raw_transcripts/lecture_5.txt

So that's best thing.

Dr. Richard Francis.

That's the they to take the word.

What is.

Hello.

Welcome back.

If you are.

My name's Sam Solomon.

It's my pleasure also to introduce to you today Professor

Elliott Carton.

He'll be taking a lot of this lecture.

So in the last two weeks, we've discussed the different

components of brains and brain function.

The purpose of this week in general is to help

you understand a little bit better to survey the variety

of techniques that are available now to study brains and

behaviour.

To set us up properly for the subsequent weeks, which

were on specific terms.

So the purpose of today is not to go into

in-depth any particular technique.

We just like to help you understand the panoply of

techniques that are available with the strengths and the limitations

of some of those techniques are and why one might

choose to employ them in particular circumstances to understand the

relationship between brains and behaviour.

This is now quite an old slide, or at least

most of it is.

I adapted it slightly on the x axis is the

timescale of which one might like to make a measurement

ranging from milliseconds through hours, days and even lifetimes.

On the y axis.

It's a spatial scale of which one might like to

make that measurement from the scale of a single snaps

of a single nerve.

So through two o columns of those whole brain areas

and perhaps even the whole brain, each of the little

things in the box are different types of techniques, most

of which will encounter some time today.

These include things that are, for example, called patch clamp

electrophysiology.

This is old school electrophysiology, incredibly powerful technique.

You put in a glass electrode, so its tip is

only 1000 or maybe 5000 millimetre wide.

You put that glass tip up against the membrane of

a single nerve cell in a slice usually, but sometimes

in a whole animal.

You suck that little membrane onto the end of the

pipette and then you break the seal from the glass

into the inside of the cell, what's called a giga

arm seal.

And that allows you to measure precisely the internal electrical

life of that.

So that's basically the smaller scale measurement techniques tend to

get through on the top.

Right, as we'll be discussing soon, is a kind of

knowledge you can gain from studying humans or animals with

lesions, maybe to large parts of the brain or to

even small parts of the brain, often formed by strokes

or other kinds of accidents.

And in between.

These are a range of different techniques that we'll be

encountering today, including electrophysiology, calcium imaging, optical imaging that Magneto

and Safflower Graham or the electroencephalogram MEG or EEG, but

also techniques that are not necessarily aimed at revealing the

functional activity of the brain, but the kind of connections

between different brain areas.

If we were to have this lecture back when the

slide was first made.

I think the slide was first made back in 1994

because there had been some profusion of techniques, including MRI,

into the field.

These techniques have now even grown more stronger and more

varied in their ability to answer these kinds of questions.

Hopefully we'll give you a flavour of that today, so

I'll hand off now to the radio to take you

through the first component, which is largely to do with

how we try and measure brain function in human beings.

Hello, everybody.

Can you hear me?

Yeah.

Hi.

It's really nice to be here.

I'm Andrea.

I'm one.

This is a professor of experimental psychology, and I mostly

do research on human cells quite nicely to telling you

about some of the techniques that we use.

And we wanted to start a little bit with.

Well, the first thing that was available for us to

study the human brain.

Right?

And that was sort of lesions in patients.

And that's something that's very much like we call neuropsychology,

although neuropsychology implies all those things as well.

And you're going to be seeing this a lot through

your modules.

And so I'm not going to go into a lot

of detail about it, but it is fair to say

that sort of the groundbreaking studies of Paul Broca were

the ones that sort of get this field going in

many ways.

So Rocco had a patient that had some language production

issues, and when that patient, that patient died and then

they examined the brain, they saw that they had a

lesion in the left hemisphere, in the frontal charges, in

the frontal cortex.

And they could also see that that happened again with

a couple of patients.

So that was the first real sort of link between

a function behaviour and sort of something localised to a

certain part of the brain.

And that was really groundbreaking for for neuroscience and for

psychology.

And it is the technique that we still use up

to today.

We're going to we're going to talk a lot about

hair and burn care during the language recognition lectures.

And we also are going to talk about a fascia

and how we still use neuropsychology to study the brain

in humans.

So as I said, it's just really good in terms

of getting this still going and making sort of relationships

between structure and function in the brain.

And what we can do now is also say, okay,

well, we have these Asian to have this lesion.

Let's try to see how they perform different tasks.

And again, you begin to see this a lot.

A classic example, the studies of our friend Amelia with

patients, H.M. and.

So yeah, really groundbreaking as well.

