Chapter 16: Cognition, Action and Self-Control

Welcome to Last Minute Lecture.

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

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

For complete coverage, always consult the official text.

Okay, let's unpack this.

For this deep dive, we are getting into something really, really crucial in the world of cognitive science.

A critical note, really, that challenges some of the core arguments and even the way people talk about Fourier cognition.

That's right.

Embodied, embedded and acted and extended.

And our main source here is Chapter 16 from the Oxford Handbook of Fourier Cognition.

It's by Sven Walter, and it's called Critical Notes on Cognition, Action and Self -Control.

And it's so interesting because he's not trying to just tear the whole thing down.

It feels more like he's trying to sharpen it.

Exactly.

His whole mission is to push the field toward a greater philosophical rigor.

He wants it to be more empirically relevant, not just a bunch of cool ideas.

So he takes these four other chapters from the same section of the handbook and kind of uses them as case studies.

Precisely.

To show what he thinks are these systemic flaws in the whole Fourier debate.

And he flags two big dangers right at the start that he thinks philosophers especially fall into.

Okay, what's the first one?

The first one is this sort of intellectual overexcitement.

He says philosophers get a little too impressed by some juicy empirical result from, say,

neuroscience or psychology.

And they forget their main job.

They forget their main job, which is rigorous step -by -step argument.

They see an interesting finding and they just leap to this huge philosophical conclusion like the mind is extended without really building the logical bridge to get there.

So they're choosing the exciting story over the airtight logic.

Yep.

And the second danger is almost the complete opposite.

It's getting bogged down in arguments that are just about words.

The philosophical concept -mongering as he calls it.

That's the phrase.

And it's a classic trap.

Spending all your time arguing about the labels.

Is this process extended or is it distributed?

Is this a case of constitution or just causation?

And while they're arguing about the words, they're missing the actual science.

They miss the empirical lessons entirely.

They're not actually advancing our understanding that the system works.

It's just word games.

So he's calling for this balance.

You have to be grounded in the science, but you also have to maintain that philosophical excellence.

And through that critical lens, he finds this huge gaping hole in the literature, especially when it comes to action.

A piece that's just missing.

A huge piece.

The topic of self -control.

Walter basically points out that four E -theorists have talked endlessly about perception and action and embodiment, but they almost never touch on volition.

Volition meaning how we actually make ourselves stick to a plan.

Exactly.

How we follow through on our long -term goals when there's a short -term temptation sitting right in front of us.

Which is something we tend to think of as the most internal process imaginable, right?

A battle of willpower happening entirely inside your head.

That's the traditional view.

Total intracranialism.

But Walter's big argument, and we'll get into this later, is that all the recent research suggests self -control is fundamentally an embodied and embedded capacity.

So you use your body and your environment to control yourself.

Yes.

And if 4E wants to be a complete theory of action, it has to grapple with this.

That's really the engine driving this whole chapter.

Fantastic.

Okay, so before we dive into his critiques, we should probably set the stage with the four chapters he's responding to.

Let's start with the first one.

It's about predictive processing and embodiment.

Right.

This is Michael Kirchhoff's work.

And to get this, you really have to understand the basics of predictive processing, or PP.

It's this really powerful, unifying idea about how the brain works.

And the core idea is about minimizing prediction errors.

That's it.

You have to imagine the brain not as a passive receiver of information, but as a massive prediction engine.

It's constantly making guesses or generating a model of what sensory input it expects to get next.

What it thinks the world should look like, or sound like, or feel like.

Exactly.

And when the real sensory input comes in, the brain compares the reality to the prediction.

Any mismatch is a prediction error.

And then it has to fix that error.

And it has two main ways to do that.

The first is to update its internal model.

Oh, I guess my prediction was wrong.

We call it perception.

The second way is to act on the world to make the world match the prediction.

And that's action.

And PP says that all of cognition is basically this constant loop of predicting, comparing, and then minimizing the error.

Okay, so the traditional view of PP would be that this all happens inside the skull.

The brain does the predicting, the body just delivers the data.

Right, what Walter calls intracranialism.

But Kirchhoff's whole point is that no, PP is actually perfectly compatible with 4E cognition.

In fact, he argues PP needs the body and the environment.

Okay, so he ties it to four key tenets of 4E.

