The misuse of behaviour terms in dog training needs to stop
What’s bad behaviour to humans? A dog couldn’t tell you. A dog is born perfectly able to be a dog, and hang out with dogs, and when we adopt him into our human family, he has an awful lot to learn. SOme of the things dogs do are reflexive, and some are instinctive. Still others are learned (such as social behaviour). Not all normal dog behaviour is considered acceptable in human society. We have to teach the dog acceptable social behaviour. Eliminating outdoors, not biting and mouthing to elicit play, not stealing food out of the trash are examples of this. If you thought about it, a dog is an English as a second language student from a completely foreign culture with absolutely no intrinsic way to know how to behave around humans. We owe it to him to make it easy for him to learn.
There are still people clinging to the notion that family dogs, or any dogs have an instinctive hierarchical pack system that makes some of them want to rule their families. These people apparently make up their own terms for this invented hierarchy and may one day publish a dictionary for their terms like “Alpha Dominant” “Beta Dominant” “Treat training” “calm energy” and their mis-extrapolations of terms used in learning like “Positive Reinforcement” training, and “Punishment”(which they of course never use, because their definition of “positive reinforcement” includes neck jabs and electronic collars). Even the term “behaviourist” is used to define a veterinarian who specializes in Applied Animal Behaviour. I’m not going to lie. This misinterpretation, creative misuse and development of entirely new, immeasurable, not quite definable, fluid, terms really bugs me.
Applied Animal Behaviour geeks hold that all animals learn in the same way. We have our own set of definitions when it comes to how dogs and other animals learn. These are not made up, fluid definitions. They come from the solid, measurable, indisputable facts that science requires. Here are the raw indisputable facts about the way dogs learn.
First, there was Ivan Pavlov. Pavlov worked with dogs and studied their digestive juices. Pavlov’s assistant would ring a bell before he fed the dogs. Pavlov noticed the dogs started to drool when they heard the bell. This little tidbit of information meant an antecedent (the thing that comes directly before the behaviour, in this case the bell) could be used to manipulate the dog’s involuntary behaviour. So even when there was no food presented afterward, the dogs still drooled at the sound of the bell.
This is called classical conditioning. It was the basis of modern behavioural theory. We still use it in classroom settings or in dog training by manipulating the antecedent so the dog can develop a positive conditioned emotional response to different environmental factors.
But learning isn’t just about involuntary responses to antecedents. Even fish can learn to use their their behaviour to effect the consequence (what happens right after a behaviour). In the 1950’s BF Skinner was studying how animals did this. He came up with Operant Conditioning learning process which was divided into 4 quadrants of behaviour based on Reinforcement (getting more of a behaviour) and Punishment (getting less of a behaviour) and Extinction (behaviour is neither reinforced or punished and after becoming exaggerated for a short time, it goes away.
- Reinforcement: Increases a behaviour
- Positive reinforcement: the animal gets something it likes and the behaviour increases
- Negative Reinforcement: Something the animal doesn’t like is taken away, and the behaviour increases
- Punishment: Decreases a behaviour
- Positive punishment: We add something the animal doesn’t like and the behaviour decreases
- Negative punishment: We take something away that the animal likes and the behaviour decreases.
Good dog trainers and dog behaviour professionals know how to make these two learning procedures work. Great dog trainers and behaviourists are adept enough at manipulating antecedents and reinforcement that they don’t ever need to delve into punishment. Imagine using shoulder/neck jabs (antecedent) to train a sea lion. You’d likely get some sort of aggressive defense response (behaviour)from that and then you’d stop (consequence). Sometimes neck and shoulder jabs get the same aggressive defense response from a dog. And so unless you are crazy, you stop, at least for a second or two.
Now you have a Behavioural law taking effect. Thorndike’s law of effect:“Responses that produce a satisfying effect in a particular situation become more likely to occur again in that situation, and responses that produce a discomforting effect become less likely to occur again in that situation.”
So you have created a situation where the dog is being conditioned to an aggressive response. The more a behaviour repeats successfully, the more likely it is to occur in a more exaggerated form. And every behaviour leaves a chemical imprint in the limbic system of the brain which changes the Synapses patterns permanently. You can never erase a behaviour forever. Once it has repeated successfully, you can imprint a new behaviour over top of it, but given the right set of circumstances, the behaviour will be back in a more exaggerated form. Sorry-I got carried away in behaviour speak there- just avoid setting your dog up to defend himself and you are golden. Got it?
Great dog trainers (the ones who win championships and titles with their dogs and/or have letters after their names) use food rewards and reward markers with their dogs because food meets a basic need, and according to behavioural science, it is the most effective reinforcer you can use, when applied in the right schedule. (I will get into this in another post). Good trainers can condition other reinforcers (toys, touch) to be almost as reinforcing as food, but food is right at the base of the hierarchy of needs and is often easy and convenient to use in training when a dog is beginning to learn a new behaviour, such as paying attention to us instead of performing unwanted behaviour that has repeated, such as barking at another dog.
An astute trainer never tries to use the food itself to distract the dog from barking. Instead, the presence of the other dog becomes the antecedent to our dog looking at us, and getting marked and rewarded for this. This happens for many repetitions and works progressively closer to a goal (such as walking right past another dog without reacting or being interested at all in the other dog. We are using both classical and operant conditioning and we always set the dog up in an environment where he can succeed. And it works. Because it is measurable. It’s indisputable. It’s science.
Dogs don’t do anything to try to be your boss or another dog’s boss. They are dogs, not hairy members of an innate corporate hierarchy, bent on upward mobility with no real reinforcement for working their way to the top. No award winning animal trainer or veterinary behaviourist believes this. This is rubbish and, for the sake of dogs and their well meaning owners, it needs to stop.