That went to a lot of studies.

We learned a lot about memory from those and that

this technique obviously, you know, has advantages which are like

and how we can link brain, um, behaviour and potentially

identify regions that are necessary and lesions that never sort

of sort of constrained to certain regions of the brain

if you want.

We don't really know what's damage.

The damage will be different between patients.

So it's a little bit, it's a kind of like

what people call sometimes a nature experiment.

Well, someone had a listen.

Let's try to understand what happened.

And I think in terms of disadvantages was that it

was that the lesions are not selective and that also,

obviously, the brain has this this capacity of plasticity of

of change in potentially structure and function to compensate for

what's happening environments is what happened in its own physiology.

And so there are a lot of cases that actually

the brain compensates over time.

So we see this a lot in patients and you

will say the cases of aphasia were potentially at the

beginning after lesion.

A patient has like several sort of potential issues with

producing or understanding language.

Let's say that as time passes, many of them recover.

So not all of them.

And that's like a real interesting or like a big

area of research trying to understand why a patient will

not recover.

Many of them do.

And that has to do with how the brain can

compensate for that damage.

However, there are lots of things that you can't really

do.

You can't really study dynamics between brain regions because you

know that lesion is already damaged.

So, you know, that's the only thing that you know.

And and I think I think for me, the biggest

one is that, well, deletion is not selective.

So you wouldn't know what exactly is damage or to

what extent.

And therefore, it's very difficult to make sort of inferences

between the structure and the function.

We can do it by studying sort of large groups

of patients.

But for example, there is currently study in Queens at

UCL where they're trying to look at that Netflix of

yeah, so people would have patients are going to be

all the brain lesions and therefore sort of language production

and perception problems and they're trying to look at what

predicts recovery and they really need to recruit thousands of

patients to be able to do this properly.

So it is, it is quite hard.

Okay.

So I'm going to talk now about some other techniques

that we can use to actually measure brain function and

structure in healthy individuals and in a non-invasive way.

And the first one I'd like to talk to you

about is magnetic resonance imaging.

And I'm just going to show you a video that

explains this very well from one of the developers of

this technique.

If you take, let's say, a human being and put

him in one of these big, very homogeneous magnetic field

magnets, then there's a tendency of that magnetic field to

line up the magnetic moments of the nuclei, the spin

of the nuclei and the hydrogen in your body, which

is in your muscle and in your blood.

Now, put on a radio frequency pulse, say 60 megahertz

or something like that.

Then you can make this magnetisation of your hydrogen nuclei.

You can turn it 90 degrees away from the direction

of the magnetic field.

Your magnetic moment will process.

If you have coils around, pick up coils, they it

will induce a signal.

If I want to see where the signal is coming

from in your body, I put on another magnetic field

on top of the very homogeneous one that's called a

magnetic field gradient.

By that I mean it makes a feel stronger in

one place and weaker in another place.

What you do in order to actually get the image

that you want, the fan resolution that you need, that

you put on magnetic field gradients of different strengths, a

series of pulses.

That's why if you get an MRI machine, you're here.

Bom, bom, bom.

It's all these pulses that we're putting out with different

strengths of magnetic field, perhaps even different directions, because we

want to get a three dimensional picture of you.

You record all of these data and when you finish,

you can now use what's called Fourier transform.

It's a mathematical technique.

You can work back to how strong the signal was

in each of these voxels.

And so this is how the image is developed.

Okay.

So that was a very quick intro to.

Right.

So am I in the wrong one?

Okay, so MRI stands for magnetic resonance imaging, and we

usually use this really big magnets, scanners to to sort

of scan people to conduct this technique.

And functional MRI refers to the kind of specific scanning

that is used for measuring, you know, something that relates

to brain function.

And I what I'm going to expand as I'm going

to explain both of them.

And so how I.

It's just plain.

Old.

Okay.

Thank you for your input on my music selection.

And so in Ryan ephemera and material I'm going to

talk about a little bit later, it really changed the

game in terms of what we could do to study

the human brain.

And in the last 25 years that these techniques have

been around, we have gained a lot of understanding of

how many functions that we can only study in detail

in humans, such as potential language or some complex decision

making or, you know, metacognition and so on, how they

work and how how, how the brain represents or like

produces its functions.

And so how does an MRI what I mean to

me, it's absolutely amazing that we can look into the

people's brain and actually look into their function without even

putting anything inside.

We can just take pictures of their brains by measuring

how hydrogen atoms move, which is why I was trying

to say here.