Let's go through those.

What's the first one?

The first is the classic embodiment and embeddedness claim.

The idea that cognitive processes aren't just in the brain.

They're realized by the whole pattern of sensor and motor activity.

The loop between your body moving and the environment you're in.

Exactly.

It's that nonlinear coupling.

Your thinking is happening in that dynamic interaction, not just in the neural firing.

Got it.

And the second tenet?

The second is about making the brain's job easier.

It says that because we have these specific bodies that can move and interact in specific ways, we don't need to build these super detailed, elaborate, internal representations of the world.

So why store a perfect map of the room in your head when you can just turn your head and look?

You got it.

The world becomes its own best model, an external memory.

It's about offloading the cognitive work.

And the third tenet brings in emotions.

Right.

Affectivity.

The idea that feelings and emotions aren't just some afterthought or a consequence of cognition.

They're part and parcel of the whole process.

They're woven into perception, action, and thinking.

Okay.

And the fourth and final one is this idea of distributed plasticity.

What does that mean?

Distributed plasticity just means that learning and adaptation aren't confined to the brain's synapses.

The whole system is plastic.

The brain, the body, and the environment.

So learning a new skill might involve changing neural pathways, but it also involves developing muscle memory in the body.

And even setting up your environment in a new way, like arranging your workshop tools so the workflow is more efficient.

That's all part of the cognitive system adapting.

So Kirchhoff is saying PP can and should embrace all four of these ideas.

Okay.

That covers predictive processing.

The second chapter Walter looks at is about joint action.

This is Tollefson and Dale.

Yeah.

And their goal is really to find a middle ground.

They call it an ecumenical approach.

Ecumenical, like bringing different churches together.

So what are the two sides they're trying to unite here?

On one side, you have what are called the intellectualist accounts.

Very traditional.

They say for us to do something together, like carry a table, we have to have these high -level abstract mental states,

shared intentions, mutual knowledge, a joint plan.

So a top -down approach.

We have to think about it a very complex way first.

Exactly.

And the other side, which is more popular with radical 4E thinkers, is the complete opposite.

It's the bottom -up account.

Which argues that we don't need all that complex mental stuff.

Right.

It says that joint action just sort of emerges spontaneously from our basic perceptual and motor processes.

We're embedded in the same environment.

Our bodies just couple up.

We dynamically adjust to each other's movements without needing a shared plan.

It's more like a dance.

And Tullifson and Dale are saying, look, it's probably a bit of both.

Precisely.

For any complex joint action, you probably need both that bottom -up spontaneous coupling and some high -level shared planning.

Makes sense.

Okay, third chapter.

Perception and specifically the sense of touch from Matthew Ratcliffe.

Why is touch such a big deal for 4E philosophers?

Well, because it seems to be the poster child for embodiment and enactment.

Think about it.

You can see or hear something passively, but to touch something, you have to be active.

You have to move your body, reach out, explore.

It inherently involves your body's surface, your movement.

It seems to be this undeniable direct link between you and the world.

So a lot of 4E thinkers want to give it this special status called the fundamental or first sense.

But Ratcliffe actually pushes back on that idea.

He does.

Walter points out that Ratcliffe concludes that, well, if touch is primary, it's not for some unique philosophical reason.

All our senses involve the body, to some degree.

He thinks if touch is fundamental, it's simply because of its incredible diversity.

Diversity.

Yeah, it just handles so much.

It's involved in emotion, in object recognition, in social interaction.

It's so varied that it becomes practically indispensable to our experience.

But that's an empirical point about its range, not a philosophical one about its special status.

Interesting.

So its importance comes from its utility, not its philosophy.

And that brings us to the fourth and final chapter Walter reviews.

Kruger's defense of direct social perception.

And this ties right into the classic other minds problem.

How can I ever truly know what's going on in your head?

The traditional answer is that I can't.

I just see you crying and I have to infer that you're sad.

Right.

It's always an inference.

But Kruger, drawing on these 4E ideas, argues that sometimes we can skip the inference.

He defends a direct non -inferential solution.

How does that work?

His claim is that emotions are, at least in part, embodied in behavior.

So the joy isn't some hidden internal thing that causes the laughter.

The laughter is a constituent part of the joy itself.

