Chapter 4: Conditioning & Learning Processes

Welcome to Last Minute Lecture.

This free chapter overview is designed to help students review and understand key concepts.

These summaries supplement not replaced the original textbook and may not be redistributed or resold.

For complete coverage, always consult the official text.

Welcome back to The Deep Dive.

Today, we are taking on a Solarn, truly foundational chapter in human psychology.

Our mission is to break down the science of conditioning.

This is so much more than just a chapter in a textbook.

We're really looking at the invisible architecture that governs how we acquire our most human traits.

You're not just talking about simple reflexes then?

No, not at all.

We're talking about your emotional responses, how you learn social skills, your attitudes, your values, even the very structure of your conscience and language.

If you want to understand why you feel or act a certain way, you really have to start here.

Let's unpack that because I think when most people hear conditioning, they immediately think of Pavlov's dogs, that classic simple lab setup.

Right, the bell and this elevating dog.

Our sources point out that while that early work was, of course, critical, it was also what they call zoo -centric.

It was focused almost entirely on animals under these very tight, very controlled conditions.

Infra -human, as the text puts it.

Exactly.

That traditional view came with a really rigid set of assumptions.

They just didn't hold up once psychologists started studying humans in the lab.

The model had to get more flexible?

Much more flexible.

It demanded a whole new approach to understanding how we learn.

What were those old assumptions that had to be thrown out when people became the subjects?

There were three big ones, really.

They relate to the response, the stimulus, and the learning curve itself.

Okay.

First, the old idea was that a conditioned response had to be a simple reflex.

A knee -jerk, a blink, salivation.

That's gone.

Human learning involves these big, complex, generalized responses.

And the second one.

The second was how you define a stimulus.

Used to be purely physical terms.

A certain decibel level, a specific wavelength of light.

Now a stimulus is defined by its salience and significance to the person.

Ah, so it's about what it means to you.

Exactly.

If a sound has emotional weight for you, that's what makes it a powerful conditioner.

Not just its objective, physical properties.

Your perception and your cognitive meaning, they factor in right from the very beginning.

Which must have completely changed the third assumption about the learning curve.

I'm guessing it wasn't so smooth anymore.

Not at all.

The traditional view was this nice, predictable, S -shaped curve.

But with humans, you get these sudden sharp jumps in performance.

Your own thoughts and cognitive factors can totally alter that learning trajectory.

So given that conditioning is this fundamental layer, how does it fit into the bigger picture of human learning?

The sources kind of lay out a few categories.

You have conditioning, which is what we're talking about, this basic association.

Then there's probability learning, motor skills, problem solving.

All the different ways we learn.

Right.

But what's really fascinating is that across all these different types, there's a consensus on the principles that determine learning efficiency.

So there are universal laws for learning well.

You could say that.

And they are, first, the critical role of motivation.

Whether that's a basic drive like hunger or something more like anxiety.

That's what energizes you.

And second, intrinsically linked to that is the role of reward and punishment.

Which we'll get into later as schedules of reinforcement.

Exactly.

Those are the environmental programs that control what you're likely to do in the future.

So before we dive into all that, let's just nail down the core definition.

When we say conditioning, what is the one process we are always talking about?

At its heart, it's a change in how you respond to something because of experience.

And that change can look like two things.

Either you start having a completely new response to something that's neutral before, or an existing response just gets a lot more efficient over time.

And the setup is always the same?

The typical experiment, yes.

A mild, neutral stimulus like a tone, followed very quickly by a significant stimulus like food.

Hashtag, tag, tag, tag.

The core mechanisms and terminology of conditioning.

All right.

Let's get the formal language down using that famous food and bell example.

The dog hears a buzzer.

It learns the buzzer means food.

And eventually it salivates just at the sound.

What are the official terms for those parts?

Okay.

So there are four key terms, the absolute foundation of classical conditioning.

First, you have the unconditioned stimulus or UCS.

That's the food.

It naturally, automatically causes a reflex.

No learning required.

And the reflex itself.

That's the unconditioned response, the UCR.

In this case, it's the salivation that the food causes.

It's the natural reaction.

So UCS and UCR, that's the pre -wired biological pairing, the unlearned part.

Correct.

Then you introduce the conditioned stimulus, the CS, that's the buzzer.

Originally, it's totally neutral, means nothing to the dog.

But after you pair it with the food over and over, it gains power.

Exactly.

It acquires the power to cause a response.

And that learned response, the salivation that happens just because of the buzzer is the conditioned response or CR.

And that CR, the learned response, is it identical to the original UCR?

Often it is, but it's not always.

Sometimes it's a bit weaker or maybe a slightly different form of the original response.

This all sounds like a simple A followed by B, but the timing, the temporal relationship, that seems to be absolutely everything.

It is.

The whole thing hinges on that simple property.

A followed by B, where B, the second stimulus, has to be the more intense or significant one.

That standard effective setup is called forward conditioning.

The CS comes before the UCS.

Right.

It creates a little window of anticipation.

And what happens if you flip it, if the food comes before the bell?

That's backward conditioning.

And the effects are, well, they're very different and usually much less profound.

You might create some kind of association, but it's often weaker.