So this is what happens.

You put the subject a the participant in a big

magnets.

The magnets and had a really, really strong magnetic field.

And to do that you need liquid helium and you

need to keep that constant all the time.

So the energy cost of doing this is quite high.

And that's why MRI is actually a really expensive technique.

And when you put a person in this big magnetic

field, as it was thought by all the sort of

elite hydrogen atoms were aligned in the same direction.

Right.

That's why you don't use trying to put them all

in the same direction.

So then when you sort of disturb them and when

a radio wave.

Right, they will sort of rotate.

When you turn that off, they're going to go back.

And when they go back, they're going to miss the

signal that you can think about sort of like a

kind of singing in a certain frequency.

You can see these electrons.

And so these hydrogen atoms are thinking in a special

frequency when they're going back to the alignment.

Right.

And that's why we're measuring now how how then can

we measure different parts of space if they're all aligned

in the same way?

That's where you introduce these gradients.

So the gradient is just changing the magnetic field slightly.

So in a way, you can think about these hydrogen

atoms sort of kind of moving and what this thing

that's going to be sent is going to be at

a different frequency.

You can think about it as thinking in a different

frequency.

So for example, if I wanted to know, you know,

how this lecture theatre was populated, but I couldn't see

it, I was in a different room.

I could just put microphones all around with different with

different frequencies and then just have a recording here.

Right.

So the guys at the very top are going to

have really low frequencies, and that's going to be a

great note here to very high frequencies.

So I'm going to go in the other room and

she's going to be recording the signals from all those

microphones when you're talking or whatever.

And by looking at the frequencies of those recordings, I'm

going to be able to exactly say who was sitting

where.

Because that's the link between frequency and space.

And that's why MRI dogs and that's why I like

to see pictures of your brain, which is amazing.

So if there's like a really strong signal at a

specific frequency, I know where it's coming from.

That part of of their image is going to look

quite bright.

And if it's a really low frequency, then it's going

to look like that.

And that's how we construct and it's really nice brain

pictures.

And so, yeah, pictures like this, for example.

Right.

So you can see and this is just a transformation

of light, really strong powers.

These things are like y by looking wide or depending

on what y your technique is, depending on the contrast.

And this the opposite for things that, you know, they

don't have that much signal.

Now this is just to get a structural image.

This just gives you a picture of the brain, right,

That doesn't link at all to brain function.

So what is it that allowed us to use functional

MRI to look at brain function?

And it is the fact that when a certain part

of the brain is active, there will be an increase

in blood flow to that part of the brain.

When there is an increase in blood flow, there's going

to be more haemoglobin, haemoglobin coming into that and that

has iron.

And it's in a sense the iron is ferromagnetic so

interferes with that magnetic field.

So by measuring how much interference there is with the

magnetic field, we can measure brain function.

But as you can see, the images and I'm sure

you hear one is the structural image to the kind

of images that we use for studying the structure of

the brain.

And this is a functional image.

It's just very low resolution because we actually want to

see.

How it changes with time.

So acquiring one of these images it takes for the

whole brain, it takes at least 5 minutes, whereas acquiring

this one could take 2 to 3 seconds depending on

what you're doing.

So we kind of compromising some of the spatial definition

to have like a better temporal resolution.

But we're still measuring blood flow, and blood flow is

quite slow.

That's why ephemeral is not a good technique for measuring

brain function.

So and you think it's no good technique for measuring

sort of Air Force temporal events is not it doesn't

have a great temporal resolution.

That being said, last week there was this really exciting

paper published where they had an MRI technique and so

milliseconds resolution now like very fast.

And it was done in animal models and it was

thought like in a single slice.

So it's not the whole brain, it's just the very

first step towards this technique.

But I think, you know, everybody in the field is

really exciting because it really looks like in a few

years, like maybe ten, 20, who knows?

We are going to have a technique that allows like

a really great temporal resolution potentially to study the human

brain as well.

And so when you get these images where you do

in the functional case and ephemera is that you take

many, many, many of them and then you compare them

across conditions.

So, for example, let's say here I will record like

lots of images do one of my goodness conditions of

my experiment, which could be it could be a memory

experiment, right?

So it could be like memorise these items.

And then I have a control condition where I just

tell people to, you know, look at them, don't memorise

them.

And then you can look at how like activity changes

across the whole brain.

Or like in here you can like look specifically into

a certain region, right?

