Understanding Visual Processing Pathways

Hello and welcome to our fourth part of the series on vision and visual processing. We'll be talking today about neural processing in the retina and the lateral geniculate nucleus. This is the starting of our visual processing abilities and our ability to detect patterns and features in the environment. So neural processing starts very early on in the visual processing pathway. And we'll talk about... this early part of that processing and then in the fifth part we will talk about cortical processing and how the visual cortex processes some visual features. So to start we're going to start with an overview of the retina geniculate striate pathway. This is the pathway that goes from the retina to the lateral geniculate nucleus to the striate cortex or the primary visual cortex and the occipital lobe. We'll talk a little bit about how along this pathway this exhibits a retinotopic organization. We'll talk about the concept of lateral inhibition, an important part of our ability to discern things like contrast. We'll talk about the MNP layers of the lateral geniculate nucleus, which are the magnocellular and parvocellular layers of the LGN. And then finally, we'll talk a little bit about receptive fields and center-surround antagonism. So first off, for an overview of this pathway, you can see here that we have... Information projecting onto the retina. The nasal part of the retina, or the nasal hemi-retina as we call it, these fibers travel over to the opposite hemisphere, so these are contralateral. So this is the right visual field. They're traveling to the left hemisphere. Similarly, the left visual field travels to the right hemisphere, so here the left visual field in the right eye stays in the right. Hemisphere. So these are ipsilateral fibers. These cross over an area called the optic chiasm. So the optic nerve exits the eye. We have the optic chiasm. Now we have all the right visual field information in the left hemisphere and the left visual field in the right hemisphere. So the nasal fibers then are contralateral. The temporal fibers are ipsilateral. They then synapse at the lateral geniculate nucleus. The ipsilateral fibers Synapse at layers 2, 3, and 5. The contralateral layers 1, 4, and 6. As we discussed in earlier lectures, the cortex has six layers, as does the LGN. So we start up setting up those six cortical layers here in the synapses at the LGN. So we start to get that layered processing that's going to become very important to columns and lamina in the primary visual cortex. So then from the LGN, they go on to layer four of the primary visual cortex. And we'll talk more about that in our next segment. So for now, we'll talk about the retina LGN portion of this pathway. At each level of the pathway, information is organized like a map of the retina. So essentially, cells respond to information that are nearby those cells in the retina, which are, of course, also representative of our visual field. So all right. what we're looking at in the environment is actually mapped out in the LGN much like it is in the retina and similarly in the cortex. Important to know that the fovea is overrepresented in this map because of course this is where all of our focal vision is. This is where most of the cones are and this is of course where cones primarily have a one-to-one, one cone per ganglion cell spatial summation, so no spatial summation and so as a result that part of the retina is over-represented in this map, which makes perfect sense. So this is, again, this idea of cortical magnification. An important property that starts at the retina and continues on in creating some of our perceptual abilities is this idea of lateral inhibition. So this is the capacity of an excited neuron to reduce the activity of its neighbors. And so... What we're going to do first is talk about how information from the horseshoe crab helped to create a lot of the knowledge we have in this lateral inhibition. Horseshoe crabs are used to study this idea, and so you can get an idea of how this functions. So this is a horseshoe crab. The horseshoe crab has very large photoreceptors that each have a single axon, and these are connected by a lateral neural network, and they're easily accessible. with their kind of compound I. So each of these is called an ommatidia. And each of these photoreceptors then has a large single axon. And so we can actually measure the activity in these axons from the horseshoe crab eye. So if we have these ommatidia receptors, they're connected by this lateral plexus. This is a model we can think about in terms of how parts of our retina actually function. So we have a single receptor of virus at a rate proportional to the intensity of its stimulus. So again, we have an intense light here and a dim light here. So these photoreceptors are firing at a much higher rate than the dim light photoreceptors. One of the things that happens is when this single photoreceptor is activated, it actually inhibits its neighbors. So it sends out an inhibitory signal to its neighbors. So within this intense light period... While they're being activated by the intense light, they're being inhibited by their neighbors. Now, the dim light, they're also being activated, but less than the intense light. But they're also sending out less