Chapter 8: Echoic Memory & Auditory Attention

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.

Welcome back to the Deep Dive.

We have a really fascinating stack of research to get through today.

We certainly do.

We are looking at something that defines your entire waking reality.

Yet, I'm willing to bet you haven't truly stopped to think about it once in your life.

It is one of those invisible mechanisms, for sure.

We are looking at the architecture of listening.

And I don't just mean the physics of sound waves hitting your eardrum.

We are going way, way upstream into the cognitive machinery.

Right.

We're breaking down chapter eight of cognitive psychology.

And specifically, we're focusing on two massive systems,

echoic memory and auditory attention.

Right.

And to really get why this matters, why this chapter is so pivotal to understanding the mind,

you have to start with the fundamental problem that the brain has to solve.

And what's that?

It's the problem of time.

This was the hook for me.

The text draws this comparison between vision and hearing that I found, well, honestly, a little unsettling.

How so?

If you walk into a museum and look at a painting, the painting is just there.

It's static.

The data is permanent, you mean.

You can look at the top left corner, scan down to the bottom right, get distracted by a tourist, look away and then look back.

And the painting hasn't moved.

The information is available in space.

It just waits for you.

But sound is different.

So different.

Sound doesn't exist in space.

It exists in time.

It is intrinsically temporal.

Think about it.

By the time you hear the end of a sentence, the beginning of that sentence physically no longer exists.

It's just vanished into thin air.

Exactly.

And if your brain didn't have a way to sort of pause that information or hold on to the ghost of that sound, you wouldn't be able to understand language at all.

You just hear a stream of disconnected noises.

It's like trying to read a book where the ink evaporates the second your eye hits the page.

That is a perfect analogy.

And that brings us to our mission for this deep dive.

We are going to figure out how the brain stitches these fleeting milliseconds of noise into a coherent reality.

How do we turn noise into And to do that, we have to start with a concept from the grandfather of American psychology, William James.

Ah, yes.

The specious present.

It's a great term.

I had to read this part twice because it's a bit of a mind bender.

He's basically saying that now the moment we are in right now isn't a razor thin instant.

No, it's a saddleback.

It's a smear.

If now we're truly instantaneous, a mathematical point in time, we couldn't perceive change or motion.

So for us to experience a melody or a spoken word, our consciousness has to span a small window of time.

It has to hold the immediate past and the breaking present in one container.

So if I say the word apple, the ah sound and the pull sound happen at different times, right?

But you hear apple as one single event because they both fall within that specious present.

If your window was too short, you'd hear ah, wait for it,

two separate events that have nothing to do with each other.

So the first system we need to unpack is the thing that creates this window.

This is what the text calls echoic memory.

This is the brain's audio buffer.

Now let's be precise here.

Is this just a tape recording?

Is my ear just recording the last few seconds of audio perfectly?

Not quite.

And the distinction is really important for students of cognitive psychology.

By the time you get an echo, the sound has already been processed by the cochlea.

So it's not raw physics.

It's physiological data.

Exactly.

But, and this is key, it is pre attentive.

It is a raw snapshot of the sound before your brain has decided what it means.

It hasn't been sliced and diced into words yet.

It's just raw texture.

Correct.

It is fine grained and it holds more detail than you can consciously process.

And we need this buffer desperately because speech is messy.

How so?

The text uses the zeal versus seal example, which I think illustrates this perfectly.

Okay, walk me through zeal and seal.

So say I utter the word zeal.

That starts with the voiced z sound.

Okay.

Now I say seal.

That starts with the unvoiced s.

Seal.

Seal.

I can hear the difference.

But the difference between those two sounds at the very beginning is incredibly subtle.

It's essentially a tiny difference in duration and hiss.

And in normal conversation, you're talking fast.

That sound flies by in milliseconds.

Right.

And here is the kicker.

Often, the information you need to identify that first letter is actually contained in the vowel that comes after it.

Wait, so I need to hear the e sound to know if I heard a z or an s.

In many cases, yes.

The shape of your mouth for the vowel helps define the consonant.

The context resolves the ambiguity.

But think about the timeline.

The z happened first.

It's gone.

That's it.

If your brain threw away that data the moment it happened, by the time the vowel arrived to help you, it would be too late.

You wouldn't have the raw data left to check against.

That's the logic.

