Visual cortical neurons can also be characterized by whether one eye or the other drives the activity to a greater or lesser degree, this is called ocular dominance. Now let's return to the organization of the visual pathways for just a moment to explain where this property arises. So, as the 2 retinas send their axons into the optic tract, representing the same side of the visual world. It might be quite natural to assume that that information is combined in the lateral geniculate nucleus. but it's not. The lateral geniculate nucleus is a layered structure. And alternating layers of the lateral geniculate nucleus receive input from one retina or the other. Actually, there are two inner layers that are a bit different from the four outer layers, and the two inner layers are divided into these two components where one layer, layer one receives input from the contralateral eye, layer two from the ipsilateral eye and then the four outer layers,are divided with layers four and six receiving input from the contralateral eye and layers three and five. Receiving input from the ipsilateral eye. So the point is, is that even within the thalamus, the information from the two eyes, while it's representing the same half of the visual world. It's still monocular at this single cell level. In fact, as the lateral geniculate nucleus projects to the visual cortex the information remains monocular, which is to say that these binocular layers of the lateral geniculate nucleus project in an interdigitated fashion to nearby regions and the visual cortex. As if my fingers are representing the inputs driven by either the left eye or the right eye. And you can see that the fingers of my left and right hands are interdigitating as this thalamic input reaches layer four of the visual cortex. This establishes a set of columns that we can see both physiologically and anatomically . And we call them ocular dominance columns. So the input layer to the cortex, layer 4, remains monocular. It's only in the upper layers of cortex as layer 4 projects to the circuitry above layer 4, layers 2 and 3 does information from the two eyes begin to get mixed for the first time. And that mixing of input from the two eyes is quite important because this makes possible the phenomenon of stereopsis. And stereopsis is that component of binocular vision that requires two eyes seeing slightly different views of the same object. And this is a perspective of depth that is possible for objects that are near within about a meter or so is where it works best. And beyond that distance or so, the lines of sight become increasingly parallel and therefore, increasingly similar. But for objects that are within about an arms length this dimension of depth is really quite rich as it's elaborated by convergence by these two monocular channels in the upper layers of the visual cortex. So, here's how this works. Imagine that we are fixating with our two eyes on a particular location in space And all objects that are closer to us and that fixation point are going to be seen by what we call non-corresponding regions of the retinas. That is, there, there are points on the retina that don't project to the same point in the thalamus so, these slightly offset regions of the retina project their axons centrally. And the slight disparity in inputs that are rising on to the same cells in the upper layers of the visual cortex from these two disparate locations of the retinas give rise to signals that allow us to appreciate the fact that an object at location a is near in depth relative to the fixation point. A similar phenomenon would occur for background objects where we might have a stimulation of non corresponding points, but now on the opposite side of the fixation point. So locations indicated by this letter A would give rise to near disparities, and locations at a distance, indicated by letter C, would give rise to far disparities. Of course, there are other cues about depth, other than stereopsis. we can convince ourselves easily of this fact. All one has to do is cover an eye. And the world does not all of a sudden seem to collapse from 3D to 2D. As we move our head around, we can see the movement of foreground objects relative to background objects. And that gives us important information about depth. but nevertheless, stereopsis can be a, a quite valuable tool. As we navigate the world around us. And especially as we interact with objects within arms length. One can demonstrate for yourself stereopsis by getting two pencils and holding them out at arms length. And then bringing them together and you can see how well you can get those tips to touch. So I'm not so bad at this. But if you close one eye and make stereopsis go away we'll see if this presents any further challenge. So I'm going to close my left eye, and now try the same, touching of the pencil tips. Whoa, I'm not so good anymore. Okay that's a little better. Let's try the, the a right eye now. So I'm closing, I'm going to close my right eye. So let's try now closing my right eye and seeing how well I do with my left. And, I seem to be worse when I try to use my left eye. Okay so, the increased performance of using your two eyes together is not so difficult to demonstrate. So next we come to the topic of parallel pathways. So we've been talking about parallel pathways in different context, first in the somatosensory system, now in the visual system so it should be no surprise to you to discover that there are parallel pathways even arising from the very same retina as visual information is processed and then projected back into the brain. If one were to look at the ganglion cells that are present in the retina we find actually a rather rich assortment of ganglion cells. They can be characterized in different ways, by size. By morphology and as investigators over the years have done so, we've adopted a nomenclature for different types of ganglion cells in different species and here's a nomenclature of that is used in the primate retina, there are small cells that we call P ganglion cells. Much larger cells that we call M ganglion cells and then an intermediate type of cell that's called the K ganglion cell. And the different morphologies that we see here are associated with different physiological properties. more so, these different types of cells give rise to axonal projections that terminate in different parts of the lateral geniculate nucleus. There are projections from those small p-cells that terminate Among the small celled layers of the lateral geniculate nucleus. These are called the Parvo cellular layers. Parvo means small. And we've illustrated them here in green. These are layers 3 through 6. meanwhile, the larger m type ganglion cells terminate in the first two layers of the laterals geniculate nucleus, illustrated here in blue layers one and two. So conveniently, the m cells project to the magnocellular layers and the p cells project to the parvocellular layer. Now the k cells project to a zone in between the layers, so there's a region Between each of these layers that we call the koniocellular layers. It's a region where there's a band of white matter or a[UNKNOWN] of white matter between these cell dense layers that have the numbers. So the k cells terminate in those spaces. These different neurons in the lateral geniculate nucleus, send distinct patterns of projection on up to the visual cortex. The Parvo-cellular layers of the lateral geniculate nucleus A synapse into the bottom part of layer four, a zone in the primate that we call layer 4c beta. The magnocellular layers from the anterior part of the lateral geniculate nucleus, send their axons into layer 