raw_transcripts/lecture_17.txt

Okay, let's let's get started.

Everyone.

Can everyone hear me?

I'm not used to speaking in such a big lecture

theatre.

Okay.

I'm Dan Bender.

I'm the associate professor of Behavioural Neuroscience.

My lab studies sleep, among other things.

And that's what I'm going to talk about today.

I do have to leave at a little bit before

11 for sure.

So if there are any lingering questions, I'll have my

email address over here and you can email me.

But my P.A. soon is operating and I can't miss

her presentation.

So we'll we'll jump right in.

So before I talk about sleep, okay, which is a

very interesting topic, I want to take something.

Go back a little more basic and ask the question,

why do animals have a brain?

Right.

That's that's probably one of the most basic things in

neuroscience.

Why do we have a brain?

And we look through that animal kingdom.

We can see that you have sort of the most

primitive species, such as sponges.

Okay.

And these actually don't have any neurones at all.

Okay.

So the first animals did not have neurones.

It happened some point in evolution.

Okay.

And you see that neurones appear around the time that

jellyfish and similar creatures developed in evolution.

And then if we get to more complex animals, it's

no longer just neurones, but we have ganglia and brain.

Okay.

So it's clusters of neurones working together to do something

that individual neurones can't.

Now, this is where it gets interesting that if we

look at our closest one of our closest invertebrate invertebrate

relatives, the starfish.

So the starfish is closer than roundworms and molluscs and

insects.

It's closer to us than all those other animals.

Starfish actually lost the ganglia.

They lost the brain.

They just have neurones that are connected together in a

ring.

Okay.

Same with sea urchins.

Next time you have uni for sushi, you're thinking you're

eating one of your closest relatives in the invertebrate kingdom.

Now, where it gets very interesting is sea squirts.

The sea squirts actually have a brain in the larval

stage, but then they lose it when they're an adult.

And if we look at sea squirts, it gives us

an answer to the question, Why do we have a

brain?

Okay.

So if we look at a larval sea squirt, it

looks a bit like a tadpole.

It has two little eyes spots.

It has a tail.

It's not a spine, it's not a vertebrate, but it

has a new record, which is, you know, very close

to what vertebrates would have.

And it has a digestive system.

And its main goal and its early life is to

find a nice rock, because once it finds a nice

rock, it's going to attach that rock for the rest

of its life.

And that rock is where it gets all its food

and acts as a filter filtering the water.

So if you place a bad rock, it kind of

sucks, right?

That's the rest of his life.

It's there.

So the thing that's so interesting is that when you're

attached to a rock, why would you need to see

it if a predator was coming?

You can't escape from it.

You're stuck to the rock.

So vision is useless.

Motion is useless, right?

You're just stuck to the rock.

So you're just floating in the water.

So what does a sea squirt do?

It eats its nervous system.

Okay.

It doesn't need it anymore.

It's useful energy.

It digests it.

Okay.

So here's the thing.

If you don't need to move, you don't need to

see.

You don't need a brain.

Okay.

So why do animals have a brain?

Animals need a brain to convert sensation to action.

A jellyfish does that and moves around the water looking

for food.

A sponge doesn't.

It just filters the water.

Doesn't need a brain.

Okay.

Now, this is where it leads to sleep.

Every animal thus far that we know of, if.

If an animal moves, it sleeps.

Okay?

Sleep is a really primitive aspect of the brain.

How do we measure sleep?

Right.

You can't measure it in a jellyfish.

It seems know when you wake it up, it snores.

You have no way of measuring it, you would think.

But it turns out that the most primitive forms of

sleep there is a lack of motion for a prolonged

period of time.

Much like us, there is an elevated threshold to react

to sensory stimulus.

Right.

So if I mention your name now, you're.

You know, you'll look around, right?

You'll hear me.

If you were sleeping in class, you might not unless

I was loud enough.

Right.

Don't worry.

I won't test that out.

The third thing, and this is a bit harder to

test, but you can test in a lot of animals.

There is a rebound after sleep deprivation.

Right.

Which.

Which makes sense.

If I.

If I keep you for three days without sleeping, you're

probably going to fall asleep really, really fast and sleep

for a longer period of time to compensate.

