- 1 Introduction
- 2 Purpose
- 3 Retina
- 4 Lateral Geniculate Nucleus
- 5 Visual Cortex
- 6 Color Vision
- 7 Visual pathway
- 8 References
I initially intended to create this as an entirely new report, but seeing as how report 2 was very detailed I decided to base this one on it. Notable changes are more detailed analysis of the retinal circuitry, as well as a much more thorough understanding of the LGN.
To gain a better understanding of the LGN which will ultimately be my area of specialization. I am currently considering investigating the temporal encoding algorithms of the LGN. I would need to explain why P cells and M cells would have to be independently dealt with. In order to be able to do that, I need a more thorough understanding of what the retina does. It has long been known that the center surround inhibition, via the horiziontal cells, increases spatial resolution. However, with the discovery of X and Y cells in the retina, what purpose would the LGN serve if its input is already temporally encoded? Why doesn't the retina just bypass it and go straight to the visual cortex? The fact that the visual cortex gives feedback to the LGN implies that the LGN serves some purpose besides repeating the input.
The retina is primary location of light processing by the rods and the cones. It has 6 groups of neurons, roughly 55 types of neurons and 10 layers, from outermost to innermost, the important 3 are:
- Photoreceptor layer - Rods / Cones
- Inner nuclear layer Bipolar Cells,Horizontal Cells,Amacrine Cells
- Ganglion cell layer - Ganglion Cells (gives rise to optic nerve fibers).
Although all three layers are composed of neurons, only the amacrine cells and the ganglion cells fire action potentials, the photoreceptors, bipolar cells and horizontal cells generate local graded potentials. The process of generating graded potentials will be discussed in the next section
|Cell||Percentage (of entire cells)||Number|
The retina turns light into discrete signals and this process is dictated by two main requirements
- a large dynamic operating range
- reduction of data to meet the transmission requirements of the optic nerve
The photoreceptor has three segments: an outer segment, and inner segment, and a synaptic ending.
- The outer segment absorbs light by visual pigments which are arranged into disks.
- The inner segment contains the nucleus, ion pumps, transporters, ribosomes, mitochondria, and endoplasmic reticulum
- the synaptic terminal releases glutamate and recieves synaptic inputs.
The process whereby light shining onto the retina is used to change a membrane potential, and generates local graded potentials, is called phototransduction and it proceeds as follows
- Inactivated (in the dark) sodium ions move into the cell and depolarize it to about -40 mV (from -65 mV)
- In the presence of light, opsin on the outer segment of the photoreceptor absorbs a photon
- This leads to several intermediary steps, and the sodium gates are closed
- In the absence of sodium the photoreceptor hyperpolarizess
- Hyperpolarization ensures less relase of neurotransmitter glutamate
The inhibition of glutamate, which can either excite or inhibit postsynaptic bipolar cells, ensures that there are now a pool of bipolar cells that are either hyperpolarized or depolarized. It is interesting to note that the presence of light actually reduces the photoreceptors response rate. Also as a note, photoreceptors do not signal colour or light intensity, only the presence of light.
Cones are primarily found in the centre of the retina (fovea) and act to distinguish light and other features present under normal lighting conditions. As would be expected, there are three different types of cones; short cones (S) react best to blue-violetish colours, medium cones (M) react best to bluish-green colours, and long cones (L) react best to yellow-greenish colours. The S cones are unique in that they are absent from the center of the fovea (foveola), and their optimal wavelength is much different than the the L or M cones. This is probably because that the L and M cones arrived later on in the evolutionary cycle, long after the rods and the S cones. The cones synapse directly onto bipolar cells, forming the bipolar cell RF center, and/or horizontal cells, which give inhibitory feedback and initiate center surround inhibition. The L and M cones are unique in that their optimal wavelengths are so similar. Cones are also connected to other cones (and rods) through gap junctions which serve to provide feedback to cones. At high illumination cone coupling is very low, increasing only with eccentricity. With low illumination cone coupling increases, indicating that coupling increases with both illumination and eccentricity. This makes sense since at low illumination levels the SNR would be very low, and spatial acuity would be traded for contrast sensitivity. Mechanisms such as horizontal cell feedback and gap junction input is a means of regulating local variations, this is a direct consequence of the distibuted processing of the mamallian retina.
