Why Does My Dog Do That?

Have you ever wondered why your dog runs to the kitchen every time you open the fridge door? Or looks under the same car on every walk? Dogs quickly learn which sounds, sights and smells predict good things. Understanding how dogs learn can help us not just understand how to train our dogs; it explains the behaviours we love, the ones we don't and the ones we've accidentally encouraged without meaning to.

This article will walk through the most common principles of

learning theory for our dogs. 

Every Action Has a Consequence (Operant Conditioning)

This is probably the most widely-recognisable type of learning in the dog training world. Dogs learn that every behaviour has a consequence. In simple terms, if the behaviour leads to something they like, they will do it more. If the behaviour leads to something they do not like, they will do it less.

Often expressed as a quadrant, there are 4 elements:

Breaking that down, if a consequence is likely to increase the likelihood of a behaviour reoccurring it is referred to as reinforcement, if a consequence is likely to reduce the likelihood of a behaviour reoccurring it is referred to as punishment.  The use of the terms positive and negative are not a statement about whether something is good or bad, right or wrong, simply that something is either added (positive) or removed (negative) from the environment and when that is done the consequence of it will make a behaviour more or less likely to reoccur.

Elements from the quadrant that I use in dog training:

Positive Reinforcement

• Definition: adding something to the environment to make a behaviour more likely

• Example: rewarding your dog with a tasty treat when they return to you when called

Negative Punishment

• Definition: removing something from the environment to make a behaviour less likely

• Example: pausing play when a puppy starts nipping at your hands 

As a reward-based, force-free trainer, I do not employ methods based on the other two elements of the quadrant, namely positive punishment and negative reinforcement. In order for these methods to be effective, they require that the dog feels a degree of aversion to the stimulus that is being added or taken away from the environment. In the case of positive punishment something aversive is added to the environment to make the behaviour less likely, like shouting at a dog to stop barking. In the case of negative reinforcement, an aversive stimulus is removed to make the behaviour more likely, like the release of pressure from a harness designed to tighten when the dog pulls.

The ABC of Behaviour

The theory is that every behaviour (B) has an antecedent (A) and a consequence (C ) - the ABC of behaviour. So if you are seeing a particular behaviour in your dog, ask yourself - what happens just before the behaviour and what are the consequences? By answering these questions you will have a better understanding of when and why your dog behaves that way.

Learning by Association (Classical Conditioning)

Dogs also learn by association. You might have heard of Pavlov's dogs. While conducting experiments into digestion, Pavlov discovered that if he preceded the dog's food with the sound of a bell, then over time the dogs would begin to salivate at the sound of the bell even though no food was presented.

In dog training we use this method to create positive emotional responses for the dog with a particular stimulus. For example, I always recommend starting recall training by conditioning the cue word. You say the word and you provide a treat (or something else the dog loves), regardless of what the dog is doing. All we are doing is building a positive association with the cue word so that every time the dog hears it they know something amazing is coming their way.

Another area where this is used is behaviour modification (more specifically counter-conditioning). This is where we aim to change the emotional response that a dog has to a particular stimulus, like reactivity towards other dogs. We use classical conditioning to build new, positive emotional responses to the thing that was triggering the unwanted behaviour.

"If Their Eyes are Open, They are Learning"

This all sounds quite straightforward - dogs will repeat a behaviour where the consequences are good and they will stop a behaviour where the consequences are bad. And by providing positive experiences in tricky situations, we will have a well-rounded, well-behaved pet. So why do so many dogs still behave in ways we find undesirable?

Every behaviour happens for a reason. If a dog barks, it is because it works for them - they are getting something from it. It may be because when they bark, you throw the ball for them. Or when they bark, you open the back door for them to go out. It may be that barking just feels good and/or provides, as a very natural canine behaviour, some sort of release for them.

 Often we have inadvertently reinforced the behaviours we do not want by giving the dog what they want or anticipate. If your dog barks at you to throw their ball, and you throw the ball, you have reinforced that behaviour. Every time you do it, you strengthen it. Now your dog barks at you incessantly to get you to throw the ball. At this point, you realise it has become a problem and so you don't throw the ball when your dog barks at you. But they just bark all the more, and start jumping up and the only way you can think of to stop it is to throw the ball. And now, as far as your dog is concerned, to get you to throw the ball they need to bark and jump. Oops!

One time, you went to the fridge, opened the door and gave your dog a bit of ham. Now, every time you open the fridge door your dog runs to the fridge with you. Sometimes they get something, sometimes they don't. It depends what's in there and what mood you are in. But what has actually happened is that you have established a variable schedule of reinforcement, one of the strongest forms of reinforcement for maintaining a behaviour¹. Oops!

See how easy it is to accidentally encourage unwanted behaviours?

The Bottom Line

The bottom line is be consistent in your treatment of your dog. If you do not want them to bark at you to get you to  throw the ball, never throw the ball when they are barking at you. If you want them not to jump up at visitors for attention when they come into the house, ensure your visitors never give them attention when they enter (if you can't control your visitors actions, you can manage your dog's opportunity to jump up when they arrive). Create positive associations with potentially tricky situations. But most importantly, be kind.