The Science of Everything Podcast - Episode 46: Vision Part 2

Episode Date: March 23, 2013

Resuming from where we left off last time, we continue our journey through the visual system by explaining the structure and function of the bipolar and ganglion cells in the retina, including their somewhat complex receptive fields. I then discuss the optic nerve and cross over of information from different visual hemifields at the optic chiasm. We end this part of our journey with a look at the lateral geniculate nucleus, and how it is structurally organised to process different types of visual information in different locations.

You're listening to The Science of Everything podcast, episode 46, Vision Part 2. And I'm your host, James Fodor. In this episode, we're going to be picking straight up from where we left off last time with Vision Part 1. So we're going to be talking about bipolar cells and ganglion cells, and how those link up together and how those have their receptive fields that are sort of the on-center and off-center construction of the receptive fields of the ganglion and bipolar cells. We'll talk about that. also talk about how the information from the ganglion cells is transmitted out of the back of the retina via the optic nerve and then the optic tract. We'll talk about the optic chyasm and how the

crossing over of information occurs there, of the different visual hemifields, and then we'll talk about how this information then comes into and innovates the lateral genucid nuclei and how the different types of information from different eyes and parts of the visual field and so on are stored effectively or at least processed in different parts of the lateral geniculate nucleus and different layers of that. And we'll end off the episode by getting up just to the stage where we're ready to talk about V1, the primary visual cortex. Recommended pre-listing for this episode is, of course, vision part one, and hence the prerequisites for that. Okay, so we've got to a stage now where we understand how the light is focused

by the cornea and by the lens, it falls onto a particular part of the retina. We understand now that the retina is comprised of rods and cones, and that these contain stacks of membranes, which in turn are studded with these rhodopsin and photopsin molecules, and we understand how when the light hits these molecules, it's basically absorbed, and that leads to a cascade of events, which ends up with less glutamate being dumped out into the synapse by the photoreceptic cells.

Where do we go from here? So how does the glutamate, how does the reduction in the amount of glutamate that's in the synapse, how does that lead to visual perception and single things sent to the brain and so on? We still haven't actually got any action potentials here yet. We need to get some action potentials before the brain can do very much. Just having the neurotransmitter glutamate being, having a higher or lower concentration, that in itself is not going to do it. So to understand how we get from the changes in the amount of glutamate to actual information sent to the brain, we need to understand the next two layers that are behind the photoreceptor molecules. So remember,

we're still in the, we're still talking about the retina here. The retina is actually composed of three main layers of cells, sort of stacked on top of each other. The first layer is basically the photoreceptor molecules. These are the rods and cones that I've just been talking about for the last over in 20 minutes or something. The two other layers, however, are called bipolar cells, that's layer two, and then on top of those, the ganglion cells. It's kind of like a sandwich. You've got the photoreceptor cells on the bottom, bipolar cells in the middle, ganglion cells on the top. Ganglion cells are the actual cells that directly connect back into the brain through the optic nerve and leading to the lateral geniculate nucleus. So the ganglion cells are actually what sends the signals

into the brain. The photoreceptor cells is what actually detects the photons and does the transduction. The bipolar cells essentially just connects them together. Helps with the processing, but they don't actually connect directly either to, they don't actually directly detect the light or directly lead to the brain. They just sit in the middle of the two. So how do the bipolar cells, how do they react to the fact that there's more or less of this glutamate neurotransmitter sitting around? The bipolar cells synaps with one or more photoreceptor cells. Now, the number of bipolar cells per photo cell actually differs. In the retina, you've basically got one photo cell to one bipolar cell to one ganglion cell. It's like

one to one. And that essentially means high resolution. Out in the very peripheral vision, you can have like one photoreceptor cell goes to like 100 bipolar cells or something like that, which in turn goes onto a bunch of ganglion cells. So the correspondence is there depend upon the region of the retina you're talking about. Obviously, the more bipolar cells you have for a given number of photoreceptor cells, the high the resolution will be. Because basically, all the brain actually see, is the output coming from the ganglion cells. Remember, that's the third layer, bipolar sits in between the photoreceptors and the ganglions. All the brain actually sees the output of the ganglion cells. So if you have one ganglion cell that actually just gets input from

one photoreceptor molecule and another ganglion cell on the edge that gets input from a thousand photoreceptor cells, the brain sort of interprets those equally. So that one photoreceptor cell that goes to the one ganglion cell is going to get a lot more processing time essentially and it's going to contribute a lot more to vision than the hundred or thousand photoseptor molecules that all went to the one ganglion cell. So in the phovi, you've basically got a one-to-one correspondence. One photoreceptor synapses with one bipolar cell. I mean, it's not quite that simple, but that's basically it.