You need the echo of the z to still be ringing in your buffer so that when the e arrives, you can retroactively process the beginning of the word.

That is wild.

It's literally microscopic time travel.

We are listening to the past to understand the present.

We are always lagging slightly behind reality in order to construct it.

And it gets even more obvious when you look at linguists call super segmental phonemes.

I highlighted that term.

It sounds terrifying.

It's just fancy talk for things like intonation, pitch, and stress.

The stuff that gives language its flavor.

Think about the difference between a statement and a question.

If I say to you, you are going.

That's a statement.

You're telling me to leave.

Versus you are going.

That's a question.

Now you're asking me.

But where is the difference in the sound?

It's at the end.

The pitch goes up ongoing.

Right.

But the sentence started with you.

That word you happened maybe two seconds ago.

How do you know if that you was part of a question or a statement when you hear it?

I don't.

Not until you finish the sentence.

You've got it.

Your brain has to hold the entire string of sound in suspension in this echoic buffer until the final cue unlocks the meaning of the whole sequence.

If the beginning faded before the end arrived, you would never understand sarcasm, questions, or emotional tone.

It would be impossible.

So we've established we have this buffer.

It's essential.

But how big is it?

Does it hold a second?

Ten seconds.

A minute.

That brings us to section two of the chapter, measuring the echo.

And this is where cognitive psychology gets really creative because you can't just ask someone, is the sound still there?

Right.

Because as soon as they focus on it, they're using active memory.

You've tainted the sample.

You have to trick the system.

We have three major experiments here that try to put a stopwatch on this echo.

Let's start with Guttman and Jules, 1963.

The white noise loops.

This one sounded like torture to be a participant in.

It wasn't the most melodic experiment, no.

They used computers, which was cutting edge at the time, to generate white noise.

Just random static.

Exactly.

But there is a trick.

And what was that?

They took a segment of that static and looped it perfectly.

So imagine a loop repeating over and over.

Now intellectually, white noise has no pattern.

It's random.

Correct.

But if the loop was short, say, half a second long, the listeners reported hearing a rhythm.

What?

A rhythm in pure static?

Yeah.

They described it as a putt -putt or a whoosh -whoosh.

So their brain found a pattern in the chaos.

Because the loop was repeating so fast, the beginning of the loop was still sitting in their echoic memory when the loop restarted.

The brain could overlay the new sound onto the echo of the old sound and say, hey, these match.

But what happens if you make the loop longer?

That's the limit.

Once the loop got longer than about one second, the rhythm vanished.

It just sounded like continuous endless hissing.

Because the echo had faded.

By the time the loop came back around, the buffer had been overwritten.

So for raw, unstructured noise, the echo seems to last about one second.

One second doesn't seem long enough to process a complex sentence, though.

It isn't.

But other experiments suggest it lasts longer if the conditions are different.

This brings us to the reading test by Erickson and Johnson, 1964.

This one is my favorite because it involves just sitting around reading a novel.

For two hours.

They told subjects to sit and read and ignore everything else.

Occasionally, a tone, a beep, would play in the room.

So I'm reading my book, a beep happens, I ignore it and keep reading.

But then, at random intervals, the reading light would suddenly go out.

And the researcher would ask, did a tone just happen?

And the variable here is the delay.

Maybe the beep happened half a second ago.

Maybe it happened 10 seconds ago.

Right.

They are checking how long that trace lingers when you aren't paying attention to it.

If you look at figure 39 in the source text, you see a really interesting curve.

Paint the picture for us.

What does the graph look like?

Okay, so imagine a graph.

On the bottom axis, you have the time delay in seconds from zero up to 10.

On the side axis, you have percent detection.

So how often they got it right?

Exactly.

When the question is asked immediately, recall is decent, about 50 to 60 % detection.

But as time passes, the line just slopes steadily downwards.

But it doesn't hit zero.

That's the key.

It doesn't hit zero immediately.

It stays above chance levels for up to 10 seconds.

10 seconds is a lifetime compared to the one second from the white noise.

It is.

But there's a nuance here that is crucial.

The researchers tracked certainty versus probability.

What's the difference?

Certainty is I definitely heard a beep.

That confidence drops off really fast.

But probability is I think the beep might have happened.

Ah, like a gut feeling.

Exactly.

That ghostly sense lingers much longer.

It suggests the visit echo fades quickly.

