Learning in the Real World
Legend Golden Retrievers
Sorry this is a bit deeper then many of my other training articles
A branch of psychology that endeavors to enlighten how a species gains knowledge and understand things can be described in a process known to behaviorists as “Learning Theory”. This consists of several diverse types of learning styles such as, operant conditioning, instinctual reactions, social facilitation, observation, conventional instruction, recall, mimicry, and classical conditioning. Learning is both art and science and additional people are finding that an understanding of learning theory helps in accepting behaviors. (Wag’N’Train)
Pavlov’s experiments made a type of learning called Classical Conditioning very famous. The basics of the experiment are that scientists, under the direction of Pavlov, offered food to dogs and then calculated how much they drooled. After this was established, he started ringing a bell right before the food was presented to the dogs. After a while, the dogs began to drool at the sound of bell, however they did not drool at first until the dog food was given. The dogs learned to connect the sound of the bell to the offering of food, which became equal to the appearance of food. They also learned that if food was taken away and just the bell sounded eventually the dog would stop drooling to the sound of the bell. (Psychology)
The use of Classical conditioning is to train an automatic response to a stimulus. (Wag’N’Train) In other words if I ring a bell I can get my dog to drool using Pavlov’s methods. Stimuli are called primary or unconditioned stimuli if the animal or person reacts without training. This includes food, water, pain and other instinctive stimulus. Secondary stimuli are things the dog has learned to like or dislike. i.e. the use of a shock stimulus.
While classical conditioning is important in learning, it is operant conditioning that has produced incredible results especially to those of us who train animals. While Thorndike, Watson and Hull were some of the early behaviorists who researched operant conditioning, B.F. Skinner did the most advanced research and is credited for his extensive study in learning theory. He is known as the behaviorist’s behaviorist. (Psychology)
The idea that reinforcement, is better than reward because when used properly it can produce lasting predictable results. Reinforcement can also be used to increase duration of a behavior. Operant conditioning can also increase the likelihood that a certain behavior will be repeated. For example, if a family member calls you, and you are genuinely delighted to hear from them, the likelihood that they will call again has probably increased. However, if that family member were to call and you accused them of never calling and not caring, you have probably decreased the likelihood that they will ever call again. This is the law of effect, which Thorndike studied and developed. (Psychology) We use operant conditioning without even knowing we are doing it. Our response can be a positive or negative reinforcer and increase or decrease the likelihood that a behavior will occur.
Timing and variable schedules play an important role in learning theory. Whenever a reinforce is used very latently, the learning is probably not associated to the correct behavior. When a reinforcer is used to early, it is called bribery, because there is not behavior yet to reinforce. Reinforcement needs to match the behavior in order to be effective. If you ask a whale to make a 22 ft jump straight up in a pool of water, a simple smelt will not be a sufficient reward to have the behavior repeated. “Larger rewards cause bigger prediction-errors and lead to faster learning than smaller rewards.” (Rose 2009)
Shaping behavior does not have to be a continuous faucet of learning opportunities. According to “The Relationship Between the Number of Training Sessions per Week and Learning in Dogs” the results of the study showed that multiple training sessions per week did not produce more learning than the dogs that were trained once a week. In fact, the dogs that were shaped once a week learned significantly more than the dogs that had many more training hours per week. The study proved that weekly training resulted in better learning performance than daily training. (Meyer 2008) This leaves you to wonder if the college concept is more conducive to learning than the high school approach.
Other factors that can affect learning are social learning and dominance ranking status. Where a dog falls in the pack can significantly affect his learning and performance. Perceived dominance ranking can have a strong effect on social learning in dogs according to “How does dominance rank status affect individual and social learning performance in the dog?”. (Pongracz 2008) The study also revealed that perceived dominance may not be the same as indisputable dominance.
Shaping behavior is breaking the desired behavior down into small understandable steps that can be perceived and results are easily obtained. Steps that are too difficult or contain too many pieces are confusing. “Raising criteria in increments small enough so that the subject has a realistic chance for reinforcement.” (Pryor pg 54) Good shaping does not have unrealistic expectations, uses variable schedules of reinforcement, focuses on one criteria at a time, and understands that there are multiple pathways to the same result. In other words, if the door is not open, a good trainer will look for an open window or another doorway. People who truly understand shaping behavior, know that training begins with understanding the individual specimen being trained.
Training is not without its challenges and conflicts and “resolving conflict involves forgiving”. (Aloff pg 109) It is obvious that some behaviors cannot be allowed, i.e. using the house as a bathroom. These behaviors cause conflict between the trainer and the trainee. Communication here is the key to success. When you are training animals who cannot speak English, you need to learn to understand what they are saying. This is private detective work at its best.
Modifying behavior is more challenging than shaping behavior that does not exist. Behavior modification means changing a learned behavior and replacing it with a new and often opposite behavior. Few trainers understand behavior modification because they only know how to shape a non-existent behavior.
Lastly, training is never stable. If you don’t use it, you lose it. Behaviors that are trained and shaped must be maintained and rehearsed without causing boredom or disdain for the behavior. Learning should always be fun and a good trainer is fun to be with. A well-trained animal is an animal that practices learned behavior, applies what has been learned and does not need continuous reinforcement or reward for the behavior to be continued. Reinforcement and rewards are different for different subjects. Just because the whale loves fish does not mean the dog will work for it. Just because one dog loves cheese, does not mean the other dog will want or even consider cheese a reward.