So if the laughter is part of the joy, then seeing the laughter is literally seeing the joy.

You are directly perceiving the mental state, not inferring it.

Okay.

So we've got the lay of the lamb.

PP can be 4E.

Join action is a mix.

Touch is diverse.

And we can directly see

Now comes Walter's job as the critic.

And he starts by going right after that last point from Kruger.

He does.

This is Walter's first big demonstration of his call for more rigor.

He immediately attacks this idea that seeing a part of something means you've seen the whole thing.

He calls it the part -whole objection, or PWO.

So the logic is, even if we grant that laughter is a part of joy, seeing the part doesn't mean you see the whole.

Exactly.

He says this principle just doesn't hold up anywhere else.

And he uses these great analogies to show it.

The first one is the MH370 wreckage.

Right.

When pieces of the plane were found, nobody said, great, we've seen the wreck.

Job done.

Of course not.

The search for the whole wreck continued.

Seeing a constitutive part, a piece of the wing, is not the same as seeing the integral object, the entire plane wreck.

That's a powerful physical example.

He also has a cinematic one.

The Lawson translation example.

He says, look, Bill Murray is an essential, constitutive part of the cast of that movie.

But if you see Bill Murray walking down the street, you can't claim you've seen the whole cast of Lawson translation.

Just doesn't follow.

It doesn't.

And his final example is maybe the strongest, the CPU of a computer.

Because Kruger might try to argue that, well, laughter isn't just any part of joy.

It's a really essential, functional part.

The CPU is the most essential, functional part of a computer.

It's the functional core.

But if I haul up a CPU chip and show it to you, have you seen my computer?

No.

You've seen an essential component, but not the whole system.

So the part -whole objection seems pretty solid.

Even if behavior is part of an emotion, seeing the behavior doesn't give you unmediated access to the whole emotion.

Right.

But then Walter takes it a step further.

He says, okay, let's pretend you overcome the PWO.

Let's just grant it for the sake of argument.

The problem of inference comes roaring right back with what he calls the super actor scenario.

Okay.

Here's where it gets really interesting.

Set this up for us.

So imagine we look at Anne who is genuinely laughing with joy.

And let's say we accept Kruger's premise.

We are directly perceiving her joy.

Now, enter Anna.

The super actor.

The super actor.

Anna is a perfect mimic.

She can reproduce Anne's laughter exactly.

Every muscle twitch, every sound, every physical detail is identical.

But she feels absolutely nothing.

She's faking it perfectly.

Okay.

So from the outside, from what you can see and hear, Anne and Anna are indistinguishable.

Indistinguishable.

But the internal state, the presence or absence of joy, is completely different.

So how could you possibly tell them apart by direct perception alone?

You can't.

If direct perception only has access to the external behavior and the behavior is identical, your perception has to be identical.

But the mental state isn't.

Therefore, you can't have direct access.

You must be making an assumption, a leap.

An inference.

An inference.

And he calls it a defeasible inference.

That's a key term.

It means it's a reasonable assumption, but it can be defeated by new evidence.

If you later find out Anne was paid to laugh, you'd immediately drop your belief that she was genuinely happy.

So the 4E account doesn't actually get rid of inference.

It just sort of rebrands it.

That's Walter's point about rigor.

Exactly.

The old school intracranial view used a causal inference.

The hidden joy causes the visible laughter.

Kruger's embodied view, at best, gives us a constitutive inference.

The visible laughter is a part of the joy, so we infer the rest of the joy is there, too.

But it's still an inference.

It's still a guess.

The promise of direct, unmediated access is broken.

Correct.

So that's his first major critique.

A lack of philosophical rigor.

Now he moves to the second one, which is more about that second danger he warned about.

Concept -mongering.

Right.

And this brings us back to Kirchhoff and predictive processing and the famous coupling constitution fallacy.

This is such a central and, honestly, a very frustrating debate in 4E circles.

So Kirchhoff's argument was, look, action is essential for minimizing prediction error in the brain.

Therefore, action and the body must be part of the cognitive process itself.

And Walter says this is the fallacy, pure and simple.

He does.

Because it blurs this really important distinction between something being causally necessary for a process being coupled to it and something being a literal component of that process constituting it.

So just because A is essential for B to work doesn't make A a part of B.

