Learning: Operant Conditioning Flashcards
Operant Conditioning
a learning process in which a behavior becomes associated with a consequence; As a result of this association, the consequence influences the probability of that behavior occurring again in the future; Must take action to learn of negative or positive consequence
If linking behavior to a good outcome, more likely to repeat that behavior
Sometimes leads to superstitious behavior, like wearing “lucky” socks that were linked to a greatly rewarded behavior
Edward Thorndike
Scientist that firs demonstrated the power of changing behavior by manipulating the consequences of that behavior; Thought to be doing the same type of learning of Pavlov, but apparently very different; The Law of Effect
Law of Effect
if a behavior is followed by a pleasurable consequence, it will tend to be repeated; If a behavior is followed by an unpleasant consequence, it will tend to not be repeated
Example of Law of Effect by Thorndike
Cat in a closed puzzle box with a food pedal to open the door that will allow them out, and they will recieve tuna in return; the desired behavior to get a reward is the cat hitting the lever to be allowed out(no scratching or meows); Cat put in box over and over again and learns to get out
eventually over shorter periods of time
this shows a context-> behavior -> consequence pattern
B.F. Skinner
Extremely influential scientist associated with further defining operant conditioning and using it to modify and control behavior; studied behaviorism and believed all should only study that because only behavior changed with reward and punishments!! Believed we had no free will or real choice
Many psychologists disagree with Skinner today
Behaviorism
behavior changes through rewards and punishments; can only know that which is directly observable
Reinforcement
the response is followed by a pleasant consequence; as a result the response is now MORE likely to occur
Punishment
the response is followed by a unpleasant consequence; as a result the response is now less likely to occur
Two ways to reinforce behavior
Response-> get something good, pleasant outcome(study for test, get an A; dog sits and gets a treat); positive reinforcement
Response-> remove something bad, taking away a burden for them(going to class to remove chance of failure, watching someone’s kids so they can have a night out); negative reinforcement
Positive reinforcement
reinforcement by application
Negative reinforcement
reinforcement by removal
Example of positive and negativer reinforcement
cat meowing at 5:30 and got food(positive); individual feeds her at 5:30 in the morning in order for cat to be quiet and go back to sleep(negative)
Two ways to punish behavior
Response -> get something bad(Yelling, spanking, shock collar); Positive Punishment: punishment by application
Response -> remove something good(Time out or getting grounded, prison; home after curfew, so no TV for a week); Negative punishment: punishment by removal
How do you reward a behavior that never occurs?
Shaping
Shaping
rewarding successive approximations of the goal behavior until the goal behavior has been mastered
Example: want to signal dog to give her a beer by saying “beer me”
Start by dog touching rope attached to fridge handle, receives treat; Starts to get harder, dog must bite down on rope, receives treat; After successfully getting a new step, is NOT REWARDED for previous steps, only this one; Continues to do this until reaching the dog bringing a beer to the individual sitting in a chair and saying “beer me”