- also information from one cone type is subtracted from another to remove redundancy. Apparently the bistratified ganglion cell does this.
Rods are found in the periphery of the retina and are primarily used at night, they outnumber cones twenty to one and function independently of cones. They are most sensitive to the blue end of the spectrum, however, because of the circuitry of the retina, they never engage in any spectral opponency and therefore do not contribute to color vision. They exist in only one form, they connect only to the ON rod bipolar cell, which connects to the AII amacrince cell. The AII cell takes the ON signal and transmits it to both the ON and OFF channels of the cone circuitry. This is due to the fact that they formed after cones did (in the evolutionary timescale), therefore, they piggy-backed on the existing cone circuitry. Multiple rods connect to only one bipolar cell increasing the overall sensitivity, however at the expense resolution. Rods also work a lot more slowly, taking on the order of 100 msecs to accumulate the light it needs, therefore they are not well suited for high temporal frequency scenes.
one means whereby the rods reach the parvocellular layer is as follows
rod bipolar cellAII cellflat midget bipolar cellOFF-midget ganglion cell
also the parasol cells (magnocellular layer) receives a substantial rod input. It has been proposed that the gap junctions from rods to cones is the pathway whereby rod information is transfered to the magnocellular layer
In the Parvocellular layer, the rod input always provides synergistic input of the same polarity as the center type.
- +L-M would recieve excitatory rod center (inhibitory rod surround)
- +M-L would recieve excitatory rod center (inhibitory rod surround)
- -L+M would recieve inhibitory rod center (excitatory rod surround)
- -M+L would recieve inhibitory rod center (excitatory rod surround)
- Why do rods connect to cones via gap junctions
Amacrine, Horizontal and Bipolar Cells
Although the connections in the retina are complicated, they follow this general layout:
- Photoreceptors connect to bipolar cells
- Horizontal cells connect photoreceptors together
- Bipolar cells connect to ganglion cells
- Amacrine cells connect bipolar cells together
- Ganglion cells converge to form the optic nerve
Bipolar cells receive their input from either rods or cones, but not both, and they are called rod bipolar cells, or cone bipolar cells, respectively. The photoreceptors, on the other hand, are not dedicated to any single bipolar cell, they are instead tapped by a large number of bipolar cells. The mamallian retina has about 12 (Masland) different types of cone bipolar cells. OFF or Hyperpolarizing bipolar cells (ones which hyperpolarize with reduction of glutamate) have an off-centre receptive field (since shining a light on its receptive field causes it to hyperpolarize) and they always connect to always connect to OFF ganglion cells. ON or Depolarizing bipolar cells (ones which depolarize with reduction of glutamate) have an on-centre receptive field (since shining a light on its receptive field causes it to depolarize) and they always connect to always connect to ON ganglion cells. When considering the temporal domain, one could look at ON cells only encoding transient increases in light intensity, and the OFF cells as only encoding transient decreases in light intensity. Direct connections from the receptors results in a hyperpolarized bipolar cell, whereas connections from horrizontal cells results in depolarized bipolar cells. Bipolar cells can also be designated Midget or Diffuse. Diffuse bipolar cells make connections to multiple cones, whereas the midget bipolar cells connect to only one cone. By the this stage we begin to see the earliest stages of center-surround inhibition, brought on by horizontal cells (see below). Also, h bipolar cells and d bipolar cells are about equal in numbers, and, like amacrine cells, outnumber ganglion cells five to one.
Diffuse bipolar cells receive inputs from both M and L cones indiscriminately, and as of yet, no significant S cone input has been observed.
Receptive first start to form at this cell level, with photorecptor input forming the center and horizontal cell layer forming the surround. The bipolar cell essentially takes a difference of these two inputs, thus removing th 'dc' component. This ensures that the bipolar cell is unnaffected by chages in global illumination, since both the center and the surround will be affected equally.
Thes cells essentially code spatiotemporal contrast in the input signal
- Do these ON/OFF cells temporally decorrelate the signal?
