Transcript powerpoint - Northern Highlands

AP Psychology Unit 6:
Learning
Learning: Definition
 Relatively permanent
 to rule out behavioral changes that result from fatigue or
motivational changes
 Exception: Short-Term Memory recall ability only lasts
about 30 seconds without rehearsal – learning happens, but
it’s not “relatively permanent”
 Change in behavior or behavior potential
 Behavioral potential is included because learning often
takes place without immediately being shown in behavior.
 learning may create the potential for behavior change when
the conditions are right – e.g. when there is an incentive
 Experience in environment
 rule out changes that result from maturation
 Complication: maturation usually works together with
experience to change behavior. It speeds up the learning
process.
Learning: Learning vs. Maturation
It is hard to separate experience from
maturation and say something is purely
learned or purely genetic.
Maturation is generally seen as preparing
a species to learn a skill rapidly
Maturation vs. Learning?
Did I learn or
just mature?
Classic Experiment by Carmichael
Do salamanders learn to swim?
Experimental Group
Salamander eggs were placed in
chloretone solution, a chemical
that prevents movement but
permits normal growth.
Control Group
Salamander eggs were placed
in tap water.
Maturation vs. Learning?
Classic
Classic Experiment
Experiment by
by Carmichael
Carmichael
Do
Do salamanders
salamanders learn
learn to
to swim?
swim?
Experimental Group
Control Group
When salamanders in the Control group reached an age when
they were swimming normally, the animals in the Experimental
group were tested by placing them in tap water and waiting for
the paralyzing chemical to wear off.
Maturation vs. Learning?
Classic Experiment by Carmichael
This experiment has been considered to be a clear
example of maturation; it is said that the salamanders
immediately started to swim normally.
But
On close examination, the results could be seen as
evidence for either maturation or learning.
Maturation vs. Learning?
Who
knows…
Classic Experiment by Carmichael
Carmichael noted that from the first twitch until normal
swimming occurred, there was a period of about 45
minutes in which the animals showed increasingly
complex swimming movements.
This could be seen as evidence for very rapid learning.
Or
It could be said the chemical was gradually wearing off
and no learning occurred.
Associative Learning
A simple form of learning in which we
comprehend that certain events occur
together
We will learn about several types:
Habituation
Sensitization
Classical or Pavlovian Conditioning
Operant or Instrumental or Skinnerian
Conditioning
Habituation and Sensitization
We sure were
annoying!
 Habituation
Yep.
Annoying.
Decreased response to a stimulus
judged to be of little or no importance
We engage in this type of learning so
we can tune out unimportant stimuli
and focus on what matters
 Sensitization
Increased response to a stimulus
when we are anticipating an
important stimulus
We engage in this type of learning so
we are prepared for dangerous
situations
Classical Conditioning
I love me some
salivating dogskis
this guy’s
a freak.
Classical Conditioning: Definition and History
 Learning in which a response naturally caused by one
stimulus comes to be elicited by a different, formerly neutral
stimulus:
1. Unconditioned Stimulus (UCS)  Unconditioned Response (UCR)
2. Neutral Stimulus (NS) + UCS  UCR
3. Conditioned Stimulus (CS)  Conditioned Response (CR)
 Ivan Pavlov
 Accidentally discovered classical conditioning
 His experiments on digestion in dogs turned into research on
learning
 Also worked with conditioning a defensive reflex
I LOVE dogski saliva!
Elements of Classical Conditioning
Unconditioned stimulus (UCS)
A stimulus that naturally and automatically
causes a specific response in an organism
And example of a UCS would be food
Unconditioned response (UCR)
The response caused by a UCS
The UCR is automatic and unlearned
An example of a UCR is salivation in response
to food
Elements of Classical Conditioning
Conditioned stimulus (CS)
A formerly neutral stimulus (NS) that is paired
with a UCS and eventually causes the desired
response all by itself
An example of a CS is the bell in Pavlov’s
studies
Conditioned response (CR)
The learned response to the CS
An example is salivation in response to the bell
Classical Conditioning: Pavlov’s Experiment
Before Conditioning
Bell
(NS)
No
Response
Food
(UCS)
Salivation
(UCR)
Classical Conditioning: Pavlov’s Procedure
During Conditioning
Food
(UCS)
Bell
(NS)
Salivation
(UCR)
Classical Conditioning: Pavlov’s Procedure
After Conditioning
Bell
(CS)
Salivation
(CR)
Classical Conditioning In Humans
The “Little Albert” experiment demonstrated a classically
conditioned fear of white fluffy things
 CS =
 UCS =
Rat
Loud Noise
 CR =
 UCR =
Fear of Rat
Fear of Noise
 NS =
Rat
Stop
touching
me, creepy
baby…
Come hither,
rattypants!