Okay.

So those there is of course, sleep is a lot

more complex.

When you talk about mammals.

But the most basic definition of sleep are these three

things.

So now you can ask the question when you move

up the food chain in evolution, how to sleep in

the animal kingdom.

So let's start with the jellyfish.

This is the upside-down jellyfish.

I believe this was a was an undergraduate project from

a bunch of students, and it got published in Current

biology.

That's the story I heard.

So they had a bunch of jellyfish in an A

tank and they tracked the activity of the jellyfish.

And they found that in the daytime it was moving

a lot.

And the Night-Time it wasn't moving so much.

Now, that does not mean that the animal is sleeping

on its own.

Right.

During this class, some of you might not be moving

a lot.

It doesn't mean that you're sleeping.

Right.

So how can you tell the difference?

So what they did is they perturb the animal so

it couldn't rest.

Okay.

There would be a gust of water coming up.

And every time the animal was sort of, you know,

resting, it would have to move again.

And what they found is that when they perturbed one

night asleep, there was a recovery where there was less

activity than you would expect prior to the perturbation.

Okay.

And even when the animal slept over here, we would

consider sleeping.

There was even less activity.

So it was almost like a like you could say

a deeper sleep.

Now, this is maybe not the same as being able

to record EEG and EMG.

You know, all the signals from the brain of humans

and monitoring their sleep.

But this is the jellyfish, right?

And we're already seeing some basic aspects of sleep here.

Okay.

Remember, they don't have a brain.

They just have neurones.

That's key.

Now, if we go up the food chain to the

fruit fly, which is obviously a lot more complicated than

the jellyfish and its nervous system, we see that the

fruit fly rests more at Night-Time.

You know, lack of movement than in the daytime, which

is already good because usually sleep doesn't have to be

12 hours a night and you're awake 12 hours during

the day.

There are lots of different ways sleep can manifest.

But you do see that there is a preferred time

for the animal to rest.

And what they can test here is also the arousal

threshold that if you, you know, lightly do an air

popper, do something that would preserve the animal.

You have to put more force to get the animal

to wake up.

Okay.

So an animal to move.

So this is one of the key things for sleep.

It's not simply the animals not moving, but it's paying

attention.

And you have to somehow break through.

It's it's an intention to to get through the animal.

Okay.

Now, once we're in the fruit fly.

There is one other thing that's quite cool that you

can start doing that this would be very difficult to

do with the jellyfish, but you can put electrodes in

the brain.

You can tell the animal in a ball so it

walks around and you can actually record neural activity and

combine this with the movement of the animal.

So now you have a way to to not just

look at sleep behaviourally, but look electrophysiological of what's going

on, to see if there are characteristics of neural signals

that that tell you, oh, the animal is sleeping because

you see this animals awake because you see this.

Okay.

Now we're going to go to a larger animal now,

the lizard and the lizard here.

You know, it's it's a vertebrate, first of all.

Right.

It's a lot closer to us.

They definitely sleep.

Okay.

And in the lizard, you start seeing it.

Sleep is not just a homogeneous pattern in the brain.

There's actually two stages.

Now, we don't know if it's exactly like the two

stages of sleep in humans, but it does seem to

be one stage where you have synchronous activity in one

stage where it's less synchronous.

And these seem to oscillate back and forth.

So it's quite different from what you see in other

mammals.

But you already start seeing neural signatures of two different

types of sleep.

And reptiles.

Okay.

Now when you get to birds and mammals, now you

have a more defined sleep pattern.

And this is typically called slow wave sleep and REM

sleep, Slow wave sleep.

You can also call non-REM sleep.

But essentially, when you go to sleep, you fall into

deep sleep.

And you have a lot of this in your early

sleep.

And then you have a little bit of REM sleep.

And then each cycle of sleep, which is about 90

minutes, you have a little bit more REM and a

little bit less slow wave.

Okay.

And by the end of the night, it's mostly REM

sleep.

Okay.

Show of hands.

Do you typically dream in your slow wave sleep or

REM sleep?

Raise your hand if it's slow.

Wave sleep.

Raise your hand in his REM sleep.

Good.

So REM sleep is where you typically dream, but you

can dream also and slowly sleep.