But a faint trace hangs around for quite a while.

Like the ripples in a pond after the stone has sunk.

That's a good way to visualize it.

And then we have one more experiment that refines this even further.

This is Pollock from 1959.

Yes, the post stimulus queuing experiment.

This one uses words, not beeps.

This is the pop quiz approach.

You hear a word buried in noise.

After the word is finished, you get a list of options.

Was the word backbone or doorstep?

And again, they mess with the timing.

Right.

If they give you the list immediately, you score high.

If they wait four seconds, your score drops and levels off.

So for speech,

the useful rich echo lasts about four seconds.

It seems so.

And the chapter also mentions rhythm perception.

To hear a pattern like long, short, long, the first beat has to still be in memory when the third beat hits.

And phrase's limit for that is about two seconds.

So we have these different numbers, one second, two seconds, four, ten.

But the principle is the same.

There's a decaying buffer.

But there was a twist in the Pollock study that I think is the most important part of the whole section.

The note taking twist.

Yeah.

Tell us about that.

In one version of the test, Pollock let the subjects write down what they heard immediately.

So I hear the word, I scribble it on a night pad.

And in that case, the delay didn't matter.

You could ask them an hour later and they'd get it right.

Well, yeah, because they wrote it down.

But think about what that proves structurally.

It proves that the echo is fundamentally different from coded memory.

Unpack that for me.

Writing it down or even repeating it firmly in your head converts the fleeting analog sound into a permanent sort of digital code.

The echo is the raw sound.

The note is the meaning of the sound.

The echo fades in four seconds.

The meaning can last a lifetime.

That's it.

So the whole goal of listening is to race against the clock.

You have four seconds to grab the raw sound from the buffer and convert it into code before it dissolves.

That is the game.

And that leaves us directly to the next problem.

We are bombarded with sound constantly.

All the time.

If we tried to convert everything into code, we'd crash.

We'd be overwhelmed.

Absolutely.

So we have to choose.

Which brings us to section three, selective listening.

Or as it's famously known, the cocktail party problem.

I relate to this on a spiritual level.

You're at a party.

It's loud.

Music is playing.

10 people are talking.

How on earth do I lock on to just your voice?

It is a staggering computational feat.

The text points out that the number one tool we use is localization.

Where the sound is coming from.

Right.

We need to know where the sound is coming from.

If two voices are coming from the exact same point in space, it is almost impossible to separate them.

After that, we use pitch and intensity.

We need that spatial anchor.

Right.

Once we have the anchor, we can start shadowing.

This is a technique developed by Colin Cherry in 1953 to test this ability in the lab.

Shadowing sounds like spycraft.

It's actually just very stressful multitasking.

Imagine you're wearing headphones.

In your right ear, you hear a news report.

Okay.

Your job is to repeat that report aloud, word for word, as it happens.

You are shattering the speaker.

So I'm just parroting the right ear.

Exactly.

Meanwhile, in your left ear, a completely different person is reading a recipe, and you are told to ignore it completely.

Talk about information overload.

But the amazing result is that people are really good at this.

They can lock on to the right ear and parrot it back perfectly.

So what happens to the left ear?

The recipe.

That's the million dollar question.

After the test, the researchers ask, what did you hear in the left ear?

What do they say?

They usually say, I don't know.

They don't know the words.

They don't know the topic.

Sometimes they don't even know if it was in a foreign language.

They are effectively deaf to it.

Functionally, yes.

But, and here's the hook, they do notice physical changes.

Oh, interesting.

Like what?

If the voice changes from a man to a woman, they notice.

If it turns into a pure tone like a buzzer, they notice that.

So the raw sound texture gets through, but the meaning stops at the door.

That appears to be the case.

And this finding triggered a massive war in psychology.

A war over what?

If the meaning stops at the door, where is the door?

Where is the filter?

Which brings us to section four, the battle of theories.

We have two main contenders here.

In the red corner, we have Broadbent's filter theory.

Broadbent was an engineer at heart.

He proposed a very mechanical model.

He said the brain has a selective filter.

Like a bottleneck.

That's it.

Imagine a Y -shaped pipe.

Two channels go in, but only one channel comes out.

And Broadbent argued that this filter sits early in the process.

What do you mean by early?

It blocks the rejected channel based on physical traits, like which ear it's coming from.

So it's a bouncer at the club door.

