Signal encoding in the lateral geniculate body and primary visual cortex
Retinal ganglion cells project their processes to the lateral geniculate body, where they form a retinotopic map. In mammals, the lateral geniculate body consists of 6 layers, each of which is innervated by either one or the other eye and receives input from different subtypes of ganglion cells, forming layers of magnocellular, parvocellular and koniocellular neurons. Neurons in the lateral geniculate nucleus have center-to-ground receptive fields, similar to retinal ganglion cells.
Neurons in the lateral geniculate nucleus project to form a retinotopic map in the primary visual cortex V1, also called “area 17” or striate cortex. The receptive fields of cortical cells, instead of the already familiar organization of receptive fields according to the “center-background” type, consist of lines, or edges, which is a fundamentally new step in the analysis of visual information. The six layers of V 1 have structural features: afferent fibers from the geniculate body end mainly in layer 4 (and some in layer 6); cells in layers 2, 3 and 5 receive signals from cortical neurons. Cells of layers 5 and b project processes to the subcortical areas, and cells of layers 2 and 3 project to other cortical areas. Each vertical column of cells functions as a module, receiving an initial visual signal from a specific location in space and sending processed visual information to secondary visual areas. The columnar organization of the visual cortex is obvious, since the localization of receptive fields remains the same throughout the depth of the cortex, and visual information from each eye (right or left) is always processed in strictly defined columns.
Two classes of neurons in area V 1 have been described that differ in their physiological properties. The receptive fields of simple cells are elongated and contain conjugate “on” and “off” zones. Therefore, the most optimal stimulus for a simple cell is specially oriented beams of light or shadow. A complex cell responds to a certain oriented strip of light; this strip can be located in any region of the receptive field.The inhibition of simple or complex cells resulting from image recognition carries even more detailed information about the properties of the signal, such as the presence of a line of a certain length or a certain angle within a given receptive field.
The receptive fields of a simple cell are formed as a result of the convergence of a significant number of afferents from the geniculate body. The centers of several receptive fields adjacent to each other form one cortical receptive zone. The field of a complex cell depends on signals from the simple cell and other cortical cells. The sequential change in the organization of receptive fields from the retina to the lateral geniculate body and then to simple and complex cortical cells suggests a hierarchy in information processing, whereby a number of neural constructs at one level are integrated at the next, where an even more abstract concept is formed based on the initial information. At all levels of the visual analyzer, special attention is paid to contrast and determining the boundaries of the image, and not to the general illumination of the eye. Thus, complex cells in the visual cortex can “see” the lines that are the boundaries of a rectangle, and they are little concerned about the absolute intensity of light within that rectangle. A series of clear and continuous studies in the field of mechanisms of perception of visual information, begun by Kuffler's pioneering work on the retina, was continued at the level of the visual cortex by Hubel and Wiesel. Hubel gave a vivid description of early experiments on the visual cortex in the laboratory of Stephen Kuffler at Johns Hopkins University (USA) in the 50s of the 20th century. Since then, our understanding of the physiology and anatomy of the cerebral cortex has developed significantly due to the experiments of Hubel and Wiesel, as well as due to the large number of works for which their research was a starting point or source of inspiration. Our goal is to provide a brief, narrative account of signal encoding and cortical architecture from a perceptual perspective, based on the classic work of Hubel and Wiesel, as well as more recent experiments performed by them, their colleagues, and many others. This chapter will merely sketch the functional architecture of the lateral geniculate nucleus and visual cortex and their role in providing the first steps of visual analysis: identifying lines and shapes from the retinal signal in a center-to-ground pattern.
When moving from the retina to the lateral geniculate body, and then to the cerebral cortex, questions arise that are beyond the limits of technology. For a long time, it was generally accepted that to understand the functioning of any part of the nervous system, knowledge about the properties of its constituent neurons is necessary: how they conduct signals and carry information, how they transmit received information from one cell to another through synapses. However, monitoring the activity of only one individual cell is unlikely to be an effective method for studying higher functions where a large number of neurons are involved. The argument that has been used here and continues to be used from time to time is this: the brain contains about 10 10 or more cells. Even the simplest task or event involves hundreds of thousands of nerve cells located in various parts of the nervous system. What are the chances of a physiologist being able to penetrate into the essence of the mechanism of formation of complex action in the brain if he can simultaneously examine only one or a few nerve cells, a hopelessly small fraction of the total?
Upon closer examination, the logic of such arguments regarding the basic complexity of the study associated with a large number of cells and complex higher functions no longer seems so flawless. As often happens, a simplifying principle emerges that opens up a new and clear view of the problem. What simplifies the situation in the visual cortex is that the major cell types are located separately from each other, in well-organized and repeating units. This repetitive pattern of neural tissue is closely intertwined with the retinotopic map of the visual cortex. Thus, neighboring points on the retina are projected onto neighboring points on the surface of the visual cortex. This means that the visual cortex is organized in such a way that for each smallest segment of the visual field there is a set of neurons to analyze information and transmit it. In addition, higher-level patterns of cortical organization have been identified using methods that allow the isolation of functionally related cellular assemblies. Indeed, cortical architecture determines the structural basis of cortical function, so new anatomical approaches inspire new analytical studies. Thus, before we describe the functional connections of visual neurons, it is useful to briefly summarize the general structure of the central visual pathways arising from the lateral geniculate nucleus.
