What is learning theory and why should I care?
Theories provide a basis to understand how dogs learn and a way to explain, describe, analyze, and predict learning. Theory helps us make more informed decisions around the design, development and delivery of training. While there are a vast array of learning theories that can be applied to dog training, the two most basic theories in dog training are classical conditioning and operant conditioning.
Classical conditioning, also known as pavlovian conditioning is where a previously meaningless stimuli (such as a word or sound), is paired with something that does have meaning to the dog (such as food) in a way that causes the dog to have the same response to the previously meaningless stimuli as they do to the meaningful stimuli. Classical conditioning was first described by Ivan Pavlov who was doing experiments on salivatory responses in dogs. Pavlov designed an automatic feeding system to feed the dogs and paired that feeding system with a bell. At first the sound of the bell meant nothing to the dogs, but over time after numerous repetitions of hearing the bell immediately before the food was dispensed, Pavlov discovered that the dogs experienced the exact same physiological response to the bell as they did to food, even when no food was present. Pavlov also discovered that in order for this response to occur, the bell had to be rung at specific times in relation to the food delivery in order for conditioning to occur.
Classical conditioning DOES NOT occur when:
- Bell was rung too far ahead of the food being delivered
- Bell was rung after the food was delivered
- Bell is rung at the same time as the food being delivered
Classical conditioning DOES occur when:
- Bell is rung immediately prior to the food being delivered
It is also important to know that the behavior that occurs as a result of classical conditioning is an involuntary response, the dog cannot choose whether or not to respond in the way we have conditioned them to respond, it happens automatically.
The basic concept behind operant conditioning is that a stimulus leads to a behavior, which then leads to a consequence. The dog makes a choice, and that choice has a consequence, and that consequence will affect how often the dog makes the same choice in the future. Operant conditioning consists of 4 quadrants that describe the relationship between actions and consequences. These 4 quadrants are:
Positive Reinforcement (R+)
Negative Reinforcement (R-)
Positive Punishment (P+)
Negative Punishment (P-)
It is important to note that in the context of operant conditioning, positive and negative do not mean bad or good but are used purely mathematically. Positive simply means we are adding something, and negative simply means we are taking something away with no emotional implications. Reinforcement means that we have caused a given behavior to be more likely to reoccur in the future, and punishment means we have caused a given behavior to be less likely to reoccur in the future.
The classic example of positive reinforcement is teaching a dog to sit. When the dog makes the choice to sit, we reward the dog for making that choice by giving them a treat. We added the treat when the dog sat (positive) to make the behavior of sitting more likely to reoccur in the future (reinforcement).
In some cases the dog might have difficulty understanding what we’re asking them to do, and it is not uncommon to see people push on their dogs butt to help encourage them to sit. This is an example of negative reinforcement, when the dog sits, the handler removes the pressure being applied by their hand (negative) which makes the behavior of sitting more likely to reoccur in the future (reinforcement)
If the dog we are trying to train really wants the treat we have in our hand that we are trying to use to get them to sit, but they are refusing to sit so we put the treat away, that is an example of negative punishment. By taking the treat away (negative) we have made the behavior of not sitting less likely to reoccur in the future (punishment)
If the dog we are training is refusing to sit and we respond by verbally telling the dog “no”, this is an example of positive punishment. By adding the word “no” (positive), we are making the behavior of not sitting less likely to reoccur in the future (punishment).
Positive (+) = Adding something
Negative (-) = Subtracting something
Reinforcement = Increases the likelihood of a given behavior being repeated
Punishment = Decreases the likelihood of a given behavior being repeated
Applying Theory to Real Life
Training theories give us a better understanding of how dogs learn so we can communicate with them more effectively, and effective communication leads to effective training. One of the most prominent ways that the two above training theories are applied in real life is through the use of conditioned reinforcers or conditioned punishers, also commonly referred to as “markers”.
What is a marker?
A marker is a dog trainer’s term for a stimulus (typically a word or sound, although it could be a visual stimulus as well) that consistently precedes reward or punishment repeatedly until the stimulus itself becomes a reward or punishment via classical conditioning. For example, if I say the word “yes” right before giving my dog a piece of food, and I do that over and over and over again, eventually my dog will have the same physiological response to the word “yes” as they do to food, even when no food is present. Once the dog is classically conditioned to our marker word “yes”, we can now begin to use that word in training. A marker allows us to pinpoint precise moments in time when the dog is correct and gives us time to produce their reward after the behavior has occurred. Teaching a dog to sit in front of us is fairly easy to do without markers, the dog sits and you give them a piece of food to reward the sit. But what happens if the dog is ten feet away from us? In order to reward the dog for sitting we would need to run over to the dog and deliver the food to the dog while the dog is still sitting. If the dog is not sitting when we deliver the food, then we are not rewarding the dog for sitting, we’re rewarding the dog for getting up from the sit. If you go running over to the dog to reward them for a good sit the dog is likely to get excited and is unlikely to still be sitting by the time you get there. If you take your time and walk over quietly to give the dog a reward for sitting, the time required to do so can cause you to “miss your moment” and reward too late. With marker words, when the dog sits we can say our marker word, in this case “yes” which causes the dog to feel as though we have rewarded them with food at precisely the right moment in time. Markers can also serve as a way to reward our dogs when we don’t have rewards on us and help us to train dogs that do not depend on having a reward present in order to listen. Markers are also a key component of successfully incorporating play into our training.
Why Should I Incorporate Structured Play Into My Training?
Playing with your dog and doing obedience training are often viewed as two separate and mutually exclusive things, but the truth is that play is an absolutely essential part of good dog training that lasts!
Effective dog training follows three basic laws:
The law of effect
The law of readiness
The law of exercise
The law of effect states that responses to a situation which are followed by a rewarding state of affairs will be strengthened and become habitual responses to that situation. Play provides us with an effective way to reward your dogs' good behaviors consistently so those good behaviors become strong habits that are resistant to fading away over time.
The law of readiness states that learning is dependent upon the learner's motivation to learn, which facilitates the strengthening of the bond between stimulus and response. Dogs are born with a natural curiosity and a drive to learn and explore through play, and play is a very natural way for dogs to learn. Play is also fun, and a dog that is trained through play is a dog that is motivated to train!
The law of exercise states that connections become strengthened with practice and weakened when practice is discontinued. Play is fun for both the dog and handler, and when training is fun it will happen more often which means the behaviors you train with play will be practiced more regularly and will be less likely to fade away with time!
Other key benefits of training through play include
Playing is fun and helps you to build a better relationship with your dog.
Play is crucial for your dog's mental health, and helps to prepare the brain to deal with stresses and challenges by providing challenges for the dog to work through and rules the dog must be aware of in a fun environment without the risk of triggering a fear response in the dog.
Play sparks the brain’s reward centers, triggering a feel-good chemical called dopamine. Higher dopamine is linked to better memory, attention, mental flexibility and motivation.
Play is good physical exercise for both the dog and the handler.
Play serves as an extremely valuable reward in training, the more the dog wants what you have, the harder they are willing to work for it!