You typically remember your dreams from REM sleep more.

Okay.

So when you're in slow wave sleep and you put

up electrodes so you can record neural activity, you can

see that there's slow oscillations which relate to cortical cortical

communication.

There's spindles which are high oscillation signals that relate to

the thalamus talking to cortex.

And there are also sharp wave ripples, which are extremely

high frequency oscillations in the hippocampus that help it talk

to cortex.

In REM sleep, you see completely different brain activity.

You see theta activity in the hippocampus and you see

tiger waves as well.

Okay.

So if you just were watching the signals coming from

the brain, you could tell if someone is sleeping and

whether it's REM sleep or non REM sleep.

Okay.

And the key thing to remember is REM sleep activity

is very similar to awake activity that is not synchronised

and sort of like all the neurones.

Imagine if the orchestra and there all every instrument is

just practising on its own.

They don't seem coordinated.

Slow wave sleep.

The whole orchestra is going boom, boom, boom together.

And that's what the slow waves are.

Okay, so you have these two different types of sleep.

Now we can look at the animal kingdom to see

how does this type of sleep?

Very right.

Do all animals sleep like us or is there a

lot of variability?

And what you find is that at least with REM

sleep, it is really, really different across the animal kingdom.

So the platypus is the king of REM sleep.

They have 8 hours a night.

We only have two.

Okay.

The dolphin is unfortunately the animal with one of the

least amount of REM sleep.

They have minutes every night.

Okay.

So we know REM sleep is really important.

The same time.

That dolphin is a lot smarter than the platypus.

So it's not really clear why all this extra sleep

a platypus is getting what it gains from us.

Now, if you look at overall sleep times across the

animal kingdom, one thing you notice is that two animals

that are genetically very similar can have very different amounts

of sleep.

So a nice example is the owl monkey below over

here compared to the human right.

We're both primates.

We actually both have the same amount of REM sleep

per night, about 2 hours.

But the owl monkey sleeps 17 hours.

We sleep well.

We're supposed to sleep 8 hours.

It's probably a lot less.

In general, you're staying up late at night, you know,

with your Twitter accounts and bingeing Netflix, like.

Like all of us.

So it's not just genes, right?

There's something else going on.

And it's very likely to be ethology, which means there

are specific constraints on animals, behaviour and habitat.

So, for instance, if you could sleep all day, right,

because you didn't have to worry about food and resources

and other things.

Yeah, maybe.

Maybe over many, many millions of years, you would start

sleeping more, right?

Another thing that's quite buried in the animal kingdom is

unit hemisphere of sleep.

So we sleep with both hemispheres.

At least most of us do.

The dolphins are famous for sleeping with one hemisphere than

the other hemisphere.

And you can see this as they will sort of

circle around when they're sleeping with one side and then

they circle the other direction when they're sleeping with the

other part of their brain.

And they have to do this because.

They need to breathe consciously, so they have to keep

part of their brain on.

Otherwise, they die.

Okay.

So it turns out that it's not just dolphins that

are cool, that there are lots of species that are

capable of this, including birds, water, animals, reptiles.

It's almost that we are the oddballs of the year.

And if you think about it, it's it's quite smart

that you always keep part of your brain on is

if a predator comes or you need to be able

to do something, you would react faster, right?

Or your brain is a bit groggy, but the other

part is awake.

And I mean, I've even heard stories of lizards that

will sleep, you know, with both eyes closed.

But if they sense that there is a predator in

the room, one eye opens and they switch to, you

know, have a third sleep.

So there's probably some flexibility, but a lot of mammals

sleep in or birds will sleep in nests and they

have sort of very safe habitats.

And over time, maybe you have a unit of hemispheric

sleep is not necessary anymore.

And it's not as efficient as sleeping with both hemispheres.

Right.

It takes the dolphin twice as long to sleep because

it have to do one side at a time.

Okay.

It's not just non-REM sleep that is due to hemisphere.

REM sleep requires both hemispheres as far as we can

tell.

So dolphins have a hard time during REM sleep.

Right.

It's like they do.

They have their little dream and then they have to

breathe.

Right.

So that's that's tough.

They're doing this at, you know, seconds or a minute

at a time.

They can hold their breath longer than us.

