Learning
n A
relatively permanent change in
•
Behavior
•
Knowledge
•
Capability
•
Attitude
n Acquired
through experience
•
Cannot be attributed to illness, injury, or
maturation
n Infants
do not ‘learn’ how to walk, as basic motor
skills and maturation govern every species
Classical
Conditioning
A type of learning in which an organism learns to
associate one stimulus with another.
n
Stimulus
• Any event or object in the environment to which an
organism responds
n
Reflex
• An involuntary response to a stimulus
n
Eye blink
to a puff of air
n
Conditioned
reflex
• A response elicited by an unconditioned stimulus with out prior learning
n
Salivate
at the sight of food
Classical Conditioning
n Discovered
by accident during saliva experiment
n Observed
salivary response occurring before presentation of food when dogs:
•
Heard footsteps of lab assistants
•
Heard food dishes rattle
•
Saw the attendant who fed them
•
Spotted their food
Classical Conditioning
n Unconditioned
stimulus (UCS)
• Stimulus that elicits a specific unconditioned
response without learning
n
Food
n
Loud
noise
n
Light in
eye
n
Puff of
air in eye
n Unconditioned
response (UCR)
• Response that is elicited by an unconditioned
stimulus without prior learning
n
Salivation
n
Startle
n
Contraction
of pupil to light
n
Eye
blink response
Classical Conditioning
n Conditioned
stimulus (CS)
• Neutral stimulus that, after repeated pairing with an
UCS, becomes associated with it and elicits a CR
n Conditioned
response (CR)
• Learned response that comes to be elicited by a CS as
a result of its repeated pairing with an UCS
n Higher-order
conditioning
• Occurs when the CS are linked together to form a
series of signals
n
Steps
leading to a blood draw at a clinic
Classically Conditioning a
Salivation Response
Changing Conditioned Responses
n Extinction
• Weakening and eventual disappearance of the CR as a
result of repeated presentation of the CS with out
the UCS
n Spontaneous
recovery
• Reappearance of an extinguished response after
exposure to the original CS following a rest period
n Generalization
• Tendency to make a CR to a stimulus that is similar
to the original CS
n Discrimination
• Learned ability to distinguish between similar
stimuli so that the CR occurs only to the original CS but not to similar
stimuli
John Watson
and Emotional Conditioning
The Cognitive Perspective
n Robert Rescorla
• Demonstrated that classical conditioning is not
repeated paring of the CS and the UCS
• Conditioning depends on whether the CS provides
information that enables reliable prediction of the UCS
• Used pairings of tones and shocks with rats
n
Only the
group where the tone reliably predicted the shock developed a conditioned fear
response
n
When the
tone provided no clue about the shock pairings did not lead to conditioning.
Classical Conditioning in Daily Life
Smell and taste are closely
associated because the smell of a particular food is a signal for its taste and
the physical sensation associated with eating it.
You can imagine how the fresh
bread smells, tastes, and its texture by viewing the picture.
n
What happens when you smell food?
• Stomach
rumbles due to
digestive processes that typically follow the
smell and taste
of food
•
Pancreas responds to counteract conditioned
rise in blood
sugar after a sweet taste on the tongue
Classical Conditioning in Daily Life
n Dental
visits
• Sound of the drills and suction
• Smell of the office
• Sight of the chair and light
n Drug use
• The CS associated with drug use lead individuals to
seek out those substances
• Counselors urge recovering addicts to avoid any cues
(people, places, and things)
n Taste
aversion
• Intense dislike and/or avoidance of a particular food
that has been associated with nausea or discomfort
n Chemotherapy
• Chemotherapy treatments can result in a conditioned
taste aversion
• Providing a “scapegoat” target can help patients maintain a proper diet
Operant Conditioning
Thorndike’s Law of
Effect
n
The
consequences, or effect, of a response will determine whether the tendency to
respond in the same way in the future will be strengthened or weakened.
n
Responses
closely followed by “satisfying
consequences” are
more likely to be repeated.
n
Responses
that have unpleasant
consequences will be avoided.
n
Puzzle Box: A cat had to press a pedal or pull a loop in order to escape the box and
receive food.
