When I am teaching new clients how to work with their dogs and address behavior issues, invariably the issue of punishment comes up. It is a natural and normal response to want to STOP the dog’s unwanted behavior by DOING something TO them in order to make them stop. I’d like to talk a bit about why I don’t use punishment, and what I mean by proactive training.
First, some science-y stuff.
The definition of punishment in the behavior analysis community is this: anything that stops a behavior. Putting any humane or ethical considerations of using punishment aside for a moment, let’s consider why punishment is difficult and often ineffective when training our dogs. First, the punishment must happen immediately as the unwanted behavior is occurring, and the dog needs to recognize that the punishment is meant for the very specific behavior you want to stop. So for instance, if you jerk your dog off of someone when they jump, or knee the dog in the chest, the dog may or may not associate the punishment with the jumping. The dog could make the wrong connection – maybe it was meant for the friendly approach, for instance. And then you’ve not punished the jumping but instead punished your dog’s enthusiastic and friendly greeting behavior. Second, the punishment should be harsh enough to need only be administered once. Very often what happens is the punishment doesn’t work, so you gradually increase the harshness hoping a stronger leash jerk will work this time. And when it doesn’t, jerk harder. But the dog becomes acclimated to each punishment given – his tolerance level goes up and up. So you jerk harder, but your dog becomes even more resistant. And you spiral down this hole of harsher and harsher punishment.
For some dogs, for punishment to be effective, you’d need to be so harsh as to unintentionally cause physical harm to them. This isn’t something anyone wants to have happen to their beloved dog.
Many people mistakenly believe that for issues like aggression, you need to get physical. The opposite is actually true, however. Especially for aggression, punishment is a dangerous road to start down. Here’re some reasons why. When you punish a dog harshly enough to get aggression to cease in the moment, you cause a dog to shut down behaviorally (and often times emotionally as well). The dog is not aggressing in that specific instance, so, ok you assume this is working. This might seem like a win in the moment but from a learning perspective, all the dog has gathered is that if they react aggressively in that one given scenario, they will get punished. In the mean time, the dog’s emotional state that was driving the aggression to begin with (most of the time it is fear) has not been addressed, and so there is a slow building pressure-cooker scenario happening in that dog’s mind. In another situation, they will aggress. Oftentimes, dogs punished for aggression will end up reacting with MORE force and greater severity, in a less predictable way (i.e. the dog is shut down, and then all the sudden reacts aggressively), and/or they will start showing other behaviors like intense fear, mouthing, submissive urination, and more.
For punishment to work here, you would need to punish the dog in every possible scenario that dog may ever be in, with every single possible stimuli (environmental trigger) that could possibly show up in their life. And that isn’t even counting all the fall out behaviors that could occur in the place of or in addition to the aggression. The dog is not learning any acceptable, alternative behavior and so flounders around trying to figure out what the heck is actually wanted or safe to do. If you think about punishment as the amusement park game where you whack an object that pops up, goes down, but then pops up in another spot, so all you keep doing is whacking and whacking but never getting anywhere, you start to see how punishment is not the most effective way to deal with problem behavior. You’ve created an unpredictable dog who will surprise you with aggression, and one who is no closer to actually learning appropriate behavior.
Another important consideration when using aggression: you can very easily create “fallout” in the form of unintended associations. For example, you are walking your dog down the street, and he sees another dog. The dog is being walked by a lady and her small child. Your dog reacts aggressively as usual to a strange dog, so you give him a strong, harsh jerk. Later, on another walk, your dog spots a child holding their mother’s hand standing by the sidewalk, and your dog – who never reacted aggressively towards children before – has an outburst. What happened is this: the leash correction you gave early was not associated with the dog’s own behavior, but rather the child that the dog saw at the time the punishment was given. Now your dog has a bad association with children and you’ve got a whole new behavior problem to address. This scenario can happen in as many different contexts as you can imagine. The behavior you meant to punish was not associated with the punishment – instead your dog zeroed in on something else in their environment and made the wrong connection.
One last consideration in the use of punishment that may be difficult to accept: because punishment often in the moment creates an immediate response in the punished dog, the punisher – again, scientifically speaking – actually becomes REWARDED FOR PUNISHING. Learning theory teaches us that rewarded behaviors increase. This means, whether you realize it or not, the action of punishing your dog becomes reinforced and you are more likely to use punishment more often in the future. This is an insidious way the act of punishing becomes a habit, and can almost become a reflex in the person doing the punishing. No one is immune from the laws of learning, and even though dog parents never consciously wish to hurt their dogs, punishment becomes so ingrained in their repertoire of behaviors they use with their dog, that it just keeps happening over and over again, in the must unintentional, even mundane ways. But punishment is never mundane to the receiver. This is a really important point to grasp and I truly hope that those reading this post can sit and think about how they may have unintentionally learned that punishment is the casual, go-to reaction to their dog’s behavior.
The great news is that we do NOT need to use punishment – positive punishment in learning theory jargon, which means we do something TO the dog to stop a behavior – is NOT necessary.
The way I work with dogs is to take a proactive approach. Instead of waiting for the wrong behaviors, I focus on what the dog is already doing right and reward those behaviors. For behavior issues, I work on teaching dogs something ELSE to do, and specifically with aggression, help the dog feel better about the thing that makes them think they have to aggress in the first place. It is much easier to teach a dog something TO do as opposed to trying to punish every unwanted behavior that pops up (whack-a-mole approach to dog training). The rewarded behaviors will increase, the dog’s attitude will improve, and before you know it, the unwanted behaviors have been pushed aside by the new ones that your dog has learned, and you have a happier dog – and a happier you.