Lateral geniculate body
The optic nerve fibers originate from each eye and end on the cells of the right and left lateral geniculate body (LCT) (Fig. 1), which has a clearly distinguishable layered structure (“geniculate” means “curved like a knee”). In the LCT of a cat, you can see three obvious, clearly distinguishable layers of cells (A, A 1, C), one of which (A 1) has a complex structure and is further subdivided. In monkeys and other primates, including In humans, the LCT has six layers of cells. The cells in the deeper layers 1 and 2 are larger in size than in layers 3, 4, 5 and 6, which is why these layers are called large cell (M, magnocellular) and small cell (P, parvocellular), respectively. The classification also correlates with large (M) and small (P) retinal ganglion cells, which send their processes to the LCT. Between each M and P layers lies a zone of very small cells: the intralaminar, or koniocellular (K, koniocellular) layer. K layer cells differ from M and P cells in their functional and neurochemical properties, forming a third channel of information into the visual cortex. In both cats and monkeys, each layer of the LCT receives signals from either one or the other eye. In monkeys, layers 6, 4 and 1 receive information from the contralateral eye, and layers 5, 3 and 2 from the ipsilateral eye. The division of the course of nerve endings from each eye into different layers has been shown using electrophysiological and a number of anatomical methods. Particularly surprising is the type of branching of an individual fiber of the optic nerve when the enzyme horseradish peroxidase is injected into it (Fig. 2). Terminal formation is limited to the LCT layers for that eye, without extending beyond the boundaries of these layers. Due to the systematic and specific division of the optic nerve fibers in the area of the chiasm, all the receptive fields of the LCT cells are located in the visual field of the opposite side. Rice. 2. The endings of optic nerve fibers in the LCT of a cat. Horseradish peroxidase was injected into one of the axons from the zone with the “on” center of the contralateral eye. Axon branches end on cells of layers A and C, but not A 1. Rice. 3. Receptive fields of ST cells. The concentric receptive fields of the LCT cells resemble the fields of ganglion cells in the retina, dividing into fields with an “on” and “off” center. The responses of the cell with the “on” center of the cat LCT are shown. The bar above the signal shows the duration of illumination. Central and peripheral zones neutralize each other's effects, so diffuse illumination of the entire receptive field gives only weak responses (lower entry), even less pronounced than in retinal ganglion cells. An important topographical feature is the high orderliness in the organization of receptive fields within each layer of the LCT. Adjacent regions of the retina form connections with neighboring LCT cells, so that the receptive fields of nearby LCT neurons overlap over a large area. Cells in the central zone of the cat retina (the region where the cat retina has small receptive fields with small centers), as well as the monkey optic fovea, form connections with a relatively large number of cells within each layer of the LCT. A similar distribution of bonds was found in humans using NMR. The number of cells associated with the peripheral regions of the retina is relatively small. This overrepresentation of the optic fovea reflects the high density of photoreceptors in the area that is necessary for vision with maximum acuity. Although, probably, the number of optic nerve fibers and the number of LCT cells are approximately equal, nevertheless, each LCT neuron receives converging signals from several fibers of the optic nerve. Each optic nerve fiber, in turn, forms divergent synaptic connections with several LCT neurons. However, not only is each layer topographically ordered, but also the cells of different layers are in a retinotopic relationship to each other. That is, if you move the electrode strictly perpendicular to the surface of the LCT, then the activity of cells receiving information from the corresponding zones of one and then the other eye will first be recorded as the microelectrode crosses one layer of the LCT after another. The location of the receptive fields is in strictly corresponding positions on both retinas, that is, they represent the same area of the visual field. In LCT cells there is no significant mixing of information from the right and left eyes and interaction between them; only a small number of neurons (which have receptive fields in both eyes) are excited exclusively binocularly. It is surprising that the responses of LCT cells do not differ significantly from the signals of ganglion cells (Fig. 3). LCT neurons also have concentrically organized antagonizing receptive fields, either with an “off” or “on” center, but the contrast mechanism is finer regulated, due to greater correspondence between inhibitory and excitatory zones. Thus, like retinal ganglion cells, the optimal stimulus for LCT neurons is contrast, but they respond even weaker to general illumination. The study of the receptive fields of LCT neurons has not yet been completed. For example, neurons were found in the LCT, the contribution of which to the functioning of the LCT has not been established, as well as pathways going from the cortex down to the LCT. Cortical feedback is required for synchronized activity of LCT neurons. Why does LCT have more than one layer for each eye? It has now been discovered that neurons in different layers have different functional properties. For example, cells located in the fourth dorsal parvocellular layers of the monkey LCT, like P ganglion cells, are able to respond to light of different colors, showing good color discrimination. Conversely, layers 1 and 2 (magnocellular layers) contain M-like cells, which give rapid (“live”) responses and are insensitive to color, while K layers receive signals from “blue-on” retinal ganglion cells and can play a special role in color vision. In cats, X and Y fibers (see section "Classification of