inhibition. So what this means is at the border between the intense light and the dim light, there's a perceived intensity. What happens is these neurons that are getting this intense light are spreading more lateral inhibition. to this next neuron that has less activity and is getting less inhibition from its neighbor. So it actually ends up with more inhibition than its dim light neighbors. And so it actually appears to be dimmer at this border. Similarly, this neuron is receiving less inhibition from this dim light neighbor. And as a result, this neuron fires at a higher rate. As a result, we get an increase in perception of intensity in the intense light side and the decrease in the perception of intensity on the dim light side. So this is one of the things that highlights borders in our perceptual ability. It also results in a number of important phenomena. So this is the neural basis of contrast enhancement. It's basically understood in terms of this idea of lateral inhibition. So at a border between different intensities of lights, we actually get This sort of magnification of this intensity difference. So we get a decrease in the neural firing on this side and an increase on this side, which highlights this kind of border. Now to show you what this looks like in our visual system, I'm going to show you a few examples of how this functions. So these are what we call Mach bands. So what you can see is when you look at the junction between these two. different areas, you see this differing intensity. So for example, it looks brighter and darker on each side of these Mach bands. You can see it looks brighter here, here, here. It looks a little bit darker on the other side. Well again, what's happening here is the brightness changes. When we get these changes in intensity, we get this change in lateral inhibition. So this Neurons are firing at a higher rate because it's receiving less lateral inhibition, and these neurons are firing at a lower rate because they're receiving more lateral inhibition relative to their neighbors. And again, when we get to this next change, we get an increase in firing rate because of reduction in lateral inhibition, and then we get a reduction in firing rate due to an increase in lateral inhibition. So we can see that here as well. And we get these kind of atmospheric effects even when we look at landscapes. You can see these kind of changes in intensity are highlighted by this idea of lateral inhibition. Other areas where we get illusions based on this lateral inhibition, and we're going to talk about the Hermann grid and how center-triangle antagonism is part of this responsibility, but when you look at any of these junctions you don't see a shadow, but at the other junctions you do. This is actually a different version of that. It makes my eyes a little crazy. We're going to talk about how lateral inhibition, in particular centers around antagonism, has a role to play in creating these periods of reduced, or these areas of reduced intensity, and therefore creating this illusion of a shadow. We'll talk more about this in our separate lecture on color vision, but when we have two different colors that are right next to one another, We get a change in color perception due to this idea of simultaneous contrast. This happens with black and white as well. And so this, again, has to do with this. Lateral inhibition based on its nearby perceptual, sorry, its nearby color. So whatever colors it's presented along with actually has an effect on the color that we perceive. And we'll talk more about that concept later, but it is part of this idea of lateral inhibition. So this starts in the retina. We then move on to the lateral geniculate nucleus. Now this is where we start to get continued division of... the visual system. So remember in the retina we had the scotopic and photopic systems. Now as we get into the LGN, the top four layers of the LGN compromise the parvocellular system. This is the system that's responsive to color, fine patterns or details, stationary or slowly moving objects, and of course its primary input comes from the photopic system, which is the cones. All of our detailed color vision is occurring. primarily in the fovea and from the cones. And so the parvocellular system then is receiving most of its input from those cones. And so you can see this is actually a slice of the LGN. It looks a little bit like a fingerprint. So these top three layers, layers three, four, five, and six, are the parvocellular system. Again, each layer at this point is monocular. Remember that's coming from either the ipsilateral or contralateral retina. So these are only coming from one eye. These top four layers are the parvocellular system. The bottom two layers are the magnocellular system. These are layers one and two. This is primarily responsive to movement, and it receives its primary input from the scotopic system. So at this point, we still have Monocular fields, or cell layers I should say, cells and layers are monocular, so each cell is only responding to one eye, so we have no depth perception at this point. Again, the parvocellular system is primarily getting its input from the photopic system, so from the cone, so this is color and detailed vision. And the magnocellular system is then primarily receiving its input from the scotopic system, so