Positive Dog Training

Home Positive Dog Training Fighting Fire Book Reviews Socialize Your Dog The Ubiquitous Labrador Retriever Potty Training Chill Out

Holly’s Den *

Fostering Peaceful Packs for Humans & Dogs 

San Antonio, TX

 

Training With Positive Reinforcement & Clicker Training

 

return to Pup & Basic 1 Lessons

 

Successful training is based on the following behavioral science laws of learning:

All animals repeat behaviors that are rewarding; behaviors that are positively reinforced become stronger and more frequent.

Behaviors that are not rewarded become less frequent and intense and may eventually extinguish.

Therefore, controlling consequences is the key to controlling behavior

Your job as your dog’s trainer is to do the following:

bullet

Proactively reward good behavior with praise and treats.

bullet

Ignore, prevent, interrupt and (when necessary) reprimand inappropriate behaviors.

 For more info: http://positively.com/positive-reinforcement/the-science-behind-the-philosophy/

HOW DOGS LEARN - Training usually includes aspects of both operant and classical conditioning.

Operant Conditioning--When a trainer provides either good or bad consequences in order to get a dog to increase the frequency of a desired behavior, or to decrease/eliminate an undesirable behavior, we call that process operant conditioning (OC). Basically, all you need to understand about this is that your dog can learn that his behavior has consequences, and that he can “operate” on his environment and make good or bad things happen as a result of his own actions.  (If I jump up, she turns her back and ignores me, but if I Sit politely for petting, she praises me and sometimes gives me a treat!) As training progresses, your dog will actually be learning that the way to get what HE wants is by doing what YOU want!

Classical Conditioning—The other common way that dogs learn is by associating one thing with another, the same way that Pavlov’s dogs’ learned to associate the sound of a bell with the subsequent dinner, so that they would start salivating as soon as they heard the bell.  Likewise, dogs learn that when you pick up the leash and your keys, that predicts a walk. This learning by association is called “classical conditioning.”  

 

More about operant conditioning-- There are 4 ways trainers can provide consequences to change behaviors:

  1. Positive reinforcement-- Rewarding your dog for good behavior.  Positive is being used in the mathematical sense of adding something good such as a food treat.  

  2. Negative reinforcement—Cessation of punishment--taking away/subtracting something unpleasant when the dog does what you desire. (For ex. releasing tension on a tight leash when the dog stops pulling).

  3. Positive punishment—Adding an unpleasant or painful consequence to inhibit or suppress a behavior

  4. Negative punishment—Taking away or withholding something good such as attention, praise and treats.

 

Out of these 4 methods, we will be placing our focus on positive reinforcement, using rewards ranging from food treats, praise, and toys, to “real life” rewards such as play or getting to go outside, or for a walk or car ride. 

Occasionally, in response to incorrect or undesirable behavior, we will also withhold rewards, use short time-outs or give a verbal reprimand immediately followed by telling your dog what he/she should do instead (for example, tell the dog  “Off" (the furniture) — "go to your mat” or “Drop it" (my shoe) — "here, chew your toy”).

 

Why positive reinforcement?  Because we don't have to hurt our dogs to train them.

 

Dog friendly training has four cornerstones:  

1.       Reward correct behavior

2.      Manage the environment to prevent problems

3.      Assume leadership through control of the resources

4.      Use gentle pain-free methods that do no harm.

 

The Case Against Physical Punishment: Punishment (being defined here as using aversive methods intended to cause fear, pain or physical discomfort) is generally counter-productive. There are several reasons to avoid the use of anger and punishment in training.

bullet

Punishment can make your dog like and trust you less:  Any good relationship, including your relationship with your dog, must be based on trust, and in order for your dog to trust you, he must feel safe with you.  Physical punishment can create a negative association in the dog's mind between the punishment and the object of the dog's fear, thus having a negative effect on your relationship. It can make some dogs aggressive, and destroy self-confidence in others.

bullet

Punishment increases stress: Any increase in stress will interfere with your dog’s ability to learn.  If your dog associates punishment with training, instead of being a happy willing worker, he is much more likely to find it unpleasant and he may even “shut down.”

bullet

Punishment is confusing: More often than not, punishment fails to provide specific information about what the dog is doing wrong or what you want him to do instead.  For example, if the a yard crew comes by and your dog starts barking and jumping at the window and you punish him, he probably doesn’t know if he was being punished for barking, jumping at the window or doing this in front of you!  Next time the doorbell rings, will he get in trouble if he barks?  If he hears an intruder at the back door and barks, will he be punished?  If he sees a cat outside and barks at the window when you aren’t home, and nothing bad happens, doesn’t that mean it’s OK to bark and jump on the window as long you aren’t around?   

bullet

Punishment can produce a dog that strikes without warning: By suppressing outward signs of reactivity (such as growling) without changing the underlying emotional state, punishment can lead to the dog inhibiting his growl until he is feels pressured enough to bite.

bullet

Punishment can create owner-absent behavior problemsthe dog thinks he is safe to do the behavior as long as you aren’t there.

