Quoted Post by Chris Jackman of Enrichmutts
As trainers when we hear the words operant and classical conditioning we typically associate it with humans applying it to dogs to reinforce or punish behaviour in training situations.
As “force free” trainers we try to actively avoid using two of the four operant quadrants which are positive punishment and negative reinforcement when we are training our dogs. These are considered unethical as they use aversion to influence behaviour.
But what about dogs? Do they use quadrants and classical conditioning?
The answer is absolutely they do.
Heres a few examples of applying operant conditioning to “Dog 1”
Positive reinforcement – Dog 1 play bows (Behaviour) Dog 2 runs (Reinforcement)
Negative punishment – Dog 1 wants to play (behaviour), dog 2 ignores dog 1 (punishment)
Negative reinforcement – Dog 1 approaches (behaviour), Dog 2 Stares until dog 1 stops (Reinforcement)
Positive punishment – Dog 1 jumps on dog 2 (Behaviour), Dog 2 barks at Dog 1 (Punishment).
So now we understand dogs can and do use operant conditioning, why is this important to understand as a “force free” trainer?
Well because I’ve seen “Force Free” trainers who are against humans using positive punishment or negative reinforcement on dogs, advise that it is correct socialisation or ethical if dogs use those two quadrants on each other.
An example I’ve seen is an adult dog with a bone, a young puppy approaches, in response the adult dog stares hard at the puppy (attempt at applying negative reinforcement) which the puppy ignores. As the puppy gets close then the adult dog snaps (positive punishment) and the puppy backs off.
Another example I often see is puppies who get quickly over aroused in interactions and jump on or hump other dogs. A common mindset is the adult should “correct” the puppy by snapping and/or barking (positive punishment).
My personal aim for socialisation is to expose my dog to other in a positive and constructive way while protecting them from negative interactions where possible, creating a positive and calm emotional response.
The above examples simply do not fit into my idea of ideal socialisation experiences as they can and do create associations of anxiety or fear.
So, in those situations what would I do? I’ll use the first example to show you:
Adult dog has the bone, puppy approaches so adult dog stares, I let puppy see this then use a positive interrupter/attention noise and reward for them disengaging. I will then follow up by using management such as separating the two dogs or putting pup on lead or marking and rewarding behaviours that are not approaching the adult dog.
What I’m doing here is teaching the puppy through classical conditioning that a dog staring is a cue to disengage. Here’s how it works:
New cue (stare) > Old cue (attention noise) = reward from trainer.
This is super smart training because not only have we managed a potentially negative socialisation experience for the puppy and adult dog, we’ve also taught our puppy that “staring” (which is often followed by a more adverse behaviour) is a cue to disengage from another dog.
What we have to acknowledge is that positive punishment and negative reinforcement often increase arousal of the puppy when applied. This hinders the puppy’s ability to make good choices to rectify the interaction.
To say dogs need to use positive punishment and negative reinforcement to learn socialisation completely dismisses the other two quadrants as options.
When socialising I purposely pick dogs that are known to be tolerant and resort to less averse methods typically negative punishment.
This normally looks like the adult dog standing still, looking away and waiting for puppy to stop their behaviour. This works so well because it reduces arousal input from the adult dog making it easier for the puppy to make correct choices, take note of social cues and hear handler cues such as redirecting with an attention noise. We then use management to help the puppy regain calmness and make great choices which we can reinforce using positive reinforcement.
We can apply the same process as before:
Adult dog freezes (new cue) > attention noise (old cue) > reinforce puppy for disengagement.
Ideally, we don’t even want this situation to happen, we should be setting puppies and adults up to have appropriate interactions with each other and be using positive reinforcement to maintain and make them more likely to happen. This is easily possible by knowing the personalities of both dogs, clever planning and knowing what situations can lead to possible altercations.
My last notes has to acknowledge that life happens when dealing with two sentient beings. Things do go wrong or don’t go as we ideally planned but in my opinion we should not actively be setting these situations up in the name of socialisation.
Comentários