Basic Learning Theory and Dogs
Before taking on the rewarding task of training your pet, it might be a good idea to learn about how dogs (and any other animal, including ourselves) actually learn. Classical conditioning and operant conditioning are two terms you may have come across. They are two of the most important concepts governing learning.
Classical conditioning was made famous by the Russian physiologist Ivan Pavlov and his dogs. Perhaps you have heard of the experiments he conducted, which eventually got the dogs salivating and drooling simply at the sound of a ringing bell, because they were expecting food. How classical conditioning works is that a neutral stimulus, such as a bell, is paired with giving food. Eventually, after a number of pairings the animal mentally associates the bell with the food.
To illustrate this, Pavlov rang a bell (a neutral stimulus) every time he showed a dog a bowl filled with food. Seeing the food made it salivate, which is the unconditioned response that occurs naturally. Over time, the dog associated a ringing bell with food, as the neutral stimulus became a conditioned stimulus and predicted the meaningful event of food. The association was so strong that the dogs started to salivate any time they heard a ringing bell whether food was present or not. A more familiar example that you may be used to is your cat running to the kitchen every time it hears the sound of a can opening. The cat has paired the sound of the can opener with food, and thus runs to investigate.
You can try classical conditioning methods yourself. For example, if you pair a hand signal with a verbal command consistently and for enough repetitions, your pet will eventually associate the hand signal with the command, and will perform the command in the presence of the verbal signal, hand signal, or both at once. Try it, you might be amazed at how quickly associations can be made.
Another type of learning is called operant conditioning. The term was coined by American psychologist B.F. Skinner who suggested that the consequences of any behaviour can change the behaviour itself. The method behind this theory is that an activity that results in a positive outcome will be performed more frequently, and a behaviour resulting in a negative consequence will decrease in frequency over time. Skinner spent a lot of time studying the behaviour of rats and how they would react when they received positive (getting food) and negative (shocks/loud noises) consequences for their actions.
His experiments on rats have been applied to the training of dogs for years. For example, every time a dog sits on command, it gets a treat. That is a pleasant outcome, and the dog will start to sit more on command to get more treats. On the other hand, every time it pulls on the leash, you stop the walk and ignore the dog. This is a negative outcome for the dog, since it wants to go for a walk. It will soon learn that pulling on the leash leads to a shortened walk, and thus will start performing the negative behaviour (pulling less), to avoid the negative outcome (a shorter walk).
A few other concepts that relate to operant conditioning that anyone training their pets should know are the concepts of punishment and reinforcement. Punishment results in a negative outcome for the animal. Reinforcement, on the other hand, results in a pleasant outcome. There are also positive and negatives of both, leading to positive punishment, negative punishment, positive reinforcement and negative reinforcement. It sounds confusing at first, but in this case, ‘positive’ means that you are adding something, and ‘negative’ implies taking something away.
Following with these definitions, positive reinforcement would be adding something pleasant. For example, as the pup sits down on command, you praise it and give it a treat. The treat is the primary reinforcer, while your praise is the secondary reinforcer. Thus, the behaviour will be performed more frequently due to the pleasant outcome.
Negative reinforcement, is removing something unpleasant. This was the basis of many dog-training theories in the past, and unfortunately is still being used today. An example of this would be the choke collar. As the dog pulls, the choke collar squeezes its neck, hurting it and making it difficult to breathe. As soon as the dog stops pulling, the pressure is released. This results in the increase of the non-pulling behavious and thus, by the principle of operant conditioning, the probability of the behaviour is reduced because of the unpleasant consequence.
(Positive) punishment is introducing something unpleasant. An example of this would be hitting the dog when it has done something ‘bad’. Punishment provokes fear, which clouds the learning process and also erodes the human-animal bond. Other training methods are far more effective and humane. In fact, punishment only works if it is applied consistently (i.e. EVERY single time the behaviour occurs), within three seconds of the undesired behaviour (i.e. the trainer must have very fast reflexes), it must be brief, and it needs to be only the minimum intensity to stop the behaviour. Most owners are not quick enough, or cannot punish every time the behaviour occurs (for example, if it occurs while the owner is at work). Thus, yelling at your dog for soiling the floor hours after the accident does absolutely nothing for it. The animal is not learning anything and is likely confused and stressed by the yelling.
Negative punishment implies removing something pleasant. This seems counter-intuitive, but it can work very quickly. For example, if a pup is playing tug-of-war and suddenly nips your hand, you can give a sharp ‘yelp’ and promptly stop the game. The pup will quickly learn that a nip means that the pleasant activity of the game will be stopped, and will learn to stop biting.
There are of course other theories and ways in which we learn. Every individual is different and often a combination of approaches are needed in tough cases. This article was meant as a brief introduction to common learning techniques as applied to dog training. If you are having trouble training your dog, ask your vet or someone else that you trust for a reference for a professional dog trainer.