ganglion cells" end in various sublayers A, C and A 1, therefore, specific inactivation of layer A, but not C, sharply reduces the accuracy of ocular movements. Cells with "on" - and "off" "-center is also subdivided into different layers in the LCT of mink and ferret, and to some extent in monkeys. To summarize the above, the LCT is a way station in which the axons of ganglion cells are sorted in such a way that neighboring cells receive signals from identical regions of the visual fields, and the neurons that process information are organized in clusters.Thus, in LCT, an anatomical basis for parallel processing of visual information is evident. Visual information enters the cortex and LCT through optical radiation. In monkeys, optical radiation ends at the folded plate, about 2 mm thick (Fig. 4). This region of the brain - known as the primary visual cortex, visual area 1 or V 1 - is also called the striate cortex, or "area 17". Older terminology was based on anatomical criteria developed at the beginning of the 20th century. V 1 lies posteriorly, in the region of the occipital lobe, and can be recognized in a transverse section by its special appearance. The bundles of fibers in this area form a stripe that is clearly visible to the naked eye (hence the zone is called “striated”, Fig. 4B). Neighboring zones outside the striation zone are also associated with vision. The zone immediately surrounding zone V is called zone V 2 (or "zone 18") and receives signals from zone V, (see Fig. 4C). Clear boundaries of the so-called extrastriate visual cortex (V 2 -V 5) cannot be established using visual examination of the brain, although a number of criteria have been developed for this. For example, in V 2 the striation disappears, large cells are located superficially, and coarse, obliquely arranged myelin fibers are visible in the deeper layers. Each zone has its own representation of the visual field of the retina, projected in a strictly defined, retinotopic manner. Projection maps were compiled in an era when it was not possible to analyze the activity of individual cells. Therefore, mapping was done by illuminating small areas of the retina with beams of light and recording cortical activity using a large electrode. These maps, as well as their modern counterparts recently produced using brain imaging techniques such as positron emission tomography and functional nuclear magnetic resonance, have shown that the cortical area devoted to representing the fovea is much larger in size than the area , allocated to the rest of the retina. These findings, in principle, corresponded to expectations, since pattern recognition by the cortex is carried out mainly by processing information from photoreceptors densely located in the fovea area. This representation is analogous to the extended representation of the hand and face in the primary somatosensory cortex. The retinal fossa projects to the occipital pole of the cerebral cortex. The map of the retinal periphery extends anteriorly along the medial surface of the occipital lobe (Fig. 5). Due to the inverted picture formed on the retina by the lens, the superior visual field is projected onto the lower region of the retina and transmitted to area V 1, located below the calcarine sulcus; the lower visual field is projected above the calcarine sulcus. In cortical slices, neurons can be classified by their shape. The two main groups of neurons form stellate and pyramidal cells. Examples of these cells are shown in Fig. 6B. The main differences between them are the length of the axons and the shape of the cell bodies. The axons of pyramidal cells are longer and descend into the white matter, leaving the cortex; the processes of stellate cells end in the nearest zones. These two groups of cells may have other differences, such as the presence or absence of dendritic spines, which provide their functional properties. There are other quaintly named neurons (bibouquet cells, chandelier cells, basket cells, crescent cells), as well as neuroglial cells. Their characteristic feature is that the processes of these cells are directed mainly in the radial direction: up and down through the thickness of the cortex (at an appropriate angle to the surface). Conversely, many (but not all) of their lateral processes are short. Connections between the primary visual cortex and the higher-order cortex are made by axons that run in bundles through the white matter underlying the cell layers. Rice. 7. Connections of the visual cortex. (A) Layers of cells with various incoming and outgoing processes. Note that the original processes from the LCT are mainly interrupted in the 4th layer. The processes from the LCT coming from the large cell layers are predominantly interrupted in layers 4C and 4B, while the processes from the small cell layers are interrupted in 4A and 4C. Simple cells are located mainly in layers 4 and 6, complex cells - in layers 2, 3, 5 and 6. Cells in layers 2, 3 and 4B send axons to other cortical zones; cells in layers 5 and 6 send axons to the superior colliculus and LCT. (B) Typical branching of LCT axons and cortical neurons in a cat. In addition to such vertical connections, many cells have long horizontal connections that extend within one layer to distant regions of the cortex. The main feature of the mammalian cortex is that the cells here are arranged in 6 layers within the gray matter (Fig. 6A). The layers vary greatly in appearance, depending on the density of the cells, as well as the thickness of each of the cortical zones. The incoming paths are shown in Fig. 7A on the left side. Based on the LCT, the fibers mainly terminate in layer 4 with a small number of connections also formed in layer 6. The superficial layers receive signals from the pulvinar zone or other areas of the thalamus. A large number of cortical cells, especially in the region of layer 2, as well as in the upper parts of layers 3 and 5, receive signals from neurons also located within the cortex. The bulk of the fibers coming from the LCT into layer 4 are then divided between the various sublayers. Fibers emanating from layers 6, 5, 4, 3 