Training with reinforcement
Thursday March 31st, 2011
Positive and negative reinforcement have become buzz words for animal behaviorists, but they are also concepts that you, as your dog's trainer, can use to help your pet develop into a well-behaved canine companion--if you understand what each term means.
Keeping it simple, positive reinforcement is purely something which the dog perceives as pleasant, that occurs during, or immediately following a behavior, which makes that particular behavior more likely to occur again. An example would be: your dog lies down at your feet, you say "good dog, good down" and your dog's tail begins thumping. He has been positively reinforced, with praise, for lying at your feet.
Confusion exists over the difference between positive reinforcement versus reward. Here's an easy way to distinguish them: Rewards come after a behavior, whereas reinforcement comes during, or so closely following a behavior that the dog feels almost as if it happened during the behavior. If you ask your dog to sit and he complies, and you say "Good boy, you can have a cookie," and you trot off to the kitchen to get a dog cookie from the jar on the counter, then the cookie he receives is his reward.
Using positive reinforcement
On the other hand, if you have small food treats already in hand, ask your dog to sit, he does so, and immediately you mark that he has done the right thing by saying "yes!" and letting the treat appear, then you have positively reinforced the behavior.
When you use food treats, be sure to remember to give praise before the treat, so that your dog is being conditioned to associate the praise with food, helping ensure that he will eventually work for praise alone. Even if his primary motivator is food, it is still important to make your words valuable to your dog, as you may not always have food handy but your voice is always available.
Training treats should be small, which for an average size dog would be about the size of a Cheerio™. Use something soft and easy to eat in one bite, to avoid the problem of the treat breaking, pieces falling to the floor, and the dog "vacuuming" instead of watching you. Treats can be in your pocket, a fanny pack or an easy-to-wash carpenter's apron, which is tied around your waist before training sessions.
Negative doesn't mean bad
What about negative reinforcement? The word negative implies that something bad is happening, but actually the easiest way to understand the principle of negative reinforcement is to think in terms of "pressure." When training around distractions, we apply pressure, then release it the instant the dog is performing the appropriate behavior.
An example would be a Gentle Leader™ head halter, or a Good Dog™ plastic training collar, which applies pressure if the dog pulls, but is released the moment the dog is walking in the right place. It is self-correcting, so the dog does not necessarily perceive the negative as being applied by the owner, but instead can associate that they, the dog, are in control of the collar.
Postive + negative = results
Both positive and negative reinforcement are useful training tools. Ideally, a dog is taught the correct behavior using positive reinforcement, then after the behavior is learned, negative reinforcement may be used if needed to teach the dog to comply even under distracting conditions.
Still confused? Here is an example of using both types of reinforcement to make sure your dog understands how to walk politely on a loose leash.
1. Start by putting pup on a short 2-foot leash while in the house, positioning him on your left side in line with your left leg. Hold treats in your left hand, and use the treats like a magnet to lure your dog into walking a few steps with you in correct heel position.
Walking the dog in a hallway works well, because you can ensure that the dog stays in the correct position. As the dog looks at you while walking, reinforce positively by cheerfully saying "good" and after a few steps, lure the dog into a sit with the treat. As the dog sits, praise and give the treat.
2. Practice this for several days, until you are able to move to a different location, and eventually try it outdoors, with the dog or pup continuing to walk a few steps in correct position, looking at you while being reinforced with praise, then lured into a sit when you stop.
3. As your dog begins to understand "heel position" (the dog's shoulders in line with your left leg), you can either continue this technique, gradually walking for a longer distance and using a longer leash, or you can transition to a training collar (like the Gentle Leader™).
Determined pullers will normally learn faster if transferred to a training collar, as will dogs which might become fearful around distractions. Fearful dogs are safer on a training collar, because they can easily back out of a buckle collar if they become frightened and try to pull away.
4. Continue to use food treats, or whatever you are using as a lure, as you begin using a training collar. Your motivator can be used simply to distract the dog away from any anxiety caused by the new collar, and also to remind him of the exercise he already knows well (walking in correct position while following the treat.)
Keep in mind that if your dog pulls ahead and the collar applies pressure, you must immediately take the pressure off the instant the dog is walking in the correct position, by putting slack in the leash. (Tip: Do not stop. Continue walking so that the dog understands that the goal is walking in the right position, not just stopping and sitting.)
5. When you stop, guide the dog's head up with a treat, or lift it with your hand, which should make their rump go down into a sit. If needed, you can still lure the dog into sit position by holding a treat over his head.
With these techniques, you will use positive reinforcement to first teach the correct position (so the dog can more quickly understand where you wish them to walk) and then by using the negative reinforcement of the pressure collar, you have helped the dog understand that this position is to be maintained even in the presence of distractions.
Almost all living beings need a balance of positive and negative reinforcement if they are going to perform a behavior reliably in distracting conditions. Sure, we human beings work well for positive reinforcement (praise from the boss) and rewards (paychecks!) but how many of us would continue to go to work every single day if there were absolutely no consequences for not coming to work?
latest news & articles
offers from our sponsors