Mack was a Jack Russell terrier who barked every time the phone rang. This was driving his owner crazy until I suggested that instead of telling him, "No! Quiet, Mack!" she should praise him instead.
Yes. I told her to praise him, then grab one of his toys, tease him with it and get him to chase her around the apartment, play a quick round of tug, then tell him to take it to his bed to chew it, and finally answer the phone.
She thought I was nuts, but she tried it anyway. And after only 3 or 4 days, whenever the phone rang, Mack would either start to bark, prompting his owner to say, "Good boy, Mack. Where's your toy?" Then he'd stop barking, run to find a toy and take it to his bed. Either that or when the phone rang he didn't bark, he simply ran to get a toy on his own, with no prompting from her. In other words the phone created a Pavlovian response in Mack, and his owner's problem was solved.
But where did this zany idea come from? (To continue reading, please scroll down...)
My dog Freddie and I were in Central Park one day when he was a little over a year old, and he found a discarded chicken breast on the sidewalk near a park bench. What a treasure! As soon as he saw it, he ran over, took it in his mouth, then looked over at where I stood, about twenty yards away, and dug his paws in, getting ready to run off.
At the time I had developed the idea that everything the training books say is wrong, and I was experimenting with my hypothesis by randomly doing the exact opposite of what you’re supposed to do in any given training situation, just to see what would happen. As Fred got ready to run away with his prize I thought, “I should probably tell him no or correct him in some way. What would the totally opposite, wrong thing be?” The answer came back, praise him. So I said, “Good boy! What a good boy!” in an excited, highly emotionally charged voice.
Freddie, without knowing what he was going or thinking about it, immediately dropped his treasure and came running back to me, wagging his tail, and smiling. (Dalmatians do smile.) I praised him even more, picked up a stick, teased him with it, then had him chase me to get it. We played a little tug-of-war, with Freddie on his back legs, and his front paws braced against my body. Then I used the stick to heel him past the chicken breast. I did it three or four times in a row. Then I had him sit right in front of it, took a few steps back, and said, “Freddie, come!” and he came to me and we ran off, playing.
I thought this over, wondering why it had worked, and the only thing I could come up with was that I’d interrupted Freddie’s flow of feelings. His expectation was to run off with his lovely, tasty chicken breast and make a meal of it. All the praise did, in this way of looking at things, was interrupt his behavior, his desired expectation, or his flow of feelings. As I thought it over even more I realized that whether you correct the dog or praise him, all you’re really doing is interrupting his behavior and his flow of feelings. Both methods stop the behavior as it’s happening, but corrections carry an unfavorable association towards whomever is doing the correcting. On the other hand praise carries only positive associations. So the question is, would you rather have your dog see you as positive or negative?
A few months after working this out in my head I heard of a study done at Harvard which showed that the human brain is designed to be unhappy, or at least not to be happy for very long. This has to do with how dopamine and serotonin levels always go down after they spike, and is similar to the effects of an addict's high, and how his substance of choice tends to lose its effectiveness the more it’s used. Another way of looking at it is the new toy at Christmas phenomenon. Man, on Christmas morning that new toy is the most wonderful thing in the world, but two weeks later it’s old hat. The Harvard researchers’ theory was that the brain is designed to not allow us to be happy for very long so we'll continue to do things, to achieve new levels of civilization, or to just get up in the morning.
I wondered if a dog’s brain worked the same way, and it occurred to me that whenever a dog (or wolf) leaves the den, his hunting instinct immediately kicks in. This means that there’s some level of dissatisfaction going on in the dog’s emotional system; some internal mechanism is telling him he’s not going feel better until he finds some way to use his prey drive, whether it’s meeting another dog to hook up with and play/hunt with, or to find a squirrel to chase.
If this is true, then when Freddie found the chicken breast he was thinking, “I’ve found it! This is what will fill that empty feeling!” He wasn't really hungry. He’d eaten a good breakfast. He was simply looking for some way to satisfy his prey drive. (By the way, if you think about it scavenging is a very economical way for a dog to hunt: you go directly from search to eat, with no need to stalk, chase, grab, and kill your prey. . .)
Then, when I praised Fred, flooding him with positive emotions, which came from a person who was the most important focal point of his life, it was even more emotionally fulfilling for him than finding the chicken breast had been.
Going back to the underlying mechanism, here’s how praise works:
A dog sees a piece of food. He goes after it with the expectation of achieving satisfaction. If he’s interrupted, both his desire and his expectation of achieving satisfaction are thwarted. Since interrupting him by saying “No!” drags negative associations with whomever is saying it into the equation, the food object is rarely seen as the primary negative. Neither is the dog’s behavior. In the dog’s experience:
THE HANDLER IS PERCEIVED AS THE PRIMARY NEGATIVE
because THE HANDLER IS AN OBSTACLE TO THE DOG’S DESIRE.
Saying “No!” is an impure correction because the negative experience of being interrupted is directly attributable to the person saying “No!” (A true Pavlovian or Skinnerian would never use a verbal correction for this very reason.) On the other hand, when you use praise, the negative experience of being interrupted can’t be attributed to the handler, only the the thing the dog desires at that moment. When you praise as a correction
THE HANDLER IS PERCEIVED AS POSITIVE because HE SEEMS TO BE SUPPORTING THE DOG’S DESIRES.
When Freddie found that chicken breast his social instincts became polarized toward resistance. The food was more important to him than his need to feel connected to me at that moment. But by praising him, I instantly reversed the polarity of his emotions from social resistance to social attraction. He lost his feelings of attraction to the food. Or rather, my praise changed his feelings of attraction for the food into feelings of attraction to me. In other words I changed his emotional state, and
WHEN YOU CHANGE A DOG’S EMOTIONAL STATE, YOU AUTOMATICALLY CHANGE HIS BEHAVIOR.
After the chicken breast incident I spent the next few days praising Fred, who’d previously been a bit of a problem scavenger, whenever he found something on the street. In three or four days his scavenging totally disappeared. All because I’d pigheadedly done the exact opposite of what the dog training books (and common sense) would tell you to do.
You have to remember that when a dog goes after something on the street he does so because of an inner feeling of need or desire. He may have no real desire to eat a pizza crust, just a general sense of dissatisfaction. (Freddie and other dogs will often scavenge after eating a full meal, so they don’t do it out of hunger, but out of a need to reduce emotional tension.) When you praise a dog, you give him a feeling of connectedness which overcomes his need for the food object. When you correct the dog, you may be successful at getting him to leave the object alone at that moment, but you do so by leaving him with an unfulfilled desire, a desire which has to then find its expression through another outlet. Ultimately it’s a matter of attraction and resistance. Praise makes the dog feel more attracted emotionally to you than he does to the food object. Scolding, even though it seems successful and seems to make sense, actually makes a dog feel more attracted to the food object than he does to you!
However, this technique only works when the praise reverses the dog’s emotional polarity. When the dog is too wound up, it does no good to praise him. So you have to praise the dog before things get ratcheted up too high. If you remember in the story about Isabella, my praise didn’t stop her from snarling at me, it just softened her up alittle so that when I bounced the tennis ball right in front of here she was able to switch gears. And it’s important to know that I really praised the hell out of her. If you’d been there you'd have thought I’d lost my marbles. I was praising her in my highest, silliest voice.
So remember this isn’t a cure-all; it’s just a very helpful tool.