Unformatted text preview: 2/21/07 Psych 2
Principles of Psychology
Christopher Gade Office: 5315 Tolman Hall Office hours: MW 2:00-3:00 Email: [email protected] Lectures: MWF 3:00-4:00, 100 GPB Operant Conditioning
• The process of learning to associate a behavior with a consequence, in order to behave in a manner that maximizes reinforcing and minimizes punishing events. – Reinforcement: any event that increases the future probability of the most recent behavior. – Punishment: any event that decreases the frequency of the preceding behavior – What makes something reinforcing or punishing?
– Biologically useful – Intrinsically satisfying – Restores equilibrium * Disequilibrium principle: any behavior that leads to a return to equilibrium will be reinforcing. Edward Thorndike (1874-1949)
• • Originated the idea of instrumental learning. Studied cats and other animals learning by trial and error to escape from puzzle boxes. The Thorndike Laws Law of Effect: Behaviors followed by favorable consequences become more likely; behaviors followed by unfavorable consequences become less likely. Law of Readiness: A series of responses can be chained together if they belong to the same action sequence and will result in annoyance if blocked. Law of Exercise: Connections become strengthened with practice and weakened when practice is discontinued. Skinner attempted to expand on Thorndike’s original theories of instrumental learning. He proposed that the learning process has a very predictable response to rewards and punishments. His work set out to show how those responses to behavior influenced future behaviors (e.g. operant conditioning). The majority of Skinner’s work was done on rats and pigeons in elaborate boxes that he designed. These boxes were called “Skinner Boxes”. B. F. Skinner (19041990) 15:30 But there’s more…
Not only are there reinforcements and punishments in operant conditioning, but these responses are either positive (adding something), or negative (taking something away).
The 2x2 Matrix of Operant Conditioning:
• • • • Positive reinforcement: an introduction of a pleasurable stimulus, which will increase the likelihood of the future occurrence of the behavior (e.g. chocolate cake). Negative reinforcement: a removal of an aversive stimulus, which will increase the likelihood of the future occurrence of the behavior (e.g. nagging). Positive punishment: an introduction of an aversive stimulus, which will decrease the likelihood of the future occurrence of the behavior (e.g. spanking). Negative punishment: a removal of a pleasurable stimulus, which will decrease the likelihood of the future occurrence of a behavior (e.g. taking away your allowance). What type of reinforcement is this? What about this one? A final review of the 2x2 matrix… A review of other related behavioral concepts…
• • Extinction: behavior without reinforcement. Stimulus Generalization: responding to a stimulus that is similar to the originally reinforced stimulus. Discrimination: not responding to a stimulus that does not result in reinforcement. • Further Skinner Contributions In addition to his work on the 2x2 operant conditioning matrix, BF Skinner was able to expand on his theories of learning in order to show how complex behaviors can be learned (and taught) through operant conditioning. Shaping: a series of positive reinforcements (or punishments) to guide behavior closer to a desired goal. Chaining: something like uber-shaping. It is the process of reinforcing behaviors with the opportunity to engage in subsequent behaviors. In this process, an organism learns the final behavior, and then the next to last, and so on, until the desired sequence is reached. Different Schedules of Reinforcement/Punishment
• • Continuous reinforcement: reinforcement for every correct response. Partial/intermittent reinforcement: occasional reinforcement for a correct response.
• Fixed ratio: Reward for a behavior after “X” responses. Causes faster responders to get more rewards. Produces high rates of responding, but quick extinction when the reinforcement is removed. Variable ratio: Reward for a behavior after a variable and unpredictable numbers of responses. Gambling is a great example of this reward system. It is very hard to extinguish after the connection is made. Fixed interval: Reward for a behavior after “X” amount of time has passed. The responses are rather sparse in down time, but get more vigorous right before time X. Variable interval: Reward for a behavior after a variable and unpredictable amount of time. This causes slow, steady responding. • • • Responses x Time Diagram
Number of responses 1000 Fixed Ratio Variable Ratio 750 Fixed Interval Rapid responding near time for reinforcement Variable Interval 250 Steady responding 0 10 20 30 40 50 Time (minutes) 60 70 80 500 Effectiveness of Reinforcement
q All things being equal, most people learn fastest with immediate reinforcement or immediate punishment. Punishment tends to be ineffective except for temporarily suppressing undesirable behavior. Mild, logical and consistent punishment can be informative and helpful. Though vicarious reinforcement can be effective, vicarious punishment is often not. Positive reinforcement can sometimes interfere with learning. If an activity is intrinsically rewarding, getting a new kind of reward can diminish a person’s interest in that activity
– q q q q – – Intrinsic motivation – doing something well for the sake of the process, having a sense of ownership over one’s achievements Extrinsic motivation – doing something for some external reward The “overjustification effect” So What Did We Learn? We learned about operant conditioning. We learned about Edward Thorndike and BF Skinner. We learned about how different reward schedules can influence the rate, timing, and extinction of learned responses. We learned about learning. And now… After learning about how we take in information, we’re going to examine how we keep that information in our head. Namely, this next series of lectures is going to discuss memory, and the way the mind stores information. See you on Friday. ...
View Full Document