Positive vs Negative Pros & Cons in Relation to Training Your Dog

I’m constantly asked by people “If I reward the good behavior and punish the bad behavior, won’t I get what I’m looking for quicker?”.  The answer is NO!  I know, it seems to throw rocks in the face of common sense, but when did learning theory become about common sense?

 Check Your Morals at the Door

In operant conditioning, positive and negative doesn’t mean good and bad – it simply means the addition of a stimulus or the removal of a stimulus.   Here is a quick overview of the 4 traditional components of operant conditioning:

Positive reinforcement – the addition of something the dog likes following an event.  For example, if Shorty touches my hand with her nose, I give her a dog treat.  She is more likely to repeat the touch because she is learning that “touch” is followed by something she finds enjoyable.

Positive punishment – the addition of something the dog dislikes following an event.  For example, a dog barks at the mail delivery person, and his owner squirts him with a water bottle.  The owner hopes that the dog dislikes being squirted by water so much that he will stop barking when the mail delivery person comes.

Negative reinforcement – the removal of something the dog dislikes following an event.  For example, Shorty may bark and lunge at an unfamiliar dog that runs up to her face while she is on leash in order to get that dog to go away.  If the dog goes away, then negative reinforcement has occurred, and Shorty is more likely to bark and lunge on leash again when an unfamiliar dog runs at her face.

Negative punishment – the removal of something the dog likes following an event.  For example, a puppy is playing with his new owner, and the puppy bites his owner too hard.  The owner stops playing, gets up, and walks away from the puppy.  The puppy will learn to inhibit his bite because he would rather keep playing with his new owner than lose his playmate.

When using punishment, we want the preceding behavior to decrease whereas in reinforcement, the preceding behavior increases.

 What is wrong with Mixing?

So why doesn’t positive reinforcement and positive punishment go hand in hand? Let’s look at some examples.

Say your dog is reactive to other dogs.  So, you figure, if my dog reacts when a dog passes her, I’ll provide a leash jerk, and if my dog doesn’t react, I’ll give her a cookie.  Your dog learns that when another dog approaches, she sometimes feels pain and she sometimes feel pleasure – how unpredictable!

Most of my clients know that I love Starbucks.  However, if Starbucks only made my drink correctly some of the time, my behavior chain of entering the store, ordering a drink, and paying for it would possibly fade or become extinct.  Like dogs, people also dislike unpredictable environments – what if only a few cars stopped every time there was a red light?  What if some cars stopped when there was a green light?  How would you begin to feel when you drive or walk up to an intersection with a light, not knowing what the other drivers are going to do?  A little anxious, a little nervous, fearful even?  This is how our dogs feel when we mix certain quadrants of operant conditioning.

Here, Ivan Pavlov also pops up – in the above example, I begin to have undesirable feelings towards Starbucks (heaven forbid!) and we all begin to hate stoplights.  Similarly, your dog begins to have undesirable feelings towards other dogs approaching her.  However, by consistently pairing the approach of other dogs with something your dog likes, those feelings can change.  However, you’ll have to stay tuned for more information on classical, also known as respondent, conditioning!

Courtesy of Bad Dogs Gone Good