4c alpha, right on top of the input derived from those parvocellular neurons. The koniocellular cells of the lateral geniculate nucleus do something just a bit odd. They don't project to layer four at all. Rather, they send their axons up into layer three, where they synapse in the zones that are rich in oxidative metabolism. And you may have heard about them. If you're more familiar with visual cortical anatomy or physiology. They're called blobs. literally blobs of high enzymatic activity for oxidative metabolism. So this is an example of parallel processing. We have a small cell pathway that's concerned especially with form and with color that runs through the parvocellular layers of the lateral geniculate nucleus, and into the bottom of layer 4, 4c beta. There is a large cell pathway that runs through the magnocellular layers of the lateral geniculate nucleus. It's not particularly concerned with high special detail or color, but rather it's concerned with broad patterns of notion and this signal is terminated in layer C4 alpha just above that parvocellular termination. And what about the koniocellular layers? Well, these are the more mysterious of the lateral geniculate nucleus layers. It's, they've been not as well studied but they do devise to this very peculiar projection up until layer three. And we think these koniocellular cells May have a particular role to play in color vision. But that remains an issue for ongoing investigation. Now you may be wondering what happens to these parallel pathways once we get beyond the primary visual cortex. Well this remains a matter of active investigation, it was very attractive to assume that the parallel organization is highly conserved. And rather quite rigid. I think we now know that that's probably not the case. That there's likely to be some mixing of information from one parallel pathway to another. Nevertheless when we look beyond the primary visual cortex, it's possible to recognize two broad streams of processing that emanate out from the primary visual cortex through a set of areas that we call extrastriate visual cortex. so recall that that striavginary helps us to recognize a cortex that we call the striate visual cortex, area v1. And then beyond that we see these concentric zones of cortex almost like concentric rings that surround our primary vision cortex. And it was attractive to call them areas of ET, B3, and B4 and so on. to emphasize our presumption that they represented progressively higher levels in some kind of visual hierarchy. well that to remains an object of ongoing investigation. Nevertheless, it is safe to conclude that beyond our primary visual cortex There are rings of higher order visual areas. Now what we're looking at here on the right is a cortical representation as if it were possible to peel off the cerebral hemisphere, flatten out the cortex and then map out these visual areas. Well, in order to make that hemisphere flat, we have to make a cut along the cortex in the calcerine sulcus so that we can relax this three-dimensional shape into a two-dimensional sheet. And that's what we're looking at here. Now I want to draw your attention to a particular location. Which is found out here in the inferior border region between the lateral temporal cortex and lateral occipital cortex. It's where we find a couple of visual areas that we call area MT or sometimes we call it the MT plus region. And this is a area that was named for the corresponding location in the macaque brain, where a middle temporal visual area was discovered that is enriched in motion sensitive cells. So this region, area mt, which is roughly in this position in the human brain, likewise. Seems to be especially enriched in motion, in motion sensitive cells. And in fact, it's from the circuitry here in this area mt plus that we derive our notion of motion. As our chairman of neurobiology here at Duke University likes to say So this is a higher order visual area that represents the kind of specialization that we encounter as we get further away from our primary visual cortex. From this area, mt, these motion signals seem to inform a spatial analysis of the visual world. That is happening in the parietal cortex. So we can think about this parietal area as addressing the question: where am I looking, or where are the objects that I'm looking at that happen to be in motion? So this, this dorsal parietal stream, as we sometimes call it, is about Mapping the world. Achieving a spatial analysis of motion patterns. It's not particularly concerned with detail, with form, with color. With the recognition of faces. Or the identity of objects that we happen to be observing. Those kinds of functions happen along a ventral processing stream. That emanates from the primary visual cortex through its near higher order areas and then out into the inferior temporal lobe. So this process extreme addresses the question, what am I looking at? So, we have our where and our what pathways. Where going dorsally through the[UNKNOWN] lobe and then what. Running ventrally through the temporal lobe. Now you can imagine with this functional dichotomy that people can have brain injuries in the parietal pathway or the temporal pathway. And have deficits in one domain of visual function or another. And that is exactly what is seen People can have injury that result from let's say, stroke or tumor or trauma that affects this where pathway. And consequently, there may be deficits in understanding spacial awareness in the world around us or perhaps very particular deficits in building a notion of motion. On the other hand, there may be strokes or other kinds of brain lesions affecting the what pathway in the inferior part of the temporal lobe. Those kinds of deficits can be quite particular affecting our ability to answer this question: what am I looking at? Or in human relations, who am I looking at? There's a particular region of this posterior occipitotemporal gyrus in an area that's sometimes called the fusiform gyrus where, when damaged leads to a deficit in the Ability of people to recognize faces. It's called prosopagnosia which means face blindness or not knowing the identity of human faces. People with such injuries will have problems of recognizing the faces of even loved ones, even their spouse or their near family members. So it's really quite a, quite a tragic situation for those individuals and you can imagine how devastating that must be to their social relations. Now thankfully we acquire other cues besides the visual that assist in recognition and these individuals get along reasonably well using these alternative cues Never the less, they remain impaired in their ability to understand and recognize the visual features that we associate with the human face. Now, it's important to emphasize that these people, that they're actually they're not blind. What they fail to do is take the structure of the visual image. That is reflected off of the human face and interpret that in terms of the identity. Nearby regions of this ventral or inferior visual pathway likewise seemed to represent other kinds of, of objects that we encounter in our world, perhaps inanimate objects. So one might be able to see but not be able to recognize. Well, this concludes this tutorial and this tour of visual processing. I wish we could spend more time on this topic, I think as you've picked up today, this is an issue that has been of concern in my own academic life and I'd love to share with you more of my research, perhaps I'll have a chance to do so when we move on a little deeper into the course. So, I'll leave you here for now and look forward to speaking with you next time.
