Ever since B.F. Skinner laid out the foundations of Operant Conditioning, OC based methods have been increasingly used to teach dogs and other animals a wide variety of behaviors. Much of the original development of the training techniques occurred at marine parks, where scientists and trainers were assisted by graduate students and Skinner’s children. Then, with the aid and impetus of the Internet, groups of dog owners and trainers alike discovered the benefits of OC and the Positive Training movement began to explode. Today, there remain two fundamentally different approaches to dog training: Positive Reinforcement (Here Fifi sweetie, come to Mommy and you get this treat) and Traditional (Hey Atlas, you crazy cur, stay away from that rabbit trail or my boot is going where the sun don’t shine). The trend, however, is clearly towards OC based Positive Reinforcement training methods. Dogs trained with the new techniques have shown, in obedience, agility and other competitions, that these techniques produce results that equal or surpass Traditional correction based training methods. The most popular OC based training method, by far, is Clicker Training. Originally conceptualized by Karen Pryor at Sealife Marine Park in Hawaii, Clicker Training has even being used for performance training of human athletes.
Why you should understand Operant Conditioning
“But I have a hog hunting/blood tracking/herding Lacy Dog bred to do those things. And he does them. What do I need with some fancy technique?” Well, the funny thing is you already use Operant Conditioning with your dog, all the time, every single day. This is so important I’m going to repeat it. You are use Operant Conditioning, every day, every time you interact with your dog! Now, given you are going to do something, don’t you want to get it right? Hey, you have a Lacy Dog. You know he is smarter than 99.9% of the dogs out there, and I bet your dog watches every little thing you say and do. You picked your dog because of what he is capable of doing. So make the most of it!
And there is a specific reason why this is the right technique for a Lacy Dog, even more so than for other breeds. Reading the NLDA Breed Standard, a Lacy is to have “incredible drive and determination to work.” These are reflections of a Lacy’s high prey drive, or predatory instinct. Lacy Dogs are bred to retain as much of their wild wolf ancestor’s predatory instinct as possible and still be safe around humans. One of the things that separates wild predators from domestic animals is that wild predators cannot be successfully disciplined. You can’t discipline a killer whale or a tiger, and, most importantly, you can’t use discipline to successfully train a wolf. Yank on a wolf’s collar and all he does is yank back even harder.
Our Lacys are as close to a wolf as most dog breeds get. It makes sense that these techniques, developed to work with wild predators, work well with a dog that has a Lacy’s predatory instinct. That’s great news. It means that you can breed in all the prey drive you want. They can have range, top, bottom, grit and be the toughest dog in the woods. Yet you are still able to train that dog to do whatever you want and to be as safe with the neighbor kids as a Chihuahua. Actually, the Chihuahuas I’ve know have been feisty little monsters that hate the neighbor kids, but you get the idea.
The building blocks of Operant Conditioning
It is easy to get confused by all the conditioning lingo. Things like OC quadrants, primary and secondary reinforcers, extinction, etc. The good news is that you only need to understand all that if you really want to. If you want to make your living as a dog trainer, then you should read it. If you just want to do a good job of teaching your Lacy Dog and increase your bond, then you don’t need to read it, because the basics are really very simple:
If you reward a behavior, it will happen more often.
Pretty simple and pretty hard to argue, isn’t it? As Homer Simpson would say, “Doh!” And yet that simple statement is the basis of Operant Conditioning as it applies to dog training. Reward what you do like and don’t reward what you don’t like.
Skinner defined 4 “quadrants” of Operant Conditioning as:
- Adding a Positive Reinforcer
Adding a Punishment
Taking away a Positive Reinforcer
Ending a Punishment
One of the problems of focusing on the four OC quadrants is that they create the impression that all four approaches are equal in importance and effectiveness for training a dog. They are definitely not. The two that work best are: Positive Reinforcement and taking away a Positive Reinforcement. Also, the concept of Extinction isn’t in the quadrants, yet it is as important as anything else. If you do not reward a behavior but do reward other behaviors, the unrewarded behavior will extinguish itself. It will go “extinct.” So, we are back to that basic concept of rewarding what you like, and, just a important, not rewarding what you don’t like.
Reading Lacy Dog forums, you see statements sounding like, “She does everything I want when I’m there, but if I’m gone, watch out!” This is usually provided as backup for how smart a Lacy Dog is. But it is really an indication of a dog trained with punishment. Any dog, not just a Lacy, is smart enough to know that if you aren’t there, you can’t provide the punishment. Ah, you say, but when I get home my Lacy knows she’s done something bad and is going to be punished. Sorry, but that just isn’t true. What your Lacy knows is that when you get home, sometimes she is punished, and being the perceptive canine she is, she also senses immediately if your mood moves towards “punishment mode.” So she starts acting all sheepish to try and avoid the punishment that sometimes happens. Trust me, she has long ago forgotten about that rug she chewed to threads and has no clue why she is being punished.
One of the behaviors Skinner observed is that the most powerful conditioning technique is irregularly rewarding of behavior with Positive Reinforcement. It is even more powerful than rewarding a behavior every time. With sporadic reward, as used within the Clicker Training paradigm, your dog believes it just has to do the behavior better and next time it will be rewarded. Now flip that concept around and apply it to punishment. What you are trying to do with punishment is to STOP a behavior, right? But for the dog, that behavior is self-rewarding. Think about it. Chewing, tugging on a leash and chasing a rabbit are all things the dog wants to do or they wouldn’t bother doing them. The reward is being able to do the act itself. So, what are you doing by inconsistently punishing a behavior? You are enabling the periodic REWARD of the behavior whenever you aren’t able to provide the punishment, which is the strongest way to reinforce the behavior. This is why Positive Reinforcement training works better (faster and with more retention) than punishments.
