Q: Can you teach everything without punishment? By punishment I mean "correction" which I translated to "punishment" in my question. Maybe that's the problem? I've been thinking that most corrections, even if given gently, are "punishments." But maybe I'm wrong and they are only that, corrections?
In my case, I'm a veterinarian and I must teach a new employee to look at stool samples. I reward her each step of the way, several times, and she seems to understand. I watch her perform each step successfully—until she puts alcohol instead of sugar in the solution. What can I do?
I don't mean to belabor this, but I just want to be sure I'm not punishing the animal (dog, horse, person).
Aversive or punishment?
There's a difference between aversive events and punishment. Life is full of aversive events—it rains, you stub your toe, the train leaves without you. These things happen to all of us, and to our pets, and we don't control when or if they occur. Kay Laurence has an amusing paragraph about the aversive events that befall her Gordon setters (all of which they ignore)—falling off the bed, running into door posts, and more (read that article here).
In general, all that we learn from the inevitable aversives in daily life is to avoid them if we can.
On the other hand, a punishment is something aversive that you do on purpose. It may be contingent on a behavior, and it may stop or interrupt that behavior—which reinforces YOU for punishing, so watch out for that.
The effects of punishment
But, a punishment does NOT have a predictable effect on the future. We make false assumptions when we declare, "I really taught him a lesson." Research has shown that punishing a behavior may change that behavior in the future, may not change that behavior, and/or may change some other behavior. (Murray Sidman did a lot of this work.)
A punishment might not happen at the same time as the behavior (that's what contingent means). If that happens, the punishment is just an inexplicable aversive, which then may become associated with you more than with any behavior.
Careful with assumptions
Removing punishment from your tool kit is NOT the same thing as removing all aversives from your learner's life. Many non-clicker trainers leap to that conclusion, point out that it's impossible to remove all aversives—dogs wear leashes and get shut in crates, clicker trainers use the Gentle Leader—and conclude that clicker training is not punishment-free. But that initial assumption is wrong. While all punishment is aversive, not all aversives are punishment.
My dogs spend a lot of time sleeping under my computer, bored. Is that aversive? I suppose that sometimes they'd rather be doing something more exciting, but so what? I have my job, they have theirs, and that's life. I am not punishing them, but they are experiencing the reality, sometimes aversive, of our life together.
Ending a training session
Since clicker training is so much fun, it's true that ending a session is a downer. Sigh! The solution is to end the session so that it doesn't come as a complete and unexpected deprivation, which COULD impact behavior. End with a clear "all done" signal that always means, "That was fun, and now I have to go do something else." I use the Hawaiian word Pau (finished, done) and a hand gesture, and my dogs accept that cheerfully. If you want to make the end of the session even less painful, after you give your "end of session" signal, hand over a toy or some other reinforcer to ease the moment. With dolphins, we scattered three or four extra fish around, a mini jackpot. By the time the dolphins had picked up all the fish, they probably felt better.
Ending a fun session is more of a problem with a very green, inexperienced animal. If you have thousands of training sessions under your belt, or fins, you probably don't worry, as there will always be another chance.
If your learner is GLAD the session is over, wags, romps, laughs, or plays after the session, but not before or during, you must think about that. I see a lot of performance dogs with that reaction.
Get real
If in real life you have to wade in with an aversive to stop something from happening—if you must yank a baby away from the light socket, stop a dog from grabbing the roast chicken off the table—so be it. Animals do reprimand (the official biologist term) their young and each other.
You'll have interrupted or stopped a dangerous event. Just don't kid yourself that you've taught or guaranteed any particular change for the future.
-Karen Pryor
Definition of punishment
Karen, I greatly enjoy your articles in your newsletter and on your web site. You are a port in a storm in a time of renewed energy by traditional dog trainers, who espouse a superstitious and anthropomorphic approach to animal training.
However, I am confused by your latest article. I define punishment as anything that suppresses a subject's behavior, i.e., makes it less likely to occur.
In this article you seem to be defining "punishment" instead by the trainer's intent, not by the trainee's reaction.
I understand completely what you are saying about the trainer, and that applying aversives is reinforcing for the trainer, and the misconceptions that trainers have when they are applying aversives.
Can you please clarify what you mean by punishment in this context?
Thank you once again on behalf of all those animals who must endure our attempts to modify their behaviors :-)
Cathy Toft
Davis, California
Karen said:
On the other hand, a punishment is something aversive that you do on purpose. It may be contingent on a behavior, and it may stop or interrupt that behavior—which reinforces YOU for punishing, so watch out for that.
The effects of punishment
But, a punishment does NOT have a predictable effect on the future. We make false assumptions when we declare, "I really taught him a lesson." Research has shown that punishing a behavior may change that behavior in the future, may not change that behavior, and/or may change some other behavior. (Murray Sidman did a lot of this work.)
Post new comment