Learning pt 2

We talked about
classical conditioning,

but of course, there are some

limits to classical
conditioning.

You can't classically
condition all behaviors.

One, classical
conditioning can't

teach new behaviors
to an organism,

but it can only elicit innate
or otherwise reflexive

behaviors with stimuli that

ordinarily might not
have these effects.

For example, I gave

an example where my cat

came running to
the sound of food.

Seeking food is an
innate behavior

that exists within cats.

The only thing are
all creatures.

The only association
that I created

was the sound of a can opening

with this already
innate behavior.

I couldn't have
taught her to run

towards the sound of

a bicycle because cats don't

necessarily have that
ability innately built in.

Another element of
classical conditioning

can be outlined by these
three important factors.

The important factors
must be met in order

for classical conditioning
to actually work.

First is cognition.

The second factor is

the evolutionary history
of the organism,

and third is the ecology
of the organism.

We'll go through each of

these individually on
the next few slides.

First, there's cognition
and evolution.

Robert Rescorla
demonstrated that

conditioning rats to fear

a tone stimulus didn't

work unless the tone reliably
predicted the shocks,

so the conditioned stimulus.

Rescorla proposed that
the rats were engaging in

cognitive activity to determine

whether the stimulus
was a valid predictor.

All of the creatures that
we classically condition,

whether it's dogs, cats, rats,

or humans, they
engage in cognition

in order for classical
conditioning to work.

In this experiment, the
rats only became afraid of

the tone if the tone reliably
predicted the shocks.

Here, the rats are conditioned

to fear response when the tone

is sounded because every

time there's a tone,
there's a shock.

For these rats in Group
2, the tone appears,

they get shocked, but then

they get shocked even when
the tone isn't there,

the tone appears,
they get shocked.

Rats aren't conditioned to
fear response when the tone is

sounded because the tone

isn't always paired
with a shock.

Thus, the rats are engaging
in some cognitive activity to

determine that the
tone is either

predictive or not
predictive of a shock.

When it comes to
cognition and evolution,

John Garcia overturned
the general

process behaviorist doctrine
of equipotentiality,

which proposed that virtually
any neutral stimulus could

be used to condition
virtually any response.

Garcia's research showed that

the evolutionary history of
each organism prepares it

to be more or less easily
conditioned by various stimuli.

In the textbook, you'll
read an instance

where either noises
were presented

alongside the ingestion of

sugar water that
made the mice sick,

feel nauseous, or
radiation was presented

along with the ingestion of

this sugar water that made
the mice feel nauseous.

For the mice, the
noises were more

effective than radiation
to help the mice

avoid the sugar water because
the noises were something

that they could
interpret and perceive,

whereas the radiation wasn't

necessarily something
that they could perceive.

He showed that not
necessarily any stimulus

can be used to condition
virtually any response.

The creature in question
has to have the ability to

interpret that
unconditioned stimulus.

The evolutionary history of

each organism prepares
it to be more or

less easily conditioned
by various stimuli.

Then we also have the
functionalist perspective

that notes some limitations
of classical conditioning.

The functionalist
perspective proposes

that classical
conditioning occurs in

natural environments as
well as in the laboratory.

Classical conditioning
in the wild

likely evolved to help
an organism survive,

whereas the motive behind

conditioning might be
different in the lab.

Conditioning stimuli
and the natural ecology

of an animal are rarely
neutral or arbitrary.

They actually do have
some predictive validity.

It's not just random,
whereas in the lab,

the neutral stimuli
might be something that

isn't available in the
creature's normal environment.

Conditioning that
occurs in nature

may be more rapid, robust,

and less dependent upon a
very small time interval

between the presentation of

the conditioning stimulus and

the unconditioned stimulus,

and thus when the pairing of
a conditioned stimulus with

the unconditioned stimulus is

presented in the animal's
natural environment,

it's more resistant to

interference in the
conditioning process

relative to those
that are paired

or the associations that
are made in the laboratory.

Those are the limitations.

We're talking about context.

We talked about
evolutionary mechanisms

that make certain stimuli more

likely to be successfully
conditioned to a response.

We also talked about how

cognition plays a role in
classical conditioning.

Now let's talk about
operant conditioning,

which is also a different
learning that occurs.

Operant conditioning
was developed as

a response or as

an extension of
Thorndike's Law of Effect.