Whereas out in the periphery, you've got many photoreceptive cells going to a single bipolar cell and many bipolar is going to a single ganglion cell. And in the middle, you know, somewhere in between. Okay, so synapse remember, it's just the connection between two neurons. It's basically where the axon or axonal terminals of the pre-synaptic neuron, the first neuron, link up. They don't directly touch, but they certainly come very close to touching. with the dendrites or the input streams of the post-synaptic neuron.

So that's exactly what happens with the photo cells and the bipolar cells. Yeah, and the bipolar cells, they synapse with each other. And on the bipolar cells, you'll have just a number of neurotransmitter gated ion channels. So these are gaited by neurotransmitters, including, importantly, luteamate, which, remember, is the neurotransmitter that's released by the photoreceptor cells. So when there is plenty of glutamate in the synapse, so the synaptic cleft, the region between the two neurons, when there's plenty of glutameter, there, the glutamate binds to the glutamate gated ion channels on the bipolar cell and allows

those to open and therefore you get a depolarization or possibly a hyperpolarization. It actually depends on the type of cell, but we'll get to that. But you get a change in the membrane potential of the bipolar cell. And that's really all you need. To get a signal, we just need a change in the membrane potential. Again, we don't get an action potential yet. We're not quite there yet, but you do get a change in the membrane potential. When there's more glutamate around, you get more channels opening and therefore a change in the in the potential of the bipolar cell. When there's not a change in the When there's less glutamate around, you don't have as much glutamate that combined to these gated ion channels, and therefore they shut down, and you tend to get, again, a change in the membrane of the bipolar cells. It would tend to stay at the resting potential level, because you can't transmit the ions across the membrane.

Now, as I hinted at before, there's actually two types of bipolar cells. Well, there are more than two types, but two types that are most relevant to what we're talking about at the moment. There's on type, or just on bipolar cells, and off type, or off bipolar cells. So on and off. Fairly simple. Basically, the crucial difference between the on and the off types is when they are activated. On-type bipolar cells are active, or in other words, depolarized. So depolaris means remember that you get a more positive membrane potential. That's what happens when a neuron gets activated or when it prepares to fire an action potential.

It gets depolarized. On-type bipolar cells are depolarized in the light. So when the photoreceptor cells are detecting photons, the on-type bipolar cells, that are synapsing with those photoreceptor cells becoming activated or becoming depolarized. Off-type bipolar cells are the opposite. They lose their excitation or become silent in the light and are active in the dark. So both on and off-type bipolar cells respond to the glutamate that's released by the photoreceptor cell, but they respond differently.

Okay, so to understand how this works, just remember that the photoreceptor cells are a bit counter-rituitive and that they release more neurotransmitter in the dark than in the light. So it's actually slightly more accurate or sensible to say that they respond to darkness rather than light. So when it's dark, they're spitting out all of that neurotransmitter, all that glutamate. When it's light, they stop spitting out that neurotransmitter, or not as much. So how do the bipolar cells react to this? So if you're an on-type bipolar cell and you see lots of the neurotransmitter, lots of glutamate, what does lots of glutamate tell you?

It tells you that it's dark because the photoreceptor cell is not responding. It's not shutting down that glutamine output. Therefore, it's dark, so you're an on-cell, you should become hyper-polarized. You should move further away from activation. You should become deactivated. So remember, hyper-polarization corresponds to deactivation. It's not active. On the other hand, if you see lots of glutamate and you're an off-cell, and you're an off-bipolar cell,

and that tells you there's not much light, it's dark, I'm an off-cell, so I should be firing. And so off-cells become depolarized when they see glutamate. In the light, everything's just reversed. The on-cells, when the on-cells see less glutamate, that means, ah, it's light now, and so they become depolarized, they activate. When the off cells see less glutamate around, they know are. It's now light outside. I'm an off-type neuron of bipolar cell,

and so I become hyper-polarized, I become less active. Again, that's a little bit confusing, but don't worry about just if the sort of glutamate interactions are confusing. But don't worry too much about that. Just remember there are two types of bipolar cells, on and off-type. On means you're active when it's light, off means you're active when it's dark.

Again, it might be a bit of counterintuitive that you have these two types, but they turn out to be very important for visual perception. We need both of these. And, of course, before I was saying things like, or the cell knows that it's light so it should turn on or whatever, of course that's not what's happening. It's just that the glutamate binds to glutamate-gated ion channels,

and the effect that has on the cell cells differs depending on the cell type. Either glutamate causes the cells become excited or causes the cell to become inhibited. That just depends on the different type of cell and the exact interactions that are occurring there. But we don't need to know the details of that. Spiceps say the glutamate can cause depolarization or hyper-polarization depending on the bipolar cell type.

