That dog sure has you trained…
This is a common statement I hear a lot from folks who don’t approve of rewards or don’t understand what a handler is doing when they are rewarding successive approximations.
The reality is that any good trainer has been well trained by their dog—one way or the other.
Many of us go through successive approximations to achieve better and better behaviors with a final goal in mind. Along the way it can look like the handler is rewarding crappy behavior—while in fact we are rewarding better-than-before behavior knowing that it will keep getting better and better because we will gently shift our criteria for rewards to slightly improved behaviors, little-by-little.
In other words, if I want a dog to settle while their favorite thing on earth is nearby (say kids playing ball) I have two choices:
- Pay the dog for the settle in a currency that is greater in value than the kids (treats or cheese…depends on the dog) and keep the payment going in reasonable intervals that will allow the dog to make the better choice and stay in the settle (here we have positive reinforcement);
- I can keep collar correcting the dog every time he tries to reach the children and then reward the dog with the cessation of correction when he lays back down (here we have two quadrants in play—positive punishment/collar correction; and negative reinforcement/stopping the collar correction when the dog performs properly).
I would submit that both handlers are being trained by their dogs and in the end both handlers who have achieved a settle with their dog in close proximity to their favorite thing have done so because they learned what their dog required. For example, the positive reinforcement trainer knows their dogs well enough (has been trained by their dog well enough) to know how often their dog will need rewards to remain in the settle. The positive punishment/negative reinforcement trainer knows their dog well enough to know when they need to correct them and how much they need to correct them (has been trained by their dog to know what level of correction the dog needs to comply and when to stop the correction).
In the end any trainer who achieves reliable behaviors has been trained by their dog to know what that dog requires from their handler. The real question is not who is training whom—it is about the choices we make as the caregivers of these animals.
Positive reinforcement is a back and forth negotiation that allows for the dog to say, “This is too hard for me so you need to pay me a bit more.” Or, “I really don’t care for this but if you pay me well I will do it.” This is not unlike when we look at our boss and tell them we will need time-and-a-half to work on a holiday. With appropriate payments the final result will be a great settle that has not negatively effected how the dog feels about performing the well-paid-for settle in the face of the children?
Positive punishment/negative reinforcement training it is not about negotiation but about forcing the dog to comply. Here the dog can once again say, “I really don’t care for this thing,” but the only thing they get in return is the knowledge that they will face a punishment for not wishing to comply. This is not unlike the boss who tells the employees they will have to work on the holiday and if they do not, they will be fired. The final result in this instance is a dog who learns to comply but has a negative association with their handler (boss) and with the children they previously wanted to go see because whenever the dog tries to engage with the child—he/she is punished.
Everyday we train our dogs we get to make a choice about what kind of relationship we want with our dogs and what kind of relationship we want our dogs to have with the world around them.