"See how guilty he looks? He knows what he did was wrong."
This myth probably results in more unintended animal abuse than any other and often arises because a dog is routinely destructive
in the house, housesoils, or gets into trash cans. Most owners will be quick to insist that the dog "acts guilty before I
even know what it has done!" This misconception arises for two primary reasons: 1) The owner anthropomorphizes, and 2) the
owner cannot read normal dog body language. The lowered head, tucked ears and tail, and avoidance behaviors that most clients
label as guilty looking are appeasement behaviors. The dog is demonstrating submission in an attempt to turn off the anger that it is either reading
in the owner's body language or expecting from the owner because of previous experience with similar situations.
I explain this concept to owners by encouraging them to perform the following experiment: Leave, sneak back into the house,
turn a trash can upside down, spread trash around, sneak back out, and then return home later as you normally would. If the
dog has had time to discover the mess, it is virtually guaranteed that the dog will demonstrate the same behaviors it would
have demonstrated if it had made the mess itself. The dog has made the association between the presence of a mess (whether
it is feces, urine, or trash) and an angry owner. It has not made the association between its behavior and the angry owner.
If an owner chooses not to try this experiment, simply describing it can be educational.
For punishment to be effective, it must involve three principles: 1) it must be applied within one or two seconds of the inappropriate
behavior; 2) it must be applied every single time the behavior is performed; and 3) it must be potent enough that the dog
will seek to avoid it in the future but not be so aversive as to frighten the dog.25 The AVSAB has issued a position statement on the use of punishment for behavior modification in animals as well as guidelines
on using punishment (
http://www.avsabonline.org/). Dogs that experience fear or anxiety during training will not learn as well and, as already mentioned, are more likely
to learn that people are scary and unpredictable. Since most pet owners are unable to use punishment according to these three
principles, the chances are great that they will do more harm than good. In short, it is much easier to teach a dog what you
want it to do by rewarding it for appropriate behavior than it is to teach it what not to do by punishing it.
The fact is, pet owners need to be taught that dogs make associations between events that consistently occur in association with each
other. Punishing a dog for something that it did even a few minutes ago (no matter how the dog is acting) does not teach the
dog what you don't want it to do. It teaches the dog that people are to be feared.
"If you use treats to train a dog, they'll always be needed to get the dog to obey your commands."
Positive reinforcement, an excellent way to teach an animal a new behavior, requires that you give something pleasant within
one or two seconds of the occurrence of the desired behavior.26 And the animal will learn most quickly if the reward is given every time it performs the desired behavior, which is called
But once the behavior is acquired and a verbal or hand cue is attached, the behavior is best maintained when the reward is
given intermittently.27 This principle is the same one that keeps people putting quarters into slot machines. Slot machines only pay off intermittently,
but people continue pulling that handle on the chance that the next time they will get the reward. Intermittent reinforcement
is powerful, and behaviors trained with it are resistant to extinction, but if we used it to teach new behaviors, progress
would be slow.27 With intermittent reinforcement, the dog never knows for sure whether it is going to get the food. The reward is presented
and given only after the desired behavior is performed. In contrast, a bribe is shown to the dog when it is asked to perform
the behavior. The dog knows it is going to receive it.
The fact is, when used appropriately, food rewards are an excellent and effective way to teach a dog new behaviors. Once a dog learns
a behavior, these rewards should be used intermittently. One of the more common mistakes a pet owner makes is to think that
a dog knows a behavior long before it actually does. To learn more and to help interested clients learn more, refer to the
books and other resources listed in Table 1.
Table 1 Recommended Reading and Resources