 

Delayed punishment—Pat Miller, author of The Power of Positive Dog Training notes this is one most common mistakes that pet owners make and it’s one of the most unfair things you can do to your dog!  It is very important to understand that generally speaking, puppies and dogs live in the present. 

Timing is crucial in training!  In order for a dog to understand and make the connection between an action and the consequences for it, the consequences must occur immediately:

To be effective, you must give your dog feedback (whether it be a reward or a reprimand) either during the behavior or within one to two seconds of the behavior.

 If you punish a dog after the fact, the dog will think he is being punished for whatever he happens to be doing at the exact moment that you punish him.  Maybe he chewed up a shoe while you were gone.  When you come home, he runs up to greet you and he’s wagging his tail happily when you spot the chewed shoe--so you grab him, scream at him and shake him.  He will not connect your outburst with the chewed shoe, but rather with running up and greeting you. 

But, you say, I grabbed the shoe and showed him what he did wrong and the next time I came home and he had chewed something and ruined it, he knew he had done wrong because he really acted guilty.  

As Miller explains, dogs don’t feel guilt--Dogs are not humans in fur coats and they do not feel guilt over their behavior.  Romping in garbage, rolling in yucky things, digging in flowerbeds, eating any food they can forage and chewing as a stress and energy outlet are completely normal and natural canine activities.  What people interpret as guilty behavior when they find a dog hiding under the bed, hanging his head and tail, rolling belly up, or peeing, is simply the dog making submissive appeasement gestures because that’s what dogs do when they sense anger from your body language and tone of voice.  If your dog cringes when you walk in the door, it’s simply because he has learned to associate A) a mess in the house and B) your homecoming, with adding up to bad news for him!  He can learn this association as easily as he learns that the doorbell ringing predicts a visitor.  But the garbage or chewed up shoe doesn’t bother him at all, as long as you’re not there.  When you do come home and make a scene, he doesn’t have a clear idea why you’re so upset with him.  Don't unintentionally hurt your dog by punishing him when he doesn't really understand what he has done wrong!

 

Question: If you shouldn’t punish, what should you do instead? Again, trainers such as Ian Dunbar, Jean Donaldson, Pat Miller, Dr. Patricia McConnell and many others have provided good answers:

The alternative to punishment is NOT permissiveness!  There are other management and training options for dealing with problem behavior. You can:

 

bullet

Control and manage the environment: If your dog gets on your bed or gets into the garbage, shut the bedroom door and keep the garbage pail in the cabinet under the sink This is the same sort of thing parents do when "baby-proof" their house. Just as you wouldn’t leave a baby or toddler alone in your house or yard, you shouldn’t give an untrained dog more freedom or responsibility than she is ready to handle!  Until your dog is reliably trained, she should either be confined in some way (crated for periods of less than 3 hours or for longer periods in a safe room behind a door or baby gate along with her stuffed Kong, other interactive toys & Nylabone) or under your watchful eye at all times!  

 

bullet

Prevent, ignore or interrupt certain problem behaviors:  If your dog is barking in her crate, ignoring her is probably the best choice; then when she is quiet, you can praise, reward and release her.  If  your dog barks at other dogs when you are out walking, be pro-active in ways that can prevent that behavior (see Chill Out handout).

 

bullet

Give an instructive reprimand: If your dog starts to get on the sofa, rather than just screaming “No,” use an instructive reprimand and tell her what she should do instead—“Off” or “Go to your crate!”  Then reward that behavior.

 

bullet

Give a Time-out. If puppy escalates her sharky nipping behavior you can call “time out” to mark what she did wrong and follow up by giving her a few minutes in her crate.

 

bullet

Train an alternative behavior that is incompatible with the problem behavior.  Train your dog to Sit politely for petting in lieu of jumping up—she can’t sit and jump up at the same time! 