it's primarily responsive to movement. low levels of illumination, and no fine detail. When we start moving into the cortex, we'll start talking about how the parvalcellular system primarily feeds into what's called the wet pathway. And then we'll talk about the magnocellular system feeding primarily into the wear pathway. They're not entirely independent from one another, but we see this sort of separating visual pathways from retina all the way up to cortex. So one of the important components of understanding both the retina and the LGN are the receptive fields of specific ganglion cells and cells in the LGN. Hubel and Wiesel won their Nobel Prize for the discovery of receptive fields. Primarily the research was in cats, but we do have good evidence of how this these receptive fields hold up in all species. Everybody's, each species visual system is a little bit different, but this idea of receptive fields was primarily discovered by Hubel and Wiesel. Excuse me. So a single neuron will respond only to light presented in its receptive field. So the idea is that each neuron in the visual cortex, particularly in the LGN and the retina, will only respond to light that's presented within its receptive field. It's receiving input from cones in that area and then cones and rods in that area and so it responds only the light presented in its receptive field. So we use the firing rate of a neuron to measure this idea neurons fire or have slow firing rates on their own when they're not being stimulated. When they're being inhibited their firing rate of course decreases. When they're being excited their firing rate increases. So we look at the firing rate of neurons in terms of what happens when we present light in their receptive fields. So at this point we're talking about shining light into the retina and seeing what happens to neurons particularly either the optic nerve or the LGN and at this point we're really talking about the LGN. So what Hubel and Vase will find is that at each level retinal ganglion, LGN, and the lower layer 4 of the primary visual cortex, The receptive fields in the fovea are smaller than in the periphery. Yeah, not surprising there, given that there is a great deal of spatial summation in the peripheral retina. And there is more closely a one-to-one mapping of cones to ganglion cells in the fovea. They also found that the receptive fields in this area were circular and monocular, that is coming from a single eye, and included both inhibitory and excitatory areas in a center surround field. So essentially this looks like a donut and we're going to take a look at what these look like here in a moment. It's called an annulus where you have a center area and the surrounding area just like a donut or a bagel but what happens is when you shine a light in one area you get inhibition and when you shine a light in the other area you get excitation. So there are inhibitory and excitatory areas in this center surround. So we can have on-center cells and off-center cells, which also means on-surround and off-surround. So here we have an on-center cell. So this is that annulus I was talking about. And if a light's presented in the center, it will increase the firing rate of that neuron. So if we present a light in the center of the receptive field of this neuron, that neuron could be in the LGN or the primary visual cortex, it's going to increase the firing rate of that neuron. Now if we present a light only in the surround, It's actually going to decrease the firing rate of that neuron. So this is an off-surround cell. Similarly, there are off-center cells. We present a light in the center of an off-center cell that will reduce the firing rate of a neuron. When we present light into the on part of this receptive field, we get an increase in firing rate from the neuron. Now, why is this important? Well, this is one of the ways in which we're able to discern contrast. So If we shine a light over the entire receptive field of this neuron, its firing rate will not change because we will be both exciting it and inhibiting it at about the same rate. But if we have a neuron right next to it and it has a differential firing rate, then we know that that's where a stimulus sort of begins and ends. So this is how we're able to do things like see text on a page, is through this center-surround antagonism mechanism because it codes for Where things start and stop in our visual world. So this is basically what we're going to be looking at in our next lecture, is how these center-to-round antagonism cells line up to create things like perception of a line that is upright, or a horizontal line, or a vertical line, etc. And we'll talk about how these receptive fields are mapped out. We'll talk a little bit more about the perceptual attributes of these, this center-surround antagonism as well. But this is similar to the idea of lateral inhibition, where we're starting to see more complex patterns of responding. And as we move up in the cortex, we'll start to see more complex patterns of responding as well. So this is a good place to pause. think about this idea of center-surround antagonism and how it affects processing of visual information by increasing or decreasing firing rates of neurons, and then we'll take a look at how this works in the visual cortex in our next lecture.