Fortunately, when we get to the first seal, it's interesting

because the first seal has a behaviour that's more like

terrestrial mammals, but it also lives in the water.

And it turns out that when it's in the water,

you can't do very much REM sleep.

And it also has to sleep with one hemisphere at

a time.

But as soon as it comes to land, then it

says, okay, great.

I don't need to consciously breathe, you know, underwater.

I can have my REM sleep and I can sleep

with both hemispheres and it switches.

So it shows you two really important things.

First, this is the this is much more efficient to

sleep with both hemispheres and to have REM sleep.

At the same time is not essential, right?

It's hard to wrap your head around.

How does an animal get away with not having REM

sleep yet?

As soon as it has an opportunity, it will do

loads of it.

Right.

So it must have a function.

We just don't know what it is.

Okay.

So to sum up so far, all animals with a

brain need to sleep.

As far as we know.

If you discover an animal that doesn't need to sleep,

nature paper, great sleep structure and requirements vary greatly across

different species with different ethology, the habitat, the behaviour, all

of that seems to influence how the animals sleep.

So there is flexibility in what a species can do.

It's not like I'm human.

I have to sleep 8 hours.

But it might take a long time for me to

adapt to then sleep more or sleep less.

So the big question is why is sleep important?

So you have to look at sleep and say this

is the worst possible idea evolution could have come up

with.

Right.

What is your function as a species?

It is to make more of you.

To make more.

Right.

So mating.

You need to get food for energy so you can

do that.

Okay.

So foraging, hunting.

And then there are other animals that are looking for

food, too.

And if they hunt you, you're not going to mate.

Right.

So you need to avoid predation and sleep is one

third of your life and you can't do these essential

things.

So why would why would the brain evolve to take

give you a disadvantage in these things?

You must be getting something in return.

But it's we still don't really understand what we are

getting in return.

And just to emphasise how important sleep is, right.

You would probably say, well, you need food and water

more.

Right.

Well, the world record for no water is 18 days.

Okay.

No food.

74 days.

How long can someone go without sleep, you think?

11 days.

Hey, Randy Gardner.

Hey.

Now it's even possible that part of his brain was

sleeping.

There are some reports of local sleep, but I think

he had problems after this experiment that he performed on

on himself.

And you probably have tested yourself on going one day

without sleep, maybe two.

It's quite difficult.

You don't function well.

But I can tell you, as someone who has young

kids, I went through a lot of sleep deprivation.

You are not the same person.

Okay.

I'm still recovering.

And they have done this as well on on rodents.

And, you know, you put the animal basically on a

on a dish that's rotating.

So the animals not just have to move quickly, but

it it's not continuously awake and where it goes into

the water.

And within 11 to 30 days or so, all the

rats died on this experiment.

Now, one very important thing one can found is that

lack of sleep leads to stress.

And stress has a huge effect on your mental mental

health and well-being.

Okay.

So it's hard to know whether the ability to remove

stress when that's gone, we are in trouble.

And that's why we're having these problems.

Or is it sleep itself?

Okay.

And it's also a vicious circle, as many of you

probably realise that when you're stressed, you can't fall asleep.

When you can't fall asleep.

It will lead to more stress.

Okay.

So sleep is as important as mating, hunting, foraging for

food, avoiding predation.

But we don't know why.

So some ideas.

All of you have been sick at one point.

What?

What do you do?

You go to your bed.

You rest.

You sleep.

You sleep often more.

So there are some benefits to your immune system and

wound healing.

Now, why that is special.

To sleep and not just resting in bed.

There seems to be your immune system does seem to

be synchronised to your sleep cycle and there is probably

a lot more there.

But why that's the case.

Why do you have to be unconscious for your immune

system to kick in and act differently?

We don't know.

Second thing is energy conservation and protection.

So many animals are adapted for hunting or foraging at

night or during the day, but not both.

Okay.

So a good example is, you know, like the nocturnal

primates that you find with these huge eyes.

Right.

Those are great.

In Night-Time, they see much better than we do.

But if you turn the light on, they're like, Oh,

right.

They would not be able to hunt.

And in fact, they would be a danger of predation.

So the best thing they can do is just hide

right here during the day, hunt at night.

