마이클 엘리스(Michael Ellis)의 개 훈련 전문 용어 해설

involuntary and this is going to be very important we'll come back to this later but I want people to understand that the dog's response to classical conditioning is not a choice it's an involuntary response. Our next term is operant conditioning also called instrumental learning. Operant conditioning covers all of the cause and effect of dog training. This is the part of dog training where or how any animal learns actually humans as well through both classical and operant conditioning but operant conditioning governs the associations dog make with their behaviors and outcomes and they choose to do certain behaviors based on their previous experience with the outcomes. I call this the cause and effect part of dog training and there are four quadrants to operant conditioning. The four quadrants are all combinations of the words positive, negative, reinforcement, and punishment. There's no judgment implied in this so positive does not mean good and negative does not mean bad. Positive means we add something to the equation. Negative means we remove something from the equation. Reinforcement means that the dog is more likely to perform that behavior in the future and punishment means they're less likely to perform it in the future. So the four quadrants of operant conditioning. Number one positive reinforcement. This is the classical give your dog a reward. My dog sits I hand him a piece of food. The food I add to the equation the primary reinforcer the reward. The behavior that prompted me giving the dog a piece of food, sitting, is now more likely to occur in the future. Positive reinforcement. Negative reinforcement. We generally erroneously think of this as a correction or an aversive consequence. It's not. In negative reinforcement the dog stops something unpleasant from happening, some aversive experience from happening with their behavior. So the old way we used to teach sit by pulling up on a dog's collar until they sat is a form of negative reinforcement. I'm pulling on the collar. The dog sits. I stop pulling. The negative portion is I remove the unpleasant pulling on the leash. Whatever behavior the dog performed to stop that, in this case sitting, is now more likely to occur in the future. Negative reinforcement. The dog's behavior that stops the discomfort is now more likely to occur in the future. The next term in operant conditioning is positive punishment. Positive punishment is the traditional correction. It's an aversive consequence. So for instance if my dog jumped up on me and I were to knee it in the chest. Or if

my dog were to jump on a counter and I were to yell no at them. The negative consequence, the unpleasant consequence, is applied. The positive portion. My dog jumps on the counter and I yell. The negative portion, the positive portion is me adding the yelling to the equation. And the behavior, jumping on the counter, is now less likely to occur in the future. It's been punished. The final part of our operant conditioning paradigm is negative punishment. And negative punishment is simply the withholding of something the dog wants or removing the dog's ability to access something they want as a form of punishment. So let's say I'm training with a young dog with food. I'm luring the dog around and as I lift the food up the dog jumps up on me. If the dog jumps up and hits me with their feet and I take the food away, put it behind my back for instance, I'm removing something they want, the negative portion, and the behavior that prompted the removal of what they want, jumping up, has now been punished and is less likely to occur in the future. It's really that simple. Our next term is conditioned reinforcer. Conditioned reinforcer simply means we take a neutral stimuli again, some sound or visual prompt that means nothing to the dog. And by predicting something that does mean something to the dog, something of intrinsic value to the dog, in this case let's say food, the neutral stimuli comes to have the same meaning for the dog as the primary reinforcer or the rewarding substance, in this case food. We can use this and the sound or visual prompt or whatever we're choosing, in this case we're primarily going to use a sound. The sound is now been conditioned to be reinforcing through the use of classical conditioning. So our use of verbal markers or someone's use of a clicker is a classic conditioned reinforcer. So for instance I make a click, I give my dog a piece of food. I make a click, I give my dog a piece of food. I say yes, I give my dog a piece of food. I say yes, I give my dog a piece of food. If I do this repeatedly, my dog responds to the click or the yes, the same way that they would respond to food. And now yes or click has been conditioned to be reinforcing. This term is interchangeable with marker in our case or frequently you'll hear it described as a bridge as well. Our last term was conditioned reinforcer. The next term is marker. And for us, we say use verbal markers to communicate to our dog when they're right and wrong. So we have a communication system that is geared around being able to tell the dog very precisely when they're right and wrong.