So neural processing starts very early on in the visual processing pathway. And we'll talk about... this early part of that processing and then in the fifth part we will talk about cortical processing and how the visual cortex processes some visual features. So to start we're going to start with an overview of the retina geniculate striate pathway.

This is the pathway that goes from the retina to the lateral geniculate nucleus to the striate cortex or the primary visual cortex and the occipital lobe. We'll talk a little bit about how along this pathway this exhibits a retinotopic organization. We'll talk about the concept of lateral inhibition, an important part of our ability to discern things like contrast.

We'll talk about the MNP layers of the lateral geniculate nucleus, which are the magnocellular and parvocellular layers of the LGN. And then finally, we'll talk a little bit about receptive fields and center-surround antagonism. So first off, for an overview of this pathway, you can see here that we have...

Information projecting onto the retina. The nasal part of the retina, or the nasal hemi-retina as we call it, these fibers travel over to the opposite hemisphere, so these are contralateral. So this is the right visual field.

They're traveling to the left hemisphere. Similarly, the left visual field travels to the right hemisphere, so here the left visual field in the right eye stays in the right. Hemisphere. So these are ipsilateral fibers. These cross over an area called the optic chiasm.

So the optic nerve exits the eye. We have the optic chiasm. Now we have all the right visual field information in the left hemisphere and the left visual field in the right hemisphere. So the nasal fibers then are contralateral.

The temporal fibers are ipsilateral. They then synapse at the lateral geniculate nucleus. The ipsilateral fibers Synapse at layers 2, 3, and 5. The contralateral layers 1, 4, and 6. As we discussed in earlier lectures, the cortex has six layers, as does the LGN. So we start up setting up those six cortical layers here in the synapses at the LGN. So we start to get that layered processing that's going to become very important to columns and lamina in the primary visual cortex.

So then from the LGN, they go on to layer four of the primary visual cortex. And we'll talk more about that in our next segment. So for now, we'll talk about the retina LGN portion of this pathway.

At each level of the pathway, information is organized like a map of the retina. So essentially, cells respond to information that are nearby those cells in the retina, which are, of course, also representative of our visual field. So all right.

what we're looking at in the environment is actually mapped out in the LGN much like it is in the retina and similarly in the cortex. Important to know that the fovea is overrepresented in this map because of course this is where all of our focal vision is. This is where most of the cones are and this is of course where cones primarily have a one-to-one, one cone per ganglion cell spatial summation, so no spatial summation and so as a result that part of the retina is over-represented in this map, which makes perfect sense. So this is, again, this idea of cortical magnification.

An important property that starts at the retina and continues on in creating some of our perceptual abilities is this idea of lateral inhibition. So this is the capacity of an excited neuron to reduce the activity of its neighbors. And so...

What we're going to do first is talk about how information from the horseshoe crab helped to create a lot of the knowledge we have in this lateral inhibition. Horseshoe crabs are used to study this idea, and so you can get an idea of how this functions. So this is a horseshoe crab.

The horseshoe crab has very large photoreceptors that each have a single axon, and these are connected by a lateral neural network, and they're easily accessible. with their kind of compound I. So each of these is called an ommatidia.

And each of these photoreceptors then has a large single axon. And so we can actually measure the activity in these axons from the horseshoe crab eye. So if we have these ommatidia receptors, they're connected by this lateral plexus.

This is a model we can think about in terms of how parts of our retina actually function. So we have a single receptor of virus at a rate proportional to the intensity of its stimulus. So again, we have an intense light here and a dim light here.

So these photoreceptors are firing at a much higher rate than the dim light photoreceptors. One of the things that happens is when this single photoreceptor is activated, it actually inhibits its neighbors. So it sends out an inhibitory signal to its neighbors. So within this intense light period...

While they're being activated by the intense light, they're being inhibited by their neighbors. Now, the dim light, they're also being activated, but less than the intense light. But they're also sending out less inhibition.