But that still doesn't answer the question.

Why do they have to be asleep?

Why can't they just, you know, hang out in their

their cubby-hole and just wait for the right time?

The next idea is homeostasis.

So when you're learning, hopefully today you're learning or your

neurones are building new synapses, the synapses are getting stronger.

But if you do that again and again and again,

there might be some negative effects of scaling up your

synapses too much.

So during sleep, it's thought that there is a homeostatic

mechanism that rebalances things, that things don't go too high

or too low.

And while still maintaining the changes that occurred during learning.

Okay.

Fourth is what's called the washing machine hypothesis that when

you sleep, your brain is able to get rid of

all the crap that's built up.

Okay.

So what do I mean by that?

Well, if you look at amyloid beta, right, related to

dementia, humans have a higher risk.

Okay.

If your sleep efficiency goes down.

So if you're not sleeping well, you have more build-up

of these toxins, which then can lead to dementia.

Okay.

And another example, you can look at how another thing

related to Alzheimer's and there is a build-up of tell

if someone is sleep deprived.

So every time you're going out and partying all night,

you're like, I don't need to sleep.

I'm young, you're building up towel your brain, okay?

You have a big capacity, you know, for how much

you can have.

But at some point you want to sleep and clear

this stuff out so it doesn't build up over time.

And it's not surprising that as you get older, sleep

quality does decline.

And these are associated with a lot of these sorts

of problems.

Okay.

Now, if we look at the brain, this is with

MRI, you can see that there are ways of activity

happening during sleep.

And when the activity goes up, okay, this happens slowly

up and down when it goes Sorry, when it goes

down.

There is a CSF exchange which you see in blue.

So activities in red, when the activity goes down, the

CSF in the brain gets flushed out.

And with this CSF, we think there are a lot

of toxins that have built up and then it's sort

of replenishes.

So that's a bit like a washing machine, right?

It's sort of flushing the brain out.

And then you have new CSF filling up the region

and this way you're getting rid of stuff.

Okay.

If you don't sleep, maybe this can still happen, but

it seems that it's not as efficient.

And there's something about sleep that is ideal for this

to happen.

Okay.

So I talked to you about four hypotheses.

The fifth one I'll focus on for the rest of

the lecture, because that's what my lab does.

And I know more about that than the other four.

So the fifth hypothesis where there is a lot of

evidence is that sleep is important for memory.

And this idea that your recent memories need to get

backed up to long term storage.

And this has to happen during sleep.

So there are lots of experiments you can find out

in the literature where there is some sort of task.

I mean, there are really dozens and dozens of these

sorts of studies.

On the y axis, you see that there's an improvement

in the task.

And if you've taken a nap after you do the

task, you'll do better than if you didn't take a

nap.

Okay, so sleeping is beneficial At the same time, if

you sleep deprived someone, you'll see that their memory retention

is is not as good.

So sleeping is good and not sleeping.

When you're supposed to sleep, you do worse than you

should.

Okay.

Now, as I said before, not sleeping leads to stress.

And there are you know, there are other factors you

have to examine.

But even when you do these things, sleep seems to

be better beneficial for your memory.

So I told you that sleep is divided into REM

sleep and non REM sleep and mammals and birds.

So what is what do you think is more important

for memory?

Is it the REM sleep or the non REM sleep?

Ram.

Raise your hand if it's Ram.

Raise your hand if it's non-REM.

Okay.

A couple brave souls.

Good.

So here's a task.

I'll answer the question in a second.

Here's a task where someone has to remember the location

of different different dots on the screen.

And while they're doing the task, they smell this odour

in the background.

Okay.

I believe it's not a very good odour, but it's

for the purpose of getting their attention and making that

association.

And then when they go to sleep, they can present

that odour while they're sleeping.

And the idea is because the olfactory bulb has direct

connections to the campus, that smell is going to activate

neurones and remind the brain, the hippocampus over here, which

is active during sleep, reminding the hippocampus that, oh, this

is what you were doing.

Think about this memory and perhaps bias your your your

memory processing towards the task rather than all the other

things you did during the day.

So it turns out if you do this during REM

and waking, there is no effect at all on your

memory.