So this region here is a few voxels, a voxel

instead of the minimal units and ephemera you can look

at as a three dimensional pixel.

And then we can average, for example, the activity through

that.

So if you look at this, the intensity in these

conditions is going to be different.

But this is something that you can't really tell, you

know, just by looking at.

And the difference is very, very small.

It's usually around, you know, between 1% to like 4%

or 10%.

Best case.

And and yeah, it's it's really small, but it is

very consistent.

So you can measure that and you can measure how

it changes.

And that's what we're doing here.

We have like these three measurements here, for example, from

that region in red, and then we compare them to

the measurements during the second condition and averages and see

this a difference in the amount of like bull's signal

in that area.

And by doing that across the whole grain, we can

come up with these statistical maps where we're showing where

in the brain there are significant differences between conditions.

Okay.

And so as I said, FEMA has brought lots of

advantages.

I think one of the really nice things is that

we can study the dynamics of the whole brain.

And, you know, that doesn't only apply to humans, but

it's also been used in animal models for this specific

purpose as well.

And after certain spends, we can understand how different parts

of the brain interact and how they form functional networks.

So we can like sort of do more complex analysis

of network dynamics and see how reaches interact under different

situations.

And you could do things that we couldn't do for

like with patients because, you know, patients essentially have some

sort of behavioural deficits.

So like there's things that they wouldn't be able to

do that we could do now because they're healthy participants.

And I'm very nicely we can think we can study

human brains, right?

We can study things that are so specific to humans

that would be difficult to study in all that and

which models.

So, for example, I study deafness and how the brain

of deaf individuals changes.

And you know, when the congenitally deaf and continues with

deafness in many cases results in a delay in language

acquisition, which then has a lot of impact on cognitive

functions as well.

So it's not that animals are not great for answering

those types of questions, so great for studying the effects

of sensory deprivation or lack of sensory experience, in this

case, a hearing and in the development of the brain

in plasticity.

And so it not so much to study what are

the impacts on cognitive processes.

And I think the best example of how we gain

from ephemera is we can do things like communicating with

patients in a coma.

And I really recommend you to look at these videos

from Agent Owen and his group where they are using

ephemera and what we know about from right.

They have been able to communicate with and with patients

that before we didn't have any insight into their entire

health status.

And so that has been a really amazing ground breaking

discovery.

So I'm not going to go into detail here right

now of time.

And there are tons of limitations, though.

And again, I think know, we have some understanding of

what the f MRI signal is.

So is this mix of like presynaptic activity and like

sort of cells firing as well.

And like, we don't know which ones it's coming from.

So you never really know what type of activity you

measure.

And you've just seen a difference in overall blood flow

in one region to the other.

So that is quite limiting.

And and again, the spatial resolution is great, you know,

like in comparison with other things.

But still, like we're looking at a bunch of cells

like and it's going to be quite hard to tell

what individual cells or small networks of self are actually

doing.

And and again, it's like this really big magnets.

So people have to stay, you know, sort of still

they can move around and that just gives limitations as

well.

I think, as some said, there's been a lot of

advances in these techniques and now we have MRI for

humans where we can measure layer specific activity.

So different layers of the and, you know, the cerebral

cortex.

But still, you know, that's still a lot of cells

and not specific.

Now, before we pass into other techniques, I wanted to

talk a little bit about this technique called functional, and

yet infrared spectroscopy, which works a little bit like ephemera

in sense that it measures changes in blood oxygenation.

So you have different sources of light that are attached

to the skull, and then you have others that you

use for detecting them.

And again, if there's a change in blood flow because

of haemoglobin, that will that will affect how the light

is reflected.

So you can measure changes as well.

Now, the spatial resolution of this technique is way worse

than if it were high, right?

Because you just only have a few seconds.

So why am I talking about it?

What would be the advantage of having this technique?

Why do you think?

Any ideas?

Yeah.

Yeah.

So exciting.

So one of the advantages that you don't need is

a big piece of equipment and is way more mobile.

Yeah.

People.

But you were always trying to.

Are you?

So hospital?

Yeah, exactly.

They don't have to be like in a specific lab

as well, although many cases are.

That is exactly that is one of the advantages.

So you're sacrificing some spatial resolution to actually be able

to do things like, you know, study people moving around

or like study children that are not going to be

staying very still in an MRI scanner.

And I think these are the kind of decisions that

we as scientists need to make, as we want to

learn about something, we're going to discover something about specific

behaviour, about the brain.