and 2 are shown on the right in Fig. 7A. Cells that send efferent signals from the cortex may also control intracortical connections between different layers. For example, axons from a cell in layer 6 other than the LCT may also project to one of the other cortical layers, depending on the type of response of that cell 34) . Based on this structure of the visual pathways, the following visual signal path can be imagined: information from the retina is transmitted to the cortical cells (mainly in layer 4) by the axons of the LCT cells; information is transmitted from layer to layer, from neuron to neuron throughout the entire thickness of the cortex; processed information is sent to other areas of the cortex using fibers that go deep into the white matter and return back to the cortex. Thus, the radial or vertical organization of the cortex leads us to believe that columns of neurons work as separate computational units, processing various details of visual scenes and forwarding the received information further to other regions of the cortex. LCT afferent fibers end in layer 4 of the primary visual cortex, which has a complex organization and can be studied both physiologically and anatomically. The first feature we want to demonstrate is the separation of incoming fibers coming from different eyes. In adult cats and monkeys, cells within one layer of the LCT, receiving signals from one eye, send processes to strictly defined clusters of cortical cells in layer 4C, which are responsible specifically for this eye. Clusters of cells are grouped in alternating stripes or bundles of cortical cells that receive information exclusively from the right or left eye. In more superficial and deeper layers, neurons are controlled by both eyes, although usually with a predominance of one of them. Hubel and Wiesel conducted an original demonstration of the separation of information from different eyes and the dominance of one of them in the primary visual cortex using electrophysiological methods. They used the term ocular dominance columns to describe their observations, following the concept of cortical columns developed by Mountcastle for the somatosensory cortex. A series of experimental techniques were designed to demonstrate alternating groups of cells in layer 4 receiving information from the right or left eye. Initially, it was proposed to cause slight damage within only one layer of the LCT (remember that each layer receives information from only one eye). If this is done, degenerating terminals appear in layer 4, forming a specific pattern of alternating spots that correspond to areas controlled by the eye sending information to the damaged area of the LCT. Later, a stunning demonstration of the existence of a distinct ocular dominance pattern was made using the transport of radioactive amino acids from one eye. The experiment consists of injecting an amino acid (proline or lecithin) containing radioactive tritium atoms into the eye. The injection is carried out into the vitreous body of the eye, from which the amino acid is captured by the bodies of the retinal nerve cells and incorporated into the protein. Over time, the protein labeled in this way is transported into ganglion cells and along the optic nerve fibers to their terminals within the LCT. A remarkable feature is that this radioactive tracer is also transferred from neuron to neuron through chemical synapses. Ultimately, the label reaches the endings of the LCT fibers within the visual cortex. In Fig. Figure 8 shows the location within layer 4 of radioactive terminals formed by the axons of LCT cells associated with the eye into which the label was introduced Rice. 8. Ocular dominant columns in the cortex of a monkey obtained by injecting radioactive proline into one eye. Autoradiograms taken under dark-field illumination, where silver grains are shown in white. (A) From the top of the figure, the slice passes through layer 4 of the visual cortex at an angle to the surface, forming a perpendicular slice of the columns. In the center, layer 4 was cut horizontally, showing that the column consists of elongated plates. (B) Reconstruction from multiple horizontal sections of layer 4C in another monkey in which the injection was made in the ilsilateral eye. (Any horizontal section may reveal only part of layer 4, which is due to the curvature of the cortex.) In both A and B, the columns of visual dominance look like stripes of equal width, receiving information from either one or the other eye. located directly above the visual cortex, so such areas appear as white spots on the dark background of the photograph). Marker spots are interspersed with unmarked areas, which receive information from the contralateral eye where no cue was injected. The center-to-center distance between the spots, which correspond to the oculodominant columns, is approximately 1 mm. At the cellular level, a similar structure was identified in layer 4 by injecting horseradish peroxidase into individual cortical axons of LCT neurons. The axon shown in Fig. 9, comes from the LCT neuron with an “off” center, responding with short signals to shadows and moving spots. The axon ends in two different groups of processes in layer 4. The groups of labeled processes are separated by an empty, unlabeled zone, corresponding in size to the territory responsible for the other eye. This kind of morphological research expands the boundaries and allows for a deeper understanding of the original description of the ocular dominance columns compiled by Hubel and Wiesel in 1962. Literature 2. o Ferster, D., Chung, S., and Wheat, H. 1996. Orientation selectivity of thalamic input to simple cells of cat visual cortex. Nature 380:249-252. 3. o Hubel, D. H., and Wiesel, T. N. 1959. Receptive fields of single neurons in the cat's striate cortex. /. Physiol. 148: 574-591. 4. about Hubel, D.H., and Wiesel, T.N. 1961. Integrative action in the cat's lateral geniculate body. Physiol. 155: 385-398. 5. o Hubel, D. H., and Wiesel, T. N. 1962. Receptive fields, binocular interaction and functional architecture in the cat's visual cortex. /. Physiol. 160: 106-154.
