What is Operant Conditioning
Operant Conditioning is a learning theory that explains the relationship between a specific consequence and a behavior.
For example, give a dog a treat after he sits several times. The dog will eventually associate the fact that a sit will bring about a good consequence (the treat). Because the consequence for the sit is a good one, the dog will likely sit again.
Scientists have divided operant conditioning into four different types or quadrants. These four quadrants are positive reinforcement, negative reinforcement, positive punishment, and negative punishment. Before discussing these four quadrants in detail, let’s define the terms used to describe these quadrants.
Punishment
The scientific definition of the word punishment is this.
Punishment – A consequence that causes a behavior to be less likely to be repeated.
One example of a punishment would be to give a leash pop when your dog pulls. Another example of a punishment would be to withhold access to something the dog wants such as the opportunity to move forward on a walk. With either of these techniques the dog is less likely to pull on the leash.
The words punishment and correction are often confused. The definition of a correction has nothing to do with the results of the correction. A correction is simply applying something to a dog’s life that the dog doesn’t like. A correction is a punishment only when the correction achieves the desired effect of causing the dog to be less likely to repeat a behavior.
The word correction is used to communicate what you do to the dog. The word punishment is used to communicate the effects of what you do to the dog.
Reinforcement
Here is a good definition of a reinforcement:
Reinforcement – A consequence that causes a behavior to be more likely to be repeated.
The words “reinforcement” and “reward” are often confused. The scientific word reinforcement is not necessarily the same thing as a reward. Though a reward can be a reinforcement, it is not necessarily. Let’s look at the difference between the two words.
A reward is simply applying something to a dog’s life that the dog likes. A reward can be a reinforcement, but only if the reward achieves the desired effect of causing the dog to be more likely to repeat a behavior.
The word reward is used to communicate what you do to or for or give the dog. The word reinforcement has to do with the effects of what you do to or give the dog.
The Two Methods to Cause Each of These Two Types of Consequences
In addition to there being two very different types of consequences to behavior, there are also two very different methods for a trainer to cause these consequences. The trainer can either add something to the dog’s life or the trainer can take something away. When we add something to the dog’s life to cause a certain consequence, we describe the consequence to the behavior as being positive. When we take something away from the dog’s life, we say that it is negative.
There are two different definitions of the words positive and negative. These two definitions are often confused. Positive and negative can mean good and bad. However, there is another use of these words. That use is the same as how we use these terms in math. Positive means to add something. Negative means to take something away.
As we discuss operant conditioning, think of the words positive and negative in the same way you consider these words in math. A positive (+) sign in math and a positive interaction in operant conditioning both mean adding something to a situation. A negative in math and a negative in operant conditioning both mean to take something away. Positive does not necessarily mean good and negative doesn’t necessarily mean bad. These two terms (negative and positive) are very often misunderstood when it comes to dog training.
The Four Quadrants of Learning
Since there are two methods for causing consequences for each of the two types of consequences, we can group the methods and types of consequences into four different categories. These four categories are called the four quadrants of learning in operant conditioning. These four categories will become more clear as we discuss each of them.
Positive Reinforcement
The first quadrant of learning that we’re going to discuss is called positive reinforcement. Positive reinforcement is simply adding something to a dog’s life in order to make a behavior more likely to be repeated. Remember, reinforcement means to make a behavior more likely to be repeated. Also, remember that positive means to add something to the situation.
An example of positive reinforcement would be giving a dog a treat in order to reinforce a sit. To reinforce a sit means to make it more likely for the dog to sit again the next time. If you add food to a dog’s life, the dog will be more inclined to sit when he’s asked again. You would have reinforced the sit by adding something to the dog’s life (doing something positive). And again, remember, when we say we are doing something positive, it does not necessarily mean we’re doing something good. It simply means we’re adding something to the dog’s life.
Negative Punishment
Now, we’re going to turn our attention to the two words opposite from positive reinforcement: negative punishment.
Negative when discussing negative punishment means that we’re going to take something away from the dog’s life. Negative does not necessarily mean bad. Remember to think of negative in the same way that we think of it in math.
Punishment simply means to make a behavior less likely to be repeated. It does not necessarily mean that we are doing something painful or harmful to the dog. Anything that makes a behavior less likely to be repeated in scientific terms is a punishment.
Let’s talk about an example to make this more clear. An effective negative punishment for stopping a puppy from jumping would be to leave the puppy when he jumps.
In this example, you would be doing something negative because you are taking away something from the dog. The something that you are taking away from the dog is your presence. Remember negative does not necessarily mean something bad. “You” are certainly not bad. Rather you have taken yourself away from the dog.
Leaving a puppy is in this case a punishment. Why? It isn’t a correction. You aren’t doing anything that physically hurts the puppy. Leaving the puppy in this case is a punishment simply because it has the effect of making the jumping less likely to be repeated.
Therefore, leaving a puppy is a negative punishment. You are taking away something from the situation: you. You are causing the behavior to be less likely to be repeated which is the definition of a punishment.
Positive Punishment
When a trainer adds a correction to a dog’s life, they are using the quadrant of learning called “Positive Punishment”. The word positive does not mean that you are doing something good to the dog. It is positive, because the trainer added the correction to the dog’s life. Trainers call corrections punishments because of the effect punishments have on the dog’s behavior. Remember, the definition of a punishment is that it is anything that makes a behavior less likely to be repeated.
