Positive reinforcement is something that the subject will work to gain, such as food, attention or freedom. It can be anything that the subject wants. The key to using positive reinforcement effectively is in understanding what is reinforcing to one individual is relative to what that individual wants at that particular moment. Your dog may work for food or play reinforcement in your living room but may have little interest in these “gains” when outside. A good trainer will know what is reinforcing to his/her dog in any given environment or situation.
Reinforcement can be anything the subject wants to gain at that moment. As an example, I have a “studly,” all-boy, very active and agile two-year-old bullmastiff named Zachary. He would like nothing more than to urine mark every tree, bush and blade of grass he passes. I have turned urine marking into reinforcement for attentive (i.e. strong eye contact) heeling. I will discuss this further when we get to shaping behaviors. The important thing to note here is that I know what is reinforcing to my dog and I know how to use it.
Positive Reinforcement: Your Strongest Training Tool
Positive reinforcement is the strongest training tool. The great advantage of focusing on positive reinforcement is that this is what each behavior is seeking. The other great advantage is that there is always something that your dog wants at any given moment, even if it is to simply be left alone. Another benefit of positive reinforcement training is that it is easy to use this tool to focus the dog’s attention on you. All good things must seem to come through you. This builds trust, attention and a willing, eager attitude. Positive reinforcement training also builds confidence as your dog is constantly being reinforced for acting “correctly” as opposed to being broken down for incorrect actions.
Don’t Think it Works?
If you don’t think positive reinforcement training can work for your dog, you don’t understand the principles. Many trainers have jumped on the positive reinforcement, clicker bandwagon, running with a fad. Few that I have watched truly understand the principles of learning theory. The clicker is indeed a great teaching tool, if applied correctly. If not, it is very limiting. If you would like to see the real results of conditioned reinforcer/clicker/positive reinforcement training, watch some wild animal shows, or the caretakers at some of the better zoos. Have you noticed how actively involved these animals are with their trainers? Have you noticed how much these animals and their trainers are enjoying the interaction with each other? Let’s face it: no one could force these animals to perform. The birds could fly away if they wanted to, the lions could eat their trainers, the dolphins could swim away and the elephants could just push their trainers out of the way and go on a rampage. When you see happy, interactive animals it is due to positive communication skills. Wouldn’t we like our dogs to perform and be as committed to us as the animals in these shows?
HOW You Reinforce is Extremely Important
How often you reinforce a behavior has a profound effect on how strong and persistent, or how weak and inconsistent, a behavior will be. We tend to believe that if we reinforce a behavior constantly, every time, that we will have a strong behavior. This is not true. I refer to this as the “McDonald’s Syndrome.” To explain, no matter how well or poorly I work, I would still make minimum wage. So, who cares? Why work harder than I have to? “Here’s your stale hamburger, Mister, now go away.” From our dogs’ perspective, every time they come, they get a biscuit and get petted. If they come quickly or come slowly, it’s the same biscuit. Why should they work harder? The payoff is always there. Intermittency, and specifically, selective intermittency, on the other hand, builds persistence. It turns dogs into salesmen. “If this doesn’t work,” they’ll think, “I’ll try harder.”
We all know persistent, solicitous dogs who have honed their skills by pushing on when the owner first tries to ignore them. They have learned that if they just try harder, the prize is theirs. This is because these dogs have learned that if their owner ignores them, they just need to try harder. They have learned that the human will break and give in before they will. Behaviors that are intermittently reinforced do not go away easily. This is great if it is a behavior that you want, and not so great if it is a behavior you wish to eliminate. Now, if we can learn to use intermittent reinforcement selectively, we can promote optimal components of any given behavior. As an example, if I reinforce my dog only for his faster recalls (“come” commands) and not his slower ones, he will soon learn to move faster in order to win the prize.
There are those who feel that only positive reinforcement should be used in dog training, but I am not one of them. I am a strong believer in positive reinforcement and use it as my foremost training tool. However, nature has given us four tools with which to work and I believe they all have a place in training. I believe that it is best to learn about all four tools and how best to apply them. You should know their virtues and their downfalls before choosing to use them in any training program.
We have discussed the powers of positive reinforcement. This is truly the strongest tool available when teaching, whether it is used to teach our dogs or other individuals. But we must not discard the other three tools available to us.
Negative reinforcement can be explained as a stimulus that induces a specific behavior to occur. You will recall that negative represents a minus, as in “taking away.” Once the behavior occurs, the inducement, the negative reinforcement ceases. For example, almost all communication in equine sports is negative reinforcement. If I want my horse to move to the left, I would apply leg pressure to his right side. As he moves to the left away from the pressure, the leg pressure ceases. Our dogs are experts at the use of negative reinforcement. For example, a dog that continually nudges his owner’s arm until the owner responds with any sort of attention has negatively reinforced his owner to respond appropriately to his desires. In turn, the owner, by responding and giving the dog any level of attention, has positively reinforced the dog’s solicitous actions. In other words, the owner is training the dog to control when and how the dog gets his attention.