In addition to direct connections with photoreceptors, bipolar cells also receive some inputs from what are called horizontal cells, which are exactly what you might expect. They connect bipolar cells across each other. So when I said that the retinae is layered, like literally the cells are long and sort of, it's like a forest essentially. You've got the, it's just made it with like three trees sitting on top of each other.

The lowest layer are the photoreceptor cells. Then on top of that, you've got another layer of trees, because they're long and they have branches. So they kind of look a bit like trees. The next layer of trees are the bipolar cells, And on top of that, you've got the ganglion cells. These horizontal cells move across, so horizontally across the different layers. And so the horizontal cells connect together a bunch of different bipolar cells

and provide additional inputs, which sort of complicate the whole processing. But it's not fully understood what their role is, but we do know that they connect different bipolar cells to each other. The function of the horizontal cells is not completely understood, but we do know one important thing that they do. They facilitate the center and surround concentric circles receptive fields of the bipolar cells. So what on earth green mean by that? Well, it means that bipolar cells don't just respond to the neuron or neurons they're directly connected to in the center.

There's sort of two regions they respond to, a center and a surround. So you can literally think about like a donut and then the circle inside the donut. So the small circle in the center is called center and the donut around the side is the surround. This applies both to on-type and off-type bipolar cells. They both have center and surround input fields. What's the significance of these? Well, basically what it means is if you're an on-bipolar cell, it means that, that in the light, if light falls on the center of your input region, then you have increased activation.

On the other hand, if light falls on the surround of your input region, then you have decreased activation. So basically your center is on, but your surround is off. So that's why we sometimes call these on center bipolar cells, because it means that only when light falls on the center region, that little circle in the middle, does the cell increase in activation. If light falls on any of the regions in the surrounding region of the receptive field, then the activation is actually decreased. And off-centre bipolar cells is the exact opposite.

When light goes on in the center region, activation is decreased, because they respond to dark and not to light, but when you have light in the surrounding regions, activity is increased. So an on-center bipolar cell responds to light in its central region, in its central receptive field, and darkness in its surrounding field. An off-center bipolar cell responds to darkness in its central field, but light in its surrounding field. These are not action potentials, remember, these are greater potential, so you can get a combination of these.

So if you have an on-center bipolar cell, for example, the best way to stimulate that to get the maximum output is just to shine light on the very central receptive field, which probably means like a couple of neurons sitting directly, sorry, a couple of photoreceptive cells sitting directly above that bipolar cell, which only project to the sort of correspond to the central region of that bipolar cell's receptive field of view. If you just shine light on that central region and darkness in the surrounding regions, you'll get the maximum amount of output. If you shine light only on the surrounding region and darkness in the central regions, that'll get the least amount of output. If you shine light on the whole thing or darkness on the whole thing, then you sort of get a canceling out effect.

The surrounding region will be activated, but the central region won't be, and so that sort of cancels each other out a bit, and vice versa, if you put the whole thing in light or a whole thing in darkness. So the way to get the most output of these type of cells is to only shine light either on the outside or the inside part. Of course, I say it as if we're deliberately shining light on different parts of the receptive field and looking at the activation, and that's what scientists do to work it out.

But in the real world, of course, it just happens to depend. Now, in other words, so a bipolar cell is deciding, in a sense, whether it's going to fire or not. How does it make that decision? Well, it's receiving input from a bunch of photoreceptor cells, some directly and some indirectly. The direct ones synapse directly with the bipolar cell, and the indirect ones synaps transmit their information via these horizontal cells.

So remember before when I said that in the retina region, you just have one photoreceptor corresponding to one bipolar cell. That's the direct synapsing, one photoreceptor cell directly synapses with one bipolar cell. But there is still indirect connections with surrounding photoreceptic cells via the horizontal cells. In other regions, you might have 10 photoreceptic cells synapsing directly with the bipolar cell, and then even more photoreceptic cells that surrounded that initial 10 of the input, contributing to the input of this bipolar cell through the horizontal cells. So you can imagine there's like a bunch of photoreceptic cells surrounded by a donut of photoreceptic cells,

and all of these transmitting information to the one bipolar cell, the center ones directly through synapses with the bipolar cell, surrounding ones indirectly through the horizontal cells. The field of vision of any of these input photo receptor cells just corresponds to, or it leads directly to, the field of vision of the more photoreceptor cells you have going to a single bipolar cell, the wider it's field of vision. So bipolar cells in the photovia have very narrow field of vision

because they only receive input from a few photoreceptor cells. But the trade-off for that is that they have a higher resolution because they can more finely distinguish between where the input's coming from. But regardless of where you are on the retina, and the phobia, in the surrounding regions, wherever, you always have these what's called antagonistic, antagonistic circular fields. That is, you've got the central region which is either on or off,

