Clicker Logic

December 27, 2006

“Bite Inhibition” is a Fuzzy Term

I want to present a different way of looking at bite inhibition. This approach can be generalized across many, many topics in animal training. It has the potential to make your training work simpler.

Here’s a definition of bite inhibition borrowed from the Partnership for Animal Welfare (Maryland) website.

http://www.paw-rescue.org/PAW/PETTIPS/DogTip_BiteInhibition.php

Bite inhibition is a learned response in which the dog consciously inhibits the full force of his biting ability. Most dogs display bite inhibition when they are playing together, and even when engaging in a fight with another dog. If a dog does not have bite inhibition, he could injure and possibly even kill another dog.

There’s a simpler way of looking at this. Science relies on the principle of parsimony. In most cases, the simplest answer is the best answer. It gets you down to the most basic components where less guesswork is necessary. Bite inhibition is a term that complicates matters and takes you farther from what is manipulable.

“Bite inhibition” is a term that implies that there is an invisible force inside the dog that controls how hard he’s going to bite in a given situation. In behavior analysis we look for something more basic. We look for a history of biting in certain situations. Bite inhibition training implies that we can somehow put an invisible “bite softer” force inside the dog. The problem is that you can’t see bite inhibition or the force it implies. You can’t measure it. You can’t predict when or how it’s going to show itself. You can’t replicate it since you can’t quantify it in any way. If you’re relying on bite inhibition as a preventive, you’re really relying on hope.

Why is it important to be able to observe, measure and replicate something you’re training? Because without doing these things there is no way to know if the training is having any effect or if it’s having the effect you are looking for. The observation and measurement process doesn’t have to be complicated. Just test the behavior in various situations repeatedly until you see how often the behavior is happening there. Make hash marks in a notebook so you can see if he bites less on Day 2 of training than he did on Day 1. If you’re getting a lot of biting while sitting on the floor with your puppy, you know that you need to continue training in that situation. Does this biting mean your dog doesn’t have bite inhibition? No, it means your dog is biting and needs more training.

Once the training progresses and the biting goes away, you know you’re getting somewhere. Does that mean your dog now has bite inhibition? No, it means your dog has a history of training and of performing in a certain way. He’ll probably behave the same way in the future but many factors can influence that. The invisible “bite inhibition” force is still just guesswork, but the behavior you’ve observed, measured and replicated is solid.

I suggest that rather than talking about bite inhibition we train our dogs to do a variety of behaviors that are preferable to biting, and to train them in a variety of situations. In fact, this is what many people do, but they call it “bite inhibition training”. Unfortunately that term comes between you and the dog and what is really happening and sets up a fuzzy area where there is a complete lack of control. The term comes between the training and the behavior.

There is a fundamental problem with the kinds of training done in the name of bite inhibition. One is that it doesn’t take stimulus control into account. We clicker trainers are familiar with the cue as an antecedent stimulus, but the fact is that the whole world is an ever-changing series of overlapping antecedent stimuli. If you teach a puppy not to bite your hands while you’re with him on the floor in your living room, he might figure out that he shouldn’t bite your hands in other places and you’ll be in good shape. But it’s also possible that he will figure that the living room is the only place where biting softly matters.

This is especially likely if you play differently with your pup in different situations. If you allow rough play in the back yard or the family room but require soft play in the living room and bed room, what is your puppy going to choose when you’re visiting at Grandma’s or sitting on the porch or meeting a new person? It’s anyone’s guess. One thing you can do is provide him with some cues that indicate that “When this happens it ALWAYS means no biting is allowed”. You might have a certain bandana that he wears in no-bite times, for example.

Another issue is the matter of the function of the behavior. What is the biting behavior being reinforced by in the current situation? In most cases of puppy biting, the reinforcer is positive. He gets social interaction. He gets some stimulation to his teeth that results in more biting activity in the future. Here is a critical point. If Fido learns not to bite or to bite softly in situations where biting has been positively reinforced, it has no bearing on whether Fido will bite or how hard he might bite in situations where he is working to escape an aversive stimulus.

If you teach your dog in your house that play time requires soft mouth behavior and your dog softly bites a stranger who bends over him at the pet store, this does not mean your dog is showing good bite inhibition. It means he bit a stranger in the pet store. The consequences for the behavior in the different situations were different. The dog was working on a different set of reinforcers with the stranger than he was when you taught him bite inhibition at home with positive reinforcement.

There’s another important issue. You cannot safely teach a dog to bite softly in a situation he wants to escape from. The only safe bite is no bite. The risks of training for a soft bite in those situations would be enormous. What you can do is teach him that there are behaviors completely different from biting that he can do, and demonstrate to him that you will change his surroundings in response to those behaviors so that biting is not the only way he can achieve the negative reinforcement he is working for.

In situations where a dog is biting from fear, from stress, etc. he is working for the natural contingency of negative reinforcement. He is working to get rid of something aversive. The fact that you’ve taught him to bite softly during play is unrelated. If your dog happens to bite softly in a different situation, that means he bit softly in a different situation. It does not mean he’s showing good bite inhibition. It is, of course, a good thing he didn’t bite harder, but he needs immediate training on getting rid of that soft bite. To say he needs more bite inhibition is fuzzy. To say he needs more training in alternative behaviors to biting is clear.

There are some folks who encourage “bite inhibition” not “bite prohibition”. I’m not one of them. Bite inhibition means that one hopes one’s dog will bite soft instead of hard. Bite prohibition means you don’t let the dog bite at all. But a clearer and more efficient method for dealing with bites is to stop talking about bite inhibition and simply train a dog to do something besides biting and make sure doing it provides the dog with a desirable outcome. It’s not about bite inhibition. It’s about behavior.

You may already be doing all the things I’ve recommended and calling them “bite inhibition training”. I urge you to drop the fuzzy label and just talk about the behavior that you can do something about.