Amacrine cells make inhibitory synapses on the axon terminals of bipolar cells, thus controlling their output to ganglion cells. Furthermore, amacrine cells outnumber horizontal cells four to ten times over, they make up 40 percent of all neurons in the inner nucleus layer and they display the most diversity of the retinal cells, existing in around 30 forms (Masland). Amacrine cells also seem to account for the corelated firing amongst ganglion cells, and it has been suggested (Masland) that the amacrine cells are responsible for distinguishing object motion from background motion. Amacrine cell types play much more specific roles then horizontal cells (the other intermediary retinal cell) since they connect certain types of bipolar cells (for example the ON signal of a rod H bipolar cell) to a certain type of ganglion cell. They can be calssified into two groups, based on whether they fire action potentials, or generate local graded potentials. Until I know what these cells do, it'll be hard to figure out what the LGN does.
- what do these cells do?
Horizontal cells are interneurons that provide lateral inhibition, like amacrine cells, and they are connected to other horizontal cells via gap junctions. Horizontal cells are like H bipolar cells in that they hyperpolarize in the presence of glutamate, and they are the least diverse of the retinal cells existing in one of only 2 (Masland) forms; HI and HII. HI cells do not discriminate between S,M or L cones, whereas HII cells tend to give a large response to S cone stimulus, which is interesting since S cones account for only 10% of the total cone population. In sum, the horizontal cells are capable of selectively seeking out or avoiding S cells, however, they are indiscriminate towards M or L cells. In addition to this, all three inputs are hyperpolarizing, therefore the horizontal cells are incapable of performing spectral opponency. It is for this reason, that it is believed that spectral opponency is carried out by the amacrine cells.
When photoreceptors are illuminated they hyperpolarize and the horizontal cell reduces the release of GABA, which has an inhibitory affect on the photoreceptors. This reduction of inhibition leads to a depolarization of the photoreceptors. We therefore have the following negative feedback
Illuminationphotoreceptor hyperpolarizationhorizontal cell hyperpolarizationphotoreceptor depolarization
One proposed theory for facilitation by the horizontal cells proceeds as follows. Assume we have 11 photoreceptors, one H bipolar cell, and one horizontal cell. All ten photoreceptors connect to the horizontal cell, and the middle photoreceptor () connects to the bipolar cell. The surrounding cells, which represent the outer receptive field, will be designated Then we can explain an off-centre arrangement as follows. If light is shown onto the then
- is activated by light and therefore hyperpolarizes
- reduces release of glutamate
- Reduction of glutamate hyperpolarizes the H bipolar cell
- Reduction of glutamate hyperpolarizes the horizontal cell and it reduces release of GABA
- Since is still releasing glutamate, reduction in GABA is marginal
If the light is shown onto the surrounding area then
- is activated and therefore hyperpolarizes
- reduce release of glutamate
- Reduction of glutamate hyperpolarizes the horizontal cell
- Horizontal cell reduces release of GABA
- Reduction of GABA depolarizes photoreceptors
- not affected since they are strongly being hyperpolarized by activation
- is affected and therefore depolarizes
- releases glutamate
- H Bipolar cell is depolarized
To explain diffuse light, then we consider both cases together, and as it turns out, the two effects cancel each other out, and we get little to no affects. This fits nicely with the model put forth in From Neuron to Brain whereby the ganglion cells respond to a center surround inhibition. The bipolar cell recieves center input from the cones and surround input from the horizontal cells. Just like cones, horizontal cells reduce their receptive fields with increasing illumination. Interpllexiform cells give feedback to the horizontal cells which regulate the size of their receptive fields by modifying the conductivity of their gap junction. Horizontal cells generally have time constants of 2-3 times longer than cones.
- what role does this difference in time constant have on the RF of bipolar cells.
These cells recieve input from the bipolar cells and connect to horizontal cells. They are believed to regulate the receptive fields of horizontal cells by releasing dopamine which reduces the conductivity of the gap junctions in the horizontal cells, which as described earlier, reduces the cell's RF. Each interplexiform cell collects inputs from several bipolar cells over its large receptive field, and since bipolar cells code spatiotemporal contrast, the IPX cell is sensitive to both high spatial contrast (edges) and high temporal contrast (motion, flickering light). When stimulated under these conditions they will reduce horizontal cell RFs, therefore, in regions of high spatiotemporal contrast the horizontal cell RF is diminished by the IPX cell feedback.