Now, in Thorndike's
Law of Effect,

he states that for any organism,

those behaviors that lead to

a satisfying state of
affairs are stamped

in while those
behaviors that lead to

an unsatisfying
or annoying state

of affairs are stamped out.

In other words, learning among

organisms doesn't
necessarily only

occur through classical
conditioning,

but it occurs through
the gain of resources

or something that the creature

really seeks and if

any behavior leads to
negative outcomes,

and those behaviors
are stamped out.

In other words, the organism
learns as a result of

the consequences of his
or her own actions.

This is Thorndike's
Law of Effect,

which led to the development
of operant conditioning.

This is a type of learning in

which the consequences
of a behavior

affect the probability
that the behavior

will be repeated in the future.

In this instance,

in this example that
you'll see here,

the organism operates on

his or her environment
in a way that results in

specific consequences
and comes to

associate these consequences
with its behavior.

Operant conditioning
is an active process

entirely shaped by the
organism's own behavior.

Whereas when we were

talking about classical
conditioning,

that tends to be more of

a process that occurs
below conscious awareness,

whereas operant
conditioning occurs

and is an active process.

Here we see a rat and there's

a little food tray that's
connected to food pellets,

and there is a little light,

and the little rat
can press a lever.

The rat engages with the lever

and based on whether
the light is on or not,

and the rat's own behavior,

the rat learns that maybe

if the light is on
and I push the lever,

food will come out.

The rat learns that it is up to

his or her own
behavior that leads

to food being delivered
to his food tray.

How does operant conditioning

differ from classical
conditioning?

I already gave you one example.

Operant conditioning
is more active,

requires more active engagement

from the creature relative
to classical conditioning.

Again, operant conditioning is

an active process that is shaped

by an organism's own behavior,

whereas classical conditioning
is a passive process

shaped by forces outside the
control of the organism.

For example, with Pavlov's dogs,

the dogs couldn't ring the
bell and then salivate.

It was up to Pavlov
to ring the bell and

then it would cause
the dogs to salivate.

Even salivation
was a process that

occurred below the dog's
conscious awareness.

They didn't have the ability
to control their salivation.

Operant conditioning,
like I said,

it functions as a result of

an organism's own behavior

and their behavior either

leads to positive reinforcement,

negative reinforcement,
positive punishment,

and negative punishment.

Now, these terms
might be a little

confusing because when we
think of reinforcement,

we think of something
that might be positive,

something that
might be a reward,

but that's not what we're
talking about in this instance.

Also, when we think
about punishment,

we might think of something
that is punitive,

something that is negative,

something that is meant to
discourage a certain behavior.

Let's talk about what
each of these mean.

When we talk about
reinforcement here,

what we mean is that reinforcers
increase the behavior,

and positive means you add

something to increase
the behavior.

Negative means you remove

something to increase
the behavior.

When it comes to punishment,

punishers decrease a behavior,

so you add something to

decrease the behavior
with positive punishment,

or you remove something
to decrease the behavior.

With positive reinforcement,

when the frequency or
probability of the behavior

is increased as a result

of the addition or
presentation of

something following the
performance of a behavior.

In this instance, it
would be like the rat

getting food pellets when

the light is on and
it presses the lever.

The frequency or
probability of behavior is

increased as a result of

the addition or presentation of

something following
the performance

of the behavior, the pellets.

Negative reinforcement when
a behavior is reinforced

by the removal or
avoidance of something.

Positive punishment is when

the consequence of a behavior is

the addition or presentation of

something, generally
something aversive.

Let's say if the rat

presses the lever when
the light isn't on,

they receive a shock.

That would be
positive punishment.

Negative punishment is
when the consequence of

a behavior is the
removal of something.

Let's go through and look at

this table that I created to

help you all understand what
positive reinforcement,

negative reinforcement
and positive punishment,

negative punishment look like.

Students frequently
have a hard time

distinguishing the implications
of positive and negative.

The important thing to
emphasize is that reinforcers

increase the behavior and

punishers decrease the behavior.

I put that here,
you might want to

remember that for the exam.

For positive reinforcement, you

add something to
increase the behavior.

A child puts toys away
and receives a cookie,

the child learns to
put their toys away.

For negative punishment,

you take away something

to decrease a behavior.

This would be if a child
is mean to a sibling,

you send him to his room
and he learns to be nicer.

For positive punishment, you add

something to
decrease a behavior.

