What if you’re standing in a playing field and your dog won’t come to you? What if you want your dog to “sit” and he keeps looking for squirrels? Eric Brad lets you in on some methods and mindsets for getting more cooperation from your dog.
I stood in that baseball field for a good 35 minutes in the wind and rain. It was cold. As the minutes passed I was getting more miserable, but not from the cold or the wet. My two dogs were having a grand old time running and chasing each other and they were not coming back to me for anything. As if to add insult to injury, a grounds keeper came by and said, “That field is closed you know?” I did know. I saw the sign only minutes after taking the leashes off my dogs — and 30 minutes later, I still couldn’t get them to come near me to leave. “Dog trainer, huh?” I thought to myself, and I had to laugh out loud.
It’s all about recall, which seems to be the “Holy Grail” of dog training. When I say “Come!” or “Here!” my dog should screech to a halt and race back to me as if I were the coolest thing in the universe. And for whatever reason, it always seems like the ONE time we need that recall to work is the time we just can’t get them to come back (usually in front of some other dog owner). It’s hard to be the coolest thing in the universe to my dogs. The world is a pretty exciting place.
Competition is Fierce
As a dog trainer, I know the behavioural background to what happened at the park that day. It was a simple case of reward versus reward. Due to the snow and the weather, it had been four or five days since my dogs had been out for a run. Being Belgian Shepherds, running all by itself is a very reinforcing activity my dogs. So what was in competition here was the history of all the times I had reinforced my dogs for coming over to me and the opportunity to
open up and just run. They chose the running.
We deal with the issue of “competing reinforcers” all the time, even as humans. Some days certain rewards are more attractive than others. Some days watching TV sounds really good and other days it doesn’t. It’s the same with our dogs. Some days running like wild savages sounds better than coming back for a yummy treat.
It’s always a choice for our dogs, whether we like to acknowledge that or not. I know that dog trainers have used the word “Command” to describe the cues we give to dogs. But really, we’re asking our dogs to do something when we say “Come!” or “Sit!” They have a choice to do it or not. Depending on how you train them, they make that choice either because they want what you will give them in return (e.g., treats, praise, throw the toy) or because they DON’T want what you will do if they DON’T do it (e.g., yell at them, jerk their leash, smack their nose). It would be great if we had a way to influence that choice in our favour.
The Mystery of History
One thing we know as dog trainers is that the more times we repeat something, the more likely our dogs are to do it successfully. This is where our friend Pavlov and his Classical Conditioning comes around to visit us again. In the example above, the more positive experiences we have watching TV, the more likely we are to think that it “sounds good” at any given moment. The same is true with our dogs. The more times they get a yummy treat or something else they want for giving us a particular behaviour, the more likely they will be to choose that option in the future.
Behaviour analysts and animal trainers call this a “history of reinforcement”, and the more we pay into that history, the stronger the behavioural response becomes. In real terms, this means that the more you reward a particular behaviour or the more you punish non-compliance (something the dog will seek to avoid), the more likely they will be willing to offer the behaviour when requested. This is the tool we use to bias or influence our dog’s decision when we ask for a specific behaviour. The more we have built up a history of success with what we want, the greater the odds of a dog doing the familiar, easy, reliably reinforcing behaviour.
Even though we run the risk of having competition from a busy and interesting universe of sights, sounds, and particularly smells, we can prepare ourselves for these challenges with a solid training program that builds a deep history of reinforcement for the behaviours we want to do most.
The Premack Principle and an Elegant Solution
Psychology professor David Premack published a paper in 1959 that introduced the concept that a more probable behaviour can be reinforcing to a less probable behaviour. Sounds tricky, doesn’t it. It’s really not. If you have reinforced your dog 1,000 times for doing a “sit”, it’s pretty likely that when you ask for a sit, the dog will sit. He knows what to do, he does it well, and he has a long history of reinforcement. So if I ask my dog to walk under a bar, that is a less probable behaviour because I have only reinforced it a few times. If I ask my dog to “sit” after he walks under the bar, his reaction will be “Oh! I know this one!” and he will be more likely to walk under the bar next time. The opportunity to do a more probable behaviour reinforces a less probable behaviour. You can learn more about the history and research into how and why Premack’s Principle works here.