The animal might even learn to relax after the food because the bell now signals a safe period where food isn't coming.

So order matters.

What are the other non -negotiable requirements for this to work?

There are three.

The first is contiguity.

The CS and the UCS have to happen very close together in time.

How close are we talking?

A really short window, usually half a second, maybe up to a few seconds.

If the gap is too long, the brain just doesn't connect the two events.

Okay.

Precise timing.

What's the second requirement?

Relative strength.

And this is the one that's a big counterintuitive.

The CS, the buzzer, it actually needs to be fairly weak.

Really?

Not allowed attention grabbing sound?

No.

A really strong CS can actually interfere with the learning.

It might distract the animal or cause its own competing reactions.

But the UCS, the food, that has to be fairly strong and adapts and stops responding to it at all.

The dog gets bored of the food.

And the whole experiment falls apart.

Okay.

And the last one seems obvious, but important.

Trials.

It's about repetition.

A single pairing of the CS and UCS is one trial.

To get a stable, reliable, conditioned response, you often need multiple pairings, sometimes hundreds of them.

Okay.

So we start the trials, pairing the buzzer and the food.

That's the acquisition phase where Leonard response gets stronger.

But of course, a huge part of psychology is about unlearning.

So how do you undo it?

That's the extinction phase.

And it's pretty simple.

You just stop presenting the UCS.

You ring the buzzer over and over, but the food never comes.

And the condition salivation just starts to fade away.

It gets weaker, less probable, and eventually it stops altogether.

Now, this is a really important point in the sources.

Extinction is not the same as forgetting.

Absolutely not.

It's a critical distinction.

Forgetting implies the memory is just gone.

Extinction is more like suppression.

And the proof for that is this idea of spontaneous recovery.

Exactly.

So you've extinguished the response.

The dog no longer salivates to the bell.

You take it out of the lab, give it a rest for a couple of hours, and then bring it back.

The first time you ring that bell.

The salivation comes back?

It pops right back, yeah.

Usually not as strong as it was at its peak, but it reappears.

This tells us the original learned connection isn't gone.

It was just being actively inhibited by a new learning.

The bell means no food.

When that inhibition fades a little during the break, the original learning shows through again.

Okay, that makes sense.

But then there's an even weirder phenomenon that seems to completely break the rules of extinction.

It's called incubation.

Ah, yes.

Incubation is profoundly counterintuitive, and it's so relevant for understanding human anxiety.

So how can presenting the CS without any reinforcement sometimes make the response stronger?

Instead of decrementing, it increments.

It grows.

This is most likely to happen after some kind of traumatic conditioning, where the initial unconditioned response, the pain or the fear, was exceptionally strong.

So let's say an animal gets one massive electric shock after a buzzer.

From then on, hearing the buzzer alone, with no more shocks, makes it more scared over time, not less.

Precisely.

The mechanism seems to be that the conditioned fear response, the CR, is itself so awful, so aversive, that it starts to function as its own reinforcement.

Wow.

The buzzer makes the animal feel terrified, and that feeling of terror reinforces the connection between the buzzer and terror.

It creates this vicious positive feedback loop.

It's literally, as the saying goes, having nothing to fear but fear itself.

That's a perfect way to put it.

The anxiety becomes its own fuel source.

It's a terrifying psychological loop, and it explains so much about why human phobias can be so persistent and even get worse over time.

Okay, so let's shift from timing to the stimulus itself.

If we train a dog to respond to a 1 ,000 cycle per second tone, what happens when we play an 800 CPs tone or a 200 CPs one?

Now we're talking about stimulus generalization.

This is where an organism learns to respond not just to one specific thing, but to a whole class of similar things.

So the 800 CPs tone would probably get a reaction.

Almost certainly, because it's very similar to the original.

But the 200 CPs tone is so different that it would likely be ignored.

The strength of the response varies inversely with how dissimilar the new stimulus is from the original.

And that relationship has a name?

It's called the generalization gradient.

The response is strongest right at the original training stimulus, and it drops off as you move away from it.

It shows that learning is flexible.

There's a really clear example of this in the sources, figure 4 .2, with pigeons.

Can you walk us through that?

It really helps visualize the idea.

Sure.

So in this experiment, pigeons were trained to peck a key that was lit with a very specific color of light,

say a wavelength of 530 millimicrons.

There's a particular shade of green.

They only got food for pecking that exact color.

Okay.

Then in the testing phase, the researchers showed them a range of different colors, some very similar to the training color, some further away.

And the pigeons didn't just peck at all the green lights equally.

You were more specific.

Very specific.

The graph shows this beautiful symmetrical peak.

The highest number of pecks happened exactly at the 530 millimicron color they were trained on.

As the color shifted away from that hue, in either direction, the number of pecks dropped off predictably.

It's a perfect visual of that generalization gradient.

So learning is precise, but it's also adaptable enough that you don't get stuck on tiny irrelevant details.

Exactly.

So what's the opposite of that?

How do we learn to tell two very similar things apart?

That's differential conditioning, or what's also called condition discrimination.

This is where the organism learns that it's actually better or more efficient to tell similar stimuli apart.

How would you train that?

Let's say you have a red light and an orange light.

At first, due to generalization, the animal responds to both.