But if you do during slow sleep, if you present

the odour, the performance improves after the sleep session compared

to if there is no odour at all.

Okay.

So this doesn't mean that REM sleep is not important

for memory.

But it seems that slow wave sleep is where this

might this process probably starts.

Okay.

The other thing is this works If you just have

a nap.

And a nap is almost entirely slow wave sleep.

We're non-REM sleep.

Okay.

So you can do these control experiments that even without

any REM sleep, you are benefiting from having having non-REM

sleep.

Okay.

So what do you think is happening during this process?

What?

Why does this work?

So.

The brain is very complicated in lots of different areas.

We're going to focus on the neocortex, which is, you

know, the outside region, all your cortical areas for seeing

and hearing and touch and the hippocampus, which is in

your medial medial temporal lobe.

Okay.

Which is in blue over here.

So it's thought that during non-REM sleep or sorry, prior

to non-REM sleep, when you're awake, cortical activity is happening,

which is representing your whole world.

Right?

It's very much like if I stimulate part of part

of my visual cortex, I'll see a flash of light,

right?

It's not.

What is actually happening out there, is what my brain

is doing that gives me perception.

So all these signals in my brain that are giving

me my conscious experience are being transmitted to the hippocampus.

And the hippocampus is recording this.

Okay.

And when you go to sleep during non-REM sleep, the

hippocampus plays back all these memories.

The cortex, of course, is not as active.

It shouldn't be active at all from the outside world

because your eyes are shut and you're unconscious, but your

brain is actually still very active and your hippocampus talks

to you in your cortex and said, Hey, this is

what you do during the day.

Let's talk about it.

Okay.

And the analogy is that it's very much like your

computer.

You download content during the day when your computer is

on, you're working on it, and then you put your

computer in sleep mode and everything gets backed up to

the cloud.

Okay.

And you have these two modes.

When the computer is active and you're doing stuff, it's

going in one direction.

When your computer is in sleep mode, it can still

be active and can still do stuff when you don't

need the computer.

And it uses that opportunity to back up the data

if it did it while you're working at least, well,

maybe now computers are much faster.

But back in my day, it would slow things down,

right?

So it's better to have these two modes.

So what is the evidence that this is happening?

All of you know H.M.

I don't need to go through the story.

But he had his hippocampus removed to try to stop

the seizures.

And when I said the campus was removed, he suffered

from anterior grade and temporally graded retrograde amnesia for episodic

and semantic memory, which meant he couldn't form new memories,

and his old memories were partially affected.

The further back in time those memories were initially encoded,

the better his memory was, Which means the new experiences

require the hippocampus and hippocampal damage does not erase old

memories.

So if a computer example we just had.

If I take my computer.

Right.

And I don't know, it just blows the fuse.

Doesn't work anymore.

Okay.

Everything.

I backed up to the cloud I still have access

to that's still there.

I can't store new memories.

I can't download content.

I don't have a computer anymore, but I still have

access to the old memories.

And that's what we think the hippocampus is, is for.

Taking new memories and storing it somewhere else.

So the hippocampus is gone.

You just can't do it again.

You can't store new memories.

Okay.

Now, that's in humans.

What happens when we look at rats, right?

We can study neural signals in the brain.

These are experiments I do in my lab.

This video is not from my lab, but we record.

We put electrodes.

Many, many electrodes to record 5000 neurones at the same

time.

And when neurones, when you're listening to neurones, they act

very much like instruments in an orchestra.

They're all doing their thing.

And when they produce the melody together, it's very easy

to recognise they form patterns.

And the question is what are those patterns there they're

forming?

What do those patterns mean?

So.

Okay, so you can hear the sound.

Isn't the sounds of the spikes when you're recording.

What you're seeing is a colour code over here of

the different spikes.

Okay.

And what you find is that when the animals and

when a particular neurone is is responding, it likes a

particular region of the track, the animals running.

So I know I talked about the hippocampus as being

important for memory, and this will come full circle to

that.

But when you're recording in the Rat, it seems like

these neurones act like a GPS.

And so the neurones like one location, another neurone will

like a different location.

And as the animal is running, you have neurone A

and B and C and D firing.

Okay.

And I'll give you a little cartoon over here.

Here's animal running.

