[BLANK_AUDIO]. In the last video we talked about

classical conditioning. Today, we're going to talk about Operant

Conditioning. And we're going to talk about two sort of

developments in the understanding of Operant Conditioning.

We're going to talk about the Law of Effect that was talked about by

Thorndike. And we're going to talk about Operant

Conditioning as further developed by B.F. Skinner.

Now, in classical conditioning, the main thing that was learned was a new

association. A conditioned stimulus now produces a

conditioned response, where it prior was neutral.

Here we're talking about not actually learning new behaviors with Operant

Conditioning, learning a behavior that wasn't in the repertoire behaviors prior

to the conditioning. And Thorndike's case, he actually put

cats in these boxes, that if they pulled levers and turned knobs and did kind of

things, they would open the door to the, to the puzzle box and there would be

food, food that the cat couldn't reach without opening the door to the puzzle

box. And the cat was put into the box, and it,

obviously, it explores things, it doesn't just sit there.

Then, it explores and touches things and tries to get to the food.

And eventually, through trial and error, it'll actually open the, put his paw and

open that, that switch and pull the right string and the door will open and the,

and the cat gets the food. And here we have the time that the cat

spends in the box across trials, putting in the cat in the box.

And as you can see, eventually, the cat stays in the box for a very short period

of time because it opens a door and gets it's actually has learned what it takes

to open a door to get to the food. Now, he called this The Law of Effect.

If behaviors are developed that produce pleasurable outcomes, then, the effect of

that pleasurable outcome will strengthen that behavior.

The behavior of opening the, the door to the box by turning the switch and pulling

the string. The Law of Effect.

This was further developed by Skinner in using, what's often called a Skinner box.

He called it the operant chamber. An animal was placed in this box and it

has several features to it. It has a lever that the animal can press.

Press this lever, and if, if we're conditioning the pressing of that lever

then that would lead to food being released in the food dispenser.

Animal presses the lever, gets food. And they also, things in the box that can

produce other kinds of stimuli. There's a loudspeaker to produce sounds.

There are lights of different colors to, to use as stimulus control to produce the

response only when the stimulus is present.

And there's electrified grid on the floor of the box so the shock can be

administered if you wanted to use punishment.

So, if I was just training the lever, the animal again, just like the cat in the

box, is not going to just sit there. It's going to wander around, touching

things, sniffing things. And eventually, it hits its paw, its nose

against the lever, and the lever dispenses food in the food dispenser.

And of course, then, we can actually watch, just watching that, the animal

will start pressing the lever to get to the food.

we might want to say, we might want to have discrimination.

So, we'll only provide food when it presses the lever when the green light is

on. When the red light is on the lever will

not lead to. So eventually, we'll have stimulus

control and when the green light comes on, then the animal will press the lever

and get the food. Operant Conditioning.

There's two things we can do. We can reinforce the animal or the

person. That's a stimulus that actually

strengthens behavior. Like the food for the rat chocolate for

me, something that you find pleasurable, will strengthen that behavior.

The Law of Effect. We can also have, we can also weaken a

particular behavior by using punishment. And punishment is a stimulus, a negative

stimulus that weakens behavior. Pun-, reinforcement, strengthening

behavior, punishment weakening of behavior.

So, reinforcement strengthens behavior and have positive reinforcement and

negative reinforcement. Positive reinforcement is where I present

a pleasant stimulus after the behavior occurs.

The rat presses the lever, it gets food. Negative reinforcement is when unpleasant

stimulus is removed after the behavior occurs.

Again, with strengthening behavior like pressing the lever, but now if I press

the lever then some unpleasant stimulus is removed.

For example, if the animal is being shocked and now it presses the lever, and

the shock is removed, then I'm strengthening the behavior of pressing

the lever. So, positive reinforcement, negative

reinforcement. I can also use punishment that weakens

the behavior. And I can have positive punishment, where

a presentation of unpleasant stimulus after the abhorrent behavior occurs.

So, if I press the level and then I'm shocked, then it's going to weaken the

behavior of pressing the lever. And I can also have negative punishment

and it's where I can remove a pleasant stimulus after the operant behavior

occurs. So, if in fact I'm receiving a pleasant

stimulus and now it's removed after I engage in the behavior, then that

behavior's going to be weakened. I like to think of my grandchildren

playing in my house and they're having fun playing but they're also being very,

very loud. And I might say, Okay, time out.

You're being too loud. Sit over here for five minutes in the

corner. Time out.