The last layer in the retina, which collects to form the optic nerve is the ganglion cell layer. There are about 12 (Masland) different types of ganglion cells, and a total of 100 million photoreceptors converge to only 1 million of them, therefore, a good deal of encoding has taken place. When one looks at the receptive field of a ganglion cell, the area of the retina from which a ganglion cell’s activity can be influenced by light, it becomes apparent that it is more sensitive to differences within the receptive field than light intensity. On-centre ganglion cells will become excited if light is shone onto the centre of their receptive fields, and likewise off-centre ganglion cells will become excited when light is shone onto the outside of their receptive fields.
Because of the nature of bipolar cells connecting to ganglion cells of similar affinity,a change in membrane potential of the bipolar cell, causes a change in the membrane potential of its ganglion cell in the same direction. Surprisingly, rods and cones of the same area of the retina, supply the same ganglion cell but by different means. The rods indirectly connect via the rod bipolar cell-AII circuitry which finally connects to the ganglion cells. Although all ganglion cells recieve input from cone bipolar cells, most of their inputs come from amacrine cells (about 70%)
The ganglion cells can be loosely grouped into P,M and K categories, which gives rise to the parallel processing in the visual pathway. The P ganglion cells project to the four dorsal layers of the LGN, and they have small receptive field centres, high spatial resolution (higher detail, probably recieving inputs from the densley packed fovea), are sensitive to colour, and provide information about fine detail at high contrast. The M ganglion cells project to the larger cells in the two ventral layers of the LGN, have larger receptive fields, are more sensitive to small differences in contrast and to movement, they fire at higher frequencies (have higher temporal resolution), and conduct impulses more rapidly along their larger diameter axons. The final K layer makes up only 10% of output and carries the blue-yellow opponent pathway.
Just like the bipolar cells they do not convey absolute levels of illumination, because they behave the same at different background levels of light, they just relay the spatially opponent signal of the bipolar cells, probably taking a few bipolar cells and combining their signals to achieve stronger center surround inhibition.
Midget Ganglion cells are believed to form the red-green opponent pathway.
Bistratified ganglion cells are believed to form the blue ON-yellow OFF pathway. However, in this case, it is not a center-surround organized, just an overlap of the two.
Ganglion cells can also be classified based on their response to the duration of stimulation. The X (sustained) cells reach a steady firing rate with continued excitation. The Y (transient) cells only respond to the onset of a stimulus, then return to their resting firing rate.
The receptive fields are smallest/most percise at the fovea and largest at the periphery.
- The ganglion cells don't do much
- Do they temporally decorrolate the signal?
Surround inhibition starts at the bipolar level, and is refined at the ganglion level. At the horizontal cell level, inhibition of the photoreceptors removes redundency from the input by relating the intensity falling on the center of the receptive field to surrounding levels of illuminations. By generating a statistical estimate of the intensity expected at the center from the surrounding intensity, and subtracting it from the actual intensity at the center, the amplitude at the center is minimized. Lateral inhibition reduces the range of intensity that neurons need to encode, yet allows them to monitor small fluctuations. At the ganglion cell leve, center surround inhibition takes advantage of the predictibility of natural scenes, and looks for changes in illumination.
Bipolar cells receive the center input from photoreceptors, and the surround input from horizontal cells, and esentially do a difference operation of the two. This scheme strips the 'dc' background intensity from the signal and allows the bipolar cell to be sensitive to local variations in contast.
The retina is concerned with increasing efficiency, not only as an end in itself, but to meet the requirements of the limited capacity optic nerve. The foveated sampling scheme, whereby the diamaters of receptive fields of retinal ganglion cells scale linearly with eccentricity ensures the highest acuity at the center of the fovea (foveola), where one sees lower convergance, and thus a larger bandwidth requirement. Moving out toward the periphery photoreceptors are more sparsely packed, there is increasing convergance with eccentricity, and a lower bandwidth requirement. The photoreceptors themeselves locally adapt their sensitivity to light by 'washing out' their photo-pigments through pigment bleaching, increasing illumination causes the pigments to be washed out of the photoreceptors, limiting their photon catching abilities. Network feedback from horizontal cells, feedback from neighbouring photoreceptors, and chemical processes ensures that the photoreceptor response range is centered around a local spatiotemporal ambient intensity. Even the surround inhibition is an encoding scheme which reduces spatial correlation. The retina also adapts the receptive fields (of horizontal cells and therefore with it, that of bipolar and ganglion cells) with changing illumination levels to ensure a good balance between visual acuity and contrast sensitivity. As illumination levels drop so does the signal to noise ratio, the retina compensates for this by increasing the receptive field which increases the contrast sensitivity at the expense of spatial resolution. When the signal to noise ratio increases with increasing illumination, the receptive field size is decreased to attain higher spatial resolution.