Exactly.

And the ant navigation analogy he uses makes this so clear.

Ants use the position of the sun to navigate.

The sun is absolutely 100 % essential for their wayfinding mechanism to work.

If you took away the sun, the ant would be lost.

Completely.

The sun is causally coupled to the navigation system.

But would anyone seriously argue that the sun is part of the ant's cognitive mechanism?

That it's a component of the ant's brain?

No, of course not.

It's an external resource the ant's internal mechanism uses.

Precisely.

It causally enables the navigation.

And an intracranialist can happily agree that the body and the world are essential causal enablers for cognition without ever having to say they are constitutive parts of cognition.

I see.

So the fore proponent is making a much stronger constitutive claim but only providing evidence for the weaker causal claim.

That's the fallacy.

But then Walter does something really interesting.

He pulls back and offers a meta critique.

He basically asks,

even if the distinction is clear in theory, does arguing about it in practice actually get us anywhere?

And he thinks it doesn't.

Because it's often impossible to tell the difference.

It's often empirically underdetermined.

We can't run a test to decide.

He says, look, some cases are easy.

The CPU is a constituent part of my computer.

The power plant that generates the electricity is just a causal contributor.

But what about the stuff in the middle?

What about the power cord?

It's external, but it's physically necessary for the system to function.

Is it a part of the computer or just a cause?

What about a USD stick that holds the only copy of the operating system?

If you unplug it, the whole thing stops working.

It feels like it's part of the system then.

It does, right.

And now, apply that ambiguity to cognition.

Let's go back to Otto and his notebook.

Is the notebook entry a constituent part of his belief?

Or is his brain's process just causally dependent on it?

And what Walter is saying is that either way you describe it, the functional outcome is identical.

Exactly.

That's the killer point.

Both causation and constitution predict the exact same thing.

They both entail the same counterfactuals.

If you change the notebook, Otto's behavior changes.

If you change the body, the cognitive process changes.

So since both labels, causally dependent and constituted by, make the same predictions about what will happen.

Arguing about which label is correct is a waste of time.

It's not a real scientific debate.

It's concept -mongering.

The interesting question isn't what we call it, but whether our theory, like PP, can actually explain and predict the behavior of the whole brain -body environment system.

Stop arguing about the boundaries and focus on the dynamics.

That seems to be the message.

It is.

And that skepticism about boundaries leads him right into his third critique, which is another flavor of concept -mongering.

The messy line between extended and distributed cognition.

Okay, so this comes out of his review of Tolson and Dale.

Remind us of the distinction they try to make.

They define two kinds of systems.

The first, they call sullipsistic systems.

This is basically extended cognition.

It's a single individual, like Otto, whose cognitive process extends into a non -social object, like his notebook.

The mind is Otto's.

And the other kind.

Collective systems or distributed cognition.

This is where the cognitive process is spread across multiple people and tools, like a cockpit crew flying a plane, or a team of scientists working on a problem.

Here, the cognition belongs to the group, the collective system itself.

That seems like a pretty clear, useful distinction.

Otto versus a flight crew.

Where does it get messy?

It gets messy the second the external resource becomes social.

So, Tolson and Dale argue that if Otto relies on his notebook, it's extended cognition.

The belief is Otto's.

But if he relies on his wife, who accesses external memory and a constant back and forth, the system becomes distributed.

So, the belief is no longer Otto's.

It belongs to the Otto and his wife system.

Yeah.

Why is that a problem?

It's a huge problem for explaining behavior.

Because the whole original reason for the extended mind thesis was to give us a straightforward way to explain an individual's actions.

We can say, Otto went to 53rd Street because he believed Momo was there.

It attributes agency and belief to the individual.

But if it's a distributed system?

Then you can't say that anymore.

You can only say, the collective Otto -come -wife system believed Momo was at 53rd Street.

You lose your explanation for Otto's individual behavior.

It undermines the very thing the theory was supposed to help with.

So, maybe the belief should always just be attributed to the individual, even when they're interacting with other people.

Well, that's one option, but that doesn't work either.

Because Walter points out, there are clear cases where the cognitive state must belong to the group.

Give us one of those.

The U .S.

Senate passing a law.

That act, that decision, can't be attributed to any single senator.

