How Dogs Learn - Part II

Friday, April 01, 2016 12:30 AM | Front & Finish (Administrator)

Written by Connie Cleveland-Nolan

Last month, in Part I of this series, I presented the flow chart illustrating the principles I use when training dogs and my beliefs about how dogs learn.  I believe that because dogs are problem solvers, and they learn by trial and error, it stands to reason that sometimes they will respond correctly and sometimes they will respond incorrectly.

Dog Makes the Correct Choice

When your dog makes the correct choice, how should you respond?

Handler Uses Positive Reinforcement

The definition of reinforcement is that it increases the likelihood that a behavior will increase in both frequency and intensity. That’s our goal – correct behaviors more often and with more enthusiasm.

Praise that Can Be Followed by a Toy or Treat

How many ways are there for you to give positive reinforcement? The left side of the flow chart says “praise followed by a toy or treat.” The reason is obvious; you need a dog that responds to your praise, not just a treat, or a toy.

However, there’s more to it than that. Praise must be sincere. If you look at the dog heeling next to you, making eye contact, or the dog that is still sitting where you left him after you have been out of sight for three minutes, you need to demonstrate how much you appreciate the effort he’s making. Your praise must be genuine and expressed with sincere enthusiasm.

What makes a dog want to be praised? Any behaviorist will tell you that it’s all about pairing the praise with something the dog enjoys. That’s true, but there is more to it than that. Dogs (unlike a lizard or a snake) have a limbic part of their brain that interprets emotion. Your dog knows when you’re sad, sick, or upset. He also knows when you are genuinely, 100% thrilled, out of your mind excited.

When someone compliments you, dopamine is released in your body and it makes you feel good. Dogs are the same. When you convey that kind of excitement, he will have a dopamine release and with time, start loving the activities that cause you to praise him.

I can’t leave the subject of praise without pointing at what it is not.

      Praise is not obligatory. Do not say “Good Dog!” unless you mean it, if you don’t feel it, don’t waste your breath. Your dog will know that you are not sincere.

      Praise does not start wanted behaviors. Praise rewards existing behaviors. A cheerful hurry, hurry, hurry!” does not cause a lagging dog to speed up, it tells him it is okay for him to lag.

Promise me that you will only praise when you sincerely, unequivocally feel genuine enthusiasm for the dog’s performance. If that is the case, your dog will respond to your praise, even in the absence of the toy or treat.

A Conditioned Reinforcer Must Be Followed by a Toy or Treat

Another form of positive reinforcement is a conditioned reinforcer. Remember Pavlov’s bell? He rang the bell and the dogs drooled. The bell is a “conditioned reinforcer.” Desired behavior is reinforced or strengthened when a conditioned reinforcer is associated with a primary reinforcer (e.g., food).

A conditioned reinforcer is often a noise - it can be a clicker or a distinct sound or word. I use a verbal conditioned reinforcer to “mark” specific behaviors. I chose the word “Yes!”. I am distinct and exude an enthusiasm that my dogs recognize when I say “Yes!” More importantly, “If I say it, I pay it!”

In the flow chart, I state that if the dog makes a correct choice, the handler responds with positive reinforcement. Why use a conditioned reinforcer instead of simply handing the dog a treat?

Scientific studies show that a distinctive sound (a conditioned reinforcer) that precedes delivery of food (a primary reinforcer) cause dopamine and other chemicals to release in the pleasure and memory centers of the brain. These chemicals cause enjoyment and pleasure. When a behavior consistently produces these chemicals, the subject starts to enjoy the behavior  that leads to the toy or treat.

Read that sentence very carefully. Dogs (and people!) are pleasure seeking. By using a conditioned reinforcer (“Yes!”) paired with a primary reinforcer (e.g., food) you can cause a dog to enjoy the aspects of obedience that are not self-rewarding such as fronts, finishes, and pivots.

Furthermore, a conditioned reinforcer marks and rewards specific behavior. For example, if you teach your dog to run to a bed, and consistently “mark” the exact moment he puts his fourth foot on the bed, the dog will come to understand that it is the act of getting on the bed that causes the reward to happen. If you simply praise and feed him, he may not clearly understand what it is that you are rewarding because several things are happening at the same time. You are walking toward him to a deliver the treat with a pleased look on your face and he is turning on the bed, wagging his tail and waiting for you. It’s simply more difficult for a dog to ascertain exactly what behavior caused the reward. For this reason, the chemical release does not occur or is not as intense.

Be aware of some important points when using your conditioned reinforcer. Excited by your dog’s response to the distinctive sound, you may start to randomly use your verbal conditioned reinforcer and not deliver a treat. I watched a trainer heel a dog around the ring chanting “Yes!” as if she were keeping rhythm with the sound of the word. This will cause the sound to become meaningless to the dog. Thus the term, “If you say it, you must pay it!”

Don’t bother using your conditioned reinforcer for activities the dog finds naturally rewarding. If your dog enjoys jumping, let jumping be the reward, you do not need to add anything more to the activity. Instead, use your conditioned reinforcer for the details that are hard to communicate and harder still for the dog to understand, like sitting attentively in heel position, pivoting and then keeping his head up, straight fronts, and holding the dumbbell properly while sitting in front of you.

As dog trainers, we say, and often believe, the dog only does it “for the treat.” However, when you use a conditionedreinforcer properly, you will discover that the sequence of events that lead to the reward becomes enjoyable for your dog. Doing obedience with a dog that enjoys the activity is the most fun of all, for both of you.

Dog Makes an Error (Incorrect Choice)

If you believe that dogs are problem solvers, and they learn by trial and error, then sometimes they will make errors. How should you respond to his mistakes? Not all errors are the same. Sometimes a dog makes an effort error, and sometimes he makes an error due to a lack of effort.