And we condition sounds to have meaning to the dog. So marker is interchangeable with conditioned reinforcer. In this case, the sound we're using happens to be a verbal sound that we make, a word. We use yes a lot, but it doesn't really matter what it is. So a marker is simply interchangeable with a conditioned reinforcer or a bridge. We're using our voice. Engagement is our term for sustained focus and motivation. So one of the things that we want to see in our dogs before we start trying to teach them things is that they're paying attention to us. And we call the process of teaching dogs that paying attention to us is rewarding engagement. So a dog that has sustained focus and wants something from us, wants a reward of some type from us, is what we call engaged. And a dog that is giving us a portion of its attention or checking out to look around and do other things in between rewards is not engaged. And engagement is a prerequisite for all of our training. Our next term is luring. And luring is one of the physical tools we use to manipulate dogs when we're teaching them new behaviors. So we have a variety of different ways to move a dog around and manipulate their behavior in order to teach them to do things. And luring is one of the simple and straightforward ways we do this. And luring in a nutshell is simply a dog following my hand. So I put a piece of food in my hand and I teach the dog that following my hand around with their head is useful. And then I can manipulate them by moving my hand around and having their body follow behind. We'll talk about some of the basic rules around luring in this DVD. But luring is really as simple as that. A dog that will follow your hand around with their head. Spatial pressure is another one of the physical tools we use to manipulate dogs behavior. We use spatial pressure to teach the dog certain things as well, just like we did with luring. And spatial pressure is simply if I move into my dog, my dog moves out of my space. So that I can use my body to push or pull the dog. And when I'm trying to move them into certain positions and teach them new behaviors. Our next term is shaping. And we've thrown in what we call a successive approximation. Our trainers call successive approximation. Shaping is simply the creation of a behavior. We're working on shaping the dog into doing a certain type of behavior. Some behaviors we capture really quickly. And some behaviors are more difficult. They take multiple steps to capture. The act of shaping or teaching the dog or manipulating the dog into doing a behavior.

Can be broken down into steps that are frequently called successive approximation. Where I teach the dog or reward steps on the way to the finished behavior. That process is called successive approximation. So we'll use shaping and successive approximation to capture some of the obedience behaviors we're looking to teach the dog. Fading is simply the elimination of help in our training. So we have certain tools we use to manipulate a dog into doing behaviors. We lure them. We use body language. We use our communication system to tell them when they're right and wrong. And over time, we help the dog do certain behaviors like sit. As the dog becomes fluent in these behaviors, we need to remove the help. And the process of slowly removing the help is called fading. Anytime we want to remove a signal that's prompting a behavior and eliminate the help, we call this fading. Our next term is hunger drive or motivation for food ultimately. So in training, especially when we're training with food, a lot of our dog's success is dictated by their desire for the reward, their motivation. And hunger drive is simply the dog's natural motivation for eating. And dogs have a wide range of hunger drives naturally. This can be manipulated to some degree by how much we feed the dogs and what types of food we're using when we train. But the truth is, some dogs are more hunger motivated than others. And some dogs are more motivated by other things. We can manipulate this, but we need to pay attention to it in our early stages of training with food. Our next term is prey drive. And prey drive is a term that you'll hear dog trainers use a lot, but seldom hear behaviorists and other people use. In a nutshell, prey drive is simply your dog's desire to chase and grab things with its mouth. And we manipulate a dog's prey drive to teach it how to play or to motivate it. So some dogs are highly motivated to chase and bite and grab things with their mouth. And so we can use this as a reinforcement system or a reward system in our training. Our next term is what we call acquisition-based behaviors. So certain behaviors, the acquiring of something the dog wants, whether it's searching, chasing, those sorts of things, are intrinsically reinforcing to dogs. We call them self-reinforcing. So we break our behavior chains down into certain types of behaviors. Some behaviors, like sitting to get a reward, aren't inherently reinforcing to the dog. Sitting has no real meaning to a dog. But chasing or searching frequently does. And the types of behaviors that dogs find reinforcing in trying to acquire a reward,