This phrase may be confusing. We so often think of positive in terms of being good and punishment in terms of being bad. But again, remember that positive does not mean good in dog training. It means that something is being added to the dog’s life.
If you think about the definition of a punishment, you can see that all corrections are not necessarily punishments. Why? Because, if the correction does not bring about the desired result of making the behavior less likely to be repeated, it is not by definition a punishment.
Also, you may be thinking that all punishments are painful. That is not necessarily the case. ANYTHING that you add to a situation to cause the effect of reducing the likelihood of a behavior being repeated is a positive punishment. Even adding something as painless as a quiet verbal “no” can bring about the desired effect. The “no” can indeed be called a positive punishment.
Negative Reinforcement
So, if positive reinforcement is the addition of something that will reinforce a behavior, what is negative reinforcement? By now, you probably have a pretty good grasp of the term negative in operant conditioning. It means to take something away from the situation. How do you reinforce a behavior by taking something away. This is best explained with an example.
Many obedience trainers and even more trainers who train hunting dogs will teach what is called a forced retrieve. In the forced retrieve, the trainer will pinch a dog’s ear and continue pinching it until the dog takes the dumb bell or bumper or whatever object the trainer wants the dog to hold onto. When the dog does what the trainer wants, the trainer stops the pinching.
This training method makes use of the quadrant of learning called negative reinforcement. It is negative because the pain of the ear pinch is taken away when the dog takes the object. It is a reinforcement because the dog will be more likely to repeat the desired behavior. He will most likely take the object the next time he feels an ear pinch because he will want the ear pinch to be taken away.
Operant Conditioning and the Balanced Trainer
The balanced trainer uses all four quadrants of learning because all four quadrants do indeed successfully modify behavior. Some balanced trainers use more positive reinforcement. Others less. A balanced trainer is simply a trainer that uses all four quadrants. Some balanced trainers are almost reward based trainers using mostly the positive reinforcement quadrant. Other balanced trainers can be down right harsh. Then, there are balanced trainers in between the two.
Operant Conditioning and the Reinforcement Based Trainer
The Reinforcement Based trainer only uses two quadrants of learning believing that the other two quadrants are not only unnecessary but also can potentially be damaging to the dog on the inside because of unseen side effects. For clarification, reinforcement based trainers sometimes refer to themselves as Purely Positive or Force Free trainers. We think that Reinforcement Based is a better description of who we are and how we choose to train.
Which two quadrants do you think the Reinforcement Based trainers use? How about those that refer to themselves as Purely Positive?
Based on the name, you’d think it was Positive Reinforcement and Positive Punishment. Both have the word positive in the name of the quadrant.
However, this is not the case because the word positive in the term Purely Positive terminology has strayed away from the scientific definition of the word positive.
Reinforcement Based Trainers Use Only the Two Quadrants of Learning that Dogs Enjoy
The two quadrants that positive reinforcement trainers use are the two quadrants that involve the giving and taking away those things that a dog likes or enjoys. Those quadrants are positive reinforcement and negative punishments.
Positive reinforcement is self-explanatory. However, the term negative punishment must be thought about to understand it fully. On the surface, this term sounds like the most undesirable of all the quadrants. However, when you consider the scientific meaning of the terms, you’ll see that that is not the case at all.
Punishments and the Reinforcement Based Trainer
Punishment simply means that a consequence causes a dog to be less likely to repeat a behavior. All trainers, even Reinforcement Based trainers, need to address undesirable behaviors. According to the definition of the word punishment, a punishment is anything (good or bad) that causes a behavior to be less likely to be repeated. Therefore, ALL trainers need to punish poor behavior. However, by using negative punishment, reinforcement based trainers are given a means for decreasing the likelihood of poor behavior without adding corrections. They simply take away (thus the word negative) something that the dog enjoys. Remember our example of taking ourselves away from the puppy who was jumping?
Summary to Help You Keep the Terms Straight
Because the scientific terminology and the general use of words such as positive and negative have taken on two different meanings, the labels for the four quadrants of learning have become confusing. However, you can remember them by considering that the scientific words for positive and negative are not good and bad. These terms simply refer to whether you are adding something or taken something away from a dog’s life. That something can be either good or bad. In behavior modification, positive can be good or bad. The word negative can also be good or bad.
Also, the terms correction and punishment can be confused as well as the terms reward and reinforcement. To keep these terms straight, remember that a correction and a reward have to do with what you are doing to or for the dog whereas the terms punishment and reinforcement have to do with the behavioral effects of what you’ve done to or for the dog.
The Effects of Operant Conditioning
Operant conditioning is a very valuable tool for addressing behavior. It provides a tremendous tool for increasing good behavior and decreasing bad. All four quadrants will affect behavior.
However, depending on which quadrant of learning you are using, operant conditioning will sometimes bring about a side effect or side benefit by creating what is called conditioned emotional responses (CER’s). Therefore, we’ve chosen to use only those quadrants that produce positive CER’s. We’re going to discuss how these CER’s are created in our next chapter on classical conditioning.