and then the surrounding region, which is the opposite of that. So if you have on-center, you have off-surround, and if you have off-center, you have on-surround. So they're always opposite responses. So now moving on from bipolar cells, we'll talk about ganglion cells in some more. in somewhat more detail. Ganglion cells, again, are the third level of visual processing behind the photoreceptors and the bipolar cells. The bipolar cells synapsed directly with

ganglion cells, and so ganglion cells receive their input, predominantly from the bipolar cells through direct synapses. And also, similar to the horizontal cells, which allow bipolar cells to receive input, not only from the photoreceptor cells directly above them, but also from surrounding ones, hence you get the center and the surround receptive fields, ganglion cells also receive input directly from bipolar cells that they synaps with, but also from surrounding bipolar cells through horizontal cells, except they're not called horizontal cells. They're called amicrine cells, but they basically fulfill the same function. They sort of exist horizontally along covering many ganglion cells and allowing them to

link horizontally with each other, as opposed to receiving input just directly from the bipolar cells sort of underneath them. So like bipolar cells, ganglion cells also have the on-center and off-center antagonistic center field and surround field, visual field. So this is preserved from the bipolar cells through to the ganglion cells. There are about one to one and a half million ganglion cells in the human retina. So this means on average each ganglion cell receives inputs from about 100 rods or cones. Generally, a particular ganglion cell will only receive inputs either from rods or from cones and not from a mixture. I don't know if that's a completely strict rule, but generally that's the case.

But again, so that 100 photoreceptors per ganglion cell is an average. As with the bipolar cells, there's a dramatic difference between, say, the phobia, where generally it's a single ganglion cell will communicate with as few as five photoreceptors. So generally, around the phobia, you'll get one photoreceptor cell going to one bipolar cell,

and then you'll have a few of those going to one ganglion cell. But in the periphery, a single ganglion cell, will receive information from dozens or even hundreds of bipolar cells, which will in turn receive information from dozens or hundreds of photoreceptors, so that a single bipolar cell will receive input from thousands of photoreceptors. So it's sort of hierarchical in the periphery, but much less so in the phobia. But still, as you move up the hierarchy from photoreceptors to bipolar cells to ganglion cells,

you do have fewer and fewer cells at each step. The key difference between ganglain cells and bipolar cells, and also photoreceptors, is that ganglion cells fire action potentials. So ganglion cells receive input, as I said before, from bipolar cells, both directly below them and from the surrounding area by the amicrine horizontal cells. And they receive graded potentials from these input cells, so just like a regular neuron, they'll have their postsynaptic membrane potential changed as a result of neurotransmitters from the presynaptic neuron opening up ligandigated ion channels,

and they're by changing the potential at the local membrane area. and the cell sort of adds up all of the local potentials along the membrane, and if the total potential change exceeds a certain threshold, an action potential is fired because the voltage-gated ion channels are triggered, and then the ions flow into the cell and becomes depolarized, and get an action potential. If you don't understand the details of that, again, go back and listen to the episode on neurons and synapses,

because I cover that in detail there, and we don't really have time to go through that again now. But, yeah, so basically, it's only an issue of greater potentials with the bipolar cells and photoreceptors, but when ganglion cells come into the picture, it's just the normal, greater potentials eventually leading to action potential if the threshold is surpassed. So ganglion cells fire action potentials,

and those action potentials are transmitting information into the brain. Now, there are a few different types of ganglion cells. This hasn't been fully mapped out, yet there are three main types that have been sort of documented in the literature, but relatively small types that represent about 90% of the total population of ganglion cells in the retina, are called parvocellular, or just type P ganglion cells. A larger type are called magnoselular, type M ganglion cells, about 5% more,

and the remaining 5% are coniocellular ganglions, or type K. So we've got type P, type M, and type K. Type P are the most common, but the others are important as well. Now, the difference between the ganglion cells, well, there's a number of differences, as I said, they differ in size, and so the morphology is somewhat different, but also they tend to have different receptive fields. Type M ganglion cells have larger receptive fields,

which means they'll receive input from more photoreceptic cells from a larger section of the visual field, and they also tend to propagate action potentials more quickly along the optic nerve. In contrast, the P cells tend to have smaller receptive fields and tend to seem to be more sensitive to detecting, like, finer shape and details. So it's thought that the type M and the type P cells send basically different types of information into the brain. The larger type M ganglia itself, or the magnicellular,

