Download Learning - WW Norton & Company

Sarah Grison • Todd Heatherton • Michael Gazzaniga
Psychology in
Your Life
FIRST EDITION
Chapter 6
Learning
© 2014 W. W. Norton & Company, Inc.
Section 6.1
How Do the Parts of Our Brains
Function?
6.1 What Are the
Three Ways We Learn?
• Learning: A change in behavior, resulting
from experience
– Central to almost all areas of human
existence
We Learn From Experience
• Behaviorism: a formal learning theory from
the early twentieth century
– John Watson: focused on environment and
associated effects as key determinants of
learning
– B. F. Skinner: designed animal experiments to
discover basic rules of learning
We Learn From Experience
• Critical for survival
• Adapt behaviors for a particular
environment
– Which sounds indicate potential danger?
– What foods are dangerous?
– When is it safe to sleep?
We Learn in Three Ways
1. Non-associative learning
2. Associative learning
3. By watching others
We Learn in Three Ways
• Non-associative learning
– Information about one external stimulus (e.g.,
a sight, smell, sound)
– Habituation: A decrease in behavioral
response after lengthy or repeated exposure
to a stimulus
– Especially if the stimulus is neither harmful
nor rewarding
• See Figure 6.2a next slide
We Learn in Three Ways
• Non-associative learning: information
about one external stimulus (e.g., a sight,
smell, sound)
We Learn in Three Ways
We Learn in Three Ways
• Non-associative learning
– Sensitization: An increase in behavioral
response after lengthy or repeated exposure
to a stimulus
– Heightened preparation in a situation with
potential harm or reward
• See Figure 6.2b next slide
We Learn in Three Ways
We Learn in Three Ways
• Associative learning
– Understanding how two or more pieces of
information are related
We Learn in Three Ways
• Associative learning
• Classical conditioning: learn that two
stimuli go together
– Example: music from scary movies elicits
anxiousness when heard
• Operant conditioning: learn that a behavior
leads to a particular outcome
– Example: studying leads to better grades
We Learn in Three Ways
• Learning by watching others
– Observational learning
– Modeling
– Vicarious conditioning
The Brain Changes During
Learning
• Long-term potentiation (LTP)
– The strengthening of synaptic connections
between neurons
– Recall that “cells that fire together, wire
together”
– Exposure to environmental events causes
changes in the brain to allow learning
Section 6.2
How Do We Learn by Classical
Conditioning?
6.2 How Do We Learn by
Classical Conditioning?
• Familiar example: association between
scary music in movies and bad things
happening to characters
Through Classical Conditioning,
We Learn That Stimuli Are Related
• Pavlov: Nobel Prize in 1904 for research
on the digestive system
• Observed dogs began to salivate as soon
as they saw bowls of food
– Salivating at the sight of a bowl is not
automatic
– Behavior acquired through learning by
association
• See Figure 6.3b next slide
Through Classical Conditioning,
We Learn Stimuli Are Related
Through Classical Conditioning,
We Learn That Stimuli Are Related
• Classical conditioning
– A type of learned response in which a neutral
object comes to elicit a response when it is
associated with a stimulus that already
produces a response
Through Classical Conditioning,
We Learn That Stimuli Are Related
• Pavlov’s Experiments Reveal the
Four Steps in Classical Conditioning :
1. Present unconditioned stimulus: evokes
unlearned response
2. Present neutral stimulus: no response
3. Pair stimuli from Steps 1 and 2: learned
response (conditioning trials)
4. Neutral stimulus alone will trigger learned
response (critical trials)
Through Classical Conditioning,
We Learn That Stimuli Are Related
• Pavlov’s Experiments Reveal the
Four Steps in Classical Conditioning
– Step 1: presenting food causes salivary reflex
– Unconditioned stimulus (US): A stimulus
that elicits a response that is innate and does
not require any prior learning (Food)
– Unconditioned response (UR): A response
that does not have to be learned, such as a
reflex (Salivation)
Through Classical Conditioning,
We Learn That Stimuli Are Related
• Step 2: clicking metronome is neutral
stimulus
– Neutral stimulus: anything seen or heard;
must not associate with the unconditioned
response
Through Classical Conditioning,
We Learn That Stimuli Are Related
• Step 3 (conditioning trials): start of
learning
– Dog begins to associate US (food) and
neutral stimulus (metronome)
Through Classical Conditioning,
We Learn That Stimuli Are Related
• Step 4 (critical trials): Association learned
– Metronome alone, without food, makes dog
salivate
• See Figure 6.3 next slide
Pavlov’s Experiments Reveal the
Four Steps in Classical Conditioning
Through Classical Conditioning,
We Learn That Stimuli Are Related
• Conditioned stimulus (CS): A stimulus
that elicits a response only after learning
has taken place
• Conditioned response (CR): A response
to a conditioned stimulus; a response that
has been learned
• See Scientific Thinking: Pavlov’s
Experiments Reveal Learning by Classical
Conditioning next slide
Learning Varies in
Classical Conditioning
• Animals adapt via conditioning
– Learning to predict outcomes leads to new
adaptive behaviors
Learning Varies in
Classical Conditioning
• Acquisition
– Acquisition: The gradual formation of an
association between conditioned and
unconditioned stimuli.