He checks your ID.

If you're from the left ear, you don't get in.

And crucially, you don't get processed for meaning.

The bouncer stops you before you can explain yourself.

Figure 40 in the text visualizes this brilliantly.

It's like a flow chart, you know.

Describe it for us.

You have these parallel lines coming in, the input channels, like your ears.

They all funnel into this one box labeled selective filter.

And then critically, only one single line comes out of that filter and goes to the limited capacity channel.

It's an all or nothing switch.

Exactly.

It's clean.

It's logical.

It explains why we don't remember the words from the rejected ear.

But.

There is always a but.

There's a problem, isn't there?

There is.

It's called the own name effect.

Everyone has experienced this.

You're at a party.

You're deep in conversation, ignoring the room.

And then from 20 feet away, someone says your name and your head snaps up.

But if broadband was right and the bouncer blocks everything from the rejected channel, how did I hear my name?

Exactly.

If the filter is a brick wall, your name should have bounced off just like every other word.

Morey showed this in the lab in 1959 with the shadowing task.

Yep.

If the rejected ear whispered subject, stop now, the person kept shadowing.

But if it said your name stop now, a huge percentage of people stop.

So the bouncer has a VIP list.

That's one way to put it.

Or the filter isn't as solid as broadband thought.

And then Anne Treisman came along in 1960 and really broke the theory.

With the mahogany table experiment.

I love this one because it's so sneaky.

It is.

So you're shadowing the left ear.

The left ear says sitting at a mahogany.

And the right ear, the one you're ignoring, says table with her head.

But right in the middle of the sentence, Chiesman swapped them.

So the left ear suddenly starts spewing nonsense and the right ear finishes the sentence with table.

And what did the subjects do?

They followed the sentence.

They said mahogany table, effectively jumping channels to the rejected ear to finish the thought.

They didn't stay in the left ear.

They followed the meaning.

Which proves that the rejected ear wasn't fully blocked.

The brain was secretly analyzing the meaning of the background noise.

And when it found a match, it grabbed it.

So broadband's brick wall is crumbling.

It is.

So Treisman proposed theory B, the filter amplitude or attenuation theory.

Attenuation just means turning down the volume, right?

Correct.

She argued there is no brick wall.

Instead, the filter just dampens the signal.

The rejected ear isn't silenced.

It's just a whisper.

Okay.

So how does that explain hearing your name?

She introduced the concept of dictionary units.

Think of every word in your brain as having a threshold, a sensitivity setting.

Okay.

Rare words like platypus have a high threshold.

You need a loud, clear signal to activate them.

But my own name has a super low threshold.

It's hair trigger sensitive.

So even if the signal is attenuated, even if it's a whisper, it's enough to trip the wire and grab your attention.

And table was primed by mahogany.

So its threshold was temporarily lowered.

Exactly.

It lowers the resistance.

It's a very elegant theory.

It sounds perfect, but our author, Nyser, he's not buying it, is he?

No, he is not.

He raises a really interesting subjective point.

When you are ignoring someone, does their voice actually sound faint?

No, not at all.

If someone is annoying me on the train, they sound loud.

I'm just trying not to listen.

Right.

Nyser quotes Tischner here.

We have to distinguish between intensity, which is loudness, and vividness, which is clarity or focus.

And the attenuation theory confuses the two.

Yes.

Nyser argues we don't turn down the volume knob.

We do something much more active.

This leads us to the aha moment of the chapter, section five, analysis by synthesis.

This is Nyser's alternative, and it completely flips the script.

It really does.

Broadbent and Traseman were treating attention as a passive filter, like a sieve that catches gold and lets the dirt fall through.

But Nyser says no.

He says no.

Attention is an active act of construction.

Construction.

Like building.

Yes.

He proposes two stages.

First, you have the pre -attentive process.

This is a passive global scanner.

It's always on.

It monitors the whole auditory field.

So this is the security camera scanning the crowd.

That's the idea.

It handles localization, where are sounds coming from, crude segmentation.

It can spot simple patterns like your name or a baby crying.

Okay.

And the second stage?

The second stage is focal attention.

This is the construction crew.

Nyser argues that to really hear and understand speech, we don't just receive it.

We internally re -synthesize it.

Wait, so I'm not just downloading your words.

I'm rebuilding them in my own head.

In a sense, yes.