And you see the first cell fire, then the the

second, the third, the fourth.

And if you were listening to the brain and you

saw blue, green, red, orange, you say, are the animals

running along this track?

Hey, I know what the animal is doing.

But then the animal goes to sleep and you see

the same neurones fire in the same sequence.

This is a phenomena called replay.

It happens much faster, but it's in the same sequence.

And because this sequence is very unique to both the

track and animal's behaviour and can decode this and say,

we know the animal is thinking about the track and

its sleep.

I don't know if this is a dream.

You can't ask the animal what it's what's the conscious

experience.

Right.

But we can see these patterns are very unique.

They don't happen by chance very easily.

And this seems to reflect what has happened.

So just so you can hear how that sounds.

And remember, that replay event is 5 to 20 times

faster.

So it sounds like all the neurones are fired at

once.

We'll look at the video again.

You're going to see a neurone fire on the bottom,

in the middle, on the left side, on the track.

And when it gets to the corner all the way

on the left, one is rounding.

That's where the replay event will occur.

And I'll I'll tell you when that happens.

So.

So it's running the blue cell fires.

And another the science cell.

Another blue cell.

And here, that was a replay event.

Okay.

And if you saw the cell activity that we play,

there are actually cells active in front of the animal

where it's planning to go.

Okay.

Now.

So the animal is not asleep here.

Okay.

This happens when the animal's awake as well, when it's

sort of tuned out and not running.

You see these replay events, but when the animal goes

to sleep and you're listening to the brain, you hear,

Shh, shh, shh.

And these are all replay events.

Okay.

Yes.

Dear me while I was.

Is the replay as fast.

Yeah.

So when it's asleep and when it's awake.

There are probably differences in the replay, but we don't

really understand those differences.

But the speed is similar.

During REM sleep.

There's one study finding replay.

Okay.

And it's the same speed as the actual experience.

So maybe this is closer to what you have as

a as a as a dream.

The other thing is that during slow wave, sleep is

very fragmented.

So it's like part of the track boom animal replays.

And then one second later, another part of the track,

boom, there's replay.

Okay.

So it's not this.

Imagine a dream is you have this whole storyline that

doesn't seem to be happening here.

Maybe REM sleep does.

Okay.

So you have a behavioural episode.

Animals running along, you know, looking for food on the

track.

This is all getting encoded in your cortex.

It's getting sent to the hippocampus to store.

And then when the animal is not paying attention, you

know, sort of zoning out or sleeping.

We call these offline periods and there's a process of

memory consolidation where the hippocampus backs up this information, the

cortex.

And one reason for this is that the hippocampus learns

things quickly.

Okay.

And it has to decide what are the most important

things to store.

You can't store everything you experience in your brain.

The hippocampus has to make that choice.

So one thing we study in the lab is a

phenomena of memory triage.

Just.

Just like a Annie, right?

You can't help everyone.

You have to rank.

What are the most important cases to see first?

So what are the most important memories?

And during sleep, those get replayed.

So what happens if we disrupt replay so we can

do this by putting electrodes in the brain near the

anterior comissioner?

So the connection between the hippocampi and when we see

this high oscillation, which is called a sharp wave ripple

where replay events occur, we'd give a little pulse and

disrupts neural activity for 100 or 200 milliseconds in the

brain.

And if we time it right, we prevent replay events

from occurring.

Okay.

So we can train the animal on a memory task

involving different arms that are baited.

And when we disrupt replay, we find that the rate

of learning is slower in those animals.

They're still able to learn, but the rate is slower.

And you have to keep in mind that you're blocking

replay just for an hour after the task, and then

the animal can sleep in its own cage after that

normally.

So you're able to disrupt some, but it's not the

perfect experiment where you've blocked all replay that will ever

occur, you know, during the sleep session.

But this is already a hint that replay is important

for learning a memory.

So an experiment that I did a while back rather

than disruption, where you always wonder, maybe I have screwed

something up in the brain by shocking.

Can we do something that is positive?

And so where we improve behaviour or we can bias

activity towards certain memories.

So I train rats on a task where they had

to run to one side if they heard a high

frequency sound, like high notes on the piano, and they

ran to the other side or they heard lower frequency

sounds okay.