– Strongest conditioning occurs when CS is
presented slightly before US
• See Figure 6.5a next slide
Acquisition, Extinction, and
Spontaneous Recovery
Learning Varies in
Classical Conditioning
• Extinction
– Extinction: A process in which the
conditioned response is weakened when the
conditioned stimulus is repeated without the
unconditioned stimulus
• See Figure 6.5b next slide
Acquisition, Extinction, and
Spontaneous Recovery
Learning Varies in
Classical Conditioning
• Spontaneous recovery
– Spontaneous recovery: A process in which
a previously extinguished response
reemerges after the conditioned stimulus is
presented again
– Can occur after only one pairing following
extinction
– Response will weaken if CS-US pairings do
not continue
• See Figures 6.5c, 6.5d next slide
Acquisition, Extinction, and
Spontaneous Recovery
Learning Varies in
Classical Conditioning
• Generalization, discrimination, and
second-order conditioning
– Stimulus generalization: Learning that
occurs when stimuli that are similar but not
identical to the conditioned stimulus produce
the conditioned response
– Animals respond to variations in CS
Learning Varies in
Classical Conditioning
• Generalization, discrimination, and
second-order conditioning
– Stimulus discrimination: A differentiation
between two similar stimuli when only one of
them is consistently associated with the
unconditioned stimulus
• See Figure 6.6 next slide
Generalization, Discrimination, and
Second-Order Conditioning
Learning Varies in
Classical Conditioning
• Generalization, Discrimination, and
Second-Order Conditioning
– Second-order conditioning: second CS
becomes associated with first CS; elicits CR
when presented alone
– Neither US nor original CS present
– Example: pairing black square (second CS)
with metronome (first CS) so black square
produces salivation (CR) on its own
We Learn Fear Responses
Through Classical Conditioning
• Phobia
– Acquired fear that is very strong in
comparison to threat
We Learn Fear Responses
Through Classical Conditioning
• The Case of Little Albert
• Classical conditioning demonstrated in
phobias:
– Watson showed “Little Albert” various neutral
objects (e.g., white rat, rabbit, dog, monkey,
white wool)
– Paired rat (CS) and loud clanging (US) until
rat alone produced fear (CR)
– Fear generalized to all similar stimuli
• See Figure 6.7 next slide
The Case of Little Albert
We Learn Fear Responses
Through Classical Conditioning
• Counterconditioning
– Counterconditioning: exposing subject to
phobia during an enjoyable task
– Systematic desensitization: exposure to
feared stimulus while relaxing
– CS -> CR1 (fear) connection replaced with CS
-> CR2 (relaxation) connection
Adaptation and Cognition Influence
Classical Conditioning
• Pavlov’s belief: Any two events presented
together would produce learned
association
• By 1960s, data suggested that some
conditioned stimuli more likely to produce
learning
Adaptation and Cognition Influence
Classical Conditioning
• Evolutionary Influences
– Certain pairings more likely to be associated
– Conditioned taste aversions: easy to produce
with smell or taste cues
– Auditory and visual stimuli: value for signaling
danger
• See figure 6.8 next slide
Adaptation and Cognition Influence
Classical Conditioning
Adaptation and Cognition Influence
Classical Conditioning
• Cognitive Influences
– Through classical conditioning, animals
predict events
– Easier when CS before US rather than after
US
– Easier when CS is more unexpected or
surprising
• See figure 6.9 next slide
Section 6.3
How Do We Learn by Operant
Conditioning?
6.3 How Do We Learn by
Operant Conditioning?
