BEHAVIOR
how dogs learn
classical conditioning
Classical conditioning (think Pavlov's salivating dogs) is automatic. It is emotional, out of the control of the dog, and is the environment having an effect on the dog:
The presence of a strange person on your walk causes the dog's heart to beat faster and the dog starts pulling because they feel nervous.
The rustling of a treat bag causes the dog to salivate in anticipation.
You can't ask the dog to stop this reaction. However, you can help to create a new association, with time. This is why dogs sometimes stop enjoying a treat that they used to love—especially if we use it to lure them into a situation they do not enjoy.
When we train skills such as “drop it”, classical/emotional learning is always at the foundation of the skill and is happening. That is why we work to create happy feelings (with treats) about sharing resources, instead of relying on the dog just tolerating the experience of sharing resources. If the dog has happy and neutral feelings about most things in their world, they are safer and more reliable companions.
trigger
A stimulus that creates a strongly conditioned negative emotional response is a "trigger": sudden sounds outside the house may be a trigger for your dog. Another trigger may be visitors to your home. Often when we see dogs react intensely to a trigger, it is because trigger stacking has occurred: this is when more than one trigger occurs with minimal time for recovery in between the triggers. This often happens on walks and other adventures, because the dog does not have the chance for down-time. When we train, we build down-time into the plan so the dog can make positive associations instead of melting down over time.
We can condition your dog to think the sounds outside are awesome by associating wonderful things with the sound, but this has to be done carefully. As soon as the sound starts, the treats start and as soon as it stops, the treats go away. When there is no noise, it's boring. Your dog will soon be anticipating the next opportunity for a visitor treat party and will feel happy when the noises start, as long as we can find a ‘therapeutic’ exposure level (distance, intensity and duration all matter here).
counterconditioning
This is slow exposure to something the dog already knows about in a controlled way so that the conditioned emotional response of the dog is 're-written' in a way that creates a different emotional response (usually, a 'negative/unhappy' feeling is replaced with a 'positive/happy' one). We use distance (as a way to manage intensity) and high value/frequency of rewards and timing of rewards to help the dog feel positively about something they used to feel negatively about. A dog will have an emotional response of fear when they see things that remind them of past scary experiences. If we systematically pair those things with yummy food, the dog will slowly be counter conditioned to be happier around those things. And we have to make sure that the dog makes the association: appearance of the scary thing CAUSES amazing treats to appear... the dog has to feel safe to be able to learn this lesson.
You cannot always control the triggers on walks, and for some dogs this means that they are overly stressed on walks. This is why we may recommend that you minimize or even refrain from walks at the start of your training journey. We will always work with you to find replacement activities to meet your dog’s needs.
If you want to take a look at an example of counterconditioning, look at this blog entry.
operant conditioning
An action the dog takes can change the environment: for example, maybe the dog knows growling makes the person move away; that sitting means they get a treat. This is a way that dogs can predict and control the environment. This is the type of learning in which we can eventually use a clicker (or marker word) for, marking the exact behavior that they can engage in to get something they want (such as praise, food, or play). This type of learning can only occur when the fight-flight system is not ACTIVATED. This is why we need to start training with our dogs in a location that is far enough away from the trigger that they are in the “thinking and learning zone” (an operant state of mind). With training and practice, your dog learns to do certain things in order to earn a reward when you ask (you say SIT, your dog sits down, you give a treat). Your dog can also do certain things on their own to gain access to a reward (stranger approaches, your dog growls, person moves away).
proofing behavior
Once your dog knows what they can do to get things they want (sit = door opens; down=treat; loose leash=movement; come when called=release to play), we can start to help our dog learn how to be more reliable with the behavior. This means we need to slowly increase the difficulty of where and how they do the skill. We break difficulty into three categories:
*distance | how far away are you from your dog/are you looking at your dog)
*duration | how long can they sustain the skill before you mark and reinforce)
*distraction | how many distractions are present when your dog does the skill)
If we increase difficulty too quickly, the skill falls apart. If we increase difficulty slowly and one D at a time, the dog is usually very successful and generalizes the skill to many different locations and contexts.