What's what is the best approach for doing this, given

the options that I have?

So in cases where you want to study brain function

in children, for example, or for example here where you

want to sort of like study people interacting and moving

around the environment, then obviously it's is going to be

a better technique that if I'm right.

Um, okay, so there are other techniques that are kind

of the opposite of around some that they have great

temporal resolution but not so good spatial resolution.

And those are magnetic photography and electroencephalography and energy and

energy.

So EEG have been around for a, for ages and

that was kind of the main technique that was used

to study the human brain in sort of healthy individuals

before any CnF MRI arrive other than just behavioural studies,

of course.

And so how do they what that means?

Just yeah, Okay.

So if you have a group of neurones, let's say

a line like it will be the case in the

cortex, they will have like went in different conditions but

actually they will have these currents moving in one direction.

Right.

And that is what generates EEG signal.

That's like the actual electrical activity.

And at the same time there will be a magnetic

field form surround that current that has gone in one

direction.

And that's why you can detect with energy.

And I think where you can see the main difference

in this really retro slide that I call it, because

it just happens quite clearly is that if you have

this current here, like in the case of ETI, it

doesn't go like and in sort of predictable way into

the skull because there is like, you know, sort of

bones and other tissues and liquid and so on, it

kind of deteriorates in a way that is quite difficult

to understand where it's coming from.

So you can still measure it here, but it's hard

to say where it's coming from.

And that's why EEG has a pretty bad spatial resolution

for energy.

This is much better because it is somehow more predictable

but still difficult to understand where exactly in the brain

this these signals are generated from.

And there's been a lot of work in trying to

do that.

And there have been some advantages based like kind of

I mean, when you talk to people that try to

understand this and try to sort of generate models of

where signals are coming from an EEG, and they basically

say this is an impossible problem to solve.

Like, you know, we can have like a really good

like closed solution as possible, but there's always going to

be a few different options.

So you're going to have to make decisions in terms

of how that now vanishes.

There's that that because I mentioned electrical activity, basically, you

know, this is direct recording of electrical activity, of neurones

firing or the magnetic field is generated and the temporal

resolution is excellent.

So again, you might want to try to understand what

experiments you want to do and whether, you know, this

technique is the best.

One of these techniques is by a technique that MRI,

for example.

And so I think I just want to show you

again, you can see there's a real difference in set

up here as well.

So energy, again, has like this really big machine, huge

magnetic field.

There has to be sort of maintain you could helium.

So again, quite expensive, but it now has to stay

really, really still whereas EEG is quite portable.

You know you can have a cap it's way cheaper

as well.

So trust me, you took my meds, you were like

the v0y is the best thing that I could do.

And then you have limitations of what your budget is

as well.

So, you know, that comes into it as well.

And so sometimes you might want to use one or

the other.

And I kind of did this sort of comparison here

to show you what are some of those things that

you will have to take into account.

So they both have excellent temporary solution, but similarly, they

both have problematic spatial resolution, although as I say, energy

is way better, right?

However, it is way more expensive and participants have to

stay very, very still.

I would say that applies to each year to some

extent as well, but potentially less and the sense.

US.

I'm in this big machine and that sort of how

the people haven't thought.

So you can't really move around.

So you need this special lab as well.

Whereas EEG is way more but more mobile and that

spatial resolution is was.

And I think it's also is good to take into

account that the signals come from different places as well.

So I'm not going to go into that.

But there are differences in where they're generated as well.

And and I think what is really nice as well

is that right now here at UCL in Queen Square,

in collaboration with all the labs in the UK, there

are developments to do with some portable e.g. such as

have like a few sensors that potentially measure brain activity

in a specific part of the brain so that people

can be more mobile and potentially have this much better

temporal resolution, so much better spatial resolution with this really

good temporal resolution as well.

And so I think this is the main so summary

of things that you can use for measuring brain activity

in humans.

I would really recommend you to watch all of these

videos, professors from our department talk more about these techniques.

And I think more once there's this is not an

extensive coverage of all the techniques that you can use

to study brain behaviour is more just to give you

a flavour of why the possibilities are and what are

some of the considerations that you have to take into

account when choosing one of those techniques?

So I'm going to pass to some now.

So it's going to be nice.

So the structure of this session is a little bit

maybe obscure for you.

Just to make clear why Bailey is talking about some

things that I'm talking about.

Other things is a video works on humans, particularly if

humans.

