Positive vs Negative Reinforcement
This post contains affiliate links.
In the book How Dogs Learn, they talk about science behind dog training. I'm going give some definitions then talk about them.
Operant Conditioning - is the part of the science of behavior that explains the functional relationship between environment events and behavior. It is the key component in explaining how all organisms learn.
Classical or respondent conditioning - explains the reflexive behavior, like salivation. It does not explain voluntary behaviors.
Law of effect - responses that produce rewards tend to increase in frequency.
Conditioner reinforcer - depends on some conditioning taking place. In other words, you have to associate it with something pleasant before it will reinforce a behavior.
Primary reinforcers - are those reinforcers that are related to biological drives, like food.
Secondary reinforcer - becomes reinforcing by being paired with primary reinforcers.
Conditioned reinforcer - (secondary reinforcer) is a previously neutral stimulus that begins to function as a reinforcer after being paired a number of times with an established reinforcer.
Negative reinforcement - means removing an unpleasant event as soon as a desired response has been preformed, in order to increase the frequency of the response. The probability of a behavior occurring in the future is increased when the behavior is followed by the removal or avoidance of a negative stimulus.
Reinforcement - occurs when a behavior, followed by a consequent stimulus, is strengthened, or becomes more likely to occur again.
Reinforcer - is a stimulus that, when presented following a behavior, causes the behavior to be more likely to occur again in the future.
Clicker Training - is one application of conditioned reinforcement. The clicker is a neutral stimulus that if followed by a primary reinforcer, and eventually becomes a conditioned reinforcer.
Positive reinforcer- is a stimulus that, when presented following a behavior, makes it more likely that type of behavior will occur in the future. examples: food, praise, toy.
Escape Conditioning - escape comes after the averse event has occurred.
Avoidance Conditioning - avoidance occurs before the adverse event takes place.
The first four definitions that I wrote about have to do with Pavlov and when he rang a bell before a dog was given food it started to salivate. In the dogs mind, it hears the bell, knows food is coming and starts to salivate. This all ties into clicker training which is a conditioner reinforcer. This means that you have to associate the clicker with something the dog finds desirable. I almost always use food. When I want the dog to associate the clicker with food, I tell my clients to "charge the clicker". How they do that is have about 5 to 10 treats in one hand and the clicker in the other. You don't tell the dog to any command. You just click the clicker and follow with a treat. After a few days the dog will associate the clicker with something good.
Primary and secondary reinforceres paired together help a dog learn better and faster. The food is a primary reinforcer and the clicker is the secondary reinforcer. The food is primary because it is something the dog can't live without. It already has a positive meaning to it. So when you give a dog a treat after it has done something good, you are reinforcing the behavior. The clicker is the secondary reinforcer. This is because if you just click a clicker with out "charging" it first. It has no meaning to the dog. Once you pair the clicker with food, the clicker becomes a positive reinforcer as well. When training with clicker and treats it helps the dog understand that what they just did was rewarded double. They will be more likely do the behavior correctly again and have the behavior stay in their minds faster.
Negative Reinforcement, simply put, gets the dog to do the behavior more often but by using an unpleasant event first. I know this sounds confusing and it was for me when I first learned it but let me try to explain it. When you are walking a dog on a pinch or choke collar, if they pull ahead or lag behind, you pinch the collar to get them to walk next to you. So with negative reinforcement, the dog wants to avoid getting pinched so when they hear the sound of the collar they get into heel position so they don't get pinched. You are using a negative stimulus, ie. pinch collar, to make a behavior occur more often. I personally do not use this method. I have been trained on this method but feel dogs learn better if they are able to think about how to get the treat. I will go more into this later.
Escape conditioning is part of negative reinforcement. Again I do not practice theses types of training. With this type of training the dog is given a negative reinforcement first and once the dog does what is expected, the negative reinforcement goes away. Here is an example. I have never used one but a bark collar. There are two different kinds, one that zaps a dog when they bark and one that sprays a dog with citronella when they bark. So when a dog barks they get zapped or sprayed. When the barking stops the negative reinforcement goes away. As long as the dog doesn't bark they won't get zapped or sprayed. I teach a different technique to get a dog to stop barking but I'll go into that on another post.
Avoidance conditioning is again negative reinforcement. With this method you would give a warning sound then the negative reinforcement is given. So the example that I am about to give is something I have done but I did not use any pinch or choke collars. When I would run with my dog Riley he always liked to pee every 5 minutes or stop and sniff everything. I wanted to keep going and keep a decent pace. When he would start to pee or sniff I would say leave it and pull the leash to let him know that we needed to keep moving. Eventually, I would say leave it and he would just keep running with me. After a while he didn't stop to pee or sniff very much at all because he knew what was expected of him.
I'm going to talk about my opinion about the different reinforcers. With positive reinforcement I feel dogs learn a lot better with this method. It lets them think about what will get them the treats and will work to get the treat by doing the desired behavior. They are happier learning this way and won't get scared or intimidated by their owner. When a dog gets to use their mind, they are more likely to remember the command. When using negative reinforcement or even punishment, dogs tend to shut down and don't learn as well. I have trained both ways and from my experience dog just learn better using positive reinforcement. It is important going in to training to know what your dog will respond too. I know there are a lot of people who like to use the shock collars because they feel like their dog will respond better to that type of reinforcement but you need to make sure that you dog will not shut down when they are zapped. Another reason people like to use shock or pinch collars instead of positive reinforcement is dog learn faster when using negative reinforcement. People what to see the results now and don't want to wait for the result. Here is an example, I had a client who wanted her dog to be trained on the invisible fence. This is a good concept for people who don't want to spend a lot of money on a solid fence but it is not good for all dogs. So when I started training her dog we would walk the fence line where she had the flags up. The dog would go close to the fence line and you are supposed to call the dog back to you once you start to hear the beeping. The beeping comes before the zap. Well, they had the fence were you don't put the wire around your yard. You put a base inside your house and it sends a signal outside that makes a diameter around your house. Well, the line moves a lot and it really is never in the same place twice so the dog got zapped. He didn't like it at all because he tried to bite. He was just going after what ever was cause him pain. He was the sweetest dog you will ever meet and you know that this was not a normal reaction for him. After that he didn't want to go outside or do anything. He shut down. This is why it is important to be able to read your dog when using negative reinforcement. If your dog doesn't respond well to the reinforcement stop what you are doing. I know a lot of people have their own opinions on this matter but that is mine and I have a lot of experience with both positive and negative reinforcement. All I ask is if you are or thinking of using any type of negative reinforcement for dog training consult a trainer before you do it. You don't want to damage your dog. Dogs can become more aggressive if you use negative reinforcement instead of positive. Think about it. Say you have a dog that barks or lungs at other dogs while you are walking. If you have a pinch collar or zap collar on your dog and every time you walk past another dog he gets pinched or zapped, what type of message do you think are are sending to your dog. The will associate other dogs with pain. The more you do it the worse it will get and you will have a dog that can't go around other dogs.
So to conclude, Positive and negative reinfocement both have thier positives and negatives. I will only use positive reinforcement in my training but if you want to use negative reinforcement, just do your research first.
I know this is a long post with a lot of info on it but I feel very passionate about this topic and what to talk about it. I am going to do other blogs about the How Dogs Learn book later covering different topics.
Here is a link to the book that I am referencing.