ganglians are thought to be sort of more responsible for detection of motion and movement, whereas the type P cells are thought to be more responsible for detection of form and shape and details. And the type K cells are thought to be, much less is known about them because they're a more recent discovery, but they're thought to be related to detection of color. And I'll talk a bit more about those things. Later on, we'll talk about the processing of these different types of inputs into the brain. But it's important to understand that these different types of ganglion cells are not just structurally different, but they seem to be functionally distinct as well, that is they have different roles and their information is kept separate

in the neural pathways, as we'll see later. However, they all still share the same basic properties as having, you know, those circular visual fields with the on or off centre and then the antagonistic surround. So regardless of the type, the P's and the Ems and the Ks all have that antagonistic receptive field and they all fire action potentials and they're all located, you know, in the same basic place in terms of the retina at the same level. It's just their functions are a bit different and their morphologies differ. Okay, so that's about all we have to say about the retina. So just a quick recap, the retina consists of several layers,

well, three layers, specifically, of light-sensitive tissue that's located on the inside of the eyeball, all around the inside, basically. The region that has the highest concentration of photoreceptors, the cells that are responsible for actually transducing the light signal into electric signal are located at the phobia, with the highest level of contrast and highest resolution. The two different main types of photoreceptors are called

rods and cones. Rods are basically responsible for detecting low levels of light and for sort of black or light vision. Cones are responsible for, are responsible for color vision and are used in bright daylight settings. Both, however, function more or less in the same way that when they detect a photon, the photoreceptor molecules in the cells are altered in such a way that leads to a sort of cascade of chemical reactions and protein changes, which culminates in a change in the amount of the neurotransmitic glutamate that's released and that therefore is, therefore that affects the bipolar cells. When this is changed, the bipolar cells will either have their post-inaptic potential changed, either increased or decreased depending on the receptive field. This will in turn change

the post-snaptic potential for the ganglion cells with which the bipolar cells synapse. So the signal is essentially sent from the photoreceptor cells through the bipolar cells, maintaining, staying as a greater potential the whole time. And finally, it's received by the ganglion cells, which receive the greater potentials from a bunch of different bipolar cells, integrate those, and then fire action potentials or not, depending on the level of input. Actually, I should say that ganglion cells will spontaneously fire action potentials all the time, that they have a base rate. What they do is they increase the rate of fire when they're excited,

and so an on-center type, a ganglion cell, for example, will increase the rate of it firing its neurons when the center of its receptive field has a light incident upon it, or the photoreceptor cells that correspond to its center of the center of its visual field. When they have light shining on them, then this ganglain cell will increase the rate at which it fires action potentials. Conversely, if you've shown light on the surrounding

area of its visual field, so it's the donut shape surrounding the center area, then since this is an on-center ganglion cell, it would be inhibited, and therefore it would reduce the rate to which it applied action potentials. And the three different types of ganglion cells, the P and the M and the K-types seem to be responsible

for somewhat different types of information. They receive input from different bipolar cells and hence different photorecept cells. But that's the essence. of the retina. It's comprised of these three layers, which together are able to transduce and prepare the information received directly from photons, and send it into the brain. So we move on from stage two of visual processing, the retina, onto stage three, where we talk about moving the information into the brain, and we cover the optic nerve,

the optic chasm, and we'll talk about the lateral genicular nucleus. So first of all, the optic nerve. So ganglion cell axons all sort of run together in the same direction, towards a single area, basically, of the retina, a circular region, which is called the optic disc, and they bundle together to form a big nerve, basically, a nerve is just a bundle of axon, so it literally is a nerve called the optic nerve. So all of the axons from all of the ganglons come together in this one region of the retina, sort of go out in the same, through the same hole in the retina, and this is called the optic nerve. Now, at this point in the retina, there aren't any photoreceptors because essentially the optic nerve is in the way. It's taking up

the spot. So no light is able to be detected in this. region and hence called the blind spot. So if you have visual information that falls on that particular part of the retina, you won't be able to see anything. And it's quite interesting. You can demonstrate that this blind spot exists because, of course, we don't see a big black dot on our visual field. What happens is our brain fills that information in automatically. But if you stare at a, so you can get like little visual, essentially a little diagrams with dots on them. And if you maintain a fixed gaze and move it just to the right position, you'll see the dot can disappear and your brain just fills in the color on either side of it. It's kind of cool, actually.