To train discrimination, you would consistently follow the red light, CS1, with food.

But you present the orange light, CS2, all by itself, with no food.

So one is reinforced, the other isn't.

Exactly.

The response to the red light stays strong, while the response to the orange light extinguishes.

The animal learns to differentiate.

This is how we fine -tune our behavior, learning the difference between a real opportunity and a false alarm.

Measuring and evaluating conditioned responses.

When we talk about measuring these responses, it's not always so simple, is it?

An air puff to the eye is one thing, but the sources make a point that a stimulus, like an electric shock, produces this huge generalized pain fear reaction.

That's a crucial point.

A shock doesn't just trigger one response.

It triggers this massive, complex cascade of both skeletal and autonomic reactions.

You get thrashing, screaming, your heart rate goes through the roof, you start sweating.

Goose bumps, even.

Pylorection, yeah.

It's this whole -body emergency response.

And for practical reasons, most studies just pick one or two of those things to measure, like heart rate or the skin's electrical response.

But we have to remember, it's part of a much bigger holistic pattern.

And it's not just external stimuli, right?

There's this fascinating area of research on conditioning that happens entirely inside the body.

Interreceptive conditioning, yes.

This is where one or both of the stimuli, the CS or the UCS, are applied directly to our internal organs, the viscera, your stomach, heart, intestines, instead of our external senses.

Can you give us an example?

The Soviet research on this sounds pretty wild.

It was pioneering work.

They had human subjects, for instance, who swallowed a tube.

The neutral CS might be an external word, like white.

The UCS would be an injection of warm water through the tube, which causes an internal vascular response called vasodilation, the widening of blood vessels.

And they could condition that internal response.

They could.

After pairings, the word white alone could start to trigger that internal vasodilation.

It proves that conditioning isn't limited to what we see and hear.

Our bodies can learn these associations on a deep physiological level that we're not even aware of.

Which could explain a lot about things like tannic attacks, where an internal feeling, like a heart palpitation, gets conditioned to some external cue.

It's a very powerful model for understanding psychosomatic issues, yes.

So when it comes to the hard data, how do psychologists actually quantify the strength of a conditioned response?

Well, the simplest way is just frequency.

On any given trial, did the response happen or not?

But beyond that, we look at the specific characteristics of the response.

How big was it?

That's the amplitude.

How long did it last duration?

And when did it happen?

The latency.

The sources have a diagram, figure 4 .3, that lays out the timeline of a single trial.

It really helps clarify where the learning part shows up.

Right.

That schematic is really helpful.

It plots the level of activity over time.

You see the baseline and the CS comes on.

And what you're looking for is the anticipatory CR.

The learned response starts to rise and reaches its peak before the UCS is even delivered.

So the key is that it happens in advance.

That's the whole point of the learning.

The diagram then shows the UCS onset and the UCR, the unconditioned response follows that.

But that CR happening before the UCS is the proof that the organism has successfully learned to predict what's coming.

And maybe the most compelling measure of all is how efficient or adaptive the response is.

Figure 4 .4 with the eyelid response illustrates this perfectly.

It really does.

It's a tale of two responses.

In part A of the figure, you see a small, poorly timed CR.

The eyelid barely flutters.

So when the air puff, the noxious UCS hits, the eye is wide open.

The response is basically useless.

It doesn't protect the eye at all.

Not at all.

But then you look at part B, here you see a big, robust, perfectly timed response.

The learned blink is so well -placed that the eye is fully closed at the exact moment the air puff would have hit the cornea.

That is a successful adaptive response.

It's the definition of it.

The organism has used the signal, the CS, to completely neutralize the threat.

This is exactly what Pavlov was getting at.

Conditioning allows an organism to deal more efficiently and safely with its environment.

Okay, so we have the foundation of class of conditioning down.

Now we need to put it in context with the other two major paradigms.

Let's start by recapping the core of classical.

The core mechanism is that the CS acts as a signal.

It creates an expectancy, allowing the organism to anticipate what's coming.

And crucially,

the subject's role is totally passive.

The experimenter controls everything, and the subject's response doesn't change the outcome.

They just react.

But that all changes with instrumental conditioning.

It completely changes.

Now the subject is active.

They're free to act on their environment, to modify it.

Their behavior has consequences.

They can work to get a reward, or more commonly in these studies, to avoid something unpleasant.

The sources use that really powerful real -world example from World War II, the air raid sirens in London.

It's a perfect illustration.

The siren is the CS, the bombs are the UCS, which causes intense fear, the UCR.

After learning, the siren alone causes the fear, the CR.

But the instrumental part is what you do next.

You run for the shelter.

Right.

And that behavior running is reinforced because it allows you to escape the immediate danger or avoid the trauma altogether.

Which brings up that key distinction in instrumental learning.

Escape versus avoidance.

Right.

An escape response is when you're already in the bad situation, the shock is on, and your action makes it stop.

An avoidance response is much smarter.

Your action prevents the bad thing from ever happening in the first place.

You hear the siren and get to the shelter before the bombs fall.

The text describes this classic, if pretty brutal, experiment with a dog in a hammock to show how that learning happens.

It's a very vivid setup.

