"Don't become a mere recorder of facts, but try to penetrate the mystery of their origin." -Ivan Pavlov
I'm sure at some point you've heard at least one reference to Ivan Pavlov's theory for structural/associative learning. The story of his experiments with dogs and their feeding has transcended animal behavior circles to human behavior and psychology as well. I can even vaguely recall hearing references to Pavlov in numerous movies and television shows throughout the years prior to ever knowing who he was. What exactly is Pavlov’s theory? In summary, the theory involves pairing a neutral stimulus with an unconditioned stimulus to elicit a conditioned response. In other words: associative learning or "X" happens which then means "Y" happens.
Pavlov realized that when he was feeding his dogs, they would naturally salivate with the presentation of their food (the unconditioned behavior). By presenting the dogs with a ringing bell (a neutral stimulus) prior to receiving their food, he noticed they would begin to salivate upon hearing the bell (conditioned response) after enough time and repetitions. The dogs began to associate the ringing of the bell with the presentation of their food. After the response became conditioned, Pavlov could ring the bell and the dogs would then begin to salivate even when the food wasn't present.
Since dogs learn primarily through association rather than deductive reasoning, we can leverage Pavlov's observation through repetition and consistency. When we train recall by overlaying the command "here" (the neutral stimulus or "cue") with the dog coming back to mom or dad (the conditioned response or "behavior") and the giving the dog kibble (the unconditioned stimulus or "reward"), the dog learns "recognize cue, execute behavior, get reward."
Keep in mind, this not only works with creating desired behaviors but it can also work in creating undesired behaviors. Often when you aren't even aware of it happening. Sticking with the same recall example, you give the "here" command and the dog doesn't listen on the first command, so you just repeat it. Then again. Then again. Now you're frustrated and when they finally come back, instead of rewarding them, maybe you angrily grab their collar or scold them. Not only have you just conditioned your dog to ignore the first 4 times you give the command, but returning to you no longer results in a reward. So, the next time you give the "here" they are even less likely to come back to you. This is a major reason why trainers so often preach the importance of being consistent and not repeating commands. We want to build the association within the dog that when they hear the cue "here" they want to come back to their owner as fast as they can because good things are about to happen!
This doesn't only apply to dogs. You can see examples of classical conditioning in our own daily lives and routines. I even used associative learning with my homing pigeons. I have a big, ugly, red bell that I found at a random garage sale years ago. I have it right by my pigeon loft and every single feeding time, I ring the bell then open the loft and pour their food in. Why do I do this? So, when I let them out to fly and get exercise, I can call them back into the coop just by ringing the bell. We hear a bell. They hear "DINNER!" and rush back to the loft to get their fill. Essentially, I taught my pigeons to recall but they don't see it that way. All the bell means to them is that it's dinner time. Not that I want them to come back to me.
In the initial phases, associative learning is relatively simple: cue, behavior, reward, repeat. Two major pitfalls to be aware of are timing and consistency and I’m sure that most of us don’t want to spend the rest of our dog’s life as a treat vending machine, so how do we eventually move past this? First, a little physiology review: you may have heard of the “happy hormones” before that includes dopamine, oxytocin, serotonin, endorphin, cortisol, and adrenaline. There are some others, but these are the main ones and they all play a different role with their own brand of happiness. We’re going to focus on dopamine, which is sometimes called the “reward molecule” as it is responsible for reward seeking behavior. Ultimately, we are looking to swap the unconditioned stimulus (reward) from the kibble to a spike of dopamine in the dog's brain when presented with a specific cue. This is what truly leaves a lasting impression in building that conditioned response. Pavlov's dogs only cared about the bell because of the involuntary bodily response to their food. If the “reward” doesn't have value to the dog, then the “cue” before it won’t either. The dog has to truly WANT the outcome of the behavior for any of this to actually matter. This is why recognizing what drives your dog matters so much; not all dogs care about food or toys. This is unique to each individual dog.
It’s also important to know that dopamine levels spike during anticipation. In other words, for any of this to leave a lasting impression, the cue or marker must come before the payment. In the example with my pigeons, if I were to pour their food in the loft and then ring their bell after they had begun to eat, the bell wouldn't create the dopamine spike and have significance to the pigeons. This is one of the reasons that markers (words like “good” or “yes” or even a clicker) can be such a powerful tool. when used appropriately.
Consistency will further solidify the behavior and in this context, consistency signifies many things, including the manner in which cues are given, the proper sequence (cue, behavior, reward), and the frequency of the reward or payment. When first establishing a conditioned response, the payment/reward needs to be there every time. We cue the dog, they do the behavior, they get paid. Repeat. Repeat. Eventually, once the conditioned behavior has been established, we can then move into variable reward systems or even transition the reward structure to something else (from kibble to some form of play or praise). The long-term goal is to condition the behavior well enough that you don't have to give any reward. It would then be the expectation and standard, and truthfully, you will have established the reward pathway such that the dog is being intrinsically rewarded via the dopamine spike that accompanies the obedient behavior.
Once you understand the foundation of classical conditioning, it can lead to countless different rabbit holes to include reward structures, markers, types of drive, free shaping, etc., all of which are topics for another day. For now, just begin with thinking about the behaviors your dog has already established through classical conditioning, remembering that it is always occurring whether intentional or not, during formal training sessions or just being a part of your family around the house. Now that you’re here, you may start to realize how often dog's undesirable behaviors are built off their handler’s unintentional actions, daily routines, and environmental factors. What is something you do on a routine basis that your dog has developed a very predictable conditioned response without you intending to do so?
Comments