test 3 content Flashcards
punishment vs negative reinforcement
punishment decreases instances of a behavior either by adding something or taking something away whereas negative reinforcement increases a behavior by taking something (typically aversive) away
influence of contingency on punishment
the punishment must depend on the behavior and the organism must understand this (ex: I am yelling at you because you stole my pencil)
influence of intensity on punishment
effective punishment is strong, especially the initial punishment; no bitch ass shocks that aren’t that bad
influence of delay on punishment
immediate punishment is more effective (probably because it helps to establish contingency) and if punishment is delayed, the effects will not last as long
influence of schedules on punishment
continuous schedules of reinforcement are more effective; can’t punish every other and expect the behavior to go away
influence of verbal explanation on punishment
punishment is more effective if you can explain to someone why they’re being punished
influence of experience with behavior on punishment
punishment is less effective if they’ve gotten away with it in the past; will probably try again
influence of alternative responses on punishment
punishment for bad behavior is more effective if there is another way for the subject to get to what he wants without engaging in the bad behavior
influence of motivation on punishment
organisms that are highly motivated to carry out behavior will be harder to effectively punish (ex: hungry rats don’t care as much if they’re shocked when they try to eat)
spanking
creates short term compliance to punish bad behavior but doesn’t promote good behavior, time outs have been shown to be equally as effective as spanking; not effective in the long term and can make children less compliant, more aggressive, and changes them to view situations as hostile when they are not
learned helplessness
a sense of powerlessness developed from a persistent failure to succeed; Seligman studied with dogs in which they were shocked until they learned they could not escape and then wouldn’t try to escape once possiblilty of shock was removed
alternatives to punishment
response prevention (create a situation in which the bad behavior can't occur in the first place) extinction (if attention based, ignore behavior to extinguish it) time out (a form of negative punishment in which you take person out of the situation that they're worked up about and don't pay attention) differential reinforcement (reward responses that are mutually exclusive with bad behavior- reward kid for sitting instead of punishing for getting up)
primary reinforcers
based on innate physical needs, rewards such as food/water/sex
secondary reinforcers
learned through association with primary reinforcement (ex: token economy)
Premack principle
responses capable of being reinforced must occur less often than the reinforcing activities; more probable behavior can be used to reinforce less probable behaviors (ex: finish vegetable [low prob] before you can eat dessert [high prob]
generalization
tendency for learned behavior to spread to new situations
discrimination
ability to distinguish between different situations; what you learn one place won’t always work for everyone else
semantic generalization
generalization based on meaning, pairing farm-themed words with a shock meant that people would freak out over the word “tractor” even though it wasn’t in the original list
discrimination training in classical conditioning
learn the association between bell and food, then have a learned lack of association between buzz and food(bell gets you food, buzz does not)
discrimination training in operant conditioning
reward for one behavior but not another; ex: reward pigeon for pecking a green light but not the yellow
errorless training procedure
it is necessary to combat frustration in learning so keep non-rewarded stimulus at a low perceptual level and progressively increase the level of intensity so the organism pretty much can’t pick the wrong choice
peak shift phenomenon
following discrimination training, peak response to stimulus will shift slightly away from the original stimulus as predicted by the Hull-Spence theory of algebraic summation (discrimination will go even further in the direction of the training)