The dog is in a harness, electrodes on its paws, a tone, the CS, comes on for five seconds, and then it gets an intense electric shock, the UCS.

And the initial reaction is just panic.

Total panic, thrashing, screaming, and a massive autonomic response.

The text says its heart rate jumps from around 100 beats per minute to 240.

It's a huge physiological trauma.

But eventually it does something that works.

In its thrashing, it accidentally hits a panel, which immediately shuts off both the tone and the shock.

That is the first crucial escape response.

And over time, it gets more efficient.

Much more.

After about 16 trials, the dog learns that as soon as the tone starts, it can quickly turn its head and push the panel.

That action terminates the tone and prevents the shock from ever being delivered.

That is the successful shift to an avoidance response.

And to really prove that the avoidance part is key, there's another experiment in Figure 4 .5 with guinea pigs.

Yes, they compare two groups.

The avoidance group only got shocked if they didn't run in a wheel when a buzzer sounded.

Their action had a consequence.

But the non -avoidance group got shocked no matter what they did.

Running was irrelevant.

And the results were completely different.

Dramatically.

The graph shows the avoidance group quickly learned to run and reached a high level of performance.

The non -avoidance group, their performance was erratic.

They were still agitated by the buzzer.

They showed fear.

But they never learned to run consistently because it never did them any good.

The contingency, the fact that their action mattered, was everything for learning.

And this is all explained by the two -process theory of avoidance.

Right?

Which bridges the two types of conditioning.

Step one is classical.

The buzzer gets paired with the shock, so the animal learns to feel fear when it hears the buzzer.

That's the conditioned anxiety.

And step two?

Step two is instrumental.

The animal learns what to do to get rid of that awful feeling.

The action, like pressing the panel, is reinforced because it reduces the fear.

So it's not the shock they're avoiding, really.

It's the feeling of anxiety.

In the avoidance phase, yes.

The reinforcement is pure drive reduction.

You're reducing that terrible emotional state.

But this leads to that huge puzzle we talked about.

The problem of persistence.

Why do animals keep performing this avoidance behavior for hundreds of trials long after the shocks have stopped?

The fear should extinguish.

This is the core paradox of anxiety.

The avoidance response itself prevents the conditions needed for extinction.

How so?

Well, for the fear to extinguish, the animal needs to experience the buzzer without the shock.

But if it successfully avoids the shock every single time by quickly turning off the buzzer, it never gets that long clear exposure to the buzzer means nothing reality.

So the avoidance behavior protects the fear from being challenged by reality.

It keeps the conditioned anxiety perfectly intact.

So the behavior continues, sometimes indefinitely, long after the actual danger is gone.

Okay.

Now for the third and maybe most influential paradigm,

operant conditioning.

And for that, you have to talk about B .F.

Skinner.

He made this key distinction.

He said there are respondents, which are behaviors elicited by a known stimulus like a reflex, and then there are operants.

And operants are different how?

They're spontaneous.

They aren't caused by a specific trigger.

They're just behaviors that an organism emits that operate on the environment to produce some kind of consequence.

And the whole system is based on that consequence.

Exactly.

The process is called reinforcement.

You present a reinforcer right after a response, and that increases the probability that the response will happen again.

And this allows for shaping, which is how you train really complex behaviors.

Right, but reinforcing successive approximations.

You can't wait for a rat to spontaneously use a marble vending machine, but you can reward it for looking at the machine, then for touching it, then for touching the lever, and so on.

You gradually shape the behavior.

And Skinner applied this everywhere.

Everywhere.

He analyzed language as operant behavior.

He looked at economic control.

Your salary is a reinforcement schedule.

And most famously, he developed behavior modification.

Which is a clinical application.

Yes.

It completely ignores the why and focuses on the what.

What is the undesirable behavior, and what is the desirable behavior we want to reinforce?

The treatment is just changing the contingencies of reinforcement in the environment to extinguish the bad and strengthen the good.

It's direct, measurable, and often very effective.

Okay, let's dig into motivation.

The sources are clear that it's a huge factor in how well we learn.

How did early theories connect motivation and reinforcement?

The early theories were very biological.

They're all about drives.

These internal physiological states like hunger or thirst.

The idea was that a drive creates this restless, unpleasant state of activity.

And learning happens when you reduce that drive.

Exactly.

The consummatory action -eating drinking terminates that unpleasant state.

And that termination is inherently rewarding.

It reinforces whatever behavior led to it.

A nice, neat biological loop.

But it couldn't explain everything?

Not even close.

Especially when you look at more complex animals.

The evidence started piling up for motives that had nothing to do with reducing a biological deficit.

This is the beyond biological drives idea.

What kind of evidence are we talking about?

Well, things like curiosity and exploration.

Monkeys will solve complicated puzzles just for the reward of being able to look out a window at something interesting.

Rats will run on a wheel just for the activity itself.

They're seeking stimulation, not reducing hunger.

And then there's the absolutely classic Harlow study with the baby monkeys that really changed the conversation.

It was a bombshell for the simple drive reduction theory.

Harlow was studying what he called contact hunger.

He gave baby monkeys two surrogate mothers.

One was just a wire frame, but it had a bottle with milk.

It met all the biological needs.

And the other?

The other was covered in soft terry cloth, but it had no milk.