You're late to a
meeting and your boss

criticizes you in
front of everyone.

You learn to avoid being late.

For negative reinforcement,
you take away

something that increases
your behavior.

You run into the
house to get out of

a storm and you learn
to avoid the rain.

Another way to think about
this if you're having

trouble during the exam is

to think of this positive
and negative reinforcement

and punishment in the
context of substance abuse.

For positive reinforcement,
you might get social gain,

physiological responses
to the drug or the high.

When you're taking drugs,

you might become more popular
at the moment or you might

find a new group of

friends, that's
positive reinforcement.

You add something to
increase the behavior.

With negative
reinforcement, that

would be decreasing
your anxiety or

depression through self
medicating with a drug and

it might alleviate
social pressure

to actually engage
in taking the drug.

For positive punishment,

you add something to
decrease the behavior.

You might have health concerns,

so you might have dependency,

cirrhosis of the
liver or cancer,

you might also experience

legal troubles as a result of
taking the drugs so that'll

decrease your behavior of

taking drugs and
negative punishment,

you take away something
to decrease the behavior,

so you go to jail,

which takes away your freedom

or you lose your employment.

Another factor that contributes

to operant conditioning is

whether the reinforcement is
continuous versus partial.

Do you always receive
this reinforcement

or is it only
partially reinforced?

Partial reinforcement
is when a behavior is

reinforced periodically
rather than continuously.

Although behaviors
may be acquired very

quickly using continuous
reinforcement schedules,

they are generally more
enduring and difficult to

extinguish when they are

conditioned with
partial reinforcement.

The partial reinforcement is

called the partial
reinforcement effect.

There are different reinforcement
schedules that one may

follow in order to elicit
different behaviors.

One, you have a fixed
interval schedule.

This is when you provide
reinforcement for

the first response after

a specific interval
of time has passed.

For example, after a minute,

after a few hours,

after a few days,

the number of responses
really has no effect

on the timing of
the reinforcement.

You also have variable
interval schedules,

and this is where you
provide reinforcement

at unpredictable time intervals.

Every five minutes on average,

for example, you could also
have a fixed ratio schedule.

This provides
reinforcement after

a specific number of responses.

Let's say every five times you

hit the lever to
get food pellets.

Then you have variable ratio.

This provides
reinforcement after

an unpredictable
number of responses.

We talked about the limits
of classical conditioning.

Now let's talk about the limits
of operant conditioning.

One, we have Tolman's concept

of the cognitive map
latent learning.

In this concept,

they were the first
substantial challenge

to the view that all behavior

could be explained
through contingencies

of reinforcement and punishment.

What this means is that

without reinforcement
or punishment,

learning still takes place,

but it might not be
apparent in behavior.

Think about you
learning in this class,

it might not always take
punishment or reinforcement.

You might still learn.

The evolutionary
history of an organism

puts constraints on what it

may learn and how easily it

might be to learn certain tasks,

just like is the case with
classical conditioning.

The final learning
that we'll talk about

today is observational learning.

This is the ways in
which an organism

might learn by observing

the experiences of others rather

than through his or
her own experience.

This could occur
through reading a book,

reading a textbook, watching TV,

learning vicariously
through others,

but it could also
be through learning

of your parents'
own experiences,

your older siblings experiences.

The theories of
observational learning

became more important
as the influence

of behaviorism explained
through the 1960s and 70s,

and important concepts include

modeling and vicarious
conditioning.

With modeling, we
have this picture

which perfectly illustrates it.

The dad is shaving,
and of course,

a little kid starts
shaving as well.

This is when you learn
through imitation.

Observing and imitating
the behaviors of others is

an important part of the way
that human beings interact.

It's the way that we
learn the social norms

of our society.

It's the way that
we learn what is

acceptable and unacceptable,

or what are acceptable versus
unacceptable behaviors.

Imitation also facilitates
empathy between people.

It helps people understand and

care about how
other people feel.

It's putting yourself
in that person's shoes.

Vicarious conditioning occurs on

the other hand when
observation of

rewarding or punishing
consequences experienced by

others serves much
the same function as

being rewarded or
punished one's self.

For example, when you
were in high school,

you saw that people
that were ahead of you,

if they got good grades,

they got into college,

they got scholarships, you might

learn through observation
that their behavior,

studying a lot and

getting these good
grades led to rewards,

getting into college and
getting scholarships,

that might affect
your own behavior.

