Striking a Balance

“Fuzzy, COME!!!!´ yelled Agnes as her little Cairn terrier took off full speed after having an unexpected meeting with a cat on the front lawn. He braced his feet, thrashed his head and backed right out of the collar he was wearing, leaving Agnes holding a leash with an empty collar attached. The retreating cat was bolting for the road. “Fuzzy, cookie, come get a cookie! Fuzzy! COME!!” Suddenly Agnes realized that Fuzzy’s path was going to intersect with a fast-approaching Chevy Blazer and her commands became more urgent, more desperate. But Fuzzy didn’t want a cookie just then, Fuzzy wanted a cat. The brakes screeched sickeningly and slid miraculously to a halt within six inches of the little brindle terrier. Momentarily distracted, Fuzzy flinched at this huge metal monster that had come and shrieked at him. He barked back at the car, having forgotten his previous pursuit.

Agnes caught up with him and scooped him up before he could get himself into any more trouble. He hung abject and limp in her arms, temporarily submissive from being caught unawares. The woman driver, a little shaken herself, put her hand over her mouth and stuck her head out the window. Taking her trembling hand away she asked apprehensively “Is it …alright? Is it…” She couldn’t finish the sentence.

“He’s OK, but I’d like to shoot him about now.’ Agnes was visibly shaken. She was upset, relieved and frustrated all at the same time. As the woman drove slowly away, Agnes balanced Fuzzy in one arm while tightening up the collar and putting it back on his neck. Fuzzy had graduated the top in his obedience class. He’d do anything for a treat but when it came to situations like this one, his ‘obedience’ seemed to go in one ear and out the other.

She thought about her neighbor Sally, who’s Doberman was so well trained that she consistently responded to commands, even when she wasn’t wearing a collar. She remembered that “Reva” as she was called, had been trained on a prong collar and that she’d always thought that the collar looked so, well, so mean. And yet Reva was a happy dog and complied to Sally’s wishes with delight as well as speed. Sally rarely to even raised her voice to the Doberman. Agnes decided to have a talk with Sally to find out more about the type of training that Reva had had. The Dobe’s consistent reliability was hard to argue with, especially after the close call she’d just had with Fuzzy.

I am just thinking out loud here about different dog training approaches. The objective with all methods is to help a dog understand what behaviors are wanted or unwanted. One I’ll call ‘direct’ and the other ‘indirect’ dog training methods. “Indirect” learning is maintained through external management and motivation. It is primarily positively induced. As an example, in training a dog to sit, a food lure is placed just above a dog’s nose and moved back toward its ears. The dog eventually gets tired of holding its head up and sits because it is more comfortable to do so. This is when it receives the food. With patience, repetition and persistence, the dog eventually associates the word “Sit” with the desired position. The times that the dog is given the food reward is gradually reduced, but it must still be given intermittently to maintain the behavior if no direct training is done.

Direct training is compulsive. As an example, in training the dog to sit, the handler uses his hands, collar and leash to show the dog how to find the sit position. The dog is manipulated into the position and praised verbally. With patience, repetition and persistence, the dog eventually associates the word “Sit” with the desired position.  In this learning phase the cues are eventually reduced and, once the dog understands the sit command, the dog is given a choice to sit and receive praise and attention, or fail to sit and receive an aversive consequence. An aversive consequence is a correction, such as a snap on the collar that causes the dog to sit. The correction should be just unpleasant enough for the dog to want to avoid it. (If it is not firm enough, it is simply a cue and the dog will take no responsibility.) Once the dog is sitting it is praised immediately. The dog works to avoid correction and receive praise. All things being equal, understanding and obedience are two different concepts. A dog trained with both indirect and direct methods self-regulates its training on the inside.  Once taught and understood, the dog internalizes its responses to behaviors that it has learned. Doing behaviors brings the dog a sense of completion, of rest.

I’ll illustrate self-regulation with another common training problem and try to look at it from the dog's point of view. Jumping up on people- OK, so it isn't a problem from the dog's point of view but jumping is often a problem for their owners. Muddy paw prints on good clothing, or a small child or frail senior being knocked off their feet is unacceptable behavior in most households. There are a lot of ways to teach a puppy (or dog) not to jump on people without ruining the relationship between dog and handler. There are also ways to help the dog understand self-regulate and understand an acceptable, alternate way to greet people.

With the ‘indirect' approach the problem is 'managed' by 'anticipating' the jumping and substituting a desirable behavior such as requesting that the dog sit instead. There is no direct communication to a dog that helps it understand that the jumping is undesirable. I don't have any problem with teaching a dog to sit instead of jumping. In fact, it is good for the dog to know what you'd like them to do instead, but it seems that a lot of dogs just don't 'get it' about ‘NOT jumping’. Dick Russell, a notable dog-training professional from Louisiana, would say that ‘the dog does not internalize the idea’ that the sitting behavior should be substituted for the jumping behavior.

The dog will continue to 'attempt' to jump because, well because it's a dog and that's what dogs do. And for the unknowing guest who doesn't know the household 'tell-the dog-to-sit rules', they are very likely to be jumped on. So the indirect solution doesn't help the dog understand that jumping is not wanted.