So just a note on why this blind spot exists to clarify a little bit, you may have heard before that the retina is actually back to front or inside out. That is the photoreceptors I've described as layer 1 and the bipolar cells is layer 2 and the ganglion cells is layer 3, and then we know the axons come out at the end of the ganglion cells and all join up together at the optic disc and then come out of the back of the eyes through the optic nerve. The thing about that is that layer 1, the photo cells,

actually face towards the back of the eye. They are the closest to the squalera, to the outside white protective tissue around the outside of the eye. The cells that are actually closest to the front of the eye, that is, you know, as light comes through, it passes through the lens and it passes through the, it passes into the large internal cavity of the eye. The first thing that the light hits when it gets to the back of the eye, not the photo cells, the first thing that it hits are the axons coming out from the back of the ganglion cells, because essentially the whole retina is inside out. It faces backwards. The axons coming from the ganglion cells are the thing that's closest. to the lens, closest to the front of the eye. And then as you sort of move further backwards,

you'll move through the ganglion cells, and then through the bipolar cells, and then finally you'll get to the photoreceptor cells at the very back far away from the front of the eye, and then you move through a couple extra layers, and you get to the sclera on the very outside. So the eye really is back to front. If this wasn't the case, then we wouldn't have a blind spot, because essentially the axons could just be pulled behind,

and you could still have all of the photoreceptors existing in front of them. Basically, the photoreceptors between the axons, and the actual brain. If you think about this, because the front of the eye is away from the brain, the back of the eye is closer to the brain, and we're saying that the axons from the ganglion cells are closer to the front of the eye than the back of the eye, and in between the ganglion cell axons

and the actual brain, when we need to do the processing, are the photoreceptor cells. So in order for the ganglion cell axons to get back into the brain, they have to get past the photoreceptors somehow. The only way that's possible is essentially to have a little hole in the photoreceptors where there aren't any photoreceptors, and that part of the retens is called the optic disc,

and because there aren't any photoreceptors, receptors there, we have a blind spot. So it's a little hard to explain that with that a diagram. It's immediately clear when you see a diagram. So if that's a bit unclear, just Google optic nerve or blind spot or something like that, and hopefully you'll see what I'm talking about. But that's a very weird quirk of evolution. I believe the octopus eyes are not like that. They're the right way around. It's very strange. I don't know if anyone knows exactly how that happened. Okay, so anyway, once the axons from the ganglion cells all bunched together into the optic nerve and the optic nerve runs out of the back of the eye, where does it go?

Well, most of the axons from the optic nerve terminate in the lateral geniculate nucleus, where the information is in turn relayed onto the visual cortex. So the lateral genucleiculate nucleus, or LGN, is a region of the brain that is located in the thalamus, and I've mentioned it before, but we'll come back to it in a moment, because there's a little bit, there's mentioning things that happen to the optic nerve before it gets to the LGN. Specifically, that is the optic nerve, well, there's two optic nerves, remember, one from each eye, and the nerves sort of both project backwards and sort of in towards each other,

and they meet at a place called the optic chayasm, or optic chayasma, Kai from the Greek letter, Kai basically, which looks like an X. So imagine this is an X. The two optic nerves are coming towards each other, and they cross in an X, and then move apart again, and each move to opposite sides of the brain. So there's a couple of things that are going on here. What happens at the optic chiasm is there's essentially a crossing over of information. The optic nerves don't just pass by each other, sort of like two sets,

separate roads that don't actually meet, they actually cross over with each other such that information is exchanged between them, but it's exchanged between them in a very specific way. In visual neuroscience, we have a term called hemifield. There's a left and right hemifield. The hemifield just refers to half of the field of vision that you can see. Each eye, so each monocular field of vision, basically, has its own two hemifields. It's left hemifield and its right hemifield. So you've got the left hemifield of the left eye, the right hemifield of the left eye,

the left hemifield of the right eye and the right hemifield of the right eye. The hemifield, just half of the area that that eye can see on the left and on the right. Now, obviously, when the optic nerve comes out from the back of the eyeball, it's carrying the information from both left and right hemifields of its given eye. Obviously, it has to, because it has to pull all the information from that eye. However, the way the brain processes things is that the information from the left visual field is processed on the left side of the brain, and information from the right visual field is processed on the right side of the brain.

but information from the left visual field comes from both eyes, the left hemifield of the left eye and the left hemifield of the right eye. So what happens at the optic chasm is that the information is switched over. So if you've got an X, you've got the two top parts of the X that meet together and then two bottom parts of the X. The two top parts of the X you can think of as the initial optic nerves that come from the eye, so they've got each of them has both hemifields for its given eye. What comes out, the two bottom parts of the X, on one side you'll just have both left hemifield. the information for both left hemifields, left and right eye, the hemifields of both of the

eyes, and then the other one will have the right hemifields for both eyes. So basically what happens is they just swap hemifields. It's like, okay, this is the left, this is, one output takes both left hemifields, the other output takes both right hemifields. So instead of having, for the two inputs, one input has both hemifields from one eye, the other input has both the hemifields for the other eye, both of the outputs have inputs from both eyes, but they only have one half of the hemifield each. I hope that was clear. It's a little bit hard to explain. Again, something that's much easier explained with a diagram. If my explanation wasn't clear, just Google Optic Chiasm, and you'll get a good diagram for that.

