Download 7 - Wofford

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Behavioral modernity wikipedia , lookup

Motivation wikipedia , lookup

Abnormal psychology wikipedia , lookup

Symbolic behavior wikipedia , lookup

Observational methods in psychology wikipedia , lookup

Learning theory (education) wikipedia , lookup

Psychophysics wikipedia , lookup

Neuroeconomics wikipedia , lookup

Thin-slicing wikipedia , lookup

Transtheoretical model wikipedia , lookup

Attribution (psychology) wikipedia , lookup

Theory of planned behavior wikipedia , lookup

Sociobiology wikipedia , lookup

Theory of reasoned action wikipedia , lookup

Parent management training wikipedia , lookup

Applied behavior analysis wikipedia , lookup

Verbal Behavior wikipedia , lookup

Descriptive psychology wikipedia , lookup

Classical conditioning wikipedia , lookup

Psychological behaviorism wikipedia , lookup

Social cognitive theory wikipedia , lookup

Insufficient justification wikipedia , lookup

Behavior analysis of child development wikipedia , lookup

Behaviorism wikipedia , lookup

Operant conditioning wikipedia , lookup

Transcript
Operant Conditioning

Operant conditioning





Association between
VOLUNTARY behavior and
its consequence
Behavior has consequences
ABC’s: Antecedents-BehaviorConsequence
Thorndike’s Law of effect
“Sit, Houston!”

Make millions training dogs!

Dog tricks
Influencing behavior: Not just for dogs




B.F. Skinner
“The major problems of the world today can be
solved only if we improve our understanding of
human behavior” About Behaviorism (1974)
Author of “Beyond Freedom and Dignity”
Study rats and pigeons in a “Skinner box”
Classical vs. operant conditioning
Rat in a Skinner Box
http://www.youtube.com/watch?v=PQtDTdDr8vs
Effect of Stimulus (Consequence) on Behavior:
Positive:
If Response
Then Stimulus
Negative:
If Response
Then NO Stimulus
Type of Contingency
Reinforcement:
Increase Probability
Positive Reinforcement
Stimulus: usually
appetitive
Punishment:
Decrease Probability
Punishment
“Punishment by Addition”
“Positive Punishment”
Ex: “Thank you!”
“rewards”
Stimulus: usually
aversive
Ex: Spanking,
Blessing Out
Negative Reinforcement
Omission
Stimulus: usually
aversive
Ex: Avoidance,
Escape
“Punishment by Subtraction”
“Negative Punishment”
Stimulus: usually
appetitive
Ex: Grounding, DRO,
suspended license
Identify the contingency for each:
Car buzzer stops after you put on your seat
belt.
 If I wear that, my friends will think I look like a
slut.
 Given a ticket by a cop for speeding.
 I press cancel on the microwave so I won't
hear it beep.
 If you buy me that diamond, I'll make you very
happy.
 I won't eat dessert because I don't want to
gain weight.
 If you make 3.5 or higher, you can have a trip
for spring break.

Shaping behaviors




How do you train a pigeon to
turn in a circle?
Shaping = differential
reinforcement of successive
approximations
After each response ->
stricter criterion for
reinforcement
http://www.youtube.com/watch?v=I_ct
JqjlrHA&feature=related
Applications?




ADHD
Autism
Sports, music, skills, parenting
http://www.youtube.com/watch?v=reNVH1zd1to
&feature=related
Rat Basketball
http://www.wofford.edu/psychology/content.aspx?id=4844
Cumulative recorder
Schedules of reinforcement







Continuous (CRF) vs.
Partial Reinf. Schedules:
Fixed-ratio
Fixed-interval
Variable-ratio
Variable-interval
Effect of extinction?
PREE!!!!
reinforcement
reinforcement
reinforcement
http://www.youtube.com/watch?v=I_ctJqjlrHA&feature=related
Spoiled brat
Schedules of reinforcement
Sample cumulative records of 4 simple
reinforcement schedules
Identify the reinforcement schedule:
In a ____ schedule, reinforcement is
delivered only after some specific number
of responses.
 In a ____ schedule, reinforcement is
delivered only after responding after
some specific period of time, not before.
 In a ____ schedule, reinforcement is
delivered only after a varied number of
responses.
 In a ____ schedule, reinforcement is
delivered only after responding after a
varying period of time.

