Learning and Behavior
The Learning Triangle
S->S*-R (Pavlovian Learning, occasion setting, operant conditiong, habit learning
Learning and Adaptation
The main way animals can adapt through experience
Fixed action patterns
Behaviors that are triggered by stimuli known as releasers or “sign stimuli”
Do not depend on feedback once triggered
Smiling is a fixed action pattern?
Innate Behaviors
Very hard to prove... subtle learning can be going on with 'innate behaviors' that make them not so innate after all
artificial selection
Select for animals that show specific patterns... does it change them?
Innate behaviors can be modified by learning, or triggered by things which by themselves are not releasers
Habituation- strength of a response declines due to repeated exposure.
Learning has clear benefit in long term adaptation, but innate learning is important for stimuli in which the first time could be deadly
The Law of Effect
|
| Good S* | Bad S* |
| Produces S* | Reward (Behavior up) | Punishment (behavior Down) |
| Prevents S* | Omission (behavior up) | Avoidance(bad s* doesn't happen) Escape(get away from already happening bad s*) (behavior up) |
Reinforcement
those behaviors which make the likeliness of a behavior increase.
Shaping
Reinforce increasingly close approximations of behavior
Can occur haphazardly in environment
Adaptation in classical Conditioning
Signals for food
Pavlovs dog
Prepares animal for digestions
Taste aversion learning
learn things that signal bad foods
Conditioned with signals for danger
Freezing
Adaptive response, that conditions from innate stimuli to others
Analgesia... don't notice being hurt
Drugs as S*
associating the S* drug with the S cues from the environment
Conditioned compensatory response... resilience against the drug
May be the cause for drug tolerannce
Impressive amount of evidence for role of conditioning (Siegel, 1975), (Crowell, Hinson, & Siegel 1981), (Mansfield and Cunningham, 1980)
(Paulos, Wilkinson, and Cappell, 1981)
(King, Boutoun, & Musty, 1987)
Sign tracking
|
| Good S* | Bad S* |
| S predicts S* | Approach S | Withdraw from S |
| S predicts no S* | Withdraw from S | Approach S |
Other parallels between signal and repsonse learning
Believed that both follow general patterns or rules
Extinction occurs in both instrumental and classical conditioning
Crucial in shaping and creating behaviors
acquisition... opposite of extinction, describes both instrumental and classical
Exposure therapy
expose the client to a fear stimulus without an aversive consequence
Spontaneous recovery
extincion does not get rid of learning, merely makes it less likely response
What an animal does does not show what it knows
The timing of S*
Occurs after only a short delay from either R or S
Learning is designed to uncover “causes” of S*
Size of S*
Large positive S*s lead to stronger overall behavior
The stronger an aversive S* is, the stronger the punishment
Preparedness
the same combination of events are learend more readily than others
Animals behave as if evolution has “prepared them to associate certain events or stimuli(Rozin and kalat, 1971), (Breland and Brleand, 1961)
The nuts and bolts of conditioning
Basic Pavlovian Learning
CS ->CR
US->UR
UR=CR
Higher order conditoing- same procedure, but old US is used as new CS
sensory preconditioning
two US are paried, then one is associated with S*
Methods for studying classical conditiong
Eyeblink conditioning in rabits (gormezano, kehoe, and marshall, 1983)
easy to measure, typically considered innate
Fear conditioning in rats
condtioned suprresion/conditioned emotional response(CER)
Suppresion ratio... how much an animal stops responding due to fear
Autoshaping in pigeons
pigeons peck keylights which signal food
Taste aversion learning in rats
Things that affect the strength of conditioning
Time
Delay conditiong.
