I’m going to train my new pup, Chip, myself, but the more I read on the subject the more confused I get. Most trainers seem to agree that some form of reinforcement is necessary to teach a dog a certain behavior, but there doesn’t seem to be much agreement on what that reinforcement should be.

When selecting a method to train their dogs, most owners automatically gravitate toward one that meets their own needs first, and their pet’s needs second. Those who believe that sparring the rod will spoil the child often opt for methods that advocate punishing a pet for misbehaving as the best way to teach it to obey. Such approaches recommend hitting the dog with rolled up newspapers, or subjecting the animal to irritating sounds, shocks, or scents. At the opposite end of the spectrum, we find owners who prefer a kinder gentler approach that rewards pets for doing the right thing rather than punishing them for their errors. Advocates of these methods typically use food treats and/or gushing praise. In both cases, though, and some owner opinion to the contrary, we can’t say the dog is trained unless it displays the proper behavior in the absence of any kind of any reinforcement. Although most people agree that treats and hugs are the more humane training option, our increased knowledge about animal thinking suggests that both kinds of human responses may create problems for Chip and his owner.

How can that possibly be? Studies indicate that, like humans, animals given the opportunity to choose the right behavior internalize the lesson much faster than those not given this opportunity. And because they opt for the right behavior themselves, no reason exists either to punish or praise them for it. For example, let’s consider a common canine misbehavior, bolting through the door and dragging the owner along. Typical reinforcement training would focus on making the dog sit and wait at the door, either to avoid punishment or to get a reward. However, we know that dogs who bolt don’t do this to irritate their owners. They do it this because, barring our conscious attempts to establish ourselves as leaders in our human-canine packs, the dog’s deeply rooted instinct to protect its territory (which includes the owner under these circumstances) compels the animal to go first.

If Chip’s owner punishes him for displaying this behavior, Chip finds himself in the distressing position of trying to protect someone who’s hurting him. On the other hand, if the owner gives Chip a reward for waiting at the door, this doesn’t address the underlying leadership problem that caused the behavior.

Now let’s set Chip up so that he can choose the right behavior. In this scenario, Chip’s owner gives him the sit command once (and only once) when he’s approximately 2 feet from the door. The owner then turns her back on the dog and opens the door about 2- 4″. The instant she senses him move, she slams the door—hard!—and looks nonchalantly at the ceiling or fixes herself a cup of tea. A few minutes later, she repeats the process until Chip chooses to sit and wait for her to go through the door first (which usually takes about 3-4 slams). Once he sits and waits quietly, she goes through the door, then invites him to join her in a low-keyed tone of voice.

In this situation the owner has created a loud monster on the other side of the door which doesn’t bother her in the least but makes Chip at least hesitate if not out and out jump. Being a smart dog as well as aware of the message of calmness and confidence—i.e., leadership—that his owner exudes compared to his anxiety, he decides to wait and let her go first.

Admittedly setting up our dogs to make choices does require a knowledge of canine behavior and it doesn’t provide the rush that dispensing punishment or treats does. On the other hand, to surreptitiously watch canine gears turning as our dogs decide not to charge out the door, jump on people, or pester guests on their own (with a little creative behind-the-scenes help from their owners) generates a sense of awe and respect for canine intelligence no power-play can begin to match.