Whenever we put a leash on our dogs we are preparing to use negative reinforcement, or unaware of the consequences, preparing our dogs to use negative reinforcement on us. When a dog pulls on the leash and the owner responds by moving into the pressure, the dog has succeeded in negatively reinforcing the owner to follow his direction. In contrast, if when the dog pulled the owner stopped, and held steady until the dog stopped at about her slack in the lead, it is possible to use negative reinforcement to teach attentiveness to the owner’s directive.
Negative reinforcement definitely has its place in teaching. However, if used to excess, or without balance, it can often cause for resistance and an inattentive attitude toward the owner and toward learning. If we keep in mind that all recurring behavior must be gainful to the animal in some way, it is best when using negative reinforcement to induce a response, to pair your training with positive reinforcement once the appropriate response is exhibited.
Punishers as a whole are something averse that the subject will work to avoid. Punishers do not have to be physically painful. A loss of food, comfort, or possibly social interaction could be viewed as an adverse action to some individuals. Just as with reinforcement, punishment is subjective to the individual at each moment. What is averse to one subject may actually be reinforcing to another.
For example, my dog loves petting and attention. This can be very reinforcing to him. However I have met many dogs that display shy and distrustful temperaments. Such dogs may find petting and attention from a strange or unfamiliar person to be averse. It is important to acknowledge that an action, in order to be considered a punisher, must be averse to the individual subject to which it is being applied. If the dog does not work to avoid this situation in the future one may assume one of two things: first that the action is not averse to that dog; and second, that the application of the punisher was poorly timed. If timing is poor, the dog may not associate the action with the specific behavior that the owner/trainer wants to diminish.
There are two types of punishers: positive and negative.
Positive punishment indicates adding an offensive or adverse stimulus. A good example of this method would be electric shock collars used in training or electric fences and containment systems. As the dog displays an unwanted behavior, such as approaching the border of the containment system, he receives a shock. On average, in teaching containment, a dog will be shocked three times before making the association between approaching the training flags and the application of the shock. If the positive punishment is used correctly, which is very difficult to do in the real world, the subject learns the association between the behavior and the consequence within three to five applications. This may sound like a perfect way to eliminate unwanted behavior. However, in order for a positive punisher to accomplish the goal, it must be applied at a level that will absolutely deter the subject. It must also be properly timed at the onset of the behavior, as opposed to when the behavior is being performed or has ended. Another consideration is that the punisher must be applied each and every time the dog attempts the behavior.
Positive Punishment Can Have Serious Consequences
Positive punishment also has serious side effects. A positive punisher will never teach the dog the appropriate behavior; only tell the dog what not to do. Given the fact that your dog must be doing something at any given moment, it is always to our benefit to focus on empowering them with what to do in any given situation as opposed to breaking them down for what they’re doing incorrectly. We must also keep in mind that the use of positive punishment can, and often does, contribute to an increase in distrust, fear, frustration, anxiety and aggressive behavior. Also any thing or any one who is associated with the punisher can become a part of the dog’s fear and aversion.
Recently I met with a man who inadvertently taught his dog to distrust children. His dog, a lovely one-year-old golden retriever, would often leave his unfenced yard to run across the cul-de-sac to play with the neighboring children. To deter the dog from doing this the owner fit his dog with a remote control shock collar. As the dog ran to the children the owner would shock the dog until it stopped. After just three occurrences the dog stopped running across to play with the children. However, what this man did not plan on was that his dog associated the children with the painful shock to his neck. Now he had a dog that ran in fear from any child who approached, including his own. The more one learns about positive punishment, the greater the realization that although positive punishment may have a place in teaching, that place is very limited, must be planned well in advance of its use and must be executed with skillful and knowledgeable application.
Negative punishment, although it sounds like the worst of all tools, can be a productive tool if used properly. Negative punishment can be explained as the systematic and well-timed subtraction of reinforcement. For example, if I walk into a home and the resident dog seeks me out in solicitation for attention, I will calmly and completely ignore the dog until he has settled into a lying position on the floor. Although I must walk into many homes on a daily basis due to my work, I rarely get jumped on, and if so, not for long. As a dog is seeking my attention, and in some cases seeking the ability to control me, my neutral and inattentive attitude quickly diminishes the dog’s attempts. The behavior has not provided gain for the dog. Therefore, the dog ceases wasting energy on an unproductive endeavor.
Negative punishment paired with positive reinforcement becomes a very strong and influential training combination. This combination creates a wonderful balance of gain and loss that can easily guide a dog to a self-confident, self-calming and self-controlled individual. It is through this combination of tools that capable animal trainers guide their wild students to such beautiful performances. This is very much like the behavior contracts that parents use with their children. We teach our children to become responsible for their own emotions and actions in order to control their own consequences. They learn quickly under such guidance that if they choose to do one thing, they can gain something they like, but if they choose to do another, they will loose something. Children very quickly learn to take control of their destiny when parents – like good trainers – are fair, consistent, and calm about applying consequences, whether desirable or not.
