Wednesday's WTFact: Conditioning- Why Training Works
There once was a man named Ivan Pavlov who was a Noble Prize winning scientist in the late 1800's. He discovered that initially, all the dogs in his experiments would salivate when they saw their food coming (as do I). He began to precede their feeding by sounding a buzzer. The dogs quickly linked this specific sound to their food coming, and began salivating simply at the sound of the buzzer, even before their food was every presented. This was to become the foundation training of what we call "Classical Conditioning".
What this means is that animals can learn that every time X happens, Y follows. After an extended period of time, neurons in the brain then wire and intertwine those two factors together. Every time there is a buzzer noise, food follows. Therefore buzzer equals food, even though a buzzer on it's own doesn't have anything to do with food. Though conditioning is often somewhat of a slower process it can also happen immediately in traumatic situations. Veterans of war often suffer from this effect of neurons wiring and firing together. For example, hearing a sudden loud noise equating to life threatening danger even after they return home. It is the brain using this conditioning for survival mode. Conditioning has many different implications. It can reinforce good behavior, curb bad behavior, and in a lot of ways be the underlying cause of unwanted behavior.
Using Conditioning to Reinforce Good Behavior
The secret to reinforcing any behavior is all in the timing. Being able to read your dog's body language so that at the exact moment they make the decision to do what you've asked of them, you reinforce it with a reward. This can be verbal praise, a treat, a toy, or a pat on the head. If you wait too long (more than a few seconds even) the X and Y factor will not get paired together in their head and they can get confused with which thing they did you seem so excited about. Was it sitting when you asked, or getting up to come see you after they sat for a moment? They may be thinking to themselves "OK, she wants me to sit for a second, then get up and go to her. Got it, 'sit' means get up and go to mom." The way to simplify this process is by utilizing something called a bridge. This literally bridges that exact moment in time when they did what you want with the promise of reinforcement. When you are calling your dog to come back to you, the moment they turn their body around (this physical change in body movement clearly indicates their intention to come to you) you can bridge that moment in time by saying "GOOD!" or using a clicker, or buzzer. If you wait until they come all the way back to you to say anything, you have missed bridging the moment they made the right decision. This will also reinforce their decision and encourage them to complete the task of coming all the way to you because they know a reward is waiting and they are sure they have done the right thing.
The precursor to this of course is pairing that bridge (the word 'good' or the sound of the clicker) with something positive, like a dog treat. The bridge ("Good") becomes the X factor, the reinforcement (treat or affection) becomes the Y factor. So every time they hear the word, they will pair it with something positive and come back to you.
Using Conditioning to Curb Bad Behavior
Utilizing the same principles, the word "No" becomes the bridge to a consequence for their action. This can be anything from redirecting them with a leash, to shaking a can of coins, making a startling noise, or going to time out. When a dog is jumping on a house guest, if every time you say "OFF!" it is immediately followed by redirecting them down with a leash, the "Off" command becomes the X factor, and a leash correction becomes the Y factor. But X always has to mean Y for those two things to start wiring together for the dog. Once wired together through consistency, you will not need to use the correction as much because they will anticipate it through this classical conditioning. However, it needs to be sporadically reinforced with the Y factor consequence for them to keep the cause and effect wired together long term.
Conditioning As An Underlying Cause Of Bad Behavior
A lot of times we are unknowingly reinforcing bad behavior. If every time your dog is scared and growling at a stranger, you tell him "it'll be alright" and pet him reassuringly, what this actually does is reinforce the fear by eliciting a positive reinforcement from you. In your head it's calming him down, to the dog it's justifying that his fear is appropriate. X factor is the fear, Y factor is being pet and given ample affection.
When a dog choses to ignore your reinforcement what that is telling you is there is something else happening that is more rewarding than what you are offering. For example. A dog that loves to chase cats is always going to pick the cat over the cookie. So if he gets past you at the front door and a cat runs by. Your offer him a dog cookie to come back in the house instead of running after it. The dog is going to pick the cat every time because the value of the reinforcement you are offering is not greater than the instinctual act of getting to chase and possibly even catch a real live moving cat! Therefore what you chose to use as a reinforcement, both positively and negatively has to be appropriate to the environment and situation at hand.
Self reinforcing behavior is something to consider as well. Things like barking, jumping, and digging are self rewarding. Nothing from you needed! The only effective way to stop these behaviors is by implementing a reinforcement that is greater in value to the dog then their unwanted action. Much like jumping on people when they walk in the door, your reaction has to be immediate and consistent. This can be difficult with behaviors that often happen when the dog is alone. Not giving them the opportunity to make bad decisions can be a more manageable solution. Utilizing crate training, blocking them from being able to see things to bark at, or not allowing free access to a yard to dig when no one is home can all be used in the conditioning process as well.
For more stories, tips and tricks on training or if you have any questions please visit our website at www.dogtrainingredefined.com or email Andrea at Andrea@theanimaldept.com