But I was always it's not quite always exclusively.

I grew up as a scientist working on primates, non-human

primates.

Both macaque monkeys and marmoset monkeys.

And I then moved when I moved to UCLA.

One of the reasons I moved here is that you

is basically the world centre of focus.

But for trying to understand the massive small world they

introduced in the last week.

Writing about half of France €70 million.

So the question I would like to pursue is what

I want you to think about.

Why would I choose to study animals?

And if I'm studying animals, including mine, which seems to

be so different from.

Why that I'm in the experimental psychology department.

Not only can I keep ahead of the problem.

So I'd like to then explain to you why I

find this such a beautiful experimental technique.

It's really exploded over the last ten years or so.

I'm going to introduce you to some of those techniques

that are really appropriate for the last ten years, and

we will be going into those in more detail.

We've actually.

But basically the reason we use animals or the reason

we turned out in some kind to understand why we

make based from them.

By that I mean that we can and.

Make small holes in their brain and introduce devices that

we can then use to measure cellular function in those

apple.

We can never be able to cover.

Study.

Study.

So.

But why do my people wear what I love about

Merhi?

Even in the best case scenario, one includes something like

$100,000.

But we can look at the active.

You need to be in the range and find out

how that could be.

That single macro might relate to cognitive functions.

If the animals, the kinds of things that do have

probably going to be very.

For example, we're in the very early parts of the

program.

People that critics are similarly enamoured with.

That makes me wonder.

I mean.

And I think we can learn a lot about how

we see or hear, or at least the signals that

come from that is or is at the centre of

the brain.

It's less clear how much we can learn about cognitive

structure thought, for example, by studying animals.

So it is when meant when we're measuring in invasively

would not be very limited circumstances that if we have

time we'll get back to at the end of this

lecture able to make these recordings from humans and he's

already introduced electroencephalogram or EEG.

That's a scalp based measurement, something that's non-invasive.

All the other things here, invasive measures, we might measure

the electrocardiogram, this liquid surface dura between the scalp and

the brain measuring almost directly neurones line underneath the electrodes.

Or we could use a little invasive, like a little

piece of silicon.

Let it be inserted into the brain first.

Slowly.

And located a particular point in the brain.

These have different spatial scales of measurement.

Electrocardiogram from each of the electrodes from the surface of

the brain, intermediate between B, g, and something else.

So maybe it measures that mm accumulated the brain issue.

The local feel and the electric car they find at

the end of one of the electrons moving about the

activity of about maybe half of their brain tissue.

And then if your electrodes, these little wires on fit

have been made well enough, you can actually start to

see the.

Speaking to people there for 2022 that you that in

the next slide we call those things spikes.

So this might be the signal that you would you

would acquire from one of the white actor to.

And the top.

This is 3 to 3 traces of the same trace.

On the top is what we would estimate as the

local fuel potential.

That's the sum that at the end of the start

to activity between the records and the ground, my whole

area.

And you can see that normally it's a local film

that was varying fairly slowly as the red line traces

out some of the slow fluctuations.

It's traced maybe a second or 3 seconds long.

And you can see that it is dominated by these

two things that are going.

But it would be very gratifying to deal with these

fluctuations if we then are able to look at those

little rapid events and those are indicated by the red

dots here.

We will see that they are the extracellular signal of

the action that we discussed in the last lecture.

How next?

Looks like from outside the city mirror.

There's not quite.

Now because his electrode is still quite large and there's

many cells around the end of those action potentials will

come from one another.

It doesn't really look.

Really?

Okay.

This better here.

Okay, I'll do that.

Thank you.

So the point to hear the difference down here, sir.

Thank you for that.

Each of each of those made.

It is like five or ten nerve cells around the

tip of the electrode.

Each of them are producing action potentials occasionally.

Often you cannot distinguish between the shapes of the action

potentials.

Just find your own one on your own.

Two on your own.

Three on your own.

For.

So we can combine these action potentials and say, well,

they're coming from single nerve cells.

It's just a few of them.

Five or ten of those may be in the local

area.

And therefore, we would call this activity the multi-unit activity.

So there's multiple units.

And by the way, you will come across this word

unit quite frequently in this course.

That just means a nerve cell in the brain recorded

in this way.

But sometimes if the electorate really is really nice and

it's really close to the so some of her neurones

in the brain tissue, you see a reproducible waveform.