It offered only comfort.

And the monkeys chose the comfort?

Overwhelmingly.

They spent almost all their time clinging to the cloth mother and they ran to it for security when they were scared, even if the wire mother was the one that fed them.

It was a powerful demonstration that needs like affection and security can be even stronger motivators than basic biological survival.

So if motivation can come from all these different places, what's the general rule about how it affects learning?

The general conclusion, which is pretty robust, is that the higher the drive level, the speedier and better the learning.

More hunger, a more intense shock.

It just energizes the whole process and leads to faster acquisition.

But this brings up a classic debate.

The learning versus performance problem.

Does motivation just flip a switch on a behavior you've already learned?

Or does it actually help you learn it in the first place?

That's a deep theoretical question.

A lot of theories are built on the idea of homeostasis that we're always trying to get back to a state of balance.

In that view, drive reduction, getting rid of that pain or hunger or anxiety is the fundamental reinforcing event in all of learning.

But Skinner, as usual, had a much simpler, more external view.

He did.

Skinner's perspective on reinforcement is completely non -theoretical about what's happening inside.

For him, a reinforcer is just an environmental event that follows a response and makes it more likely to happen again.

That's it.

If it works, it's a reinforcer.

It's a purely operational definition.

Purely operational.

And he divided them into two types.

Primary reinforcers are things that are innately rewarding, like food or water.

And secondary reinforcers are things that get their power through association.

Like money or praise.

Exactly.

Money is the classic example.

It has no inherent value, but it becomes incredibly powerful because it's associated with all the primary reinforcers it can buy.

And there's a great study that shows this in action with chimpanzees.

Wolf's study, yes.

He taught chimps to work for poker chips.

They would perform tasks to get these tokens, and then they could take the tokens and put them in a machine to get the primary reward, which was food.

So the poker chips became a form of money for them.

They did.

The chips became powerful secondary reinforcers, and the chimps would work hard just to accumulate them.

It's a perfect model for how generalized social reinforcers, like attention and approval,

shape so much of our complex human behavior.

Hashtag, tag, tag, tag, tag six.

Schedules of reinforcement and neural mechanisms.

So it's not just if you get reinforced.

It's when and how often.

This is the whole field of schedules of reinforcement.

And these rules are incredibly powerful for controlling and predicting behavior.

The basic distinction is between continuous reinforcement, where you get a reward after every single response.

And intermittent or partial reinforcement, where you only get it sometimes?

Right.

And that leads to this really counterintuitive finding called the partial reinforcement extinction effect, or P -R -E -E.

It's a huge deal for understanding why bad habits are so hard to break.

And figure 4 .6 in the sources shows this perfectly with the conditioned eyelid response.

It does.

The graph shows what happens during the extinction phase.

The groups that got 100 % continuous reinforcement during learning, they extinguish really fast.

The moment the reinforcement stops, their response rate plummets.

They learn that no puff of air means the trial is over.

Right.

But the group that only got reinforced 50 % of the time during learning, they are far more resistant to extinction.

Their response rate stays higher for much longer.

So getting less reinforcement makes the behavior stronger.

Why?

It's all about expectancy.

If you're used to being rewarded every time, the first time you're not, it's a huge signal that the rules have changed.

But if you're used to long periods of no reward that are eventually followed by a reward, you learn that persistence pays off.

You just keep going, expecting it will eventually come.

Let's break down the main types of these intermittent schedules.

OK, so first you have ratio schedules, which are based on the number of responses.

A fixed ratio, or FR schedule,

gives you a reward after a set number of responses.

FR10 means you get rewarded after every 10th lever press.

Like getting paid for every 10 widgets you make.

Exactly.

It produces a high rate of work with a little pause right after you get the reward.

And then there's the one that explains gambling.

The variable ratio, or VR schedule.

This is where the reinforcement comes after an unpredictable number of responses that averages out to a certain number.

You might get a reward after one press, then 50, then 5.

You never know when it's coming.

And that's why it produces the highest, most steady rates of responding, and is incredibly resistant to extinction.

Slot machines are the perfect real -world example of a VR schedule.

OK, so now let's shift gears completely and look at the brain.

What did Olds and Milner find in 1954 about the neural basis of reinforcement?

This is a groundbreaking discovery.

They set up what they called a do -it -yourself situation, which is shown in Figure 4 .7.

They implanted electrodes in a rat's brain,

and the rat could press a lever to deliver a little jolt of electricity to that spot.

Brain self -stimulation.

Right.

And what they found was astonishing.

When the electrodes were in certain areas, particularly these midline regions like the hypothalamus, the rats would press the lever compulsively.

How compulsive.

We're talking rates between 205 ,000 presses per hour.

They would choose the stimulation over food, water, even sex.

They would cross painful electrified grids just to get to the lever.

It was more rewarding than basic survival.

That's incredible.

It's like they found the brain's pleasure center.

That's essentially the hypothesis.

That the stimulation was directly activating the neural circuits that are normally associated with the satisfaction of our most basic drives.

It was a huge step in understanding the biology of reward.

All right, so all this detailed work with animals seemed to promise these universal iron -clad laws of learning.

