Many people train dogs without understanding exactly why what they are doing works. My toddler trains our dogs constantly without knowing anything about training techniques or the science of behavior. For example, he has trained them to hold a down-stay (i.e. stay for a long period of time in a down) under his chair whenever he eats. How has he achieved this feat at just 18 months of age? By intermittently dropping food on their heads, of course! Granted, this was not an intentional training of the animals, but it was training nonetheless. Understanding why even toddlers can train dogs can help all of us be better and more thoughtful trainers. The four core principles of behavior are:
- Reinforcement (positive and negative)
- Punishment (positive and negative)
- Stimulus Control
Today I’m going to delve into the world of positive reinforcement. This is a term that is used constantly in dog training- people are often termed “positive reinforcement trainers.” But what does “positive reinforcement” actually mean?
Definition: Positive reinforcement is when you add something to dog’s environment that increases the future frequency of the behavior. Adding something is what makes it “positive,” and the increase in behavior is what makes it “reinforcement.”
In other words, when my son drops food on my dogs’ heads, he is adding something (food) to the dogs’ environment that is increasing the length of their down-stays in the future. In short, he is positively reinforcing their down-stays. We can give our dogs loads of cookies for just about anything, but that isn’t enough to say we are “reinforcing” their behavior, unless it actually increases the future occurrences of the behavior.
How does this impact dog training? Well, if you are trying to train your dog to sit, and giving them a Milk-Bone each time they sit, but they are not actually getting better at responding to the “sit” cue, it could be that the Milk-Bone is not actually reinforcing the sit- it’s just a tasty treat they’re eating. Maybe by switching the treat to something stinkier like freeze-dried liver, you might have more success. If each time the dog sits on cue, you give it a piece of liver, and the dog starts responding to the cue more frequently, you will know the liver is actually reinforcing the behavior.
This means that positive reinforcement is in the eye of the beholder. If your dog LOVES tennis balls, asking for a down and then throwing a ball (adding the tennis ball to your dog’s environment) could be positively reinforcing. If your dog LOVES to tug, asking for a sit and then letting them tug could be positively reinforcing for a sit. If, on the other hand, your dog doesn’t care for toys, giving them a ball or tug in exchange for a behavior won’t increase the future frequency of the behavior and is therefore NOT reinforcement. This shouldn’t discourage you from playing with your dog; it just might not be the best way to train a specific behavior.
Although clicker training is a form of positive reinforcement training, I call myself a behavior analyst rather than a “positive reinforcement trainer,” because my training techniques are not limited to positive reinforcement. Positive reinforcement using treats or toys is often the best and most enjoyable way to train your dog, but it is only one of numerous humane techniques you can use to change your dog’s behavior. I will discuss more of those techniques in the upcoming posts.