Operant Conditioning
Biography
Burrhus Frederic Skinner was born March 20, 1904, in Susquehanna Pennsylvania. Burrhus received his BA in English from Hamilton College in upstate New York. After some traveling, he decided to go back to school, and earned his masters in psychology in 1930 and his doctorate in 1931, both from Harvard University., and stayed there to do research until 1936.
In 1931 he moved to Minneapolis to teach at the University of Minnesota. There he met and soon married Yvonne Blue. In 1945, another move took him to the psychology department at Indiana University, where he became department chair. In 1948, he was invited back to Harvard, where he remained for the rest of his life. He was a very active man, doing research and guiding hundreds of doctoral candidates as well as writing many books.
August 18, 1990, B. F. Skinner died of leukemia after becoming perhaps the most celebrated psychologist since Sigmund Freud.
Skinner accepted the model of classical conditioning as originated by Pavlov and elaborated on by Watson and Guthrie, but he thought this type of conditioning only explained a small portion of human and animal behavior. He thought that the majority of response by humans do not result from obvious stimuli. The notion of reinforcement had been introduced by Thorndike, and Skinner developed this idea much further.
Skinner's Theory: Operant Conditioning
B. F. Skinner's system is based on operant conditioning. The organism, while going about it's everyday activities, is in the process of “operating” on the environment. In the course of its activities, the organism encounters a special kind of stimulus, called a reinforcing stimulus, or simply a reinforcer. This special stimulus has the effect of increasing the the behavior occurring just before the reinforcer. This is operant conditioning: “the behavior is followed by a consequence, and the nature of the consequence modifies the organism's tendency to repeat the behavior in the future.” A behavior followed by a reinforcing stimulus results in an increased probability of that behavior occurring in the future.
Skinner's observations can be divided into independent variables which can be manipulated by the experimenter, and dependent variables, which can not be manipulated by the experimenter and are thought to be affected by the independent variables.
Independent variables:
Type of reinforcement Schedule of reinforcement
Dependent variables (measures of learning):
• Acquisition rate- how rapidly an animal can be trained to a new operant behavior as a function of reinforcement. Skinner typically deprived his lab animals of food for 24 or more hours before beginning a schedule of reinforcement. This tended to increase acquisition rate.
• Rate of response- this is a measure of learning that is very sensitive to different schedules of reinforcement. In most cases, animals were given intermittent schedules of reinforcement, so they were called upon to elicit the desired response at other times as well. Rate of response is a measure of correct responses throughout a testing schedule including the times when reinforcement is not provided after a correct response. It appears as if test animals build expectations when they are given rewards at predictable times (Animals which are fed at the same time each day become active as that time approaches, and a dog whose master comes home at the same time each day becomes more attentive around that time of day.) Also, Skinner found that when fixed interval reinforcement was used, the desired behavior would decrease or disappear just after a reinforcement, but when it was almost time for the next reinforcement, the animal would resume the desired responses.
• Extinction rate- The rate at which an operant response disappears following the withdrawal of reinforcement. Skinner found that continuous reinforcement schedules produced a faster rate of learning in the early stages of a training program, and also a more rapid extinction rate once the reinforcement was discontinued. A behavior no longer followed by the reinforcing stimulus results in a decreased probability of that behavior occurring in the future.
Types of reinforcement:
1 Primary reinforcement- instinctive behaviors lead to satisfaction of basic survival needs such as food, water, sex, shelter. No learning takes place because the behaviors emerge spontaneously
2 Secondary reinforcement - the reinforcer is not reinforcing by itself, but becomes reinforcing when paired with a primary reinforcer, such as pairing a sound or a light with food.
3 Generalized reinforcement - stimuli become reinforcing through repeated pairing with primary or secondary reinforcers. Many are culturally reinforced. For example, in human behavior, wealth, power, fame, strength, and intelligence are valued in many cultures. The external symbols of these attributes are generalized reinforcers. Money, rank, recognition, degrees and certificates, etc are strongly reinforcing to many individuals in the cultures that value the attributes they symbolize.
Reinforcers always follow a behavior and could be pleasant or unpleasant (noxious) and could be added to or removed from a situation. The following table summarizes the various combinations:
Add to a Situation After a Response:
Pleasant = Positive Reinforcement- Reward. Increases the probability of the same response occurring again. (Example: praise, monetary reward, food)
Noxious = Punishment- Administering a painful or unpleasant reinforcer after an unwanted response. Decreases the probability of the same response occurring again.(Examples: corporal punishment, electrical shocks, yelling)
Remove from a Situation After a Response:
Pleasant = Punishment - Decrease the probability of the same response occurring again (Example: punishing a teenager by taking away his cell phone or car keys.)
Noxious = Negative Reinforcement - Removing or decreasing an unpleasant or painful situation after a desirable response is produced. Increases the probability of the same response occurring again (Example: time off for good behavior)
Schedules of Reinforcement:
• Continuous reinforcement - reinforcement is given every time the animal gives the desired response.
• Intermittent reinforcement - reinforcement is given only part of the times the animal gives the desired response.
• Ratio reinforcement - a pre-determined proportion of responses will be reinforced.
• Fixed ratio reinforcement - reinforcement is given on a regular ratio, such as every fifth time the desired behavior is produced.
• Variable (random) fixed reinforcement- reinforcement is given for a predetermined proportion of responses, but randomly instead of on a fixed schedule.
• Interval reinforcement- reinforcement is given after a predetermined period of time.
• Fixed interval reinforcement - reinforcement is given on a regular schedule, such as every five minutes.
• Variable interval reinforcement - reinforcement is given after random amounts of time have passed.
In animal studies, Skinner found that continuous reinforcement in the early stages of training seems to increase the rate of learning. Later, intermittent reinforcement keeps the response going longer and slows extinction.
Skinner specifically addressed the applications of behaviorism and operant conditioning to educational practice. He believed that the goal of education was to train learners in survival skills for self and society. The role of the teacher was to reinforce behaviors that contributed to survival skills, and extinguish behaviors that did not. Behaviorist views have shaped much of contemporary education in children and adult learning.
Learning Theory Bibliography
Boeree, C. G. (1998). B. F. Skinner. Retrieved September 19, 2003 from http://www.ship.edu/%7Ecgboeree/skinner.html
Lefrancois, 1972
Santrock, 1988
Merriam & Caffarella, 1991