we call acquisition behaviors. Our next concept is what we call reinforcer versus reward. Now technically, they're slightly different. But in our system, we tend to use them kind of interchangeably. Normally when we say reinforcer or reward, we mean the same thing. But remember our discussion of operant conditioning. And they are technically slightly different. So if I'm pulling on my dog's leash and it's uncomfortable and he does something, and I stop pulling, the stopping of pulling is inherently a reinforcer. It's reinforcing. The dog's more likely to do the behavior that made it stop. So it's not necessarily a reward. But most of the time when we say reinforcer, we mean reward. We mean giving the dog something they want for having done a behavior. Our next concept is what we call creating a reward event. So in our early stages of training, we used to think that the item was the main driving force in rewarding a dog. What we gave them was the most important part. Over time, we've evolved into trying to create what we call an interactive reward event. Meaning the dog and I do something together during the process of rewarding the dog that the dog finds both valuable and motivating, but also sort of a bonding experience between the dog and I. So if I'm rewarding the dog in an interactive way and the dog likes what's happening, not only are they reinforced for the behavior that they're performing, but also they start to pay better attention to me, like working with me more. So there's an overall improvement in their attitude about rewards. And we call these little sequences of rewards, reward events. Our next term is verbal cue. Verbal cue or verbal prompt really just means a command. So any kind of sound we make, a word we use, a verbal sound that we make that prompts behavior. So a prompt or a cue is something that just simply signals the dog to do a certain behavior that they've learned. A physical cue is simply that, a physical move or motion that we make that prompts behavior. And so in the same way that a command is a verbal cue or verbal prompt, a physical prompt might be luring the dog into a sit, body language, leaning over, any of those things. Anything physically we do to prompt the dog to do a certain behavior. Reinforcement schedules are frequently how often we reward a dog for a certain behavior. And so by manipulating a dog's reinforcement schedule, we can radically manipulate their behavior. So it might be the frequency of rewards in terms, in time, or it might be how often we reward a given behavior.

The general paradigm in training is we start out on what we call a continuous reinforcement schedule, where we reward the dog for every repetition of a certain behavior, and what we call a high rate of reinforcement. We reinforce the dog very frequently to hold their attention. Over time, we vary and randomize the reinforcement schedule, meaning we don't reinforce the dog for every single repetition. And we go longer in between rewards to improve performance over time, and to eliminate the necessity of constantly rewarding the dog for every behavior. You'll hear us talk a lot about the term arousal. And arousal just means the dog's excitement or stimulation level. So when a dog gets aroused to varying degrees, excited, motivated, worked up, whatever you want to call it, they're excited. Then it can either help a certain behavior or hinder a certain behavior we're trying to capture. So when we talk about manipulating a dog's arousal level, we're talking about manipulating their excitement level to make the training that we're doing more productive. So sometimes higher arousal rates are better, and sometimes lower arousal rates are better. Another key concept is a concept called overshadowing. And overshadowing is simply if we give a dog two prompts or two signals or two inputs at the same time, the dog will pay attention to the more relevant one and ignore the other. So for instance, if I give my dog a verbal prompt and a physical prompt at the same time, for dogs, physical inputs or physical prompts are always more relevant than verbal prompts. Dogs aren't verbal creatures. We talk around them frequently. Most of it means nothing to them. And so they pay more attention to physical cues. So if I do them at the same time, the physical prompt overshadows the verbal prompt. And we think the dog's paying attention to the verbal prompt, but they're really only paying attention to the physical prompt. So if I give them only the verbal prompt, they don't respond. But if I give them the physical prompt, they do. This concept is going to be an integral part of all of our training going forward. And this gives people lots of problems in their training. Our next term is motivation. And motivation simply means the dog's desire for some activity or some item. So my dog, we can talk about my dog's desire for food or play or chasing as their level of motivation. And we're simply describing the level and intensity of their desire for a certain activity.

What is motivation for those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those of us as those