• Operant conditioning
– A learning process in which the
consequences of an action determine the
likelihood that the action will be performed in
the future
Animals Learn through the
Outcomes
of Their Actions
• Thorndike’s Experiments Reveal the
Effects of Action
– Thorndike’s puzzle box: challenged fooddeprived animals to find escape
– Trap door would open if animal performed
specific action
– Animal quickly learned to repeat behavior to
free itself and reach the food
• See Figure 6.10 next slide
Thorndike’s Experiments Reveal the
Effects of Action
Animals Learn through the
Outcomes
of Their Actions
• Thorndike’s general theory of learning
– Law of effect: any behavior leading to a
“satisfying state of affairs” likely to be
repeated
– Any behavior leading to an “annoying state of
affairs” less likely to reoccur
Learning Varies in
Operant Conditioning
• B. F. Skinner’s learning theory based on
the law of effect
– Animals operate on environments to produce
effects
– Reinforcer: stimulus occurs after response
and increases likelihood of response
reoccurring
– Consequences determine likelihood of
behavior in future
• See figure 6.11 next slide
Learning Varies in
Operant Conditioning
• Shaping
– Shaping: operant-conditioning technique;
reinforce behaviors increasingly similar to
desired behavior
• See Figure 6.12 next slide
Shaping
Learning Varies in
Operant Conditioning
• Reinforcers can be conditioned
– Reinforcers that are necessary for survival,
such as food or water, are called primary
reinforcers
– Events or objects that serve as reinforcers but
do not satisfy biological needs are called
secondary reinforcers
Learning Varies in
Operant Conditioning
• Reinforcer potency
– Premack principle: more valued activity can
reinforce performance of less valued activity
– Example: “Eat your spinach and then you’ll
get dessert”
Reinforcement and Punishment
Influence Operant Conditioning
• Reinforcement and punishment have
opposite effects on behavior
– Reinforcement: behavior more likely to be
repeated
– Punishment: behavior less likely to occur
again
Reinforcement and Punishment
Influence Operant Conditioning
• Positive and negative reinforcement
– Both positive and negative reinforcement
increase likelihood of a given behavior
Reinforcement and Punishment
Influence Operant Conditioning
• Positive and negative reinforcement
– Positive reinforcement: The addition of a
stimulus to increase the probability that a
behavior will be repeated
– Example: feeding a rat after it has pressed a
lever
– Negative reinforcement: The removal of a
stimulus to increase the probability that a
behavior will be repeated
– Example: taking a pill to get rid of a headache
Reinforcement and Punishment
Influence Operant Conditioning
• Positive and negative punishment
– Both positive and negative punishment
reduce likelihood that behavior will be
repeated
Reinforcement and Punishment
Influence Operant Conditioning
• Positive and Negative Reinforcement
– Positive punishment: The addition of a
stimulus to decrease the probability that a
behavior will recur
– Example: electrical shock, speeding ticket
– Negative punishment: The removal of a
stimulus to decrease the probability that a
behavior will recur
– Example: loss of food, loss of privileges
• See figure 6.14 next slide
Reinforcement and Punishment
Influence Operant Conditioning
• Schedules of partial reinforcement
– Continuous reinforcement: behavior
reinforced each time it occurs
– Fast learning, uncommon in real world
– Partial reinforcement: behavior is occasionally
reinforced
– More common in real world
Reinforcement and Punishment
Influence Operant Conditioning
•
Schedules of partial reinforcement
– How reinforcement given by how
consistently given = four common schedules
1. Fixed schedule: predictable basis
2. Variable schedule: unpredictable basis
3. Interval schedule: based on passage of time
4. Ratio schedule: based on number of
responses
Reinforcement and Punishment
Influence Operant Conditioning
• Schedules of partial reinforcement
– Fixed interval schedule (FI): Reinforcing the
occurrence of a particular behavior after a
predetermined amount of time since the last
reward
– Example: paycheck
Reinforcement and Punishment
Influence Operant Conditioning
• Schedules of partial reinforcement
– Variable interval schedule (VI): Reinforcing
the occurrence of a particular behavior after
an unpredictable and varying amount of time
since the last reward
– Example: pop quiz
– More consistent response rates than fixed
interval
Reinforcement and Punishment
Influence Operant Conditioning
• Schedules of partial reinforcement
– Fixed ratio schedule (FR): Reinforcing a
particular behavior after that behavior has
occurred a predetermined number of times
– Example: paid by the completed task
– Often yields better response rates than fixed
interval
Reinforcement and Punishment
Influence Operant Conditioning
• Schedules of partial reinforcement
– Variable ratio schedule (VR): Reinforcing a
particular behavior after the behavior has
occurred an unpredictable and varying
number of times
– Example: slot machine
• See Figure 6.15 next slide
Schedules of Partial Reinforcement
Reinforcement and Punishment
Influence Operant Conditioning
• Schedules of partial reinforcement
– Partial-reinforcement extinction effect:
behavior lasts longer under partial
reinforcement than under continuous
reinforcement
• To condition behavior to persist:
– Use continuous reinforcement initially
– Slowly change to partial reinforcement
Operant Conditioning Affects Our
Lives
• Parental Punishment Is Ineffective
• To be effective, punishment must be
– Reasonable
– Unpleasant
– Applied immediately
– Clearly connected to the unwanted behavior
Operant Conditioning Affects Our
Lives
• Parental Punishment Is Ineffective
• Punishment can cause confusion
– Wrongly applied after desirable behavior
– Leads to negative emotions (e.g., fear,
anxiety)
– Fails to offset reinforcing aspects of the
undesired behavior
• Reinforcement teaches desirable behavior
• See Figure 6.16 next slide
Parental Punishment Is Ineffective
Operant Conditioning Affects Our
Lives
• Behavior modification
– Behavior modification: operant conditioning
replaces unwanted behaviors with desirable
behaviors
– Token economies: opportunity to earn tokens
(secondary reinforcers) for completing tasks
and lose tokens for behaving badly
– Tokens later traded for objects or privileges
– Gives participants sense of control
Biology and Cognition Influence
Operant Conditioning
• Behaviorists believed conditioning
principles explained all behavior. In reality,
reinforcement explains only a certain
amount of human behavior
Biology and Cognition Influence
Operant Conditioning
• Dopamine activity affects reinforcement
– Dopamine has a biological influence on
reinforcing value
– Drugs that block dopamine’s effects disrupt
operant conditioning
– Drugs that enhance dopamine activation
increase reinforcing value of stimuli
Biology and Cognition Influence
Operant Conditioning
• Biology constrains reinforcement
– Some animal behaviors hardwired
– Difficult to learn behaviors counter to
evolutionary adaptation
– Conditioning most effective when matched to
animal’s biological predispositions
Biology and Cognition Influence
Operant Conditioning
• Learning without reinforcement
– Tolman argued that reinforcement impacts
performance more than knowledge acquisition
– Ran rats through complex mazes to obtain
food
– Cognitive map: A visuospatial mental
representation of an environment
Biology and Cognition Influence
Operant Conditioning
• Learning without reinforcement
• In Tolman’s study three groups of rats
traveled maze
– Group 1: No reinforcement
– Group 2: Reinforcement every trial
– Group 3: Reinforcement only after first
10 trials
• See Figure 6.19 next slide
Learning Without Reinforcement
Biology and Cognition Influence
Operant Conditioning
• Learning without reinforcement
– Latent learning: Learning that takes place in
the absence of reinforcement
– Group 1: Slow, many wrong turns
– Group 2: Faster, fewer errors each trial
– Group 3: Before reinforcement, similar to
Group 1. After reinforcement, better than
Group 2
Biology and Cognition Influence
Operant Conditioning
• Learning without reinforcement
– Insight learning: A sudden understanding of
how to solve a problem after a period of either
inaction or thinking about the problem
Section 6.4
How Do We Learn by Watching
Others?
6.4 How Do We Learn by
Watching Others?
• Behaviors we learn by watching others
– Mechanical skills, social etiquette, situational
anxiety, attitudes about politics and religion
• Three ways we learn by watching
1. Observational learning
2. Modeling
3. Vicarious conditioning
Three Ways We Learn
through Watching
1. Observational learning: The acquisition
or modification of a behavior after
exposure to at least one performance of
that behavior
– Examples: foods safe to eat, objects and
situations to fear
– Powerful adaptive tool
• See Figure 6.20 next slide
Three Ways We Learn
through Watching
Three Ways We Learn
through Watching
• Bandura’s research reveals learning
through observation
– Observation of aggression: Bandura’s Bobo
doll study
– Group 1: watched film of adult playing quietly
with Bobo, an inflatable doll
– Group 2: watched film of adult attacking Bobo
– Viewers of aggression were more than twice
as likely to play aggressively
Three Ways We Learn
through Watching
• Learning through modeling
• Modeling: The imitation of behavior
through observational learning
– More likely to imitate actions of attractive,
high-status models similar to ourselves
• See Figure 6.22 next slide
Learning Through Modeling
Three Ways We Learn
through Watching
• Learning through vicarious conditioning
• 3. Vicarious conditioning: Learning the
consequences of an action by watching
others being rewarded or punished for
performing the action
– Rewarded behavior more imitated
– Punished behavior less imitated
• See Figure 6.23 next slide
Learning Through
Vicarious Conditioning
Watching Others Results in
Cultural Transmission
• Meme: shared piece of cultural knowledge
– Similar to genes, selectively passed across
generations, can spread much faster
– Animals also show this kind of knowledge
sharing
• See Figure 6.24 next slide
Watching Others Results in
Cultural Transmission
Biology Influences
Observational Learning
• Mirror neurons
– Fire in your brain and other person’s brain
every time you watch them engaging in an
action
– Does not always lead to imitation
– Scientists are still debating mirror neurons’
function