Looks the same every time you identify in the recording

and that we would call single unit activity the activity

of a single nerve cell, the same nerve cell on

each turn, so able to record with this piece of

wire or multiple pieces of wire.

The activity of these nerve cells in the whole living

brain, in animals that are awake, behaving, even moving around

all environments.

So we can.

I think that this technique, which has been around now

for several decades, try to recall the activity of nerve

cells in the brain.

Oops.

This video shows you a video from the laboratory of

Torsten Wiesel and David Hubel trained maybe 50 years ago.

They won the Nobel Prize for these recordings.

These videos are hard to find, so I'd like to

show it here.

You're going to hear you can actually play those phrases

I showed you through an audio speaker and you can

actually listen to the action potentials that are being fired

by neurone.

In this case, they're recording from the visual cortex of

a cat.

And they're able to work out what this neurone is

saying by presenting different visual stimuli.

And you'll be able to see that during the course

of this video.

Well, you can hear the.

As I continue to continue to find.

What they're doing is they're going to the frontier now

where it's going to have to be a little crazy

for these things to be generated.

Reporter But this is a receptive field.

Getting into that.

And this particular unit is selected for the director most

of the time and.

Academic Congressional.

Important political orientation of the.

So that video just shows you you can use this

technique to see that these nerve cells in the visual

cortex are selective with the orientation, the motion direction of

of a visual stimulus.

And that's why they won the Nobel Prize.

By using this technique, you would not have been able

to determine that if you looked at the MRI signal,

which is something over hundreds of thousands of nerve cells.

You have to look at individual nerve cells to be

able to work that out.

Oops.

So the ultimate goal of all this new activity is

to control behaviour.

And I said that there were some limited circumstances in

which we could make these measurements in humans, including people,

for example, who are suffering from epilepsy, where we need

to measure from the brain to work out whereabouts the

surgical intervention could take place.

It also includes from people who in this case are

suffering from paraplegia, automatically paralysis.

Where the hope is that we recording the activity of

ourselves.

Cells will be able to help them regain function of

a robotic limb.

You have to effectively take what their brain, which is

still intact, is sending the signals that are sending and

use that to control something that we can then help

them move around the world.

In this particular case, these devices, which are often called

utile rays, which we have used in animals, I've never

used them in humans.

They were developed from being used in humans, have been

implanted in little part of the brain for the motor

cortex, which is important in generating movements.

And we'll get that in about two weeks time.

I just wanted to show you this video briefly because

it's incredibly evocative, like the Parkinson's video we watched the

other week and say to you, when we know how

we can interpret the signals and nerve cells, how powerful

that can be.

In this paper to people with Tetra and as two

people who were unable to move their arms or legs

in any functional, useful way, were able to control a

prosthetic or a robotic arm simply by thinking about the

movement of their own paralysed hand.

And they did that using the investigational BrainGate neural interface

system.

So they thought about using their own arm and hand

as though they were reaching out themselves with their own

limb.

And the robotic arm moved much the way their own

arm would have moved.

One of the longstanding questions not only in neuroscience but

in neuro rehabilitation, is whether the cells in the motor

cortex and other parts of the brain, whether they continue

to function the same way years after that original injury.

It is possible for people to use their thoughts to

control devices, either a computer or a robotic arm.

The way that happens is that we implant a tiny

sensor just about the size of a baby aspirin just

into the surface of the brain, and that sensor pick

up the electrical impulses from a bunch of neurones.

And each of those neurones are like radio broadcast towers

putting out impulses.

And when they get to the outside, the computer translated

converts the pattern of pulses into something that is a

command.

One of our participants was able to do something that

when all of a sort for the first time gave

us all pause.

She reached out with the robotic arm.

She thought about the use of her own hand.

She picked up that thermos of coffee, brought it close

to her, tilted it towards herself, and sipped coffee from

a straw.

That was the first time in nearly 15 years that

she had picked up anything and been able to drink

from it solely of her own volition.

There was a moment of true joy, true happiness.

I mean, it was beyond the fact that it was

an accomplishment.

I think an important advance in the entire field.

So I want to say that because actually that work

has, as we'll go into it next week, based on

the foundation of recording from these individual nerve cells or

small groups of them, in this case, maybe 50 or

100 nerve cells from the motor cortex.

And because we know what those individual nerve cells are

doing, and because we can record them simultaneously with these

kinds of devices, we can infer what the signal is

that she was trying to send her now arms at.

And now she's now disconnected from because she has paraplegia.