Extinction: Partial v Continuous
The PREE!
(Blue – Red)
Partial
reinforcement
Continuous
reinforcement
Operant conditioning applications

Animal training



Quiet the class; Produce good grades
Superstitious behavior


http://www.cesarmillaninc.com/videoplayer/index.php
Teaching techniques


“Don’t shoot the dog: The new art of teaching
and training” K. Pryor
Behavior and consequence linked (variable-ratio)
Understand disorder and create therapy


Depression; Anorexia; Drug abuse
ADD/ADHD and Autism
Mechanism of Depression: Seligman, 1975
“Learned helplessness”
then
Training



Escape shock by pressing within
10s after light
 Red: can avoid shock
 Blue: can’t change outcome
Move to shuttle box
 Red: learns task
 Blue: becomes passive
Abused wives stay?
Parenting: You do “behavior modification”
every day -- should you know what works?
Positive punishment vs. negative
punishment
 Effect of types of punishments?
 Limitations of positive punishment




Doesn’t teach correct response
May increase aggression
Other forms of punishment



Time-out
Penalty or fine
Correction
Modeling behavior






In operant conditioning: organism learns the
consequences of its own behavior.
Observational learning: organism learns the
consequences of another’s behavior.
Sports: You are much more likely to perform well
if you see how “experts” do it.
Arts: learning to paint well, dance well, sing well,
nearly all benefit from modeling.
Parenting? Did the manual come with the baby?
Bandura et al. (1963) – Learning aggression:
Bobo doll -- Observe how behavior of others is
reinforced or punished
Modeling behavior

Observational learning


Rather than only trial-error
Influenced by presence of significant role
model

Imitate, model, behavior of significant others
Modeling behavior

Observational learning

Bandura et al.(1963) – Bobo doll



Learn how behavior of others is reinforced or
punished?
Learn what is fun from
others
Learn how to do things
Think about your own life:


What is a behavior you want to increase?
What is a behavior you want to decrease?

For each, explain how each of these would
influence your target behaviors:
Positive reinforcement
Negative reinforcement
Punishment by addition
Punishment by subtraction

What schedules of reinforcement would you use?




Extra slides
To review: Operant conditioning



What is operant conditioning?
What are real examples of consequences of
behavior (pos/neg reinforcement/punishment)?
What are the schedules of reinforcement?
What is the effect of each on learning and
extinction?
To review: Classical conditioning

What is classical conditioning?



Real examples?



Association between stimuli and NATURAL response
Stimuli signal outcomes
What is US and UR
What is CS and CR
What variables can you vary?






Number of reinforced trials
Number of unreinforced trials
Order of CS and US
Stimulus generalization: test different versions of the CS
Second order conditioning – test new CS to cue first CS
Test after extinction – present stimulus after long break
Thought paper
What is classical conditioning versus
operant conditioning?
 What is an example of classical
conditioning in your every day life
 What is an example of operant
conditioning in your every day life?
 How can you use learning theory in the
future?

Extra slides
Radial-arm maze
Place learning
Olton & Samuelson (1976)








Rats explore maze: 15min/day for 9 days
Free choice: one piece of food in each arm (10
days) – could make 16 choices
Within 5 days all animals were consistent in
choosing an average of 7 arms within the first 8
choices – didn’t repeat choices!
Odor: odor added so can’t leave “trail” (3 days)
Added odor didn’t influence accuracy
Switch: after 3 choices, switched location of two
arms
Rats continued to choose arm based on spatial
location (repeating the arm from first 3 trials)
Use cues outside the maze in the room!
Corwill & Rescorla, 1985

Method:




Results:




Lever press = food pellet
Chain pull = sugar water
Later add toxin to sugar water
Taste aversion to sugar water
Continued to press lever, not pull chain
Avoidance conditioning (vs. escape conditioning)
Conclusion:


“Act-outcome representation”
Use method to see what tastes animals can
differentiate