So they can do the task and it takes a

while.

Humans are a bit faster.

Rats can learn this in a week to two weeks,

but they learn.

Hey, I hear the sound around this.

I hear that sound runs the other side and they

go to sleep.

And what you would expect if you're recording brain activity

is you would see replay of running to the right

and replay of running to the left right.

These are relevant memories for the animal to learn.

It was getting a reward.

It has to figure out the task.

So you're going to see these replay events.

Now, the question is, if I play sounds associated with

the task, the sound telling the animal to the left,

the sound telling the animal to go right.

Does that affect the replay?

Right.

Just with the like, the old factory cue I showed

you before that influenced the memory.

So potentially, since the brain is still able to hear

sounds, maybe not in exactly the same way you still

hear sound when you're sleeping.

Maybe I can sort of tease the hippocampus to replay

what I wanted to replay.

And that's what I found, that you can you can

have a measure called the mean rate bias, which is

basically you have more activity for these neurones.

If the neurone likes running to the right.

Are those neurones more active for sound?

Are compared to sound.

The sound telling it to go right.

So the blue bar is higher, which means the neurones

are more than neurones unlike to run right respond in

animal runs.

Right are more active when you play sound.

Ah and the neurones that like going left are less

active when you play sound are.

So there seems to be a preference for what replays

when you play a sound.

And what's interesting is that this bias over here persists.

So you might think I play a sound.

I get a replay of just like that.

That's not what happens.

Okay.

You play the sound and you create a state in

the brain for almost 10 seconds where if a replay

event occurs, it is more likely to be in the

direction that is associated with that sound.

Okay.

So you're not causing more replay events.

You were just biasing what will happen when a replay

event occurs?

But as soon as you play another sound, the fact

goes away.

So it seems to be very, very intolerant to say

lots of different things you're doing.

Supplying a sound cue caused a greater reactivation of play

cells associated with that sound for more than 10 seconds,

which was the max tested.

So the summary so far is if you train the

animal to run left for.

A high pitched sound and run rate for a low

pitched sound.

Playing the high pitched sound will cause replays that left

at a low pitch Sound will cause replay to the

right.

If you do this in humans, humans, you don't have

to train them to run right or run less for

just two sounds.

You can do things that are more complicated so you

can have 50 objects.

Each object is the subject has to remember the location

the object should be.

So they're touching on the screen.

Oh, I see the cattle.

I need to press this part of the screen.

And each object is also paired with a sound.

Okay, so you have 50 pictures, sound pairings that the

subject has to remember and during a nap.

So that's only slow wave sleep.

They present half the sounds to the subjects while they're

sleeping.

Okay, So the sounds they present are called the cued

stimuli and the unseen stimuli are they still have a

sound, right?

But those sounds were not played.

Then they test the subjects with acute versus on cued.

Okay.

Importantly, they do not present the sound in the test.

It's only the visual objects and the person has to

remember the location.

And when they do this they see there is less

errors in the cued sound.

So there is learning, there is an improvement in the

cued versus on cued.

So presenting a sound during sleep from the right literature

suggest that we can actually bias what is replaying in

the brain.

And then on top of that, in humans, if you

have presumably more replay of one memory, you have a

better memory than if you're replaying it less.

Okay.

So.

What's the take home message when you study?

Maybe you should have some sounds in the background.

I don't know how robust this is for normal sleep

conditions outside the lab, but if you have test specific

sounds, not music per se, but test specific sounds so

like you have a word you have to remember for,

you know, anatomy.

And then you say, okay, every time I'm thinking about

this word, I'm going to hear the sound that if

you present that sound when you're sleeping, you'll probably remember

the word pairing better.

And so if yes, if you present those sounds while

you're sleeping, you should have a better test performance than

if you just sleep on it on your own.

Now, even more important, if you don't sleep, your performance

will definitely be worse.

Okay.

If there is one take home message, you need to

sleep.

And.

This.

This leads us to evidence that more replay leads to

better memories.

This phenomenon of replay, and that's something that we study

in the lab now of how do you manipulate the

amount of replay in a rats or mouse?

And how does this change the performance of the animal?

Okay.

So thanks for your attention.

I have time for a few questions, but.