But when researchers shifted their focus to humans,

that strict stimulus response view started to run into some big problems.

A huge challenge, yes.

And the problem is the overwhelming role of our subjective experience.

Our thoughts, our feelings, our expectations.

Animal studies often use severe deprivation or pain, which kind of overrides the cognitive factors.

But with humans, where you can only use mild discomfort, cognition becomes a huge player in the game.

Human subjects are not passive.

We're thinking about the experiment.

We're absolutely not passive.

We're trying to figure it out.

We verbalize what we think is happening.

We adopt these cognitive sets, a positive or negative attitude, and we develop expectancies.

You simply cannot ignore what the human subject thinks is going on.

And this cognitive influence isn't the same for all types of responses, is it?

No, and that's critical.

Our autonomic responses, heart rate, sweating, are very strongly affected by what we think.

If you tell me I'm about to get a shock, my heart rate will shoot up immediately, even since it's not true.

But simpler reflexes are less affected.

Right.

A simple blink response to an air puff is a bit more automatic and less influenced by my verbal instructions to myself.

So when a person finally becomes aware of the rule, aha, the light means a shock is coming, what happens to their learning curve?

It doesn't curve.

It jumps.

You often see this sudden sharp increase in responding.

It's not the gradual acquisition you see in animals.

The moment of awareness can instantly establish or even abolish a conditioned response.

And this was tested directly in an experiment by Nichols and Kimball, shown in Figure 4 .8, where they manipulated the instructions they gave to people.

Yes, this is a great study.

They had two groups in an eyelid conditioning experiment.

The physical setup was identical for both.

The only difference was the instructions.

The facilitatory instructions group was told to just relax and let your reactions take care of themselves.

They were told to just let it happen.

But the inhibitory instructions group was told to actively concentrate on not blinking.

They were given a cognitive set to suppress the response.

And the results showed a huge difference.

A huge difference.

The facilitatory group conditioned much better and reached a much higher level of response.

What's interesting is that the two curves on the graph separate early and then run parallel.

What does that mean?

The interpretation is that there are two things happening.

There's a rapid cognitive facilitation or inhibition that gets established very quickly.

And then there's the slower, more traditional conditioned reflex that develops at the same rate for both groups.

But it's starting from a higher or lower baseline, depending on the instructions.

So this all feeds into the big awareness debate.

Do you have to be consciously aware of the connection for conditioning to happen?

The current thinking is that no, awareness is not strictly necessary.

We have good evidence, like with the interoceptive conditioning, that it can happen below the level of awareness.

However, awareness has a huge effect on the rate of learning.

It's like a cognitive accelerator or brake on the whole process.

And we can even learn fear without any direct experience at all through vicarious conditioning.

Right, or no trial learning.

This is when you acquire an emotional response just by observing someone else being conditioned.

If you watch someone get shocked every time a buzzer sounds, you might start to feel anxious when you hear the buzzer, even though you were never shocked yourself.

How does that work?

How does the fear transfer?

The mechanism seems to be these internal mediational responses.

You're creating imaginal and verbal representations of what you're seeing.

You think, wow, that looks painful, or that buzzer is bad news.

Those internal thoughts and images are what allow the emotional response to be transferred through observation.

Hashtag, tag, tag, tag, tag, eighth.

Individual differences and the persistence of fear.

So in human studies, you can't really use food deprivation to create motivation.

Instead, researchers often look at a person's baseline anxiety level.

And the sources detail two major competing personality theories that connect anxiety to how easily we're conditioned.

Right, two very different ways of looking at it.

The first is Hans Eysenck's big model, his inhibition theory.

He argued that our major personality traits, things like neurosis or criminality, come from fundamental differences in how our brains produce excitation and inhibition.

And this ties directly into his theory of extroversion and introversion.

It does.

His theory is that extroverts have brains that generate central inhibition more quickly and strongly.

They get bored or desensitized to stimulation faster, which leads to impaired learning and their more sensation -seeking disinhibited behavior.

Whereas introverts are the opposite.

Introverts, he postulated, have a chronically higher level of cortical arousal and generate less inhibition.

They're more vigilant, more easily aroused, and therefore they should be more easily conditioned.

And figure 4 .9 shows an experiment testing this.

It does.

It compares introverts and extroverts in an eyelid conditioning task.

Under favorable conditions, a strong air puff, reinforcement on every trial, everybody conditions pretty well.

But under unfavorable conditions, a weak puff, partial reinforcement, the extroverts are significantly impaired.

Their high inhibition seems to shut down the learning process.

But the introverts keep on conditioning.

They do.

They condition well under both sets of conditions, which supports the theory that their higher baseline arousal makes them more susceptible to conditioning.

Okay, so that's Isenc's theory.

What's the other major one?

That would be Spence's anxiety drive theory.

He took the opposite approach.

He saw anxiety or emotionality as a positive drive that energizes behavior.

So more anxiety means more drive.

Exactly.

He measured anxiety using the Taylor -Manifesta anxiety scale, and his prediction was simple.

Anxious people should condition more rapidly than non -anxious people because their high internal drive energizes the whole learning process.

And there's data to back this up in figure 4 .10.

A lot of data.