• Cat learned how to open the door almost immediately
after many trials
Operant Conditioning
n A type
of learning in which the consequences of behavior are manipulated in order to:
•
Increase or decrease the frequency of a response
•
Shape an entirely new response
n Operant
•
Voluntary behavior that accidentally brings about a
consequence
n Reinforcer
•
Anything that:
n Follows a
response and strengthens it
n
Increases the probability that it will occur
Process of Operant Conditioning
n Generalization
• The tendency to make the learned response to a
stimulus similar to that for which the response was originally reinforced
• A pigeon trained to peck a yellow disk will peck
similarly-colored disks
n
The less
similar the color the lower the rate of pecking will be
n Discriminative
stimulus
• A stimulus that signals whether a certain response or
behavior is likely to be rewarded, ignored, or punished
• Children misbehave with a grandparent because the
discriminative stimuli (parents) are not present
Variations in Operant Conditioning
n
Skinner box
• A sound-proof chamber with a device for delivering food to an
animal subject
• Food pellets or water are delivered after correct response(s)
• Records are kept on a cumulative recorder
Shaping Behavior
Shaping
n
Consists
of gradually molding a desired behavior (response) by reinforcing any movement
in the direction of the desired response
n
Gradually,
responses are guided toward the ultimate goal
n
Learning
a piano concerto is done by teacher praising your ability to
• Read musical notes
• Play individual notes
• Play the notes continuously
• Play one hand of notes according to the music
• Play both hands according to the music
• Play the concerto slowly
• Play the concerto up to tempo
• Play the concerto correctly
Process of Operant Conditioning
n
Successive
Approximations
• A series of gradual steps, each more similar to the
final desired response than the one before
• Reward disruptive children for very short periods of
good behavior, then expecting them to gradually work for longer and longer
periods
n
Extinction
• The weakening and eventual disappearance of the
conditioned response as a result of withheld reinforcement
•
Shaking a
vending machine that fails to deliver soda or candy before giving up and
walking away
Reinforcement
n
Any
event that follows a response and strengthens or increases the probability that
the response will be repeated
• The series of behaviors involved in using an ATM
• Paying bills on time avoids steep late-payment fees
Reinforcement
Reinforcement
Negative reinforcement
• Termination of an unpleasant condition after a
response
• Increases the probability that the response will be
repeated
n
Turning
on air conditioning to avoid the heat
n
Getting
out of bed to turn off a leaky faucet
n
Heroin
addicts will do almost anything to get
another fix and avoid the pains of withdrawal
Negative Reinforcement
Reinforcement
n Primary Reinforcer
• A reinforcer that fulfills
a basic physical need for survival and does not depend on learning
n
Food
n
Water
n
Sleep
n Secondary
Reinforcer
• Acquired or learned through association with other reinforcers
n
Money
n
Praise
n
Applause
Schedules of Reinforcement
A systematic process for administering reinforcement
n
Fixed Ratio
• Reinforcement given after a fixed number of correct responses
• An effective way to maintain a high response rate
n
The
faster people respond, the more reinforcers they
receive
n
A
migrant worker is paid for each bushel of fruit picked
•
When
large ratios are used:
n
people and
animals tend to pause after each reinforcement
n
Then
return to the high rate of responding
Schedules of Reinforcement
n
Variable Ratio
• Reinforcement given after a varying number of correct
responses
• Reinforcement can occur anywhere within the given
ratio
n
If a
30-response set is established, reinforcement could occur after 10 responses
and then again at 50 responses
n
Impossible
to predict exactly which response will be reinforced but it will occur within
30 responses
n
Gambling
is a classic example!
Schedules of Reinforcement
n
Fixed interval
• Reinforcement given following a correct response
after a specific time interval has passed
• Workers paid by salary, rather than hourly rate
n
Paid
$XXX.00 per month rather than by number of hours worked
• Responding pauses or sharply declines immediately
after reinforcement
• Responding increases rapidly just before the next reinforcer is due (scalloping effect)
Schedules of Reinforcement
n
Variable Interval
• Reinforcement given after the first correct response
that follows a varying amount of time
• Based on average time
• Reinforcers could be given at 30-, 90-, 45-, or even 72-second
intervals rather than every 60 seconds
• Maintains a steady and uniform rate of response,
though typically lower than ratio schedules
• Random drug testing in the workplace
Two Examples of Variable-Ratio Reinforcement
Punishment
The removal of a pleasant stimulus or the application
of an unpleasant stimulus, thereby lowering the probability of a response
Punishment
n
The
difference between punishment and negative reinforcement:
• Punishment
n
Adds an
unpleasant condition
n A teen is grounded for not
cleaning their room.