How to use Operant Conditioning
Clicker Training has evolved through millions of hours of practice, by a large number of trainers, in a wide variety of uses. And it has proven to be the single most reliable method of applying Operant Conditioning to dogs. There are many articles that cover why this is true. The key points are that (a) it is precise and thus less subject to interpretation by the dog and (b) the click sound is more likely to imprint on the primitive portion of the dog’s brain (the Amygdala) than your voice. Behaviors that reside in the Amygdala are much more likely to be permanently retained and consistently performed than those that reside in the cerebral cortex and thus require thought. They are almost hard-wired, just like jumping in reaction to a loud noise is hard-wired.
Clicker Training can be confusing to the human when you are first attempting it. Not only is there this stupid little toy to deal with, but it also uses terminology you may not be familiar with. But it can be greatly simplified to one single concept:
The instant your dog does something you want, click and treat. Repeat.
The click does two things. First, it promises your dog that it will get a treat. By the way, this makes the click a “conditioned reinforcer,” thereby ensuring it bypasses the cortex in favor of the Amygdala. Don’t ever break your promise! Because the click also lets the dog know that what it was doing, at the instant it heard the click, is what is getting it a treat. That means you need to very careful about when you use the clicker and only click when you want to give a treat for a behavior.
It’s really that simple, but as with many things, it takes practice. You have to learn to click at the right time (you are almost never too early, almost always too late) and you have to learn to have the clicker in one hand, treats in the other and swap them back and forth frequently.
The next step is learning to “shape.” Even extremely complex behaviors can be broken down into small pieces that, strung together, form the complex behavior you want. And each individual behavior can, and must, be trained. Getting your dog to do a specific behaviors and then chaining them together is called Shaping.
One of the key principles of Clicker Training is that you do not show your dog what to do. You want your dog to think for herself, and you want her to constantly be thinking up things she can do to “please you,” which in dog terms really means “get you to give her a treat.” Let’s face it, your Lacy really isn’t trying to interpret your inner state of mind in an effort to make you happy, she is trying to get you to give her what she wants. You need to teach her that, to get what she wants, she needs to figure out, with your help, what it is you want her to do. So you create the opportunity for her to learn what you want her to do and then reward her when she does.
Using Operant Conditioning in the real world
Here is an example, with real world application for hunting or tracking Lacys. Let’s say you decide it would be nice if, no matter what was going on around you (other dogs, strangers, the dog knowing a hunt was about to happen), you could reach for the latch on your Lacy’s crate in the back of the truck and he instantly sits down, waiting for you to open the door. You open the door and your dog still sits, eagerness all over his face, waiting. You casually get out your trailing harness and lead, or your cut collar and transmitter, and get your dog ready. Of course now he is beside himself with anticipation, but he stays there, sitting obediently in his crate. You collect your gear and talk to the hunter that wounded a deer, or to your fellow hunters to scope out the area, then, finally, you look at your Lacy and give him a release command, at which point he bounds out of the crate and comes right up to you, ready to go.
Do you know how long it took me to shape this behavior in my Lacy Dog Cannon (bearing in mind he already had a fairly solid “stay”)? Less than 10 minutes. I didn’t do it again for a week. When I tried again, do you know how much he forgot? Nothing. What happens if I don’t have the clicker or a treat? Exactly the same things as when I do.
That’s the power of Operant Conditioning, of Positive Reinforcement, of Clicker Training.
Here is how Cannon’s behavior was shaped. I walk up to Cannon in his crate, put my hand on the latch and waited. He knows he is supposed to do something, but he doesn’t know what it is. Pretty quickly, he realizes that I frequently reward him for sitting, so he tries that. Click and treat! I open the door to give him the treat, then close it and back off. He stands up. I touch the latch again. After a few seconds, he sits again. Click, open the door, treat, close it and back off. He already gets it, so he immediately stands up. I touch the latch, he sits, he gets the door opened and his treat. He repeats this as fast as I can do it. I have become his treat machine. In his mind, I’m no longer rewarding him. HE is getting ME to give him treats. That is the “Aha!” moment trainers talk about and strive for. It is time to up the ante, so I gradually increase the time I have the door of the crate open before I click/treat. When he understands that he is supposed to stay there until I release him, I start walking further and further away, until I can go talk to the instructor and he stays put. Voila.
It’s worth noting that Cannon has always been good at “waiting” until released. He knows that waiting is the quickest way to get what he wants. If he doesn’t wait, he will just have to wait again, to get it right, before he gets a treat. And he trusts that, at some point, I will release him and good things will happen. There is no fear of punishment, as I have never punished him for not waiting. And there is no dependence on my having a treat. The behavior has been conditioned into him so now that it is just what he does.
So give Operant Conditioning and Clicker Training a try. What do you have to lose? I doubt there is anything you want your Lacy to do, that she is capable of doing, that can’t be trained using a clicker and a good treat.
This article was written by Jim Browning, a member of the National Lacy Dog Association living in Arizona. Jim and his Lacy Cannon have done search and rescue work and are currently training for agility.