 

bullet

 Don’t make the mistake of ignoring good behavior!  Teaching your dog the correct behavior and rewarding him for it is easier and more efficient than trying to catch and punish him for every wrong thing he can do.  Supervising him and teaching him what is OK to chew-- his stuffed Kong or Nylabone—and praising and rewarding him for that, is far easier than going at it from the other direction and trying to teaching him everything in the world it’s not OK to chew--socks, shoes, towels, books, pillows, the newspaper, the rug, plants, furniture, etc.  while constantly punishing him for making mistakes. Always remember the number one training principle--animals repeat/increase behaviors that are rewarded!  Be on the lookout to “catch your dog doing something right” and praise and reinforce/reward him for it!

ABCs of Training

Getting Starting--Before you can mark or reward a behavior such as Sit, you first have to get the dog to display or offer the behavior--you have to get the dog to Sit.  There are several ways to do this:

1) Luring—Let the dog smell a treat in your hand and target it with her nose.  As she follows the treat with her nose, you can lure her into certain body positions or actions and then reinforce/reward her.

2) Catching or Capturing is what happens when you “catch” a behavior and reinforce it—the dog spontaneously Sits, you “catch” the behavior by immediately marking it with a click or praise, and rewarding it with a treat. As the dog repeats the behavior, you begin to name it--right before your dog's rear touches the ground, you say "Sit!"

3) Shaping by approximation is a process in which the trainer reinforces/rewards each small action the dog makes in the right direction toward the desired goal. For example, if you are trying to teach your small, fast moving dog to go Down but he only follows your food lure part way down before popping up again, you can go ahead and reward his partial down movement with a click or praise & a treat, then encourage him to go down a tiny bit further before giving the next click and treat.  By very gradually raising your criteria for reward, you can encourage him to eventually go all the way down.

 

Clicker Training: Clicks & Treats--What's It All About

*Clicker training is an option available to anyone who trains with me.

A clicker is simply a fun little gadget for communicating with your dog in a way he can readily understand.

When teaching a dog a new behavior, it is important to let him know when he gives the correct response.  Most pet owners do this by using verbal praise. However, because timing is so crucial for clear communication to take place, marine animal trainers found it was more effective to communicate with whales and dolphins by using a whistle or clicker to “mark” correct behaviors at the precise moment they occur, and subsequently other animal trainers found the concept worked for them as well.

Marking a correct behavior followed by giving positive reinforcement can speed along the learning/training process.

The click is a behavior marker but it is also a reward marker – the click tells the animal that it did the right thing AND that it has earned a reward, usually a piece of food,  that always immediately follows.

Here’s how it works:  First we use classical conditioning to teach the dog to associate the sound of the click with a treat—we do this simply by clicking and treating 15-20 times until the dog catches on that a click predicts a treat. 

 

Now when the dog performs a desirable behavior, by clicking the moment it happens, we are “marking” the behavior and telling the dog 1)  “YES, that’s correct,” and 2) You have just earned a reward!  What gives the click its power is that it is always followed with a treat. 

 

These are the rules for clicker training:

1) Click during the desired behavior

2) Always follow a click with a treat, and treat as soon as possible after the click.

Treat delivery--Although the click marks a correct behavior/action, when you deliver the treat you will also be reinforcing anything your dog is doing at that moment.  Therefore it is important to time or pair your treat delivery to coincide with good behavior. 

Fading the clicker—The clicker is most useful when teaching a new behavior. When your dog is very solid on a behavior it is time to fade out the clicker because you no longer need to mark that behavior with precision.

How-- Give your dog his usual cue/command, for example, tell him to Sit, and if he responds correctly, don't click but continue to praise him warmly and give him an occasional treat.

 

Can I use my voice instead of a clicker?  Yes--It is your choice - you can train your dog with or without a clicker. You can of course use verbal praise to tell your dog that he did the right thing and pair your praise with rewards like treats and play.  Another option is to use your voice in lieu of a clicker by choosing a special word such as “YES” or “RIGHT” and teaching your dog to associate that word from you with a treat/reward.   However, the same rules that apply to using the clicker must apply when using your click word-- first train your dog to associate the special word with getting a treat, then say it the micro-second the behavior happens and always follow with a treat.

So why use a clicker at all?  The clicker seems to work especially well as a behavior/ event/reward marker 1) because most dogs are highly responsive to the distinctive consistent sound, and 2) because using a clicker helps your own timing to be more precise.  

 

© Copyright 2006 Beverly Hebert 

 

Click on Holly to follow her Home or

return to Pup & Basic 1 Lessons

Holly's Den
Dog and Puppy Training
San Antonio,TX
contact via email

Professional Member
Association of Pet Dog Trainers.
Association of Pet Dog Trainers
APDT member #6125
© Copyright 2002 – Beverly A. Hebert. All rights reserved