Maps of visual fields in the lateral geniculate body
Functional layers of LCT
Cytoarchitecture of the visual cortex
Incoming, outgoing pathways and layered organization of the cortex
Separation of incoming fibers from LCT in layer 4
This is a subcortical center that ensures the transmission of information to the visual cortex.
In humans, this structure has six layers of cells, as in the visual cortex. Fibers from the retina enter the chiasma opticus, crossed and uncrossed. The 1st, 4th, 6th layers receive crossed fibers. The 2nd, 3rd, 5th layers are uncrossed.
All information coming to the lateral geniculate body from the retina is ordered and the retinotopic projection is maintained. Since the fibers enter the lateral geniculate body like a comb, there are no neurons in the NKT that receive information from two retinas simultaneously. It follows from this that there is no binocular interaction in the NKT neurons. The tubing receives fibers from M-cells and P-cells. The M-path, which communicates information from large cells, transmits information about the movements of objects and ends in the 1st and 2nd layers. The P-path is associated with color information and the fibers terminate in layers 3, 4, 5, 6. In the 1st and 2nd layers of the NKT, the receptive fields are highly sensitive to movement and do not distinguish spectral characteristics (color). Such receptive fields are also present in small quantities in other layers of the tubing. In the 3rd and 4th layers, neurons with an OFF center predominate. It is blue-yellow or blue-red + green. The 5th and 6th layers contain neurons with ON centers, mostly red-green. The receptive fields of the cells of the lateral geniculate body have the same receptive fields as ganglion cells.
The difference between these receptive fields and ganglion cells is:
1. In the size of the receptive fields. The cells of the external geniculate body are smaller.
2. Some NKT neurons have an additional inhibitory zone surrounding the periphery.
For cells with an ON center, such an additional zone will have a reaction sign coinciding with the center. These zones are formed only in some neurons due to increased lateral inhibition between NKT neurons. These layers are the basis for the survival of a particular species. Humans have six layers, predators have four.
Detector theory appeared in the late 1950s. In the frog's retina (in ganglion cells), reactions were found that were directly related to behavioral responses. Excitation of certain retinal ganglion cells led to behavioral responses. This fact allowed us to create the concept according to which the image presented on the retina is processed by ganglion cells specifically tuned to the elements of the image. Such ganglion cells have specific dendritic branching, which corresponds to a certain structure of the receptive field. Several types of such ganglion cells have been discovered. Subsequently, neurons with this property were called detector neurons. Thus, a detector is a neuron that responds to a specific image or part of it. It turned out that other, more highly developed animals also have the ability to highlight a specific symbol.
1. Convex edge detectors - the cell was activated when a large object appeared in the field of view;
2. Detector of moving fine contrast - its activation led to an attempt to capture this object; corresponds in contrast to the objects being captured; these reactions are associated with food reactions;
3. Blackout detector - causes a defensive reaction (the appearance of large enemies).
These retinal ganglion cells are tuned to secrete certain elements of the environment.
Group of researchers who worked on this topic: Letvin, Maturano, Moccalo, Pitz.
Neurons of other sensory systems also have detector properties. Most detectors in the visual system are concerned with motion detection. Neurons increase their reactions when the speed of movement of objects increases. Detectors have been found in both birds and mammals. Detectors of other animals are directly connected to the surrounding space. Birds have been found to have horizontal surface detectors, due to the need to land on horizontal objects. Detectors of vertical surfaces were also discovered, which ensure the birds’ own movements towards these objects. It turned out that the higher the animal is in the evolutionary hierarchy, the higher the detectors are, i.e. these neurons may already be located not only in the retina, but also in the higher parts of the visual system. In higher mammals: in monkeys and humans, detectors are located in the visual cortex. This is important because the specific method that provides responses to elements of the external environment is transferred to higher levels of the brain, and each animal species has its own specific types of detectors. Later it turned out that during ontogenesis the detector properties of sensory systems are formed under the influence of the environment. To demonstrate this property, experiments were carried out by Nobel laureate researchers Hubel and Wiesel. Experiments were carried out that proved that the formation of detector properties occurs in the earliest ontogenesis. For example, three groups of kittens were used: one control and two experimental. The first experimental one was placed in conditions where mainly horizontally oriented lines were present. The second experimental one was placed in conditions where there were mainly horizontal lines. The researchers checked which neurons formed in the cortex of the kittens in each group. In the cortex of these animals, it turned out that 50% of neurons were activated both horizontally and 50% vertically. Animals raised in a horizontal environment had a significant number of neurons in the cortex that were activated by horizontal objects; there were practically no neurons activated when perceiving vertical objects. In the second experimental group there was a similar situation with horizontal objects. Kittens of both horizontal groups developed certain defects. Kittens in a horizontal environment could jump perfectly on steps and horizontal surfaces, but were poor at moving relative to vertical objects (table legs). The kittens of the second experimental group had the corresponding situation for vertical objects. This experiment proved:
1) formation of neurons in early ontogenesis;
2) the animal cannot interact adequately.
