Celebrating 25 years -
Making tails wag since 1996!
3400 N. Monroe St. - Spokane, WA
Why is a Cue not a Command?
There is a difference between a Cue and a Command.
A cue is a promise that if the behavior is performed well, a reward may be available. A command is an order. Do it, or suffer the consequences. A command is a threat that if the required behavior is not followed correctly, a correction will result.
Dogs follow cues to earn rewards. Dogs follow commands to avoid correction.
The emotional response to a cue vs a command is different. Dogs are excited when they hear a cue - it signals the opportunity to earn a reward. Dogs are relieved when they avoid a correction. Either dog may appear "happy" at the result: "Woo hoo, I got a piece of chicken" or "Yay, I avoided getting my neck popped." Both methods will get results. You must decide what relationship you want in your training partnership with your dog. Do you want a dog who performs because he is worried about what might happen if he doesn't? or excited about the prospect of what might happen when he gets it right? Some trainers might say you need "both" - but you don't need to administer corrections to have reliable and accurate behavior. Read this article for more info.
When your dog discovers that a specific behavior earns a reward, he wants to repeat that action. He is eager to hear you to say "yes" or the sound of the "click" because those sounds mean that by performing that specific movement, he has earned access to the reward. Guess correctly, and rewards happen. Guess incorrectly and nothing happens. Offer a sit, get chicken. Look! I'm doing it again, can I have another bit of chicken?
At first, any variation of a sit gets rewarded until the dog has clearly identified the basic behavior you are looking for. How do you know? Because he eagerly offers it again and again without hesitation. Next, we will perfect it.
Polishing and perfecting
Once the dog realizes that putting his rear on the ground makes rewards happen, you can tweak his performance. By rewarding what you like and ignoring what you don't like about the way he sits, you can guide the dog toward perfection. The dog will repeat what is working. This way of sitting apparently gets more cookies than that way, so the dog chooses to perform the behavior the way you have selected instead. You decide how you want the behavior to look and reward the variations that more closely match that picture. Shape the behavior you want to see until the dog is accurately performing it the way you want.
The quality goes in before the label goes on.
Notice that up until now, we have not added a visual signal or verbal cue to prompt the specific behavior. We don't want to name the behavior until it really is the behavior we want, otherwise we might actually get 'hesitate and then flop' instead of 'quickly put your butt squarely on the floor.' When you like the way it is being done and the dog is reliably offering the behavior the way you want it, you can attach the cue.
Timing is everything; insert the cue just as the dog is about to offer the behavior. You know when he's 'about to' sit because he's done it many, many times already. Maybe he looks at your face, perks his ears, lifts his nose just a little, right before he starts to shift his weight and bends his hind legs. This time, when he perks his ears and starts to raise his nose, issue the cue. AS his butt hits the floor, mark it with "yes" or "click". You have carefully inserted the verbal or visual signal (not both, just one) in the perfect spot so the sequence is "cue >> behavior >> marker >> reward. " He starts to notice that you always do 'that thing' right before he gets rewarded for this behavior. And you do 'that other thing' right before he offers that other behavior. There appears to be a connection!
The cue is a question: "Do you know what this means?"
You execute your perfect hand signal for sit. The dog studies your hand signal, and chooses to put his rear on the floor. Your marker is like a game show bell that tells him he's won the prize. "Ding! ding ding!" The crowd goes wild. "You answered the question correctly and you win this piece of chicken!"
When you issue the cue out of the blue, at first, his response will be hesitant. He's seen that gesture or heard that word before. He knows it goes with one of the behaviors you like. He's just not completely sure which one. He's still guessing. When he guesses correctly, he wins. If he offers an incorrect answer or doesn't offer any answer when we present the cue, we simply show him what we wanted. We help him, we don't correct him or sound disappointed. We help him get it right and then give him another chance to try again. Because he isn't afraid of being wrong, the next time he is shown the cue, he is eager to give it a try. Maybe he sees the down cue and starts to lie down but doesn't make it all the way, we praise his partial guess and help him finish. The next time you give the cue, he is a little more sure of the answer His confidence grows. "Is this what you wanted?" His responses get quicker, his tail wags faster when the cue is given and you and he celebrate his success in solving the puzzle.
The process of clarity in responding to the cue for a dog who is just learning might look like this:
You give a signal to lie down and ...
the dog stares at you, puzzled. You lure him into position, mark "Yes!" or "Click" as his elbows touch the floor and reward him.
the dog sees the cue, drops his head slightly, but doesn't follow through. You lure him into position, mark as his elbows touch and reward him.
the dog sees the cue, hesitates and then slides slowly into a down without needing help. Mark and lots of praise and reward.
the dog sees the cue, studies the hand and then drops into the down a bit more quickly. Mark and celebrate as you reward.
the dog sees the cue, eagerly plops into the position, you mark with a "Yes!" or "Click!" and BIG PARTY!
He gets faster and more animated as his confidence grows.
But what if he "knows it" and chooses not to? Shouldn't you correct refusal?
The reason most dogs fail to respond to learned cues is not because they are dominant or stubborn, but because they are unmotivated or confused. Another reason might be that their training is not fluent enough for the situation you have put them in. "Come" across the backyard is not the same as "Come" 40 yards away in the park when he's spotted a squirrel. He is simply in a situation that he has not been trained to deal with yet.
So, how do you get him to respond when he's in over his head? Old school trainers might set him up to notice a squirrel, shout the command and then up-end the squirrel-chasing dog on a long line for his failure to respond, or zap him with an electronic collar. Positive trainers spend more time in the park practicing short recalls around squirrels at gradually increasing distances with high powered rewards to perfect his performance in the new environment in the face of increasing challenges. He can't think about food in the presence of squirrels? We'll work at a greater distance. We might reward the dog with access to running to the tree the squirrel is in or a game chasing his favorite frisbee. Either way, he will become reliable. But the second option will avoid the chance of injury to his body, his psyche, or your relationship.
You are teacher and student, not master and slave.
Old dog training manuals stated that "the dog must know you are the master." How sad that this definition of the relationship made the dog the slave and the subordinate. It is much better to be his teacher, his mentor, and his best friend.
"Properly trained, a person can be a dog's best friend!"
This handout may be reprinted in its entirety for distribution free of charge and with full credit given:
© CAROL A. BYRNES "DIAMONDS IN THE RUFF" Training for Dogs & Their People -
ditr_training @ hotmail.com - http://www.diamondsintheruff.com