So basically, you go from having the inputs from the eyes separate to the inputs of the two eyes combined, but instead they're separated on the basis of whether it's the left hemifield of the vision or the right hemifield of your visual field. And so it's separate. The information becomes separated on the basis of the hemifield that the information is from, and not the eye that it's taken from. Okay, so after the optic hyacism, the bundles of axons continue to travel through a bunch of loops and ends up in the axons terminate in the lateral geniculate nucleus. As I said, this is a region of the thalamus, which is a very important part of the brain, sort of around the middle of the brain, sort of around the middle nearest to the bottom.

I mean, it's not a bit hard to explain, but it's not right on the edges of the brain, and it's not on the back of the head either. The primary visual cortex, which we'll talk about in a moment, is at the back of the head. But the thalamus is more in the middle. Anyway, so the lateral genuculate nucleus is part of the thalamus. The name, so lateral just means side, basically. So it's on the side. Genuculate is a Greek, I think it's Greek word, that essentially means bent.

Well, it means knee, but in this case it's referring to like a bended knee. And nucleus means, in this case, it means basically a group of neuron cell bodies that are located together at some point in the brain. So lateral genuculate nucleus is literally just a part of the thalamus, and the thalemones. is just a region of the brain, as I said, that's on the side of the thalamus, and that it's a bunch of neuron cell bodies that are altogether in a kind of a bent shape. And the lateral genuculate nucleus, like you can stain it and observe what it looks like with all the cell bodies there, all the somer of the neurons. And it sort of looks like a stack of pancakes, except it's a bent stack of pancakes.

It's like someone pushed it up in the middle, and it's drooping on either side. There are six layers in the lateral genucleiculate nucleus, and they're very distinct. You can see them very easily, so it's kind of like there are six pancakes, again, bent in the middle. Each layer is made up of a bunch of cells piled four to ten, roughly, you know, four to ten high. Again, when I say that the information from the optic tract goes to the lateral genucular nucleus, that's a little misleading because there are actually two lateral genuculate nuclei, one on the left side of the brain, one on the right side of the brain. So pretty much everything that I'm going to be talking about from now on is happening

simultaneously both on the left and the right sides of the brain, not necessarily exactly symmetric, but generally largely so, because the two hemispheres of the brain are fairly symmetrical. So there's a lateral geniculate nucleus on the left hand side of the brain and on the right hand side of the brain. The information from the left visual field goes to, goes to one, and the information from the right visual field goes to the other. Now, remember I said there are six layers in the lateral genucleate nucleus, kind of like the pancakes pile on top of each other, six layers of cells? Well, it turns out that different information goes to different layers in this pancake. In particular, the M cells, remember the magnoscellular ganglion cells that have a larger

visual field and seem to be responsible for motion, the axons from those ganglion cells project to layers 1N2, whereas input from the parvocellular ganglion cells, the P cells, which are responsible for more finer details and structural information, project to layers 3 through 6. And the third type, the K cells project to the regions in between the main number layers, because it turns out that there are still cells in between the main number of layers. They're a bit harder to see, but they are still there. So basically you've got six layers and then the six main layers and the spaces in between those. Two of those layers, M cells, synaps, with those, and so their information goes to layers

one and two. Four of the layers, P-cell axons go to there, and so the P-cell information goes to layers three through six, and then the spaces in between the main layers, the K-cell axons go to there and transmit their information. So the information from the different types of. types of cells is kept separate, which is interesting. Also, it turns out that information is also segregated on the basis of which eye it came from. So remember, in each of the lateral geniculate nuclei, you're going to have input that comes from both eyes.

So some input will come from the Ipsilateral eye, which means the eye on the same side, on the same side of the body as that nuclei is in. So for the right geniculate nuclei on the right side of the brain, the Ipsilateral eye would be the right eye, and the contralateral eye would be the left eye, because that's the eye on the other side of the brain. It turns out that information from each eye goes to different layers of the lateral genucleine nucleus. So in particular, axons from the ibsalateral eye, so the same side eye synapsed with layers on cell layers 2, 3, and 5, whereas input from the contralateral eye projects to layers 1, 4, and 6. And so essentially we can combine this with the specificity in terms of different types of ganglion cells synapting with different layers,

so that each layer essentially only receives input from a particular eye and a particular type of ganglions in that eye. So, for example, layer 1 will only receive input from M cells from the contralateral I. Layer 2 will only receive input from the M cells from the Ipsilateral eye. Layer 3 will only receive input from the P cells of the Ipsilateral eye, layer 4 will only receive input from the P cells of the contralateral I, and so on. I mean, it's not important to remember the exact numbering and ordering of all that. The point is that the information is segregated both in terms of the eye that it came from

and in terms of the type of ganglion cells that it came from. There's two eyes that it could have come from and three different types of cells. So that's six different combinations that are possible. And that's very interesting. And it turns out this happens quite a lot in the visual system. We'll see this more in V1, where basically, the type of information is kept separate based on where the information came from. And remember, each LGN is only receiving input from one hemifields. So one half of the visual field, left side from the left hemifield, right side from the right hemifield.