So what this means is at the border between the intense light and the dim light, there's a perceived intensity. What happens is these neurons that are getting this intense light are spreading more lateral inhibition. to this next neuron that has less activity and is getting less inhibition from its neighbor.

So it actually ends up with more inhibition than its dim light neighbors. And so it actually appears to be dimmer at this border. Similarly, this neuron is receiving less inhibition from this dim light neighbor.

And as a result, this neuron fires at a higher rate. As a result, we get an increase in perception of intensity in the intense light side and the decrease in the perception of intensity on the dim light side. So this is one of the things that highlights borders in our perceptual ability.

It also results in a number of important phenomena. So this is the neural basis of contrast enhancement. It's basically understood in terms of this idea of lateral inhibition. So at a border between different intensities of lights, we actually get This sort of magnification of this intensity difference. So we get a decrease in the neural firing on this side and an increase on this side, which highlights this kind of border.

Now to show you what this looks like in our visual system, I'm going to show you a few examples of how this functions. So these are what we call Mach bands. So what you can see is when you look at the junction between these two. different areas, you see this differing intensity. So for example, it looks brighter and darker on each side of these Mach bands.

You can see it looks brighter here, here, here. It looks a little bit darker on the other side. Well again, what's happening here is the brightness changes.

When we get these changes in intensity, we get this change in lateral inhibition. So this Neurons are firing at a higher rate because it's receiving less lateral inhibition, and these neurons are firing at a lower rate because they're receiving more lateral inhibition relative to their neighbors. And again, when we get to this next change, we get an increase in firing rate because of reduction in lateral inhibition, and then we get a reduction in firing rate due to an increase in lateral inhibition. So we can see that here as well. And we get these kind of atmospheric effects even when we look at landscapes.

You can see these kind of changes in intensity are highlighted by this idea of lateral inhibition. Other areas where we get illusions based on this lateral inhibition, and we're going to talk about the Hermann grid and how center-triangle antagonism is part of this responsibility, but when you look at any of these junctions you don't see a shadow, but at the other junctions you do. This is actually a different version of that.

It makes my eyes a little crazy. We're going to talk about how lateral inhibition, in particular centers around antagonism, has a role to play in creating these periods of reduced, or these areas of reduced intensity, and therefore creating this illusion of a shadow. We'll talk more about this in our separate lecture on color vision, but when we have two different colors that are right next to one another, We get a change in color perception due to this idea of simultaneous contrast.

This happens with black and white as well. And so this, again, has to do with this. Lateral inhibition based on its nearby perceptual, sorry, its nearby color. So whatever colors it's presented along with actually has an effect on the color that we perceive.

And we'll talk more about that concept later, but it is part of this idea of lateral inhibition. So this starts in the retina. We then move on to the lateral geniculate nucleus. Now this is where we start to get continued division of... the visual system.

So remember in the retina we had the scotopic and photopic systems. Now as we get into the LGN, the top four layers of the LGN compromise the parvocellular system. This is the system that's responsive to color, fine patterns or details, stationary or slowly moving objects, and of course its primary input comes from the photopic system, which is the cones. All of our detailed color vision is occurring.

primarily in the fovea and from the cones. And so the parvocellular system then is receiving most of its input from those cones. And so you can see this is actually a slice of the LGN. It looks a little bit like a fingerprint.

So these top three layers, layers three, four, five, and six, are the parvocellular system. Again, each layer at this point is monocular. Remember that's coming from either the ipsilateral or contralateral retina.

So these are only coming from one eye. These top four layers are the parvocellular system. The bottom two layers are the magnocellular system.

These are layers one and two. This is primarily responsive to movement, and it receives its primary input from the scotopic system. So at this point, we still have Monocular fields, or cell layers I should say, cells and layers are monocular, so each cell is only responding to one eye, so we have no depth perception at this point.

Again, the parvocellular system is primarily getting its input from the photopic system, so from the cone, so this is color and detailed vision. And the magnocellular system is then primarily receiving its input from the scotopic system, so it's primarily responsive to movement. low levels of illumination, and no fine detail.

