Cognitive [Learning, Language, Memory, Thinking] (Psychology Subject) Flashcards
Learning
*the relatively permanent or stable change in behavior as the result of experience
Classical conditioning (associative learning)
- Ivan Pavlov; Pavlovian conditioning
- pairing a neutral stimulus with a not-so-neutral stimulus; this creates a relationship between the two
Unconditioned Stimulus (UCS)
*not-so-neutral stimulus
- in Pavlov’s dog experiments, the UCS is the food
— without conditioning, the stimulus elicits the response of salivating
- unconditioned because they don’t have to be learned
— reflexive or instinctual behaviors
Unconditioned Response (UCR)
*naturally occurring response to the UCS
- in Pavlov’s dog experiment, it was salivation in response to the food
- unconditioned because they don’t have to be learned
— reflexive or instinctual behaviors
Neutral Stimulus (NS)
*a stimulus that doesn’t produce a specific response on its own
- In Pavlov’s dog experiment, this was the light/bell before he conditioned a response to it
Conditioned Stimulus (CS)
*the neutral stimulus once it’s been paired with the UCS
- has no naturally occurring response, but it’s conditioned through pairings with a UCS
- in Pavlov’s dog experiment, the CS (the light) is paired with the UCS (food), so that the CS alone will produce a response
Conditioned Response (CR)
*the response that the CS elicits after conditioning
- the UCR and the CR are the same (i.e., salivating to food or a light)
Simultaneous Conditioning
*the UCS and NS are presented at the same time
Higher-order conditioning/second-order conditioning
*a conditioning technique in which a previous CS now acts as a UCS
- in Pavlov’s dog experiment, the experimenter would use the light as a UCS after the light reliably elicited saliva in the dogs; food is no longer used; light could be paired with a bell (CS) until the bell alone elicited saliva in the dogs
Forward conditioning
*pairing of the NS and the UCS in which the NS is presented before thee UCS
- two types:
— delayed conditioning
— trace conditioning
Delayed conditioning
*the presentation of the NS begins before that of the UCS and lasts until the UCS is presented
Trace conditioning
*the NS is presented and terminated before the UCS is presented
Backward conditioning
*the NS is presented after the UCS is presented
- in Pavlov’s dog experiment, they would have been presented with the food and then with the light
- proven ineffective
- only accomplishes inhibitory conditioning (later the dogs would have a harder time pairing the light and food even if they were presented in a forward fashion)
Taste aversion learning
*occurs when food or drink becomes associated with an aversive stimulus (nausea, vomiting), even if the food or drink itself didn’t actually cause the nausea/vomiting
- type of classical conditioning but differs in:
— the response usually takes one pairing vs. longer acquisition
— the response takes a long time to extinguish vs. beginning when you remove the UCS
- evolutionarily adaptive so human/animal doesn’t eat poisonous food and die
Law of effect
*a cause-and-effect chain of behavior revolving around reinforcement
- E. L. Thorndike
- precursor of operant conditioning
- “connectionism” because learning occurs through formation of connections between stimuli and responses
- “Puzzle Box” experiment (cats learning complex tasks through trial and error)
Theory of association
*grouping things together based on the fact that they occur together in time and space
— organisms associate certain behaviors with certain rewards and certain cues with certain situations
- Kurt Lewin
- forerunner of behaviorism
School of behaviorism
- John B. Watson
- everything could be explained by stimulus-response chains
— conditioning was key factor in developing these chains - only objective and observable elements were of importance to organisms and psychology
Hypothetico-deductive model
*designed to dry and deduce logically all the rules that govern behavior
- Clark Hull
— created equation involving input variables leading to output variables; included intervening variables in between that’d change the outcomes
Radical behaviorism
*school of thought where it’s believed that all behavior, animal, and human, can be explained in terms of stimuli and responses, or reinforcements and punishments
- no allowances for how thoughts/feeling might factor into the equation
Operant conditioning (associative learning)
*aims to influence a response through various reinforcement strategies
- B. F. Skinner
- Skinner Box (rats repeated behaviors that won rewards and gave up on behaviors that didn’t)
— shaping (differential reinforcement of successive approximations) process rewarded rats with food pellets when near the lever and after touching lever
- also known as instrumental conditioning
Primary reinforcement
*a natural reinforcement; reinforcing on its own without the requirement of learning
- i.e., food
Secondary reinforcement
*a learned reinforcer
- i.e., money, prestige, verbal praise, awards
- often learned through society
- instrumental in token economies
Positive reinforcement
*adding something desirable to increase likelihood of a particular response
- some subjects are not motivated by rewards because they don’t believe/understand that the rewards will be given
Negative reinforcement
*reinforcement through the removal of a negative event
- i.e., taking away something undesirable to increase the likelihood of a particular behavior
- NOT punishment/delivery of a negative consequence
Continuous reinforcement schedule
*every correct response is met with some form of reinforcement
- facilitates the quickest learning, but also the most fragile learning; as soon as rewards halt, the animal stops performing
Partial reinforcement schedules
*not all correct responses are reinforced
- may require longer learning time, but once learned, behaviors are more resistant to extinction
- types:
— fixed ratio schedule
— variable ratio schedule
— fixed interval schedule
— variable interval schedule
Partial reinforcement schedules
*not all correct responses are reinforced
- may require longer learning time, but once learned, behaviors are more resistant to extinction
- types:
— fixed ratio schedule
— variable ratio schedule
— fixed interval schedule
— variable interval schedule
Fixed ratio schedule
*reinforcement delivered after a consistent # of responses
- ratio of 6:1 –> every 6 correct responses = 1 reward
Variable ratio schedule
*reinforcements delivered after different #s of correct responses
- ratio can’t be predicted
- learning less likely to become extinguished
- one performs a behavior not because it’s been rewarded but rather cause it COULD be rewarded on the next try
Fixed interval schedule
*rewards come after the passage of a certain period of time rather than the # of behaviors
- i.e., if fixed interval is 5 minutes, then rat will be rewarded the first time it presses the lever after a 5-minute period has elapsed, regardless of what it did during the preceding 5 minutes
Variable interval schedule
*rewards are delivered after differing time periods
- second most effective strategy in maintaining behavior; length of time varies, so one never knows when the reinforcement is just around the corner
- slow and steady learning
Token economy
*artificial mini-economy usually found in prisons, rehabilitation centers, or mental hospitals
- individuals in the environment are motivated by secondary reinforcers (tokens)
- desirable behaviors reinforced with tokens, which can be cashed in for more desirable reinforcers (i.e., candy, books, privileges, cigarettes)
Stimulus
*any event that an organism reacts to
- first link in a stimulus-response chain
Stimulus discrimination
*the ability to discriminate between different but similar stimuli
- i.e., doorbell ringing vs. phone ringing
Stimulus generalization
*make the same response to a group of similar stimuli; opposite of stimulus discrimination
- i.e., not all fire alarms sound alike, but they all require the same response
Stimulus generalization
*make the same response to a group of similar stimuli; opposite of stimulus discrimination
- i.e., not all fire alarms sound alike, but they all require the same response
Undergeneralization
*failure to generalize a stimulus
Response learning
*form of learning in which one links together chains of stimuli and responses
- one learns what to do in response to particular triggers
- i.e., leaving a building in response to a fire alarm