New Learning Based on Old:
Higher Order Conditioning
 Once a neutral stimulus becomes a conditioned
stimulus, it may function as an unconditioned
stimulus to elicit new learning.
 For instance, in Pavlov’s experiment, once the
bell produced the salivation response in the
dogs, it could be paired with a new neutral
stimulus, such as a red light, until the dogs
learned to salivate to the red light alone.
Classical Conditioning: Key Variables
 In order for Classical Conditioning to work the
following variables must exist:
 STRENGTH - Stimuli (UCS, NS) must be noticeable
enough to provoke a response.
 TIMING - UCS and NS must be paired close
together so that an association is made between the
two. Best case: NS precedes UCS.
 FREQUENCY - UCS and NS must be paired
together many times so that an association is made
between the two and the NS can come to elicit the
same response as the UCS.
Classical Conditioning in Humans:
Class Demonstrations
I’d say
something, but,
sadly, I have no
mouth.
 Demo 1: “A Day at the Beach”
 How does this demonstrate classical
conditioning?
 UCS, UCR, NS, CS, CR?
 Demo 2
 Lick your finger and dip it into your cup of
lemonade powder, but DO NOT EAT IT.
 When you hear the tone, immediately eat the
powder on your finger, and then dip your
finger back into the cup to prepare for the next
trial.
 You must eat some of the powder
immediately after each tone, but not any other
time.
 After several “learning” trials, you will be
instructed to simply listen to the tone without
eating the powder.
 What happens? Label the UCS, UCR, NS, CS
and CR in your notes based on the demo.
Classical Conditioning in Humans:
Class Demonstrations
 “Day at the Beach”
 UCS
 Shark Attack (in
Jaws)
 UCR
 Fear
 NS
 Music
 CS
 Music
 CR
 Fear of Music
 Salivating Students
 UCS
 Eating Powder
 UCR
 Salivation
 NS
 Tone
 CS
 Tone
 CR
 Salivation to tone
Classical Conditioning: Predictability
 Blocking – previous learning prevents
conditioning to a second stimulus when the two
stimuli are presented together
Kamin’s Work – rats that learned to fear a tone which
had been followed by a shock. When the tone was
simultaneously paired with a light and followed by the
shock, the rats failed to fear the light alone.
How is this different from Higher Order Conditioning?
 Backward Conditioning – NS may come
before OR after UCS, but not before AND after.
Rescorla’s Work – tone could either mean be fearful or
relax, depending when it occurred with shock.
Classical Conditioning:
Extinction and Spontaneous Recovery
 Extinction –
 After a period of time passes when CS is not paired with UCS, CS returns
to being an NS
 e.g. Baby Albert would eventually cease to be afraid of white fluffy things
after they were not paired with a horrible and frightening noise (Mary
Cover Jones)
 Spontaneous Recovery –
 Just because extinction occurs, does it mean that the learning is gone?No!
 After extinction, it is not unusual to see the recurrence of the conditioned
response
 This proves the learning never disappeared; it was just obscured by new
learning - like interference
Classical Conditioning:
Generalization and Discrimination
DUDE! Get that
white furry evil
away from me!
 Generalization –
Which
one
yields
food?
An organism may learn to respond
not only to the CS, but also to other
stimuli that are similar to the CS.
e.g. Baby Albert was conditioned to
fear a white rat, but also feared
cotton balls, rabbits, white sweaters,
Santa’s beard, etc.
 Discrimination –
Organisms can also learn to decipher
between similar stimuli when only
particular stimuli are paired with a
UCS.
Classical Conditioning in the Real World:
Taste Aversion and the Garcia Effect
Dude…I’m
never eating
pumpkin
seeds again.
No more
livestock?
Now what?
 Some learning mechanisms are so
powerful they do not require
frequency of pairings.
 Taste Aversion –
 Occurs when organism becomes ill
following consumption of a particular
food.
 Organism may never be able to eat the
food again.
 WHY?