And because we were able to look at those individual

nerve cells, we can interpret that signal in ways that

is able to involve a prosthetic limb.

Now, when we when that device came out into the

market about 15 years ago, it seemed like a game

changer move from one electrode to 100 electrodes.

Well, recently and this is work from Nick Steinmetz and

colleagues from just across the road here.

We've developed new devices that can actually record from thousands

of neurones.

At the same time.

These are sometimes called neuro pixel devices.

We use these routinely.

Most people, many people use they'll now use these routinely.

We can record from multiple brain areas.

Sometimes the individual nerve cells, in each case a groups

of them within each of these brain areas.

And I don't want you to try and understand what's

on this line in terms of signals and measuring.

But the point is that we can start to interpret

how individual nerve cells across the brain are working together

to make cognitive decisions in these small animals lives which

share some of our brain capacity.

The other technique that you'll be exposed to over the

next few weeks is called calcium imaging.

I won't again, I won't take you through this in

any great detail.

It relies on the fact that every time an action

potentially is produced, I've talked to you about how sodium

ions come flexing into the cell.

There's also a little thing called calcium line to become

flexing in the cell.

And by designing particular molecules to sense the presence of

those calcium lines and make line signals dependent on how

many calcium lines are present, we can actually engage with

light the internal activity of so many cells at once.

We make injections or use animals of being transgenic engineered

to express these little proteins in many cells in the

cortex.

And we can use funky little microscopes or very large

ones, actually very expensive ones, to try and image the

activity of those cells in the brain without, in this

case, electrodes rather, using light to record the activity of

these nerve cells.

And this particular technique has a visual advantage with electrical

recordings.

We can't work out which cell we were recording from.

We know it was a single cell, but we don't

know which side it was.

But with calcium imaging, with imaging the brain, we actually

know which cell produced the activity.

We can take the brain down the animal after the

experiment, having killed the animal and then remove the brain.

And then we can process the brain tissue in particular

ways that allows us to look at the anatomical structure

of that brain circuit so we can take this functional

activity is activity recorded in, say, a hundred neurones in

a particular part of the brain and relate it to

the structural connections between those neurones and between those neurones

and the rest of the brain is not to bring

together.

Finally, how does individual nerve cells work together in a

circuit to provide cognitive function?

So that's why I'm interested in studying animals, because I

think that it's only by looking at the real structure,

the size structure of new activity in these circuits that

we'll understand the structure of cognition.

Now, I think they might disagree and say that animals

don't have interesting enough cognitive functions to allow us to

make much inference about humans.

But I hope that there will be some intersection between

those two things where we can actually find some cognitive

function in animals that we find interesting as humans, and

where we can understand the neural circuits that actually allow

that cognitive function.

So that's the span of the different techniques that we

might come across in this course.

The next election will be about how we can apply

some of these to, for example, rehabilitation.

And then after that, from next week onward, movies are

about specific systems, specific pathways for the brain.

We better wrap it up there as as was last

week, and you lead by this talk as well so

we can.

Thanks, everyone.

During.

Hi.

Hi.

Yeah.

Yeah, I know.

I worked that out this morning.

You think?

Yes, I'll do that.

Thanks for reminding me.

Like mom and.

Dad.

Oh, fantastic.

Love to see that.

Yes.

Good.

All right, well, I'll send those things to you today.

Anything else?

Not at the moment, I think.

Thanks.

I know at the amount of money.

Tucker Carlson get.

At the end of last week, they spoke briefly and

say how he's doing.

Yeah, I just want to know that we know from.

I can't blame the people.

Who are involved in the rescue issue.

In 1945.

Clinical Psychology.

I think having different people on the floor.

If you if you get.

I don't get the story.

I didn't.

I didn't.

Pretty promptly.

It's getting better.

I found last week because I'm given a lecture and

I live next to my two or three years.

Pacing was completely wrong.

Yeah, I know that.

She goes.

Is getting inadequate lecture fees?

Yes.

Down here.

Yeah.

I knew about luck.

Because the natural instinct is to end up that way.

Yeah.

Last week Hobson did it, and it's a week away

from doing it.

That's right.

There's no reason show will be the best.

But.

Oh, come into right now.

He's like, friend.

We think.

That there.

41.

Former.

Better.

It's been more than a game.

She.

Can.

But.

You.

Life.

Yeah.

So that was the thing that I think everybody.

That.

Great.

And you?

I want to point out.

Thank.

I didn't.

So.

So.

Okay.