Now, I said before that after passing through the optic chasm, the optic nerves then terminate in the lateral genucleine nucleus and synapse with their input there. So the lateral genucleine nucleus has its own neurons that synapse with the axons that come directly from the ganglion cells. So when you have a ganglion cell fire, that will lead to firing of the cells, firing of action potentials of cells in the appropriate layer of the lateral genucleine nucleus. However, not all of the inputs of the lateral genuculate nucleus come through the optic neufth and the optic chiasm. In fact, most of the inputs from the lateral genucleineucleus actually comes from the visual cortex, which is kind of weird because most of the output of the lateral genucleine nucleus goes to the visual cortex. So in other words, the LGN sends a bunch of axons to V1, and then it gets a bunch of axons back from V1.

So it's kind of circular in that sense. And I don't know what the percentage is, but it's more than half of the input, I think it was like 80% or something, of the input to the LGN are coming from the visual cortex. And it's really not clear what's going on there. It may be some aspect of top-down processing, where essentially what you're expecting to see has an implication on how it's processed it earlier on in the system. I'll talk a bit more about that later,

but it's really not clear what that sort of backward sending of neural information is doing, but it's definitely doing something there. And another important thing to note is that not all of the axons from the optic nerve synapse in the lateral genoculate nucleus, only about, so about 80% of all of the axons from the gangloid cells end up synapsing in the lateral genucin nucleus. Some of them go off and connect with other regions of the brain, like the brainstem and so on, and are responsible for maintaining things like the sleep, wake cycle and other things like that. But we're not terribly interested in those, because those other regions are not responsible for conscious perception of vision.

They need light input for other things, like, for example, for maintaining our internal circadian rhythm with external cues of light and dark. but that's not very interesting from... That doesn't really have anything to do with visual perception, so we're not going to worry about that. So most of the relevant output from the axons in the ganglion cells goes into the lateral genucleine nucleus, and the lateral genucingulate nucleus, in turn,

gets a whole chunk more input from the V1, the primary visual cortex. Most of the, as I said before, most of the axons that leave the LGN go straight to V1, the primary visual cortex, even though, as I mentioned before, the LGN also receives a bunch of axons back from V1, but we won't talk any more about those

because we don't really know what they do. So, but this is another, example of what I said before, it's not all one-directional, it's not linear, it's, there's a lot of backward propagation of singles and so on, and we're not entirely clear what those do. Okay, so that's it for this episode. Next episode, in part three, we'll be completing our tour through the biological aspect of the visual system by talking about V1, the primary visual cortex, and then higher cortical areas, V2, V3, V4, the IT area, and so on. So look out for that. That should be coming in a couple of weeks. Also, just something that I'd like to,

to announce are... You'll notice that this is episode 46 of the show and episode 47's basically, well, mostly ready, so that'll be coming out soon. So we're fast approaching episode 50. Now, I wanted to do something special for episode 50, something a little bit different and something a little bit fun, but still related to, you know, science, obviously. So, I mean, I had a few different ideas. I was thinking about something like the physics of time travel or faster than light travel, or, you know, the physics of Star Trek, or common science mistakes in movies, or medical errors in movies and television, or, you know, something along those lines. If anyone likes any of those ideas or has something similar that I'd like to put forward, give me an email, FODs12 at gmail.com, or you can also post a suggestion to our Facebook page. Just go to Facebook and search for The Science of Everything podcast, and you'll be able to find us there and give us a like.

Any other suggestions and comments and feedback that you have about the show would also be appreciated. if you could also jump onto iTunes and give the podcast a review, hopefully a positive one, and a rating that would be much appreciated. Now, I've got a few ratings, but it seems to be the way the iTunes podcast rankings work is that you have to keep getting new ratings and also new subscriptions, but particularly new ratings in order to stay high up in the ranking level. So although I've got a few there, I need more, so keep doing that. Seriously, even one or two new ratings can make a big difference. So I'd really appreciate if you could do that if you enjoy the show.

So that's enough from me. Thanks again for listening and I'll talk to you next time.