cs comes then ends with presentation of of US
Trace conditionining
CS and US seperated by a gap
Simultaneous conditioning
Presented at the same time... hard to measure hwo much subject learns
Backwards conditioning
CS follows, rather than preceds, the S
Can act as a conditioned inhibitor
Spaced trials are better than massed trial
Time in CS and time between trials equal
Novelty of CS and US
The more novel, the more learning that occurs
Preexposure effect
Latent inhibition
Intensity of CS and US
The more salient the more effect
if they are too strong, they can elicit their own responses
Pseudoconditiong
sensitization
Conditioned inhibition
CS associated with Absence of US = inhibitor
How to test
Summation
amount of responding is decreased or increased
Retardation-of-Aqcuisition
inhibitor cannot be converted into an excitor easily
How to produce
Discriminative inhibtion
One CS is paired with US, one paired with no US, second one is inhibitor
conditioned inhibition
A is paired with B, AX is paired with no B... x signals no B
Information value in conditioning
CS-US contingencies in classical conditioning
Pairing alone is not enough to induce conditioning
CS has to predict an increase in the probability of the US (positive contigency)
If it predicts a decrease, it becomes an inhibitor(negative contingency)
Blocking and Unblocking
learning only occurs when CS provides new information about US... redundant information is not learned
Relative Vailidity
Subjects learn about the best predictiors of the US
Theories of Conditiong
The Rescorla-Wagner Model
Suprisingness... learning occurs on a conditioning trail only if the US is suprising
Overexpectation effect
The role of attention in conditioning
the amount of attention a subject will pay to a cs depends on how well CS predicts US
Mackintosh model
More attention given to better predictors
Pearce hall model
More attention given to CS who's meanings are not understood
Short term memory
Priming of a stimuli reduces surprise
Habituation
Caused by background context (long term memory)
Caused by self generated priming (short term memory)
SOP model
A1 A2, inactive
AESOP model
Sensory vs. emtoional US nodes
Emotional nodes move slower between A1-A2-inactive
Elemental vs. configural CS nodes
Configural elemnts replace old elemnts
Whatever happened to behavior Anyway
Memory and learning
How well is conditioning remembered?
Conditioning can be retained supringly well over time
Some forms of conditiong are forgotten more quickly than others
Memory reactivition... reexposure to original situation
Causes of Forgetting
Trace decay
interference
retrieval failure
Remembering, forgetting, and extinction
interference and retrieval may play an important role
renewal ffect
shows extintion does not necessarily mean old memory is gone
The modulation of behavior
occasion setting
a cue that provides information about whether another cs will be paired with a US
Three properties of occasion setters
They modulate the behavior that is otherwise evoked by the target
they are not affected much by changing their direct associations with the US
they do not always affect responding to ew stimuli
Behavior systems
sets of behaviors that are organized around biological functions and goals
Hierarchically organized
different “modes” of responding
affecting panic attacks
Preparatory vs. panic (pre-encounter vs. circa strike)
Taste aversion learning
The “hedonic shift”
good foods elicit different behaviors than bad foods
Verbal instructions can have a powerful effect on learning
Schedules of reinforcement... to maximize different things
The matching law
The quantitative law of effect
The more positive choices you have, the less likely you'll make a negative choice
Impulsiveness and self control
Pre-commitement strategies
Distraction
behavioral economics
substitutes-increase to decrease other behavior
independents
complements-decrerase to decrease other behavior
Theories of reinforcement
The premack principle
If a consequent behavior is more preffered than it's antecedent, it will be a reinforcer
Behavioral regulation
If a behavior is below it's preferred level, it will be a reinforcer.
State dependant learning
only take food if you know that food fills hungry void
learning to calibrate based on need
Learning to like certain things
learning to dislike certain things?
Anticpating reward and punnishment
negative and positive conrast effect
Initial explaratory effect
Frustration reaction
Overjustification effect
happens if:
Tangible
announced ahead of time
not dependant on performance (expectations and interpersonal context)... self determination theory
Partial reinforcement extinction effect
robert eisenberger-learned industriousness
sequential theory
transfer of control (inhibtory CS's to the state change the actual response, as to excitatory)
Using Anchors... how to create inhibitory responses?
Consumatory vs. prepatory responsee (go vs. know, motivational)
Occasion setting
Opponent process theory
emotions happen, then afterwords have the opposite reaction
Is it learned?
Self generated vs. retrieval generated priming
Massed trials create temporary wanting, spaced trials create long term wanting
Drug users may be wannting something before they need it
There is a difference between wanting and liking
No comments:
Post a Comment