• Negative reinforcement
n
An
unpleasant condition is terminated or avoided
n
The
probability of a desired response is increased by removing an unpleasant
stimulus when the correct response is made
n The teen is grounded until the
room is clean.
Punishment
Potential Problems
• Punishment
does not extinguish an undesirable behavior
n
Suppresses
the behavior when the punishing agent is present
n
The
behavior is likely to continue when the threat of punishment is removed or in
settings where punishment is not likely
• Repeat offenders in the criminal justice system
•
Punishment indicates that a behavior is unacceptable
but does not help people develop more appropriate behavior
n
Punishment
should be used in conjunction with reinforcement or rewards for appropriate
behavior
• Punish a child for hitting with time-out then reward
them later when they are playing appropriately
Punishment
Potential Problems
•
The person who is severely punished often becomes
fearful and feels angry and hostile toward the punisher.
n
May be
accompanied by a desire to retaliate, avoid, or escape from the punisher or
punishing situation
• Teens run away from home
n
Loss of
privilege is more effective than physical punishment and engenders less fear
and hostility
•
Punishment frequently leads to aggression.
n
May
model aggressive behavior
• Demonstrates aggression as a means for solving problems and
discharging anger
n
Children
of abusive, punishing parents are at greater risk than other children of
becoming aggressive and abusive themselves
Making Punishment Effective
n
Punishment
is necessary to stop destructive behavior.
n
Factors
that influence the effectiveness of punishment: its timing, intensity, and the
consistency of application
• It is most effective when applied during the
misbehavior or as soon afterward as possible
n
Interrupting
the misbehavior diminishes its rewarding aspects.
n
The
longer the delay between the response and the punishment, the less effective
the punishment is in suppressing the response.
• Don’t kick the dog today for
what it did yesterday – it won’t connect the punishment
with the misdeed.
n
If delay
is necessary, the punishment should remind them of the incident and explain why
it was inappropriate.
Making Punishment Effective
n Punishment
should be of the minimum severity necessary to suppress the problem behavior.
• Unnecessarily
severe punishment leads to adverse side effects.
• Purpose
of punishment is NOT to vent anger but to modify behavior.
• Punishment
meted out in anger is usually more intense than needed to bring about desired
result.
• If too
mild, it will have no effect.
• Gradually
increasing the intensity of the punishment causes the perpetrator to adapt and
the unwanted behavior will persist.
• To
suppress a behavior, the punishment must be more punishing than the misbehavior
is rewarding.
n
A $200 ticket is more likely to suppress speeding
than a $2 ticket.
Making Punishment Effective
n
It must be applied
consistently.
• Parents can not ignore
misbehavior one day and punish the same act the next day.
• Both parents should react to the same misbehavior in
the same way.
• An undesired response will be suppressed more
effectively when the probability of punishment is high.
n
Most people will not
speed when a police car is in the rear-view mirror.
Escape and Avoidance Learning
n
Escape learning
• Performing a behavior because it prevents or
terminates an aversive event
n
Taking
an aspirin to avoid a headache
n
Avoidance learning
• Engaging in behaviors to avoid an aversive
consequence
n
Coming
to the dinner table when mother’s voice
reaches just the right pitch
Learned Helplessness
A passive resignation to
aversive conditions learned through repeated exposure to inescapable and
unavoidable aversive events
n Overmeier and Seligman Experiment
• Dogs in a harness received electric shock
• Later, same dogs placed in a box with two
compartments separated by a low barrier
• Warning signal indicated shock was coming in floor on
one side
• Dogs did not jump the barrier to avoid the shock
•
Dogs in control
group jumped the barrier
n
Humans
who suffered painful experiences they could not avoid or escape may simply give
up and react to disappointment in life by becoming inactive, withdrawn, and
depressed.
Applications of Operant Conditioning
n Shaping
the behavior of animals
• Shaping
trains circus, zoos, animal parks
• Instinctual
drift
n
Animal eventually resumes instinctual behavior
n Biofeedback
• The use
of sensitive equipment to give people precise feedback about internal
physiological processes so that they can learn, with practice, to exercise
control over them
n Behavior
modification
• A method
of changing behavior through a systematic program based on the learning
principles of classical conditioning, operant conditioning, or observational
learning.
