How dogs learn – a brief overview

Classical conditioning is where dogs learn through association, connecting two or more stimuli that occur almost simultaneously or in quick succession. It can be thought of as ‘robotic, automatic and involuntary’ responses to stimuli, where a response happens without any volition from the individual.

This type of learning relies on:

• contingency – one stimulus reliably predicting the other
• contiguity – the two stimuli being linked occurring very close together in time.

Classical conditioning happens without the dog’s owner necessarily intending it, and without the dog being consciously aware of it. A common example is a dog who associates their owner picking up their lead with immediately being taken outside for a walk. A dog who really enjoys going outside might quickly begin to get excited as soon as their owner lifts their lead, without having been specifically encouraged to do so. They are classically conditioned to respond to their lead being picked up.

Another example is a dog who has become classically conditioned to feel anxious when entering the veterinary waiting room because this has become so strongly associated with a negative outcome, such as pain or distress. The stress response here is involuntary, the dog is conditioned to feel worried and to behave accordingly. Think about anyone you know who has a phobia about spiders – they might be classically conditioned to feel fearful and even jump up in a panic at even the mention of a spider nearby, without any choice or control over this physiological and behavioural response!

Operant conditioning is where dogs learn through the consequences of their own behaviour. This learning arises through voluntary behaviours, where the dog might ‘try out’ various behaviours and repeat them in future, should they result in a good outcome.

It is important to note that not responding is also a behavioural choice – a dog might choose to do nothing for the moment and wait to see if anything else happens.

If the consequence doesn’t work out well for the dog (from their perspective) then they will be unlikely to repeat that behaviour again in response to the same or a similar situation.

So, the dog learns which behaviours work out well for them and which don’t, and this informs their future behaviour. For the most part, dogs learn through repetition and therefore patterns of behaviour develop through a dog experiencing the same outcome from repeated behaviour.

It’s important to note that it is the individual dog who decides whether or not the consequence of their behaviour is reinforcing (perceived as beneficial so will more likely be repeated) or punishing (perceived as bad so will less likely be repeated).

A common example is an owner who verbally reprimands their dog for jumping up because they view telling the dog off as punishing and hope this will stop the dog from jumping up in future.

However, if the dog finds it rewarding to receive their owner’s attention – regardless of how cross the owner might appear or the tone/volume of their voice – they can still find this interaction reinforcing and continue to jump up.

Operant conditioning is most commonly used when training a dog with reward-based teaching methods. It can be very effective if the owner can pinpoint which things the dog finds most enjoyable and rewarding – such as treats and toys.

They can then use these to reinforce desired behaviours which the dog is therefore much more likely to repeat and do on cue. The Dog Friendly Clinic Scheme advocates that every dog should be trained using reward-based methods (positive reinforcement) – which are underpinned by robust scientific evidence – and objects to the use of equipment or techniques that use discomfort, pain, or fear to train a dog.

It is also important to remember that because behaviour is flexible, how a dog feels about things can change over time, so something a dog once found enjoyable and rewarding, which helped to strengthen their behaviour, might well be found punishing at another moment in time.

For example, a dog who generally really enjoys getting treats, might not find them rewarding when they are feeling nauseous.

It is believed that dogs can also learn through watching and then copying the behaviour of others. This is likely to be beneficial for development, and, ultimately, survival.

Consider puppies imitating their mother’s or each other’s behaviour when playing, for example, which can help them develop motor as well as social skills.

One-trial learning

Sometimes just one single experience, particularly if it is emotionally overwhelming for the dog, can be enough to inform future behaviour. For example, a dog who is involved in a car accident might not want to get into a car ever again.

It is commonly understood that just one painful experience is sufficient for a dog to negatively associate the context within which it happened.