So, I'm removing a pleasant stimulus of playing with their, their their cousins

and by removing that pleasant stimulus then I just hope to reduce the noise

that's made while they're playing. It works sometimes but it, it's negative

punishment. Now, I also with operant condition, can

engage in very, very complex behaviors. You've seen animals do tricks at

circuses. Very complex behaviors can really be

controlled by using operant conditioning. But I have to do it by, it's, sometimes

these behaviors, like a rat doing a backflip, are not going to occur

naturally. And so, I have to somehow figure out how

to do that by shaping the an-, shaping the behavior.

The method of successive approximations. So, I might be I might use shaping.

Let's say I want to make the animal turn around completely, I might pre-, give the

animal food only when it turns away from the lever.

Then, it turns further away from the lever towards the back of the cage, then

I have to give it reinforcement. And if it turns around, so that it's

turned around far enough, so that now when I give it food, it completes the

circle. Eventually, I'll have the animals

circling the box to get the food. I'm shaping a complex behavior by using

successive approximations of the behavior I want to shape.

Now, I'd like you to Google rats playing basketball on your computer, and you'll

see some films of complex behavior. Rats actually going down the court In

shooting hoops that had been shaped by really classroom kids learning about

operant conditioning. So Google rats playing basketball and

watch those films. It's a lot of fun.

Just want to mention one other thing about operant conditioning.

Olds found that, if you put an electrode down in the hypothalamus of the brain, an

area below the cerebral cortex, the hypothalamus.

The hypothalamus is known to control motivation, feeding behavior, drinking

behavior pleasure, pleasurable kinds of things.

And I stimulate the hypothalamus, then the animal will engage in operant

behavior to get that stimulation. In fact, it's a very powerful reinforcer.

If, if the animal's hungry and I have one lever that goes to food and one lever

that goes to stimulation of the hypothalamus, then they'll engage in

pressing the lever to get the stimulation for long periods of time, thousands of

trials. it's a very powerful reinforcing event.

So, he actually said this has something sort of like the pleasure center of the

brain that I'm stimulating to produce very strong operant behavior.

Now, I want to talk a little bit about schedules of reinforcement because I can

have reinforcement at a particular schedule that produces particular kinds

of strength of behavior. I think I've mentioned earlier on the

pretest that if you reinforce the animal every trial, versus reinforce the animal

on the partial number of the trials, that re-, partial reinforcement will lead to

stronger behavior in continuous reinforcement.

And I can schdule in differnt ways. I can have ratio schedules where I

reinforce on the basis of how many responses are made, a ratio.

So, a fixed ratio for example, might be every fourth response I reinforce.

It's every time the fourth response pressing the lever, they get food.

Or I could use a variable ratio where it's on the average, on the average every

tenth response is reinforced, where there might be eight responses this time, 12

responses next time. Variable schedules lead to a higher rate

of responding than fixed schedules. And I'll show you that in a minute.

And we could also have interval schedules where I don't use the number of responses

but a response after some interval of time passes.

A fixed interval 10 second might be the first response made after a 10 second

interval is reinforced. Or it could be variable, on the average

the first response after some interval occurred will be reinforced.

And again, variable schedules lead to high rate of responding than fixed

schedules. And to show you this, I'm going to use a

cumulative recording. This is a pen on a piece of graph paper

that's running through a roller. And every time the animal makes a

response, it blips up one notch. So, the pen would go up one for this

response. Then it's flat.

And if the animal engages in rapid responses, and of course the slope of

this will get higher. Then, every time they get a reinforcer,

then it should, indicates that on the recording.

Let's look at that for the different schedules of reinforcement.

This is variable ratio, fixed ratio, variable interval, fixed interval.

And as you can see, the fixed schedules lead to not a straight rapid rate of

response. In the fixed ratio, for example, often

the animal will pause after reinforcement.

And then start making the responses to get to the ratio, the fixed ratio that's

required. And so variable ratio in the animal does

not do those pauses, and has a much higher rate of responding.

With the interval schedules, the same thing.

With the interval schedules, let's say it's a fixed schedule of ten seconds, the

animals start out at a very slow rate of responding, and then gradually increase

their rate of responding up to the point at which the reinforcement will occur.

With the variable interval schedule, you don't see those changes within the rate

of responding, you have a continuous high rate of responding.

So, variable schedules will lead to higher rate of responding than fixed

schedules, and all of these schedules will be, lead to more responding than if

you just reinforce in every trial. Thank you.

Lecture Video 1.2: Operant Conditioning

Introduction to Psychology as a Science 2 – Fundamentals of the Mind and Behavior

Meet the Instructors