Changing animal behavior in a changing environment. Each generation has its own set of external stimuli that produce a new set of neurons.
Specific features of the visual cortex
From the cells of the external geniculate body (has a 6-layer structure), axons enter the 4 layers of the visual cortex. The bulk of the axons of the external geniculate body (ECC) are distributed in the fourth layer and its sublayers. From the fourth layer, information flows to other layers of the cortex. The visual cortex retains the principle of retinotopic projection in the same way as the NKT. All information from the retina goes to the neurons of the visual cortex. Neurons in the visual cortex, like neurons at lower levels, have receptive fields. The structure of the receptive fields of neurons in the visual cortex differs from the receptive fields of the NKT and retinal cells. Hubel and Wiesel also studied the visual cortex. Their work made it possible to create a classification of receptive fields of neurons in the visual cortex (RPNFrK). H. and V. Discovered that the RPNZrKs are not concentric, but rectangular in shape. They can be oriented at different angles and have 2 or 3 antagonistic zones.
Such a receptive field can highlight:
1. change in illumination, contrast - such fields were called simple receptive fields;
2. neurons with complex receptive fields– can select the same objects as simple neurons, but these objects can be located anywhere in the retina;
3. super complex fields- can highlight objects that have breaks, boundaries or changes in the shape of the object, i.e. super complex receptive fields can highlight geometric shapes.
Gestalts are neurons that highlight sub-images.
Cells of the visual cortex can only form certain elements of the image. Where does constancy come from, where does the visual image appear? The answer was found in association neurons, which are also associated with vision.
The visual system can distinguish different color characteristics. The combination of opposing colors allows you to highlight different shades. Lateral inhibition is necessarily involved.
Receptive fields have antagonistic zones. Neurons of the visual cortex are able to be excited peripherally to green while the middle is excited to the action of a red source. The action of green will cause an inhibitory reaction, the action of red will cause an excitatory reaction.
The visual system perceives not only pure spectral colors, but also any combination of shades. Many areas of the cerebral cortex have not only a horizontal, but also a vertical structure. This was discovered in the mid-1970s. This has been shown for the somatosensory system. Vertical or columnar organization. It turned out that the visual cortex, in addition to layers, also has vertically oriented columns. Improvements in recording techniques have led to more sophisticated experiments. Neurons of the visual cortex, in addition to layers, also have a horizontal organization. A microelectrode was placed strictly perpendicular to the surface of the cortex. All major visual fields are in the medial occipital cortex. Since receptive fields have a rectangular organization, dots, spots, or any concentric objects do not cause any reaction in the cortex.
The column is the type of reaction, the adjacent column also highlights the slope of the line, but it differs from the previous one by 7-10 degrees. Further research showed that there are columns located nearby in which the angle changes in equal increments. About 20-22 adjacent columns will highlight all tilts from 0 to 180 degrees. The set of columns capable of highlighting all gradations of this characteristic is called a macrocolumn. These were the first studies that showed that the visual cortex can highlight not only a single property, but also a complex - all possible changes in a feature. In further studies, it was shown that next to the macrocolumns that fix the angle, there are macrocolumns that can highlight other properties of the image: colors, direction of movement, speed of movement, as well as macrocolumns associated with the right or left retina (ocular dominance columns). Thus, all macrocolumns are compactly located on the surface of the cortex. It was proposed to call collections of macrocolumns hypercolumns. Hypercolumns can analyze a set of image features located in a local region of the retina. Hypercolumns are a module that highlights a set of features in a local area of the retina (1 and 2 identical concepts).
Thus, the visual cortex consists of a set of modules that analyze the properties of images and create subimages. The visual cortex is not the final stage of processing visual information.
Properties of binocular vision (stereo vision)
These properties make it easier for both animals and humans to perceive the distance of objects and the depth of space. In order for this ability to manifest itself, eye movements (convergent-divergent) to the central fovea of the retina are required. When considering a distant object, the optical axes move apart (divergence) and converge for nearby ones (convergence). This system of binocular vision is present in different species of animals. This system is most perfect in those animals whose eyes are located on the front surface of the head: in many predatory animals, birds, primates, most predatory monkeys.