The figure clearly shows that subjects who score high on the anxiety scale show a consistently higher percentage of conditioned eyelid responses across all the trials compared to the low anxiety subjects.

It's strong support for the idea that anxiety can act as a facilitator for this type of learning.

Let's circle back to that critical problem of why conditioned fears are so hard to get rid of.

We touched on the idea of incubation, where the fear can actually grow over time.

Right, and Isenc had a cognitive explanation for this.

He suggested that with neurotic fears, the conditioned response, the feeling of fear itself, is so unpleasant that it acts as its own reinforcement for the stimulus.

You get stuck in that fear the fear loop.

And the physiological data on this from the Polkov shown in figure 4 .11 is genuinely alarming.

It's one of the starkest examples you'll ever see.

The graph shows the blood pressure response in dogs after just one single traumatic conditioning trial.

After that, they were just exposed to the CS, the tone by itself, the standard extinction procedure.

But the response didn't extinguish.

It did the opposite.

It grew dramatically.

Their blood pressure response went from a normal change of 30 or 40 millimeters of mercury up to a hypertensive crisis of 190 to 230 millimeters.

And this heightened response could last for over a year.

Just from being repeatedly exposed to the trigger without the original trauma?

Yes.

The fear response was feeding on itself and growing stronger with every repetition.

It's a terrifying model for how PTSD and phobias can become these self -sustaining pathologies.

Are there other theories for why avoidance is so persistent?

Zelligman and Johnston offered one based on safety signals.

They argue that it's not just about avoiding the fear.

It's that the successful avoidance response creates a feeling of safety.

The CS, when paired with the absence of shock because you responded, becomes a positive signal of safety.

And that feeling of relief positively reinforces the avoidance behavior.

Finally, the sources make a point of broadening what a UCS can be beyond just physical pain.

Yes, the idea of frustrating non -reward.

This is the anger and anxiety that you feel when an expected reward doesn't show up.

And that feeling of frustration physiologically acts a lot like fear or punishment.

It can act as a powerful drive to motivate behavior, which is really important for understanding human disappointment and motivation.

So if these conditioned fears are so incredibly persistent, reinforced by the very behaviors we use to cope with them, the therapeutic challenge seems enormous.

How do you actually get them to extinguish?

Well, most of the effective techniques are direct applications of the extinction principles we've been talking about, just applied in a controlled therapeutic setting.

The first and most direct is forced exposure or restraint.

Which sounds exactly like what it is.

It is.

You expose the person to the feared stimulus, the CS, and you physically or verbally restrain them from performing their usual avoidance response.

You make them stay in the situation and confront the fact that the feared outcome, the UCS, doesn't actually happen.

They're forced to test their fear against reality.

And the anxiety, after an initial spike,

dissipates pretty rapidly, once they see the danger signal is false.

A gentler approach, and a very famous one, is systematic desensitization.

Right, this is a form of counter conditioning.

The theory is that anxiety is a conditioned sympathetic nervous system response fight or flight.

The treatment aims to condition a new, opposite response from the parasympathetic system, which is deep relaxation.

So you replace fear with calm.

Gradually.

The patient creates a hierarchy of their fears, from least scary to most terrifying.

Then, while in a state of deep relaxation, they imagine or confront each item on the list, starting with the easiest, until they can remain relaxed, even when confronting their biggest fear.

And then there's the intense version of that.

Implosion or flooding.

This is the opposite of gradual.

It involves intensive, prolonged exposure to the most anxiety -producing cues you can imagine, often just through guided imagery.

You stay with that maximal fear until it naturally starts to burn out and extinguish on its own.

And finally, there are the more direct operant procedures.

This is pure behavior modification.

The therapist identifies the contingencies of reinforcement in the person's life and changes them.

They apply reinforcement for desired behaviors,

and remove it for undesirable ones, effectively shaping new, more adaptive ways of acting.

But the sources acknowledge that treating fear in humans is complicated, because fear isn't just one thing.

Exactly.

Human fear is this complex concept that involves at least three partially independent systems.

You have the autonomic system, heart rate, sweating.

You have the skeletal motor system, the actual overt avoidance behavior.

And you have the cognitive perceptual system, which you say you're feeling, your appraisals.

And they don't always line up.

That's the problem.

A person might say, I'm not afraid of flying anymore, but they still refuse to book a ticket, and their heart pounds if they see a plane.

You can get a change in one system without getting a change in the others.

Which leads to the idea of multi -system training.

Right.

The most sophisticated approaches argue that you have to target all three systems.

You might use relaxation or biofeedback to lower the autonomic arousal, use behavioral rehearsal to retrain the motor responses, and use cognitive therapy to change the verbal sets and attitudes.

You have to treat the whole fear, not just one piece of it.

Hashtag, hashtag, X.

Conditioning of attitudes, values, and language.

So conditioning isn't just for reflexes and fears.

It's actually fundamental to how we form our social attitudes and beliefs.

The Rosran studies on this from the 1940s are fascinating.

They're a powerful demonstration.

He took neutral socio -political slogans and had people rate them.

Then he paired one set of slogans with a really pleasant experience, eating a free lunch.

A pleasant UCS.

Right.

And he paired another set of slogans with a deeply unpleasant UCS -inhaling putrid awful odors.