• Autistic
children and adults stopped self-injurious behaviors
Applications of Operant Conditioning
n
Token
economy
• A program that motivates socially desirable behavior by reinforcing it
with tokens that
can be exchanged for desired items or privileges
• Poker chips or coupons are given when a patient
grooms correctly, does chores, or interacts in a socially appropriate way
• Positive behavior may stop when tokens are not given
n
Time out
• Misbehaving child is removed for a short time from
sources of positive reinforcement in an attempt to extinguish the unwarranted
behavior
Using Behavior Modification
Use
conditioning to modify your own behavior
•
Identify the target behavior.
•
It must
be both observable and measurable.
•
Gather and record baseline data.
•
Keep a
daily record of how much time you spend on the target behavior for about a
week.
•
Plan your behavior modification program.
•
Formulate
a plan and set goals to either decrease or increase the target behavior.
•
Choose your reinforcers.
•
Any
activity you enjoy can be used to reinforce an activity you enjoy less.
•
Set the reinforcement conditions and begin recording
and reinforcing your progress.
• Don’t set the
reinforcement goals so high that it becomes nearly impossible to earn a reward.
• Keep in mind that shaping takes place through
rewarding small steps towards the goal.
• Chart your progress as you work toward the target
behavior.
Cognitive Learning
n Cognitive
processes
• Mental processes such as thinking, knowing, problem
solving, remembering, and forming mental representations
n
Insight
• The sudden realization of the relationship between
elements in a problem situation, which makes the solution apparent
• Chimps who had given up in attempts to get bananas
suddenly returned with a
solution not accounted by trial-and-error
• Transferred the learning to other situations
Latent Learning and Cognitive Maps
n
Latent learning
• Learning that occurs without apparent reinforcement
and is not demonstrated until the organism is motivated to do so
n Cognitive
maps
• Mental representation of a spatial arrangement, such
as a maze
• Tolman and Honzik study of
rats in a maze
n
Rats
rewarded only after 11 days for
running the maze showed marked performance the next day and outperformed others that were rewarded
daily.
Latent Learning and Cognitive Maps
Rats in Group1 were rewarded
daily for running the maze correctly
Rats in Group 2 were never
rewarded
Rats in Group 3 were rewarded
only on the 11th day and thereafter outperformed the rats in group
1.
Latent learning occurred as the
rats had learned the maze but were not motivated to
perform it until rewarded.
Observational Learning
Learning by observing the behavior of others and the consequences of that behavior
• Often involves imitation
n Modeling
• Another name for observational learning
• Improved learning when:
n
Several
sessions of observation precede attempts to perform the behavior
n
Repeated
in the early stages of practicing the behavior
n Model
•
A person
who demonstrates a behavior or whose behavior is imitated
Observational Learning
n Modeling
effect
• Learning a new behavior from a model through the
acquisition of new responses
• Teachers showing how to solve math problems on the
board then followed step-by-step until accomplished
n Elicitation
effect
• Exhibiting a behavior similar to that shown by a
model in an unfamiliar situation
• Watching someone use appropriate silverware at an
elaborate state dinner so you, too, act correctly
Observational Learning
n Disinhibitory effect
• Displaying a previously suppressed behavior because a
model does so without receiving punishment
n
Not
belching in public
n
Copying
drinking or drug use seen in movies or TV shows
n Inhibitory
effect
• Suppressing a behavior because a model is punished
for displaying the behavior
n
Slowing
down when seeing others receive speeding ticket
n
A mother’s fear of snakes can be passed on to her toddler
Observational Learning
n Bandura and observed violence
• Bandura demonstrated how children are influenced by
aggressive models.
• “Bobo Doll” and
later studies confirmed that exposure to humans portraying aggression on film
was the most influential in eliciting and shaping aggressive behavior.
• Violent video games increase aggressive behavior.
• Violence in music, music videos, advertising, and on
the Internet also influences behavior.
• Observing consequences for aggressive acts helps
preschoolers learn that violence is morally unacceptable.
• School-age children judge the rightness or wrongness
of violence based on provocation:
n
Retaliation
believed morally acceptable even if punished by authority figures
• Individuals who watch the most violence as children
were more likely to engage in acts of violence as adults.
• Children will also imitate prosocial
or helping behavior
n Mister Rogers’ Neighborhood and Sesame Street