The other day I wrote some about how much of my therapy work has in common with some of the basic lessons I learned training my dog, Lucy, in puppy class. If you missed it, you can read here about how we tend to project intent onto dogs (and people!) depending on how we feel about something they did. You can also laugh about my “purple” incident, and read about how it’s important to make sure (vs. assuming) our meaning is understood.
Today I have some stories about how we (dogs and people) learn—what helps us learn to do the right thing, consistently, and how some modes of “teaching” end up giving a different lesson than we intend.
Reinforcement, or, I Do This Because I Want To and Not Because I’m Scared Not To
Mel’s puppy training class is based on positive reinforcement, which means rewarding the behavior you desire. Lucy sits when I say sit, and I give her a treat. She sits when I say “purple,” she doesn’t get a treat. She jumps up on me, which I don’t like, but instead of forcing her down or yelling at her I ignore her and wait for her to quiet herself. When she does, she gets a treat. Pretty soon she’s not jumping at all (no reward, only being ignored, which is no fun for a puppy!) and all four paws remain on the floor when anyone greets her.
I encourage the humans I work with to experiment with positive reinforcement. Did your husband apologize for criticizing you? Tell him how good that made you feel to hear his “I’m sorry.” Did your son pick up his room? Give him a hug and thank him. Did you get up for a run this morning? Pat yourself on the back and acknowledge how awesome you are. Ignore the three-year-old’s tantrum, and give her a small reward when she calms down. Positive reinforcement can be simple, and it doesn’t need any explanation.
In dog training, positive reinforcement works best when you set your puppy up for success instead of failure. I can, and I did, train Lucy to leave food alone when I say “leave it,” but it helps to not leave food out all the time. I found this out the day I was making some cookies and left some butter on the counter to soften. I left to run an errand, telling Lucy to “leave it.” When I returned, Lucy’s entire muzzle was slick with grease and there was no butter to be found. Our training was set back a few steps as I taught her that “leave it” means “keep leaving it alone even when I’m out of the room” and she taught me “Mama can help Lucy be a good dog by putting temptation out of reach when possible.”
Taking away the chance to goof up helped the training go much faster, as Lucy had more chances to do things right (and cement the correct response) and fewer chances to do things wrong (and weaken the association between the command and the correct response). This works with people, too; with children, with adults, and with ourselves. When we’re trying to do a new thing (say, go on a diet) it helps to set ourselves up for success (not have cookies in the house). We help children learn in similar concrete ways be removing things from their environment that could be confusing or harmful to them (we teach them to master their letters a few at a time, and put covers on the outlets until they know better not to stick fingers in there). As we grow more confident with our new behavior, we are more able to ignore the distractions, temptations, and irrelevancies that would challenge our learning at first.
A quick word about punishment as a tool in learning: it doesn’t work. At least, it won’t work consistently or well, and even if it works in a limited way, it does so at the cost of you feeling good about your relationship with your dog (or your person). Mel taught me that dogs generally don’t have the sense of future planning for the threat of punishment to keep them from misbehavior; Lucy might learn not to pee on the rug if I spank her when I see her do it, but she can’t reliably extrapolate that threat to understand she can’t pee on the rug when I’m not around. I can’t succeed by punishing her for misbehavior, because punishment only “works” when I’m around to deliver it. And more importantly, punishment communicates only “don’t do this,” it contains no information about what I do want Lucy to do (pee outside, not on the rug when I’m not around, or on a different rug, or on anything that’s not outside).
There’s a lot of kickback to this idea, but punishment doesn’t work with people, either. Though we can more reliably extrapolate threats to the future (“Dad would kill me if I took this candy from the store”), we have other ways to get around it (“But he’ll never find out”). We human beings are masters at avoiding punishment by all means necessary—rationalization, lying, subterfuge—all means, that is, except for avoiding misbehavior. (How many people actually drive the speed limit when no cop car is in sight?)
As with dogs, punishment does not motivate us to do the right thing or even communicate what the “right thing” is. Of course, there are times when it’s important to know not to do the wrong thing, and at these times it’s helpful to learn from the consequences (different from punishment) of misbehavior: If I cheat on my diet, I gain weight. If my son spends all his allowance money on candy, he won’t have anything left for that toy he’s saving up for. If Sally cheats on John, John will break up with her. Consequences are different from punishment in that they are a natural derivation from the misbehavior. Like punishment, consequences are unpleasant to experience, but because consequences don’t depend on someone to mete them out they are a more effective deterrent to misbehavior.
Over time, Lucy’s positive behaviors were so well practiced that they were habit for her and I didn’t need to constantly reward them. But Mel cautioned me to make sure to reinforce them on occasion so that good behavior didn’t become stale. We can do that with ourselves and each other, too, after we have learned new habits and gained deeper understanding in our relationships. Where constant positive reinforcement or conversations about intent and meaning would come across as condescending, periodic reward or “checking in” with someone lets them know we still care and are still connected. I’m glad to say this approach has helped Lucy remain a wonderful dog—and I hope that the human version of Puppy Training 101 will do as well for you.
Thanks again to Mel for allowing me to use her name and real-life examples from her training classes. For more information about actual dog behavior and the finer details of training, please visit Mel’s excellent website.