In other animals, the eyes are located laterally (ungulates, mammals, etc.). It is very important for them to have a large volume of perception of space.
This is due to the habitat and their place in the food chain (predator - prey).
With this method of perception, perception thresholds are reduced by 10-15%, i.e. Organisms with this property have an advantage in the accuracy of their own movements and their correlation with the movements of the target.
Monocular cues to spatial depth also exist.
Properties of binocular perception:
1. Fusion - fusion of completely identical images of two retinas. In this case, the object is perceived as two-dimensional, flat.
2. Fusion of two non-identical retinal images. In this case, the object is perceived three-dimensionally, three-dimensionally.
3. Rivalry of visual fields. There are two different images coming from the right and left retinas. The brain cannot combine two different images, and therefore they are perceived alternately.
The remaining points of the retina are disparate. The degree of disparity will determine whether the object is perceived three-dimensionally or whether it will be perceived with competing visual fields. If the disparity is small, then the image is perceived three-dimensionally. If the disparity is very high, then the object is not perceived.
Such neurons were found not in the 17th, but in the 18th and 19th fields.
How do the receptive fields of such cells differ: for such neurons in the visual cortex, the receptive fields are either simple or complex. In these neurons, there is a difference in receptive fields from the right and left retina. The disparity of the receptive fields of such neurons can be either vertical or horizontal (see next page):
This property allows for better adaptation.
(+) The visual cortex does not allow us to say that a visual image is formed in it, then there is no constancy in all areas of the visual cortex.
Related information.
External geniculate body
Axons of the optic tract approach one of four second-order perceptive and integrating centers. The nuclei of the lateral geniculate body and superior colliculus are the target structures most important for visual function. The geniculate bodies form a “knee-like” bend, and one of them - the lateral one (i.e., lying further from the median plane of the brain) - is associated with vision. The quadrigeminal tubercles are two paired elevations on the surface of the thalamus, of which the upper ones deal with vision. The third structure - the suprachiasmatic nuclei of the hypothalamus (they are located above the optic chiasm) - uses information about the intensity of light to coordinate our internal rhythms. Finally, the oculomotor nuclei coordinate eye movements when we look at moving objects.
Lateral geniculate nucleus. Axons of ganglion cells form synapses with cells of the lateral geniculate body in such a way that the display of the corresponding half of the visual field is restored there. These cells in turn send axons to cells in the primary visual cortex, a zone in the occipital lobe of the cortex.
The superior tubercles of the quadrigeminal. Many ganglion cell axons branch before reaching the lateral geniculate nucleus. While one branch connects the retina to this nucleus, the other goes to one of the secondary level neurons in the superior colliculus. As a result of this branching, two parallel pathways are created from the retinal ganglion cells to two different centers of the thalamus. In this case, both branches retain their retinotopic specificity, i.e., they arrive at points that together form an ordered projection of the retina. Neurons in the superior colliculus, receiving signals from the retina, send their axons to a large nucleus in the thalamus called the pulvinar. This nucleus becomes larger and larger among mammals as their brains become more complex and reaches its greatest development in humans. The large size of this formation suggests that it performs some special functions in humans, but its true role remains unclear. Along with primary visual signals, neurons in the superior colliculus receive information about sounds emanating from certain sources and about the position of the head, as well as processed visual information returning through a feedback loop from neurons in the primary visual cortex. On this basis, it is believed that the tubercles serve as the primary centers for integrating information that we use for spatial orientation in a changing world.
Visual cortex
The bark has a layered structure. The layers differ from each other in the structure and shape of the neurons that form them, as well as the nature of the connection between them. According to their shape, neurons of the visual cortex are divided into large and small, stellate, bush-shaped, fusiform.
The famous neuropsychologist Lorente de No in the 40s. twentieth century discovered that the visual cortex is divided into vertical elementary units, which are a chain of neurons located in all layers of the cortex.
Synaptic connections in the visual cortex are very diverse. In addition to the usual division into axosomatic and axodendritic, terminal and collateral, they can be divided into two types: 1) synapses with a large extent and multiple synaptic endings and 2) synapses with a short extent and single contacts.
The functional significance of the visual cortex is extremely great. This is proven by the presence of numerous connections not only with specific and nonspecific nuclei of the thalamus, reticular formation, dark association area, etc.
Based on electrophysiological and neuropsychological data, it can be argued that at the level of the visual cortex, a subtle, differentiated analysis of the most complex features of the visual signal is carried out (identification of contours, outlines, shape of an object, etc.). At the level of the secondary and tertiary areas, apparently, the most complex integrative process occurs, preparing the body for the recognition of visual images and the formation of a sensory-perceptual picture of the world.
brain retina occipital visual
External geniculate body It is a small oblong elevation at the posterior-inferior end of the optic thalamus, lateral to the pulvinar. The ganglion cells of the lateral geniculate body end with the fibers of the optic tract and the fibers of the Graziole bundle originate from them. Thus, the peripheral neuron ends here and the central neuron of the visual pathway begins.