Extinction is the elimination of a previously performed behavior by revoking the previously gained reinforcement to that action. For example, as a puppy, my dog attempted to jump and paw at my legs in an attempt to gain attention. No doubt, prior to my adopting him, his breeder gave attention for such behavior. As this adorable little puppy grew into a massive and powerful adult dog, this behavior could not be continued. Each and every time my dog attempted to gain my attention by displaying this behavior, I completely and calmly revoked my attention and walked away.
Each time he was calm and had all four of his feet securely on the floor, I approached him with praise, petting, treats, play, or picking him up. But the very moment he even thought about jumping again I would immediately turn away and ignore him. In less than a week my dog learned that jumping on people did not gain him the desired response of attention and social interaction. On the contrary, by using positive reinforcement for an alternate, more appropriate response, such as standing, sitting or lying still, I quickly taught my dog that by controlling himself with a calm and controlled manner he could gain the attention and play that he sought. Extinction is the elimination of the behavior by the appropriate use of negative punishment. It works best when used in conjunction with positive reinforcement being offered appropriately for the display of an alternate and desirable behavior.
We have discussed the effect of genetics, psychological (i.e. learning and emotional) growth stages, and basic learning theory. Before bringing the pieces all together, we should discuss one last area of importance, the basic principles of canine behavior. In this discussion, we will better understand how our communication and daily interactions affect our dogs’ behavior, attitude and temperament.
Shaping is the process of building a behavior by reinforcing successive approximations of the desired end behavior. Successive approximations are any behaviors that bring the individual closer to the desired behavior, or any behavior that resembles the desired behavior. Let’s look at two examples.
Shaping With Reinforcers
Earlier in this text I mentioned how I started to teach Zachary, my bullmastiff, to heel. Again let’s look at this process. In the end I was attempting to teach an attentive heel. This meant that he should walk at my left side with his front feet right at my side, matching my exact pace and looking up at me as we walked. I started teaching this behavior when Zachary was 6 months old and already quite powerful.
I began this training by first standing still and holding Zachary’s lead just long enough for him to be able to stand calmly at my side. At first he struggled and pulled, but soon he realized there would be no gain for pulling. As he settled and the tension came out of the lead, I yelled “Go!,” marking his moment of correct behavior, extended the lead and ran with him to the nearest bush to allow him to get to a tree to urine mark. We repeated this several times and he quickly caught on. The fourth time I shortened the lead, Zachary settled immediately. However, I did not release/reinforce him. As he looked up at me in wonderment, I yelled, “Go!” and offered the reinforcement. Now we have calm behavior and he will visually reference me. As we continued on, I added in for him to take one step with me, the “Go!,” and so on, until I taught Zachary that the only way he was going to get to the bushes and trees to urine mark was to first attentively walk in sync with me. This is the intentional shaping of a wanted behavior and attitude.
Shaping can also work to our detriment. If we inadvertently give attention to the wrong behaviors we can cause them to build. I have a friend who taught her Akita, Cody, to be very noise sensitive, especially to thunder. As a puppy, Cody was very bold and fearless, exploring new objects and situations with confidence. During Cody’s first spring/summer season, our region had a great many thunderstorms. At first Cody showed no concern for the storms and continued to play, eat, or rest as they progressed. However, my friend has a terrible fear of storms and would pace and shiver as the thunder boomed. To ease her discomfort, she often picked up Cody and held him tight, pretending that it was he who had the concerns.
Cody soon started to become attentive to the booms of thunder. As he would startle, my friend would caress him and try to soothe him. By the time he was one year old, Cody became very fearful of thunder and other loud sounds. At a strange or startling sound, Cody now shakes and seeks the comfort of my friend, who, in turn, reinforces him with her praise and petting. Today, Cody is a very fearful dog, who has taken his concerns for life and the world around him from my friend. She unintentionally taught her well-bred, well socialized, confident puppy to be fearful by being a weak leader, and by giving attention to the fearful behaviors that he displayed.
Read about Dog Behavior.
To learn more about my training methods, sign up for a dog obedience class or to schedule an in-home, private behavior consultation, please contact me today.
“Linda is amazing, both as a trainer and a person. Her knowledge base is unrivaled, as is her ability to translate this knowledge into helpful, usable suggestions for her clients. Her training classes not only support her clients, but also their dogs. Her classes enrich and strengthen the vital trust relationship between dog and owner. This step is missed by many less-experienced trainers. Dogs walk away from her classes with a stronger sense of self control, and composure. Clients walk away from her classes with a better understanding of their beloved canine companions, and a more fulfilling, comfortable relationship with their dogs.”
– Amy Fellner, Certified Veterinary Technician
Veterinary Behavior Technician