I recently read an article that was titled “The Punishment of Positive Only” and was very intrigued by the title. I’m always keen to learn something new and I think it is always interesting to see other perspectives. My concern with the article is that the author doesn’t really seem to understand what a punishment is or how to implement them effectively.
A punishment in dog training terms is just something or an outcome that is designed to reduce the likelihood of the preceding behaviour occurring again. In the article the author has also changed the word ‘punishment’ to ‘correction’ to make it sound a little softer to the regular audience. I like to think of myself as a reward based positive type trainer. Ultimately I am a dog trainer like every other trainer, but that is the description I think is best fitting for my style of training. I would definitely not describe myself as a positive only trainer because that is very misleading. Positive reinforcement is just one way to increase the likelihood of a behaviour reoccurring. I use it all the time but I also use punishments. However, this doesn’t mean these punishments have to be painful or scary. Here is an analogy in human terms that I like to use to help people understand how effective punishments work.
When I first moved to Wellington, I enjoyed trying out the restaurants that had been given good reviews. I won’t name the restaurant but one was noted for having great steak. I ordered a rare steak but when it came it was well done, which was the worst thing that could have happened to me at the time. I haven’t been back since that ‘incident’.
That was the biggest ‘punishment’ I could have gotten from the chef. If he came out and started shouting, screaming and arguing with people next to me, but the steak was great I would have gone back – because the desired outcome would have been worth it to me. The chef could have been the nicest person ever, yet if you give me a well done steak, I probably won’t return.
The same can apply to dogs. If you try to give your dog something they don’t want in replacement of something they do want, the chances are they are not going to be too interested in doing what you want them to do. If you call your dog away from their mates and then pop the leash on could you have given them a more effective punishment than that? Probably not. You’ve called them away and stopped their fun. The dog is never going to want to come back to you again!
The author talks about popping on the leash to stop chasing behaviour which can ‘work’ temporarily. But what if the dog is off leash and they’ve learned the leash pop can only happen when the leash is on? The dog will probably chase – so the leash popping isn’t really that effective. I like to teach a Premack type recall, and it’s the one I have found most effective. I teach the dog that in order to do something they really want to do, they have to do something they don’t particularly want to do first. For example, in order for the dog to play with other dogs (which they really want to do) they have to come back to me first (something they don’t always want to do). And I teach this by using rewards and punishments. But the punishment in this case is just removing something the dog wants. I do not add anything horrible if they get it wrong. So here, I have taught a reliable recall that I do not need a leash to pop them with – In this case, I would use a leash, but only initially and only for management, not to cause any pain, pressure or discomfort. Again, I didn’t teach that using ‘positive only’ – I did also use punishments.
The author also thinks you need excellent timing which is true to a certain extent. You do need excellent timing to teach very specific behaviours using positive reinforcement. You will however, need even better timing and consistency without fail to be successful with punishments such as leash popping. For it to be effective, you will need to leash pop every time your dog does something you don’t want them to do. And I mean every time for it to be effective.
The beauty of using rewards is that for them to be most effective you don’t reward every time. So if you forget, it doesn’t matter! Whereas if you forget to ‘correct’ your dog, it could be worth it next time for them to resort to the behaviour you don’t want as there is a chance the leash pop won’t happen. Again, your timing has to be perfect to punish at the right moment. Let’s say your dog happens to start pulling and then you leash pop 3 seconds after they have started pulling. By now, your dog is looking at a child when the leash pop happens, and there’s a chance that your dog has not associated the leash pop with the pulling but with the child. Therefore, children can predict bad things. If, on the other hand, you reward your dog with a piece of liver at the wrong time what’s the worst that could happen? Your dog thinks a little child was the reason for the liver.
Let’s not forget to talk about when your dog barks or does something that is caused by stress. How about then, is a quick leash pop okay or even beneficial? The author mentions dogs ‘correct’ each other all the time so maybe then we should do that and ‘correct’ them. Firstly, I have never seen a dog ‘correct’ another dog with a lead. I have also never seen a dog force another dog to an environment that they are uncomfortable in and lastly, I have never seen a dog ‘correcting’ another for being anxious or scared.
Seeing as we now understand that when dogs react it is usually because they are uncomfortable in an environment, does it make sense to punish or ‘correct’ a sentient being that is scared? Would we do that to a child that simply says ‘I’m uncomfortable’. No, we wouldn’t. Even if they aren’t behaving out of fear, let’s say excitement, science can tell us that hitting or yelling is counter intuitive. We need to teach them how we want them to behave, not punish them due to our lack of teaching.
The same applies to dogs. If they are scared then it is our fault for pushing them too far and not noticing them say ‘I’m uncomfortable’. If they are doing things we don’t want them to do out of excitement and not fear, then we need to educate them. We can teach them what we want them to do with rewards and we can teach them what we don’t want them to do with punishments. But do not forget, punishments DO NOT have to be painful or scary to be effective.by