It has been established that although majority fibers of the optic tract and ends in the external geniculate body, yet a small part of them goes to the pulvinar and the anterior quadrigeminal. These anatomical data served as the basis for the opinion, widespread for a long time, according to which both the external geniculate body and the pulvinar and anterior quadrigemina were considered the primary visual centers.
Currently A lot of data has accumulated that does not allow us to consider the pulvinar and anterior quadrigemina as primary visual centers.
Comparison clinical and pathological data, as well as the data of embryology and comparative anatomy, does not allow us to attribute the role of the primary visual center to the pulvinar. Thus, according to Genshen’s observations, in the presence of pathological changes in the pulvinar, the visual field remains normal. Brouwer notes that with a changed lateral geniculate body and an unchanged pulvinar, homonymous hemianopsia is observed; with changes in the pulvinar and unchanged external geniculate body, the visual field remains normal.
Likewise The same is true for the anterior quadrigeminal region. The fibers of the optic tract form the visual layer in it and end in cell groups located near this layer. However, Pribytkov's experiments showed that enucleation of one eye in animals is not accompanied by degeneration of these fibers.
Based on all of the above higher at present there is reason to believe that only the lateral geniculate body is the primary visual center.
Moving on to the question of retinal projections in the lateral geniculate body, the following should be noted. Monakov generally denied the presence of any retinal projection in the external geniculate body. He believed that all fibers coming from different parts of the retina, including papillo-macular ones, are evenly distributed throughout the external geniculate body. Genshen proved the fallacy of this view back in the 90s of the last century. In 2 patients with homogeneous lower quadrant hemianopia, during a pathological examination, he found limited changes in the dorsal part of the lateral geniculate body.
Ronne with optic nerve atrophy with central scotomas due to alcohol intoxication, found limited changes in ganglion cells in the lateral geniculate body, indicating that the area of the macula projects to the dorsal part of the geniculate body.
The above observations from certainty prove the presence of a certain projection of the retina in the external geniculate body. But the clinical and anatomical observations available in this regard are too few in number and do not yet provide accurate ideas about the nature of this projection. The experimental studies we mentioned by Brouwer and Zeman on monkeys made it possible to study to some extent the projection of the retina in the lateral geniculate body.
External geniculate body (corpus genicu-latum laterale) is the location of the so-called “second neuron” of the visual pathway. About 70% of the fibers of the optic tract pass through the external geniculate body. The external geniculate body is a hill corresponding to the location of one of the nuclei of the optic thalamus (Fig. 4.2.26-4.2.28). It contains about 1,800,000 neurons, on the dendrites of which the axons of the retinal ganglion cells end.
It was previously assumed that the lateral geniculate body is just a “relay station”, transmitting information from retinal neurons through the optic radiation to the cerebral cortex. It has now been shown that quite significant and diverse processing of visual information occurs at the level of the lateral geniculate body. The neurophysiological significance of this formation will be discussed below. Initially it is necessary
Rice. 4.2.26. Model of the left lateral geniculate body (after Wolff, 1951):
A- rear and inside views; b - rear and external view (/ - optic tract; 2 - saddle; 3 - visual radiance; 4 - head; 5 - body; 6 - isthmus)
Let us dwell on its anatomical features.
The nucleus of the external geniculate body is one of the nuclei of the thalamus opticus. It is located between the ventroposteriolateral nucleus of the optic thalamus and the cushion of the optic thalamus (Fig. 4.2.27).
The external geniculate nucleus consists of the dorsal and phylogenetically more ancient ventral nuclei. The ventral nucleus in humans is preserved as a rudiment and consists of a group of neurons located rostral to the dorsal nucleus. In lower mammals, this nucleus provides the most primitive photostatic reactions. The fibers of the optic tract do not approach this nucleus.
The dorsal nucleus makes up the main part of the nucleus of the lateral geniculate body. It is a multilayer structure in the form of a saddle or an asymmetrical cone with a rounded top (Fig. 4.2.25-4.2.28). A horizontal section shows that the external geniculate body is connected anteriorly with the optic tract, on the lateral side with the retrolenticular part of the internal capsule, medially with the middle geniculate body, posteriorly with the hippocampal gyrus, and posteriolaterally with the inferior horn of the lateral ventricle. Adjacent to the nucleus of the external geniculate body is the cushion of the visual thalamus, anteriolaterally - the temporopontine fibers and the posterior part of the internal capsule, laterally - Wernicke's area, and on the inner side - the medial nucleus (Fig. 4.2.27). Wernicke's area is the innermost part of the internal capsule. This is where visual radiance begins. Optic radiation fibers are located on the dorsolateral side of the lateral geniculate nucleus, while auditory tract fibers are located on the dorsomedial side.