it, but the rest of the time, I will offer support or suggestions when appropriate. This only makes sense. After all, my goal is to be as positive with people as I am with my dog. By my own calculations, that requires offering them positive reinforcement 85% of the time!
So, let me take this opportunity to positively reinforce everyone who comments on my blog. I appreciate you all, even those of you who disagree with me. Perhaps especially those of you who disagree with me, because it forces me to examine what I've said. Sometimes it strengthens my convictions, and sometimes it causes me rethink my position. Either way, I learn and grow as person and a trainer, and for that, I am grateful.