When it comes to
vicarious conditioning,

four steps are required
or four stages are

required for you to vicariously
become conditioned.

First, we have attention.

For observational
learning to occur,

one must first direct
one's attention to

a model's behavior
and to the rewards

or punishments that
follow from the behavior.

Let's say that this was
me when I was a sophomore

and I saw juniors and seniors.

I saw the groups of
kids who studied

a lot and got admitted to

college and got scholarships

versus the kids who
didn't study very much,

who often ditched
class and didn't

get scholarships are
accepted to college.

First, I'm paying attention to

those two groups of people.

Then you have retention.

Because observational
learning involves

behaving in the present as

someone else has
behaved in the past,

one must retain a memory
of the observed behavior.

Of course, I'm a
sophomore in high school,

so I have to keep

my behavior up through
junior and senior years,

so I make sure I don't
teach classes too much,

make sure that I study
and keep a high GPA so

that I can get to college
and get scholarships.

It requires some
retention of information.

Reproduction, you
must be physically

capable of reproducing
the behavior.

It must have been
physically capable of

attaining a high GPA to

get into college and to
get some scholarships.

Apart from that, really
important component

here is that you have
to have the motivation.

You might still be
motivated to perform it.

I still have the motivation to

perform at very high
levels in academics.

That's what led me

to get into college
and get scholarships.

There are other instances where

vicarious conditioning might not

occur because you might not be

motivated or you might
not be able to reproduce.

For example, another example
where I fail to do this is,

I always love the
summer Olympics,

especially women's gymnastics.

First, I paid attention.

I looked at the
model's behavior,

I watched all over the
summer gymnastics,

I learned how to do
cartwheels and all things on

my own because I love
gymnastics so much.

I retained the information,

so I must retain a memory of
the observed behavior I did.

That's why I went
and did car wheels

and learned all these
things on my own.

But then the problem came
down to reproduction.

I wasn't physically capable of

reproducing all of the
behavior on my own.

I required coaching.

Maybe it was 10 years too late.

Most gymnasts start early on in

their lives so that they
retain their flexibility.

Because I wasn't physically

capable of reproducing
the behavior,

it decreased my motivation,

so I was no longer
motivated to perform it.

Therefore, vicarious
conditioning was not successful.

On the next page, I'm
going to show you

really it's just such
an adorable picture.

This is illustrating
mirror neurons.

Mirror neurons describe
a neuron that fires

when an animal performs
a particular action,

and also when the
animal observes

the action performed
by someone else.

This person's mirror
neurons are being

activated along with this
little monkey's mirror neurons.

Mirror neurons allow
the brain to create

mental representations
of observed behavior,

allowing such behavior
to be learned.

This person is holding
a three day old monkey,

this baby is only three days old

and mimicry or
modeling is really

important for creatures to

know how to behave
in the wild right.

This monkey is observing

this human sticking
their tongue out,

only three days old,

but their mirror neurons
are already being

activated only at
three days old,

this little monkey learns
how to stick his tongue,

his or her tongue out too.

This is super adorable.

I'll post a video that
shows you more about

the different
behaviors that they

were able to elicit from
this three day old monkey.

The final learning, well,

I guess this follows,

modeling and
observational learning.

Observational learning
has often been

used as a core argument

for the relationship between
media violence and behavior.

Strong opinions exist
regarding the question of

effects of violent media,
violent video games,

violent TV shows,
violent websites,

et cetera, on aggression.

But the question hasn't really

been settled to
everyone's satisfaction.

There are some studies
that show that there's

a link between media violence,

and aggressive thoughts,
emotions, and behavior.

But this association isn't
always straightforward.

There is a difference
between possible effects of

violent media on aggression
and the effects of violence,

but there are all confounds
that exist within that.

Are people that already have

violent tendencies more likely
to watch violent media?

Probably, even with the
absence of violent media,

would these people engage
in violent behavior?

Maybe. It's not only

the fact that violent media
leads to aggressive behavior.

There are other factors like

personality factors,
the context,

the social environment in
which these people grow up

in that leads people to
engage in violent behavior.

It's not only
observational learning,

but other factors contribute
to violent behavior.

I will post this
video up and then

I will share the
other keep video

with mirror neurons
and the baby monkeys.

Enjoy the rest of your day
and I'll see you soon.