 Garcia Effect –
 Using principles of taste aversion, John
Garcia put this phenomenon to good use
 Sprinkled carcass of sheep with a
chemical that caused illness in coyotes
 Coyotes did not attack the livestock
following this experience
Classical Conditioning in the Real World:
Preparedness and Contrapreparedness
 Some conditioned responses
come naturally, others do not.
 Preparedness
Scary.
Conditioned behaviors that work
well with organism’s instinctive
behaviors and are easy to train
e.g. phobia of snakes or spiders
 Contrapreparedness
Not Scary.
Other conditioned behaviors go
against the organism’s instinctive
behaviors and are difficult or
impossible to train.
e.g. phobia of chairs or tables?
Classical Conditioning in the Real World:
Treating Phobias
 Many phobias are learned responses and can be unlearned
 This can be done gradually or all at once
 Systematic Desensitization
Therapist and client generate “fear hierarchy” of situations
that are increasingly threatening
Client then learns relaxation techniques
Client experiences “in vivo” therapy to directly experience
each item on fear hierarchy to gradually unlearn his/her fear
 Flooding
Client faces worst-case-scenario involving fear
If they can survive this, they have no reason so be fearful
every day
Operant Conditioning
Skinner and his Box
Skinner “Crib” containing
Skinner’s child
The Ultimate Skinner Box
Operant Conditioning
Learning in which an organism
engages in a spontaneous behavior
which is followed by a consequence a reward or punishment
Organism learns to perform behavior
in order to gain a reward or avoid a
punishment
Law of Effect
If a behavior is reinforced, it is MORE
likely to occur
If a behavior is punished, it is LESS likely
to occur
I’ll punish
you…
History of Operant Conditioning
 E.L. Thorndike
 Researched cats in a puzzle box
 Cats learned to escape from box to attain a reinforcement of food
 B.F. Skinner
 Created a device called a Skinner Box to train organisms using
operant conditioning
 Also did research on superstition (pigeons) and connected it to the
principles of operant conditioning
Elements of Operant Conditioning
Reinforcer
A stimulus or event that
follows a behavior and
makes that behavior more
likely to occur again
Punisher
A stimulus or event that
follows a behavior and
makes that behavior less
likely to occur again
Types of Reinforcement
 Positive reinforcer (+)
 Negative reinforcer (-)
Adds something
rewarding following a
behavior, making that
behavior more likely to
occur again
Giving a dog a treat for
fetching a ball is an
example
Removes something
unpleasant from the
environment following a
behavior, making that
behavior more likely to
occur again
Taking an aspirin to
relieve a headache is
an example
Types of Reinforcement
 Primary reinforcer
Adds something
intrinsically valuable to
the organism
Giving a dog a food for
shaking hands
 Secondary reinforcer
Adds something with
assigned value to the
organism
Giving a person $100
for each “A” on their
report card
Types of Punishment
 Positive Punishment (+)
Adds something
undesirable to decrease a
behavior
Spanking a child for
swearing
 Negative Punishment (-)
Removes something
desirable to decrease a
behavior
Taking a child’s toy away
for swearing
Also called omission
training
Types of Punishment
 Primary Punishment
Method of decreasing
behavior is directly
threatening to organism’s
survival
Beating a prisoner for
trying to escape
 Secondary Punishment
Method of decreasing
behavior is undesirable, but
not life-threatening
Taking away a prisoner’s
recreational privileges for
trying to escape
Complex Behaviors and Shaping
 Some behaviors are too complex to occur
spontaneously
 For these behaviors, shaping must be used
Shaping reinforces successive approximations to the
desired behavior
Organism eventually learns what the desired behavior is
in small steps
Similar to playing “hot and cold”
Our class demonstration?
Preparedness and Contrapreparedness
in Operant Conditioning
 Some changes in behavior are easily trained
 Preparedness
Conditioned behaviors that work well with organism’s
instinctive behaviors and are easy to train
e.g. Brelands’ “Dancing Chicken”
 Contrapreparedness
Other conditioned behaviors go against the organism’s
instinctive behaviors and are difficult or impossible to train.
e.g. Brelands’ raccoon
Reinforcement vs. Punishment?
Punishment not as effective as
reinforcement
Does not teach proper behavior, only
suppresses undesirable behavior
Causes upset that can impede learning
May give impression that inflicting pain is
acceptable
Effective Punishment?