It is irreducibly an act of the collective body.

Or think about the blind rage of a mob in a riot.

That rage is an emergent property of the group dynamic.

It doesn't reside in any one individual.

So we're stuck.

We need both labels, extended for individuals and distributed for groups.

But the boundary between them is completely blurry.

It's an unresolvable boundary.

Walter's point is, on what grounds do we draw that line?

How much interaction does it take for an individual using a social resource to become a genuinely integrated, distributed system?

Any answer you give is just going to be based on your preferred theory, not on any hard empirical fact.

And we're back to concept -mongering.

We're back to arguing about labels instead of understanding the function.

So he's basically cleared the table.

He's shown the field has problems with rigor and with getting stuck in semantic debates.

And that sets the stage for him to introduce the topic everyone should be talking about instead.

And that brings us to part three.

The missing piece of the puzzle.

Self -control.

Right.

Walter's argument is, you know, if 4E is really about action and how we are embedded in the world, then volition has to be at the heart of it.

And the central challenge for volition is the constant conflict between our goals and our desires.

And this isn't some rare dramatic thing.

He says we're facing this conflict almost every second.

Constantly.

The world is full of what he calls affordances or incentives.

Your phone affords scrolling.

The cake in the kitchen affords eating.

Your email affords checking.

And to get anything meaningful done long -term, you have to constantly manage and resist these short -term temptations.

You have to make sure you actually do what you decided to do.

Exactly.

And he brings in Harry Frankfurt's model of desires here, which is really useful.

Frankfurt talks about first -order desires, like I want that piece of cake right now.

And second -order volitions, which are desires about desires, I want to be the kind of person who wants to be healthy.

And self -control is about making those two things line up.

It's achieving a mesh between them, either aligning them or making sure the second -order volition wins when they conflict.

And the research he cites shows this isn't a small part of our lives.

It's a huge part.

It's massive.

He mentions a 2012 study by Hoffman and colleagues.

They found that people reported feeling some kind of urge or desire about half of their waking hours.

And crucially, they were actively trying to resist those urges 42 % of that time.

Wow.

So if you do the math.

It works out that we spend about a fifth of our entire waking life actively fighting our own desires.

It's an enormous drain on our cognitive resources.

OK, so if we're spending that much energy, we need good strategies.

And the traditional strategy is just more willpower.

That's the old view.

That self -control is this purely internal thing.

You just have to form a stronger intention or use your cognitive skills to focus or just draw on this internal faculty of will.

Just try harder.

But Walter says the empirical evidence just demolishes that view.

And the first big nail in the coffin is the concept of ego depletion.

Yeah, ego depletion research was a game changer.

It showed that willpower isn't some abstract mental power.

It behaves like a physical resource, like a muscle.

It can get tired.

It can be depleted.

And it's even linked to things like your blood sugar levels.

Can you walk us through one of those classic experiments, like the radish and cookie one?

That's the perfect example.

So researchers bring people into a lab that smells of freshly baked chocolate chip cookies.

There are two bowls on the table, one with the cookies, one with radishes.

One group of people is told they can only eat the cookies.

The other group, the unlucky ones, are told they can only eat the radishes.

So the radish group has to use a ton of self -control to resist the amazing smelling cookies.

A ton, they are actively depleting their willpower resource.

Then after that, both groups are given a second, totally unrelated task,

an unsolvable geometry puzzle.

And the researchers just measure how long they persist before giving up.

And the radish eaters give up much faster.

Significantly faster.

Their willpower muscle was already exhausted from the first task.

They had nothing left in the tank for the second one.

And this depletion doesn't just make you quit puzzles sooner.

It has much bigger, scarier effects.

It really does.

Other studies showed that depleted subjects were more likely to engage in impulsive spending.

They were more aggressive.

And most troublingly, they were more prone to risky behaviors.

They were more willing to say they'd drink before driving, for example.

It shows that this internal resource is fragile and exhaustible.

Okay, and the second big blow to the internal willpower model is situationism.

Right.

This is the idea from social psychology that our environment has this huge, often unconscious, influence on what we do.

No matter how strong your internal resolve is, the situation you're in can push and pull you in directions you didn't intend to go.

So relying on your internal strength is just a bad strategy.

It's a terrible strategy.