Effort Errors, Characterized by Confusion or Fear

Effort errors are made by dogs that are attentive, but appear confused, worried or fearful. When your dog makes an effort error, he will typically offer one of two responses: (1) He does nothing, or (2) He offers the wrong behavior.

Dog Does Nothing 

How should you respond to an inexperienced dog that stares intently at you but does not move when you say “Sit” or the dog in a sit at go-out that does not move when you say “Jump”?

Dogs, similarly to horses and goats, learn when physically moved in the correct direction.This is not true for all species. Cats do not learn by giving them direction nor do chickens or dolphins. Because your dog can learn when physically moved in the correct direction, use it to your advantage!

Be gentle, but firm as you pull up on the dog that fails to sit. When he fails to move in response to your jump command, go get him and physically guide him in the correct direction. He has made an effort error and you are providing him with information about the direction in which he should move.

Dog Offers the Wrong Behavior

Sometimes an effort error is demonstrated by a dog offering the wrong behavior. How should you respond to the dog that heads for glove 2 when sent for glove 3 or the dog that comes straight to you when told to jump over the broad jump?

Dogs have the ability to solve problems. Offering any behavior, even a wrong one, is an attempt to solve a problem. Follow the steps outlined in the article “A Simple Rule to Train By,” (http://www.dogtrainersworkshop.com/connies-corner/a-simple-rule-to-train-by/)  tell your dog he made a mistake, calmly take him by the collar and guide him back to the position in which he was last right, then start again by simplifying the task. By following this rule, you are giving your dog another opportunity to solve the problem.

Lack of Effort Errors: Characterized by Inattentiveness or Disinterest

Not all errors are committed by dogs that are attentive and trying their best. Dogs also make “lack of effort errors” when they are distracted or uninterested.

Have you ever called your dog to come, seen him look in your direction, ignore your command and return to sniffing as if to say, “Just a minute, Im busy.” Have you ever told your excited and/or distracted dog to “Sit,” only to watch him act as if he didnt even hear you? Have you ever given your dog a command and watched him respond as if he was not even trying to be obedient? These are all examples of “lack of effort errors.”

If you agree that lack of effort errors occur when our dogs are not trying, the question becomes how to respond? Dog trainers are quick to use the word “correction,” as in “I would correct the dog,” but what does that mean? The word correction has been used in so many different ways that it has become ambiguous. For example, if you were to tell me that your dog is running around the high jump and ask me, “How would you correct him?” I could not be sure whether you are asking me “How would you fix it?” or “What kind of negative response would I use to stop him?”

I use negative reinforcement when a dog is not trying. By definition, reinforcement increases the likelihood that a behavior will occur. Negative refers to something the dog wants to stop and wants to prevent.

For example, think about your seatbelt buzzer. That annoying sound increases the likelihood that you will buckle your seatbelt, so, by definition, it is reinforcement. It is negative because you want to make it stop.  Thankfully, you know how to do so. When you hear it, buckle up. You can prevent the annoying sound by buckling your seatbelt before you start the car. You are in complete control of the annoying buzz!

How about an underground fence? Smart fence owners teach their dog how to control the sound and electric stimulus the electric fence collar provides. First, they teach their dog where the boundaries of the yard are located. Next, they walk their dog around the boundary, and when the noise occurs, they pull the dog back into the yard (dogs learn by being given direction). When done properly, the dog learns how to stop the sound and electric stimulus (the "aversive") and how to prevent it from occurring again. The dog romps happily around the yard offering the desired behavior (staying inside the delineated boundaries) because he knows how to control the aversive.

A pop on the leash is another example of negative reinforcement. My dogs learn that a pop on the leash can be stopped by looking at me. They also know that they can avoid a pop on the leash by giving me their attention. My dogs know that they are in control of the pop and comfortably give me their full attention.

We have all seen trainers who become frustrated and lose their tempers. We have seen dogs that become nervous and scared when aversive techniques are used. This occurs when the dog does not know how to control the negative events from happening. None of us are interested in making our dogs nervous or scared. In fact, we are horrified by the thought of it. That is NOT what I am talking about. I NEVER do anything that my dogs consider offensive until I have, in a calm and systematic way, taught them how to stop it and how to prevent it.

Promise me that you will never do anything to your dog that he finds offensive unless:

(1) He has made a “Lack of Effort Error”; and

(2) You have taken the time to teach him how to stop it and how to prevent it.

If you make this promise, your dog will enjoy working with you in the same way the well-trained dog enjoys romping in the yard within the confines of the underground fence.

I have been training dogs for 40 years. I have had 10 Obedience Trial Champions, 4 Field Champions, and earned numerous other titles. I have taught obedience classes in Greenville, South Carolina for over 30 years and have helped countless people earn titles on all breeds of dogs. Through it all, I have not grown tired, nor do I take for granted, the skills that dogs allow us to teach them. I have trained dogs to do obedience, field work, service work, therapy work, tracking, and even some detection. No matter what I’m teaching a dog, I follow the principles outlined in this article. By presenting the task as a problem, and helping the dog solve it, I am repeatedly able to teach dogs the countless tasks presented to me. I am wowed every day by what dogs are able to learn. I hope that you are too!

To view more articles please visit our Members Page!



To view more articles please visit our Members Page!

Home     About     Advertising     FFX Awards     Join Us!     Issues     Ratings     Search     Shop     Sitemap     Subscribe     Terms of Use     Privacy Policy     Writer's Guidelines



Copyright 2018 Front & Finish, LLC
2310 US Highway 150 North
Wataga, IL 61488-9520
Questions?  Need help?  Email us at: dogs@frontandfinish.com

Please Note:  Our site is secure!  Clicking links that lead to resources outside this site may issue a security message.  This is for your protection!