It's analysis by synthesis.

You analyze the sound by trying to synthesize a matching model for it.

You're constantly generating a hypothesis of what is being said and checking if the raw sound matches your hypothesis.

That sounds exhausting for the brain.

It is.

That's why listening is hard work.

But it explains the data beautifully.

Why do you ignore the rejected ear?

Not because I blocked it.

But because you didn't build it.

You didn't allocate the construction crew to that site.

Exactly.

The raw sound hit the ear.

It sat in the echoic memory buffer for a second.

The pre -attentive scanner said, nothing interesting here.

And because the construction crew never touched it, it dissolved.

It wasn't filtered out.

It just wasn't synthesized into meaning.

Precisely.

And this explains the mahogany table switch, too.

Your construction crew was following the blueprint of the sentence structure.

When the blueprint moved to the right ear, the crew followed the blueprint.

There is a specific experiment mentioned here regarding the lag in shadowing that I found really technical.

But I feel like it's the smoking gun for this theory.

It is the smoking gun.

This was Treisman again in 1964.

She played the same message in both ears, but staggered.

Like a musical round.

Row, row, row your boat.

Pretty much.

Now, if the shadowed message, the one you are focusing on, was lagging behind the rejected message, something interesting happened.

So the rejected ear hears the word apple, and then one second later, my focused ear hears apple.

Right.

And subjects noticed the messages were identical very quickly, within about 1 .4 seconds.

Why so fast?

Because the rejected message, apple, was still sitting in the raw echoic buffer.

When the focused ear heard apple, the pre -attentive scanner looked at the buffer and said, hey, that matches the raw echo I just heard over there.

But what if it's the other way around?

What if the focused ear is leading?

So you focus on apple, and then say four seconds later, the rejected ear says apple.

In that case, the echo of the focused apple is long gone from the buffer.

Right.

But you still notice the identity up to 4 .5 seconds later.

Why?

Because you aren't matching against the raw echo anymore.

You are matching against your synthesized meaning, your short -term verbal memory of the word.

That is brilliant.

The 1 .4 seconds is the limit of the raw echo.

The 4 .5 seconds is the limit of the synthesized meaning.

The numbers fit the theory perfectly.

It shows that we have two different representations of sound, the raw trace and the constructed meaning.

This idea of synthesis that we are building our reality gets even spookier when we talk about sleep.

Section six, sleep, dreams, and inner speech.

It raises the question, does the synthesizer turn off when we sleep?

Well, we know the pre -attentive scanner stays on.

That's why I wake up if my baby cries, but I sleep through a thunderstorm.

Right.

The security camera is running.

But Nyser suggests that during dreams, specifically REM sleep,

we might actually be using the synthesis mechanism to build the dream.

There was this itch experiment mentioned.

Cobb et al, 1965.

This one is bizarre.

They waited until subjects were in REM sleep.

Then they whispered complex suggestions like, your nose itches and you will want to scratch it.

And what happened?

Later, the subjects scratched their noses.

So they processed the sentence while dreaming.

Yes.

But Nyser argues the sound wasn't just heard.

It was synthesized and incorporated into the dream narrative.

The dreamer hears the voice, but instead of waking up, the brain weaves it into the story.

Oh, that's just a character in my dream telling me my nose itches.

Exactly.

It repurposes the input to protect the sleep state.

And this connects to inner speech.

I mentioned earlier that I can't listen to you if I'm thinking really hard.

Try it.

Try to recite a poem in your head while listening to a podcast.

You can't.

It's impossible.

I lose the thread immediately.

Nyser says this is because inner speech uses the same synthesis mechanism as listening.

You only have one construction crew.

That is a profound thought.

It is.

If your crew is busy building your internal monologue, they cannot build the external speech.

So when you are lost in thought, you are functionally deaf to the world.

You are in a self -imposed dream state.

In a way, yes.

That explains so many arguments I've had with my partner.

Are you listening to me?

No, I was building a thought.

You can try that excuse next time.

Sorry, my synthesis crew was occupied.

I'm sure that will go over well.

There was one other study in this section about subliminal perception, the Pine 1960 study.

The cow and the hook.

Right.

So subjects are writing stories for the thematic apperception test.

Meanwhile, in the background, someone is reading an essay.

One group had a cow essay full of passive warm soft imagery.

The cow is warm, it is soft.