And the feeling from the lunch or the smell transferred to the slogan.

It did.

The slogans that were paired with the free lunch got significantly higher approval ratings afterwards.

The ones paired with the bad smells got lower ratings.

It was a clear demonstration of the classical conditioning of a political attitude.

That's kind of disturbing.

It suggests our political beliefs can be swayed by something simple as whether we're comfortable or uncomfortable when we hear an idea.

It shows just how powerful and subtle these associative processes are.

And stats and stats showed something similar with the conditioning of verbal meaning itself.

How do you condition the meaning of a word?

Well, you condition the emotional response to it.

They show that if you consistently pair a neutral nonsense word with negative emotional stimuli like mild shocks or unpleasant images, that word itself will start to elicit a negative emotional feeling.

The emotion gets baked into the symbol.

And those words then become reinforcers themselves.

Exactly.

Words like good or beautiful become powerful secondary reinforcers.

Words like ugly or wrong become punishers.

This is how language gets its emotional power.

And all of this supports the general principle of effective conditioning.

Which basically states that any neutral stimulus, a person, a symbol, an idea, that is consistently paired with a strong feeling of liking or disliking will tend to start eliciting that same feeling on its own.

This is the root mechanism of how biases and prejudices form.

So we're moving way beyond simple bells and lights here.

We have to.

With humans, a stimulus isn't just a physical event.

It conveys information and meaning.

And a response isn't just a muscle twitch.

It includes our personal experience and our judgment whether it's good or bad.

The most complex and important human learning happens at that level of meaning and feeling, not just physical association.

Theoretical problems and future outlook.

Okay, so given all this complexity, especially with humans, it's clear that the traditional simple models of conditioning face some big theoretical problems.

One of the biggest is just trying to unify the three paradigms we talked about.

Right.

The lines between classical, instrumental, and operant conditioning are really blurry in

There's still no clear agreement on the best way to distinguish them.

The sources suggest that maybe we need to stop focusing on the simple experimental setup and start classifying them based on the organism's strategy or the specific type of reinforcement involved.

And what about the hunt for the physical basis of conditioning in the brain?

The idea that learning creates a new physical pathway.

That specific idea of a new direct wire being formed from the sensory part of the cortex has gotten very little support.

In fact, research shows that you can form crude conditioned responses even without the higher brain centers.

The cortex seems to be necessary mostly for the fine -tuning, the differentiation and refinement of the response.

A huge challenge to the whole field came from mythology, from studying animals in the wild.

The old conditioning researchers worked from this assumption of a tabula rasa, a blank slate.

They did.

They assumed that any response could be conditioned to any stimulus with equal ease.

Anathologists just showed that's completely untrue.

Evolution predisposes animals to learn certain connections much more easily than others.

This is the concept of biological preparedness.

Exactly, from Seligman and Hager.

An animal is biologically prepared to associate a taste with nausea, because that's relevant to avoiding poison in the wild, but is very unprepared to associate a sound with nausea.

And for humans, this might explain why we seem to have these evolutionarily relevant phobias.

It could.

It's much, much easier to condition a fear of snakes or spiders or heights than it is to condition a fear of flowers or electrical outlets, even though outlets are arguably more dangerous in our modern world.

Our evolutionary history gives certain stimuli a head start in the fear learning race.

So finally, the biggest question of all.

What is actually learned?

Is it just a mechanical SR link?

The field has moved decisively away from that.

Current theories all incorporate some kind of mediation processes, things that happen between the stimulus and the response expectancy, a cognitive map, thoughts, feelings.

The organism isn't a simple switchboard.

It's an active information processor.

So the conclusion is that while the traditional zero -centric theories were essential, they just weren't enough for us.

They weren't.

Applying these ideas to complex human problems forced the field to grow.

And it became clear that you can have a satisfactory theory of learning without fully integrating the critical role of cognitive and subjective factors.

The human mind is just too complex for a purely reflexive explanation.

Hashtag tag tag outro.

So let's just kind of wrap up this deep dive.

Conditioning is really the core mechanism for how we learn so much, from our most basic fears to our complex social values.

And those three paradigms give us the framework.

Classical conditioning is all about anticipation and signals.

Instrumental is about acting on the environment to change the outcome.

And operant is about strengthening behavior through consequences.

But the big takeaway, especially for us, is that you have to layer the immense power of human cognition, our thoughts, our expectations on top of all of that.

That's the crucial step, yes.

And that distinction between the persistence of a learned fear and what it takes to extinguish it is really striking, especially that incubation effect where the fear can actually grow over time, reinforcing itself without any new trauma.

The anxiety literally feeds itself.

And that process is always running in the background, whether we're consciously aware of what's being paired with what or not.

So here's our final thought for you to take away from this.

Think about the avoidance behaviors you have in your own life.

It could be anything.

Social situations you avoid, tasks at work, maybe even things related to diet or exercise.

What previously harmless cues in your environment might have become these powerful self -reinforcing emotional triggers simply because your successful avoidance has prevented that fear from ever being challenged and extinguished.

A powerful lens through which to see our own patterns.

We hope this deep dive helps you navigate some of those unseen mechanisms.

Thank you for joining us.

We'll see you on the next deep dive.