Effective punishment must be
SWIFT
Should occur as soon as possible after the behavior
CERTAIN
Should occur every time the behavior does
SUFFICIENT
Should be strong enough to be a deterrent
CONSISTENT
Should apply to all individuals the same way
Impact of Punishment
 When punishment is given
haphazardly, learned
helplessness can result.
 Learned Helplessness occurs
when NO MATTER WHAT THE
ORGANISM DOES, it cannot
change the consequences of
behavior.
 Martin Seligman’s experiment
with dogs showed that dogs
given a series of inescapable
shocks stopped trying to escape
the shocks even when given the
opportunity to escape later.
 Another example would be
finding that whether or not you
study for your calculus tests, you
fail, so you stop trying
altogether.
Alternatives to Punishment
An alternative to punishment if known as
AVOIDANCE TRAINING
the organism is given a “warning” before
punishment occurs so it may change its
behavior in order to avoid an unpleasant
consequence like a punishment.
ex/ “Counting to three” before punishment is
delivered to provoke a child to stop
misbehaving.
Behavioral Change Using Biofeedback
Biofeedback is an operant technique that
teaches people to gain voluntary control
over bodily processes like heart rate and
blood pressure
When used to control brain activity it is
called neurofeedback
Demonstration: Word Generation
Sit with your assigned partner
Clear off your work space – all you need is
a pen and a sheet of blank white lined
paper
Decide who is #1 and #2
Wait for further instructions
Schedules of Reinforcement
 Continuous reinforcement vs. Intermittent
reinforcement?
 Interval schedules
Reinforcement depends on the passing of time
Fixed-interval schedule
 Reinforcement follows the first behavior after a fixed
amount of time has passed
 An example would be receiving a paycheck every two
weeks
Variable-interval schedule
 Reinforcement follows the first behavior after a variable
amount of time has passed
 An example would be pop quizzes
Schedules of Reinforcement
 Ratio schedules
Reinforcement depends on the number of responses
made
Fixed-ratio schedule
 Reinforcement follows a fixed number of behaviors
 For example, being paid on a piecework basis
Variable-ratio schedule
 Reinforcement follows a variable number of behaviors
 An example would be playing slot machines
 Yields the most rapid response rate
• Subject does not know when reward is coming - works
consistently
• Subject must WORK to get reward
Response Patterns to Schedules of
Reinforcement
 Which schedule
yields the fastest
response rate?
 What happened
in our class
demonstration?
Intrinsic and Extrinsic Motivation
 Reinforcement can sometimes come from within and
be intrinsically motivated
e.g. Playing on a swing set is intrinsically motivated in
children
 Reinforcements can also be extrinsically motivated, or
come from an exogenous source
e.g. Adults get paid for going to work – no pay, no
work.
 What happens when we give extrinsic reinforcements
for intrinsically motivated behavior? Overjustification
Effect!
Giving an extrinsic reinforcement for intrinsically motivated
activity decreases intrinsic motivation!
Marshmallow study
Compare and Contrast:
Classical and Operant Conditioning
Cognitive Learning
Cognitive Learning
 Sometimes learning involves more than simply
reacting to stimuli – it involves THINKING!
 Cognitive Learning
Learning that depends on mental activity that is not
directly observable
Involves such processes as attention, expectation,
thinking, and memory
 While behaviorists typically focus on learning
that is based on reactions, cognitive
psychologists explain learning in terms of
additional mental processes.
Generative Learning and Insight
Generative learning
Using what you know to figure out something
you don’t
E.g. realizing a new song is by a favorite group
of yours
Insight
After thinking about a problem for a bit, you
suddenly figure it out
E.g. Kohler’s chimps; “Genius” Chimp
Latent Learning and Cognitive Maps
 Latent learning
Edward Chance Tolman and the “Tolmaniacs”
learning that takes place before the subject realizes it
and is not immediately reflected in behavior
Taking a test on material learned over the course of a
few weeks
 Cognitive mapping
latent learning stored as a mental image
Slideshow experiment
 Demonstration of Latent Learning and Cognitive
Mapping
Latent Learning
Latent Learning: A Classic Experiment
(Tolman & Honzik, 1930)
Three groups of rats were given practice trials in a
maze, 1 trial per day.
A trial started when the rat was placed in the Start box
and ended when he entered the Goal box, after which
he was removed from the maze.
The maze consisted of a series of components shaped
like the letter
T.