Walter uses this great analogy.

He says it's like paddling upstream against the current.

It's exhausting.

Your chances of success are low.

And if there's any way to avoid the current in the first place, you should take it.

Which sounds like the perfect opening for a 4E approach.

If we can outsource our thinking to the world, why can't we outsource our willpower?

That is the absolute core of 4E self -control.

We can use our bodies and we can structure our environment to make the volitional struggle easier.

Okay, let's break that down.

Let's start with the body -based strategies.

How can your body help you resist temptation?

The research here is just fascinating.

It shows these incredibly direct links.

For example, a simple act of firming your muscles, clenching your fists, tensing your bicep, even just straightening your posture has been shown to increase your ability to withstand pain, to resist tempting food, or to take unpleasant but necessary medicine.

So the physical act of being firm actually helps your mind be firm.

It's like a physical anchor for your mental resolve.

It's a two -way street.

Another really cool finding is about physical exercise and our relationship with time.

How so?

Well, people who engage in regular physical activities show less delayed reward discounting.

That's the technical term for preferring a small reward now over a bigger reward later.

So exercise makes you more patient.

Yeah.

More future -oriented.

Exactly.

And you can see how huge the practical applications of this are.

For addiction, for eating disorders, for consumer behavior.

These are all areas where we can design interventions that leverage the body instead of just telling people to try harder.

So that's the embodied part.

Now let's talk about the embedded or extended part, the environmental strategies.

Right.

This is all about strategically managing your situation.

The guiding philosophy here is that the people with the best self -control aren't the ones with the most willpower.

They're the ones who have engineered their lives, so they have to use willpower as little as possible.

This is the famous Ulysses strategy.

The classic example.

Ulysses knows he won't be able to resist the siren song with his internal will alone, so he pre -commits.

He gets his crew to tie him to the mast and plug their own ears.

He uses the external environment, the ropes, the mast, as a constraint on his future self.

The environment becomes his willpower.

It becomes an external, stable, volitional resource.

And we do this all the time in smaller ways.

He gives these very practical examples.

Like what?

Like if you're trying to stay sober, the smart strategy isn't to go to the pub and test your willpower all night.

The smart strategy is to not go to the pub.

Or if you're on a diet, don't get the dessert menu and then torture yourself.

Yeah.

Just tell the waiter not to bring it in the first place.

Exactly.

You manipulate your environment to remove the point of conflict.

You make the bad choice harder or impossible.

The environment does the heavy lifting, so your fragile internal willpower can rest.

It's such a powerful and honestly, a much more compassionate way of thinking about self -control.

It is.

And Walter is clear that these are just the first steps.

The philosophy needs to catch up and really integrate these findings.

But the empirical evidence is overwhelming.

Self -control is not just in your head.

It's an embodied and embedded skill.

This has been such a fantastic deep dive.

We started with Walter's call for more rigor and less concept -mongering in 4E philosophy.

We saw that lack of rigor with the direct perception claim where the super actor problem showed that we're always making a defeasible inference.

And we saw the concept -mongering in those endless debates about constitution versus causation and extended versus distributed cognition boundaries that are ultimately arbitrary.

And then we got to his big positive contribution, bringing self -control into the 4E conversation, showing how the traditional model of internal willpower fails in the face of evidence like ego depletion and situationism.

And how a 4E approach, using both embodied strategies like muscle tensing and embedded ones like the Ulysses strategy, offers a much more powerful and realistic account of how we actually manage to achieve our goals.

And his final thought really brings it all together, building on that famous quote from Andy Clark.

Yeah, Andy Clark famously said, our brains make the world smart so that we can be dumb in peace.

We offload the hard computational work onto the environment.

So we don't have to do it all in our heads.

Right.

And Walter takes that and applies it to volition.

He concludes that our brains do even more than that.

They make the world firm so that we can be tempted in peace.

That's brilliant.

It's not about winning the internal battle.

It's about designing a world where you don't have to fight the battle in the first place.

That's the takeaway.

True self -mastery is a feat of engineering, not just internal strain.

An amazing way to reframe the entire conversation.

Thank you so much for walking us through these really dense, but incredibly important ideas.

It was my pleasure.

And thank you for joining us for this deep dive into the critical heart of 4E cognition.