Now, let’s go to a different park with the same dogs and a different trainer, my wife Petra. Rizzo is our 10-month old puppy. He’s young but he’s smart. So he learned very quickly that coming back to mom at the park sometimes means the leash gets put back on and the fun is over. His solution? Never come within 10 feet of mom at the park. Frustrating? Oh, you bet!
One day recently my wife happened to notice that Rizzo would wait with great anticipation while our other dog, Tira, brought her ball to Petra to be thrown. Rizzo seemed to LOVE chasing Tira after Petra threw the ball. It’s always been our habit to ask our dogs for a “sit” or a “down” or some other behaviour, and to reinforce that behaviour by throwing the ball. Petra was doing that with Tira, but she was not really working with Rizzo.
Then Premack’s Principle popped into Petra’s head. What was Rizzo’s more probable behaviour? Chasing Tira while she ran for the ball. Guaranteed, every time he was off like a shot after her. What if she asked Rizzo to “sit”, wherever he was, before throwing the ball for Tira? Bingo! In just a few trials, Rizzo was almost automatically sitting while waiting for Petra to throw the ball for Tira.
Once the “sit” was firmly established, she needed to add yet another behaviour — “stay there and let me pet you on the head before you take off”. She then added a collar grab to the pat on the head and the “sit”. The entire sit-pet-grab collar behaviour chain is then reinforced with the big chase after Tira. In the course of one hour, my wife had solved her problem! An elegant application of David Premack’s Principle of reinforcement helped her teach young Rizzo to stay and have his leash put on where he used to run away just an hour before. And no treats were involved at all!
It’s difficult for us to duplicate the conditions of the real world in our homes or backyards. But, in this case, Petra used a behaviour that had a long history — “sit” — and a behaviour that had a high probablity — “chase Tira” — to her advantage in retraining Rizzo. In a sense, “chase Tira” had a long history too, although it wasn’t necessarily a trained behaviour. But it was clearly reinforcing. So we have an opportunity to develop solid reinforcement histories on common behaviours that we can then use in some other situation to our advantage in the future.
Let’s go back to wet and cold and embarrassed Eric at the park with his two wild dogs. I’m sitting here writing this so obviously I didn’t have to end up living at the park with my dogs. I eventually got them to come back, and you may be wondering how that happened. Well, it took me a while to work through the very human emotional fog of frustration and embarrassment, but eventually my brain engaged again and the dog trainer re-emerged.
I stood up and walked toward Tira, my seven year old, and said “Wait!” This is a command she has hear thousands of times. She’s been reinforced with nearly everything I could think of over the years — food, toys, tug play, access to the outdoors, being let into the car for outings, her dinner, her breakfast. History of reinforcement? This was nearly archaeology! We’ve been rehearsing this one since she was four months old and we still practice it every day! Tira stopped dead and looked at me as if to say, “What’s up, dad?”
While I was internally smacking myself for not thinking of it sooner, I walked up to Tira, clipped on her leash and told her what a good girl she was and gave her a few treats for waiting. After that, gathering up Rizzo was a breeze. We got back to the van and everyone got treats. And I couldn’t help but shake my head at my own human-ness in forgetting what I knew in an emotional moment.
Don’t Forget to Think While You’re Feeling
The recall “Come!” is perhaps the most emotionally charged thing we teach our dogs. We ask them to come to us and we feel so betrayed when they find other things more interesting and important than us. That can be hard to take. But our dogs, like us, have their preferences and their moods. Sometimes we’re just not the most interesting things around.
We can use “history of reinforcement” and Premack’s Principle to help us, but sometimes the world can be the stronger competitor on any given day. So take that into consideration and prepare yourself for how likely your recall command will be in a given circumstance. The stronger your history of reinforcement and the relative dullness of the environment, the stronger thing will be biased significantly in your favor. But if that recall doesn’t work, take it as an indicator that you have some work to do and not as a sign of rejection by your dog. Remember that dogs are loving, forgiving animals that just happen to have different priorities than you in that particular moment.
Until next time, have fun with your dogs!
All photos © Petra Wingate 2008-2011