When we start moving into the cortex, we'll start talking about how the parvalcellular system primarily feeds into what's called the wet pathway. And then we'll talk about the magnocellular system feeding primarily into the wear pathway. They're not entirely independent from one another, but we see this sort of separating visual pathways from retina all the way up to cortex.

So one of the important components of understanding both the retina and the LGN are the receptive fields of specific ganglion cells and cells in the LGN. Hubel and Wiesel won their Nobel Prize for the discovery of receptive fields. Primarily the research was in cats, but we do have good evidence of how this these receptive fields hold up in all species.

Everybody's, each species visual system is a little bit different, but this idea of receptive fields was primarily discovered by Hubel and Wiesel. Excuse me. So a single neuron will respond only to light presented in its receptive field. So the idea is that each neuron in the visual cortex, particularly in the LGN and the retina, will only respond to light that's presented within its receptive field.

It's receiving input from cones in that area and then cones and rods in that area and so it responds only the light presented in its receptive field. So we use the firing rate of a neuron to measure this idea neurons fire or have slow firing rates on their own when they're not being stimulated. When they're being inhibited their firing rate of course decreases.

When they're being excited their firing rate increases. So we look at the firing rate of neurons in terms of what happens when we present light in their receptive fields. So at this point we're talking about shining light into the retina and seeing what happens to neurons particularly either the optic nerve or the LGN and at this point we're really talking about the LGN. So what Hubel and Vase will find is that at each level retinal ganglion, LGN, and the lower layer 4 of the primary visual cortex, The receptive fields in the fovea are smaller than in the periphery.

Yeah, not surprising there, given that there is a great deal of spatial summation in the peripheral retina. And there is more closely a one-to-one mapping of cones to ganglion cells in the fovea. They also found that the receptive fields in this area were circular and monocular, that is coming from a single eye, and included both inhibitory and excitatory areas in a center surround field.

So essentially this looks like a donut and we're going to take a look at what these look like here in a moment. It's called an annulus where you have a center area and the surrounding area just like a donut or a bagel but what happens is when you shine a light in one area you get inhibition and when you shine a light in the other area you get excitation. So there are inhibitory and excitatory areas in this center surround. So we can have on-center cells and off-center cells, which also means on-surround and off-surround.

So here we have an on-center cell. So this is that annulus I was talking about. And if a light's presented in the center, it will increase the firing rate of that neuron. So if we present a light in the center of the receptive field of this neuron, that neuron could be in the LGN or the primary visual cortex, it's going to increase the firing rate of that neuron. Now if we present a light only in the surround, It's actually going to decrease the firing rate of that neuron.

So this is an off-surround cell. Similarly, there are off-center cells. We present a light in the center of an off-center cell that will reduce the firing rate of a neuron. When we present light into the on part of this receptive field, we get an increase in firing rate from the neuron. Now, why is this important?

Well, this is one of the ways in which we're able to discern contrast. So If we shine a light over the entire receptive field of this neuron, its firing rate will not change because we will be both exciting it and inhibiting it at about the same rate. But if we have a neuron right next to it and it has a differential firing rate, then we know that that's where a stimulus sort of begins and ends.

So this is how we're able to do things like see text on a page, is through this center-surround antagonism mechanism because it codes for Where things start and stop in our visual world. So this is basically what we're going to be looking at in our next lecture, is how these center-to-round antagonism cells line up to create things like perception of a line that is upright, or a horizontal line, or a vertical line, etc. And we'll talk about how these receptive fields are mapped out.

We'll talk a little bit more about the perceptual attributes of these, this center-surround antagonism as well. But this is similar to the idea of lateral inhibition, where we're starting to see more complex patterns of responding. And as we move up in the cortex, we'll start to see more complex patterns of responding as well. So this is a good place to pause.

think about this idea of center-surround antagonism and how it affects processing of visual information by increasing or decreasing firing rates of neurons, and then we'll take a look at how this works in the visual cortex in our next lecture.

Transcript for:Understanding Visual Processing Pathways

Transcript for:
Understanding Visual Processing Pathways