Training 101
When that adorable bundle of energy and puppy breath arrives, you may wonder: how do I even begin training him for Schutzhund? After all, this is more than just a basic sit, down, and come. It is three different phases that all require intense focus, precision, speed, accuracy, and enthusiasm. Fortunately, whether it is downing on the article on a track, retrieving a dumbbell in obedience, releasing the sleeve in protection, or simply learning "sit" for the first time, the basics of teaching these behaviors are the same!
There have been many books written on the subject of how dogs learn. But learning can be simplified to a very basic principle: dogs repeat behaviors that they find rewarding. Our job as trainers is to:
Make the desired behaviors rewarding for our dogs
Ensure that they are not being rewarded for behaviors we don’t want
Show them that the fastest way to get what they want is by doing what we ask of them
Enjoy the process!
So how do we do this? Through the principles of operant conditioning! While learning theory and operant conditioning (a la psychologist B.F. Skinner) has flaws and limitations, it is still useful for understanding how to teach dogs to do things. Just keep in mind that real life is much more complicated and messy than a laboratory!
Reinforcement vs. Punishment
In Skinner’s theory of operant conditioning, reinforcement means strengthening behavior, while punishment means decreasing or stopping behavior. There is no emotional or moral value assigned to either of these terms; ‘punishment’, while having a negative connotation of being bad, simply means that the behavior has decreased or stopped. So, closing your hand over a piece of food to stop the dog from snatching it is technically punishment; it stops the behavior. Under Skinner’s paradigm, reinforcement cannot stop or decrease behavior, and punishment cannot strengthen behavior.
Positive vs. Negative
In operant conditioning, positive means something has been applied, while negative means something has been removed. Again, these two terms have no value of being good or bad, even though we tend to assign emotional responses to them (positive as being good, negative as bad. That’s not how it works!). Positive and negative are then combined with reinforcement and punishment to describe the four quadrants of operant conditioning:
Positive reinforcement: something is applied to strengthen a behavior (applied, strengthens)
Negative reinforcement: something is removed to strengthen a behavior (removed, strengthens)
Positive punishment: something is applied to decrease or stop behavior (applied, decreases)
Negative punishment: something is removed to decrease or stop behavior (removed, decreases)
When it comes to training dogs in real life outside of the laboratory setting, things rarely are so simple. Learning often occurs with two or more quadrants interacting, like two sides of the same coin. For example, a correction—which is usually considered positive punishment—can have the effect of actually increasing, or strengthening, a desired behavior (but not the behavior for which the correction is applied!). So it punishes one behavior but then reinforces another, making it act as both punishment and reinforcement in the span of a split second. Let’s take the bark and hold as an example. The dog comes barking into the blind, lunges toward the helper, gets a little ‘check’ on the line, and then immediately settles into even stronger, more forceful barking in correct position. The check, or line pressure, is positive punishment for lunging at the helper, but it becomes negative reinforcement when the dog then barks more strongly after the pressure is removed. It decreases the lunging, but increases the barking behavior.
Photo courtesy of: Oggy Brett
The takeaway here is that behavior and training do not always fit exclusively in one of those quadrants ALL OF THE TIME. What we think may be reinforcing for a dog may actually be punishing, and vice versa. Or it may be reinforcing in one situation, but punishing in another. So we must always pay attention to the dog’s behavior, attitude, and end result. It is the dog who tells us how this action (or lack of action) was perceived, based on his resulting behavior and attitude.
The tendency for some handlers is to get caught up in the language of dog training. Resist. Instead, look at your training in terms of what you want to reinforce in your dog (including both behavior and attitude/enthusiasm), how you want to reinforce it, and what you can do to prevent unwanted behavior. Did you like the behavior? Then mark and reward it. Is the behavior unwanted? Then prevent the dog from doing it again and develop the wanted behavior. Seems so simple, but it's not always so easy!
Now we get into the nuts and bolts of teaching behaviors. When training in Schutzhund, most of the foundation work is motivational, or taught through positive reinforcement (with some interaction from the other quadrants). In order to teach your dog specific skills or tasks through reinforcement, you must have a way to mark and then reward the correct behavior.
Marking Behavior
A marker is something that lets your dog know it performed a behavior correctly. The marker clearly communicates that “Yes! That’s it, right there! You did something right!” Because of this, the marker is a very important tool in teaching behavior! Without it, there is no clear communication that the dog did what you wanted at the moment it happens. Can a dog be trained without a marker? Yes, but progress is much slower, and with greater confusion. Utilizing a marker gives the trainer the opportunity to give immediate feedback on correct behaviors, or even correct components of behaviors, allowing for faster progress and learning.
Examples of markers are a clicker, a verbal “cluck”, or even a verbal cue like “Yes”. When the dog performs the right behavior, you “click” the clicker (or verbally mark the behavior) and then reward the dog. The marker only has value because of the reward that follows it, which means the first thing you must teach is that this neutral stimulus (marker) signals a reward is coming.
It is best to teach the dog both types of markers—the clicker, and a verbal cue. You won’t always have a clicker on you, but you at least have your words! Your marker should always be used to mark the exact behavior you want to reward, before you reward it and before you even make a move to reward it. Don’t just throw the ball or tug for the dog without marking which behavior it was that earned it! The better your timing, the more clearly you communicate and the faster the dog learns.
When training behaviors, the sequence should always be: Correct Behavior > Marker > Reward.
Examples:
Dog sits > “Yes!” > Ball is thrown
Dog pushes into hand to get food and moves into proper heel position > “Click!” > Feed dog
Later, when the behavior is linked to a cue, the sequence becomes: Cue > Correct Behavior > Marker > Reward.
Examples:
“Sit” > Dog sits > "Yes!" > Ball is thrown
“Fuss” > Dog swings into heel position > “Click!” > Feed dog
Reinforcers/Rewards
Reinforcers are anything that reinforces or strengthens the behavior, and are commonly referred to as rewards. There are primary reinforcers like food, water, air, and sex, and secondary reinforcers like toys and play. Many Schutzhund dogs will choose secondary reinforcers over primary reinforcers when in drive. Some would much rather have a ball or a toy to play with than take food in training, even if it is delicious food. The job of the handler is to figure out what their dog values, at what level, and when.
The value of reinforcers can change based on the dog’s emotional state, environment, and drive level. A food-driven or even toy-driven dog may not care about food or toy if in the protection mindset, for example, because the value of the helper is much higher than anything else you have to offer. It is impossible to be the most interesting thing to your dog ALL THE TIME (despite what people tell you about having to be the most interesting thing in your dog’s life. Guaranteed that the dog will find the helper more interesting than you!). So in situations like this, we Premack it: teach the dog that the way to get what he wants is to do what we want first.
The helper can be a very powerful reinforcer! Photo: Louise Jollyman
In Schutzhund training, we tend to use five different types of rewards: praise, interaction with the handler, food, toy/play (with a ball or tug), and interaction with the helper. Food is used as a primary teaching tool in obedience and tracking; toys are usually used to build drive, speed, enthusiasm, and handler interaction/bonding. Interaction with the helper is primarily used in protection. Praise and interaction with the handler should be used throughout each phase!
Every dog is different. Find what the dog values as reinforcers, and use that to your advantage. Not all dogs prefer the ball. Some prefer tugs. Some prefer food. Others prefer floppy Frisbees or soccer balls or just physical interaction with the handler. I’ve even known a dog that would work for wood chips. His handler would throw the wood chip as a reward, he would run and hunt for it, then pick it up and bring it back. Find what works for your dog!
Next blog will focus on finding what works for your dog and making the most of motivation!