Latent Learning
Latent Learning: A Classic Experiment
(Tolman & Honzik, 1930)
GOAL
...
i
T
T
T
T
T
T
START
When the rat went up
the stem of the T, he
reached a choice point.
If he turned one way, he
came to a dead end.
If he turned the other
way, he came to the
entrance of the next
component.
Latent Learning
Latent Learning: A Classic Experiment
(Tolman & Honzik, 1930)
GOAL
...
i
T
T
T
T
T
T
START
Each time the rat turned
into the dead end, it was
counted as an error.
The measure of
performance (dependent
variable) was the number
of errors on a trial.
If learning occurred, the
number of errors should
decrease as more and
more trials were given.
Latent Learning
Latent Learning: A Classic Experiment
(Tolman & Honzik, 1930)
GROUP 1: On every trial, these rats received food
when they reached the goal box.
GROUP 2: These rats never received food. They
were simply removed from the maze when they got to
the goal box.
GROUP 3: These rats got no food on Trials 1 to 10.
But on Trial 11, and every trial afterwards, they
received a food reward.
Latent Learning
GR 1 —
GR 2 —
GR 3 —
4
6
8
The day-to-day decrease in
errors represented a “relatively
permanent change in behavior”
that resulted from practice.
This was clear evidence for
learning.
2
0
Average Errors
10
Latent Learning: A Classic Experiment
(Tolman & Honzik, 1930)
1
10 11
Trials (1 Trial per Day)
17
Latent Learning
GR 2 —
GR 3 —
6
8
GR 1 —
4
Group 2 got no food but still improved slightly.
Removal from the maze was a small reward.
There was little evidence for learning.
2
0
Average Errors
10
Latent Learning: A Classic Experiment
(Tolman & Honzik, 1930)
1
10 11
Trials (1 Trial per Day)
17
Latent Learning
GR 2 —
GR 3 —
8
GR 1 —
4
6
Getting no food on
Trials 1 – 10,
Group 3 performed
like Group 2
through Trial 11.
2
0
Average Errors
10
Latent Learning: A Classic Experiment
(Tolman & Honzik, 1930)
1
10 11
Trials (1 Trial per Day)
17
Latent Learning
GR 2 —
GR 3 —
6
8
GR 1 —
4
On the next trial,
Group 3 matched
Group 1, and then
did even better!
2
0
Average Errors
10
Latent Learning: A Classic Experiment
(Tolman & Honzik, 1930)
1
10 11
Trials (1 Trial per Day)
17
Latent Learning
Latent Learning: A Classic Experiment
(Tolman & Honzik, 1930)
Interpretation
Group 3 learned the route to the maze on Trials 1 to
10 but didn’t show it because there was no
motivation to perform.
They outperformed Group 1 because the shift from
no reward to reward made the reward seem larger
by comparison. This is called “positive contrast.”
Latent Learning
Latent Learning: A Classic Experiment
(Tolman & Honzik, 1930)
Conclusion
We must observe a change in behavior to say that
learning has occurred, but if no change occurs, we
can draw no conclusion. Learning may be present
“beneath the surface.”
This supports a distinction between learning and
performance.
Learning Sets and
Trial and Error Learning
Learning sets/Learning to Learn
refers to increasing effectiveness at problem
solving through experience
organisms “learn how to learn”
Figuring out how to study best
Trial and Error Learning
Learn by your mistakes
Class demonstration - “Blind Maze” - what
happened?
Wired for Imitation: Mirror Neurons
 Mirror neurons are specialized neurons in the frontal lobe
that not only fire when we are engaged in certain actions,
but ALSO when we watch others perform those actions.
 These neurons form the basis of imitation, empathy and
our social nature.
Yawning, smiling, laughing, etc.
Monkey study? Same neural activity doing and watching!
Impact of Vicarious reinforcement and punishment?
What happens
to your brain
When you watch
Cheesy Lifetime
movies
Learning by Observing
Why,
Bandura
WHYYYY?
 Social Learning Theory or
Observational Learning Theory
focuses on what we learn from
observing other people
Albert Bandura’s Bobo Doll
experiment
Children imitated adult role model - adult
models behavior and child imitates
Non-human animals have even been
shown to learn through observation!
Modeling refers to process of observing
and imitating a specific behavior
Evil Bobo Doll
 Prosocial and Antisocial
behavior?
Learning by Observing