With a direct approach, there is a cause and effect relationship between the actual jumping behavior and the immediate consequence of that behavior. This realization is established directly through a correction. In this case, a correction is an immediate consequence to a behavior that the dog finds unpleasant. This could be an unseen step on a dragging leash that inhibits the jump, a 'reflexive' knee bump to the chest or a startling snap on that line from behind the dog by an owner ‘at the ready’.

This is done with no yelling or scolding of the dog. No negative emotion or animosity is directed at the dog. Emotionally-detached corrections are key to success. The dog will associate corrections with the jumping because that is what happens every time he jumps. Over a short period of time, practicing in different locations with different people, the dog pretty quickly realizes a direct cause and effect relationship between jumping and the startling consequence. Dogs don't like surprises, so they learn to avoid jumping to avoid the consequence. I usually talk very pleasantly to a jumping dog while my knee (or shin or ankle with a smaller dog ) ‘reflexively’ bumps into the dog’s chest, just enough to startle and make the jumping ‘unrewarding’. When the dog ‘lands’ and seems unsure what to do, the “sit” command can be given. The dog is given calm, soothing attention, (both praise and petting are given) as long as the dog does not jump.

As a parallel illustration, when you go the Doctor for a check up and he taps your knee with that little rubber reflex hammer, your knee goes flying. Nothing personal, just a reaction to a reflex center. Think of the dog as the reflex hammer. When a dog jumps on a person, a correction is an immediate surprise that 'startles' the dog in a way that makes them want to stop jumping. The owner makes no eye contact with the dog until it ‘lands’. When the dog lands, they tell them to ‘sit’ and then give them positive attention (praise and petting and maybe even an occasional treat) as long as they remain in the sitting position.

Let’s look at Fido the Labrador. Fido runs up to greet dad at the end of the day and dad says pleasantly, “Hi Fido, Sit.” And Fido, in his exuberance, jumps up and, “ooof’’, runs into a knee, which puts him back on the ground. Where upon dad continues without missing a beat, “What a good fellow, so glad to see you!” Once again Fido leaps for joy. Dad’s knee seems to be having the hiccoughs every time he jumps. Ooof again and as he lands, dad gives the alternative - “Sit” command. Finally Fido tries the suggestion and lo and behold, dad actually reaches down calmly to pet him. (“SO! That’s how I get him to pet me”, thinks Fido.) Fido starts to wind up again and the petting stops. (Fido thinks, “Oh no, don’t stop the petting!”) He starts to jump but then checks himself (internalized inhibition because he’s thinking about the consequences.) He thoughtfully sits back down and immediately dad reaches down once again and begins to praise and pet him. Fido quivers with delight, his choice to exhibit self-control has earned him the affection he craved. It’s been my experience that the learning curve is cut way down when the two approaches are combined.

The correction gives dogs a more direct 'frame of reference' as to why it is better not to jump. This is more immediately effective than merely re-directing their behavior. They are more quickly able to figure out an appropriate response to the combination of techniques than when either is used alone and, if done appropriately, the training ‘sticks with’ a dog much longer than when done with either technique separately.

All dog owners want to be kind to their dogs. When they come to obedience classes for training, they generally also want their dogs trained to have reliable, good behavior i.e.–come when they’re called and stay when they’re told, to be good with people, and they want to do it as ‘time and cost effectively’ as possible. A trained dog is one that has learned to take responsibility for learned behaviors. Its behavior becomes self-regulated because the dog has learned what different consequences its choices result in.

Training with food has become popular, although it’s certainly not a new concept. It is an indirect, external motivation. Besides dogs, circus and aquarium animals have been trained with food for years. Whether a trainer is dealing with a killer whale or lions and tigers, they are dealing with different social instincts in a very controlled environment. This is not the same as the situation that average family has in trying to deal with a puppy or dog in a home environment where family and visitors come and go. Most dogs respond well to food-based training but often, especially as puppies get older, other needs and interests override their desire for food. For example, the dog that generally comes running from the yard when called because it usually gets a treat for doing so. If a dog bolts out the front door, it may prefer to chase a passing squirrel or cat into the road at that given moment. This is because, at that moment, its chase instinct or prey drive may override its need or desire for food. The same dog, if distraction-proofed, (which involves teaching the dog how to respond when distracted by checking it at the end of a 15 ft. long line when it runs after distractions) will be far more likely to think over the consequences of its behavior and respond to a command. In fact, it will probably have been taught not to pass through the door without permission.

For some dog owners, the thought of causing the dog any discomfort makes it difficult for them to ‘correct’ their dogs. ‘All positive, no correction’ methods seem like a more pleasant alternative than one that uses a combination of positive and negative techniques. I can offer for consideration the thought that, although it is unpleasant for a dog to be checked on a long line, it is a controlled consequence that is far preferable to having an untrained (or ‘food trained only’) dog get hit by a truck.

A balanced approach puts some responsibility for making choices on the dog. This is a very attainable objective. You will see such decision-making when you see a well-trained Seeing-eye dog or service assistance dog. These dogs see their ‘jobs’ as self-rewarding. Performing a command gives them a great deal of satisfaction. They know that they will both avoid discipline and earn reward.

There are some activities that are innately rewarding to a dog such as hunting, herding, or tracking. But for the coercive things we require of them –sit, down, stay, don’t jump, don’t pee in the house, we can either get mad at them, hand out treats and hope, or give them compelling reasons and rewards to do or avoid behaviors.