The other group had a hook essay, aggressive cold phallic imagery, the hooks unbending arc.

And what happened to the stories the subjects wrote?

The elements bled through.

People who heard the cow essay wrote softer stories.

People who heard the hook essay wrote more aggressive stories.

So the pre -attentive scanner let the vibe through.

Perhaps.

Nyser is skeptical, though.

He notes that sometimes subjects write the opposite.

If they hear the aggressive story, they might write a super nice story just to distance themselves from the negative noise.

But it does suggest that even if we aren't synthesizing the words,

the emotional tone might leak in.

Now, we have to be fair, researchers here.

We can't just present the theory that works.

We have to look at the cracks.

Section seven, complications and challenges.

Psychology is never settled.

There are challenges to Nyser's synthesis theory.

First, let's talk about the cat, the Hernandez -Peon experiment.

This is a classic physiological study.

They put electrodes in a cat's ear right in the cochlear nucleus.

They played a sound.

The ear fired with electrical activity.

The ear is on.

Then they showed the cat a mouse.

A visual distraction.

A very high stakes one for the mouse.

And the electrical activity in the ear dropped.

So the cat actually turned off its ear.

It seems so.

This suggests a peripheral closing down or gating.

This supports Broadbent's filter theory,

actually turning off the input more than the synthesis theory.

Nyser admits this is a conflict.

So maybe we have both.

Maybe we have a dimmer switch and a construction crew.

That is the most likely reality.

Biology rarely chooses one solution when it can use two.

Then there's the three channel problem.

Triesman again.

She showed that shadowing gets much harder if you have two rejected messages instead of one.

Two people distracting you instead of one.

Makes sense.

But here's the weird part.

It's harder if the two distractors are spatially separated.

If they are both in the left ear, it's easier to ignore them than if one is left and one is in the center.

Why does that matter to Nyser's theory?

Because if we aren't synthesizing them, if we are just letting them dissolve, why should their location matter?

It implies we are tracking them more than the theory might suggest.

And finally, we have to debunk one misconception about synthesis.

Some people hear synthesis and think subvocalizing.

Right.

The motor theory.

The idea that to hear a word, your brain has to microscopically move your mouth muscles to mimic it.

Like when you see a child moving their lips while reading.

Exactly.

But there is a killer piece of evidence against this.

The simultaneous translator.

The cognitive athlete.

Absolutely.

Think about what they do at the UN.

They listen to Russian.

They speak English at the exact same time.

If they had to mimic the Russian with their mouth muscles to understand it, they couldn't possibly be speaking English words at the same time.

You can't run two motor programs on the same mouth at once.

So synthesis isn't muscle movement.

It has to be purely mental.

It is abstract, mental construction.

The brain is operating on two abstract levels simultaneously, receiving meaning in one code, Russian, and building meaning in another, English.

That really highlights the sophistication of the human attention system.

It's not just a reflex.

It's a creative act.

It is indeed.

So let's bring this deep dive in for a landing.

We've covered a massive amount of ground.

We have.

It's a dense chapter.

We started with the problem.

Sound is a ghost.

It vanishes the moment it's born.

So we discovered echoic memory, the buffer that holds the specious present, for about one to four seconds, giving us just enough time to decode the noise.

We looked at the battle of the filters.

Broadbent's bouncer.

Treisman's volume knob.

And we landed on Nyser's analysis by synthesis.

The idea that we have a passive scanner watching the world and an active builder constructing our reality from the sound.

And this synthesis model explains why we hear our names, why we can't think and listen at the same time, and how we follow a voice at a cocktail party.

Attention isn't just blocking noise.

It's the active act of creating meaning.

I want to leave the listener with a final thought.

We talked about that simultaneous translator.

Just visualize that for a second.

We usually think of attention as a spotlight focusing on one thing.

But that translator proves the brain can actually be a prism.

A prism?

How do you mean?

Well, it can take in one beam of light rushing refracted through this abstract synthesis process and project a completely different beam English simultaneously.

It proves we aren't just filters.

We are transformers.

We take the raw chaos of the world and transform it into order.

That's a beautiful image.

It really is.

Something to chew on next time you're trying to listen to this deep dive while writing an email.

Yeah.

You just try to run two construction crews at once.

And probably doing a bad job at both.

Exactly.

From the last minute lecture team, thanks for lending us your synthesis.