The retinal processing can be summed up as follows
- Foveated sampling scheme and greater convergence of cone inputs into ganglion cell outputs with increasing eccentricity reduces bandwidth.
- Photoreceptor mechanisms and network feedback (horizontal cells, rods, and cones) locally adapt the sensitivity of photoreceptors to ensure response range is centered around a local spatiotemporal ambient intensity
- Surround inhibition removes the 'dc' coponent and allows bipolar cells to be sensitive to the local variations in contrast
- The retina adapts the size of receptive fields as a function of illumination to maintain a good balance between high visual acuity and contrast detectability.
- Interplexiform layers regulates surround receptive field in areas of high spatiotemporal contrast, which sharpens spatial edges, and improves sensitivity to temporal changes.
- There are three channels of parallel processing, Parvocellular, Magnocellular and Kinniocellular.
Lateral Geniculate Nucleus
The optic nerve projects onto a structure called the Lateral Geniculate Nucleus. The LGN is structured such that it has a left and a right hemisphere, within which there are six distinct layers. Layers 2, 3 and 5 receive their inputs from the ipsilateral eye (w.r.t the left or right hemisphere) whereas layers 1, 4 and 6 receive their inputs from the contralateral eye (w.r.t. the left or the right hemisphere). There are roughly 1 million cells in the LGN, however, the optic nerve fibres diverge to connect to multiple LGN cells, as opposed to a simple one to one mapping. Not only are the cells topographically ordered (neighbouring ganglion cells project to neighbouring geniculate cells) but they are retinotopically registered along the different levels. Therefore, if a sample rod is taken tengentially from layer 1 to layer 6, we will hit the same receptive fields, alternating between one eye, and the other. Suprisingly, retinal input accounts for only 30% of the input to the LGN, with the great majority coming from V1. Historically the layers are numbered from the bottom up, and layers 1 and 2 have the larger, and much faster cells.
An interesting observation pointed out by Blitz is that ganglion cells which synapse onto interneurons in the LGN cause inhibition that spreads onto thalama-cortical (TC) neurons beyond those connected to it. Perhaps I can propose a model for the K layers in the LGN
The total number of cells recieving inputs from the contralateral eye exceeds that from the ipsilateral eye. This is because the contralateral eye projects to layers 4 and 6 and therefore is larger than layers 3 and 5. Also in the magnocellular layer the contralateral eye projects to layer 1 which is smaller, but still much thicker than layer 2.
The receptive fields of adjacent neurons overlap since neighbouring regions of the retina make connections with neighbouring geniculate cells. This overlap means that the receptive field of the LGN is even more scrutinous when it comes to diffuse light than is the ganglion receptive field. This is because the contrast mechanism is more finely tuned by more equal matching of inhibitory and excitatory areas. Therefore, whereas the ganglion receptive field does somewhat respond to diffuse light, the LGN receptive field is very poor at responding to diffuse light. There is almost no divergence on the part of the incoming retinal cells, therefore, we can conclude that little to no spatial decorrelation takes place. In fact there is a great deal of convergence, with a great deal of retinal fibers converging on one geniculate cell. Each geniculate cell center receives multiple retinal fibers from either all OFF or all ON cells, and its surround recieves the opposit retinal input (all ON or all OFF respectively).
Roughly 30% of the input to the LGN comes from the RGCs. Of these, 80% are Midget Ganlgion cells, 10% are Parasol Ganglion cells, and 10% are koniocellular cells. The other 70% comes from the primary visual cortex (V1)
Output leaves the LGN through the dorsal lateral geniculate nucleus, and projects mainly to the primary visual cortex.
These M cells, in layers 1 and 2, work much faster, however, they do not process as much information (higher temporal resolution, lower spatial resolution). The M cells are the largest, and they are the most ventral, therefore, they are closest to the optic nerve. They recieve their input from the Parasol ganglion cells, which have lower spatial frequency selectivity, and higher contrast sensitivity (contrast sensitivity= 1/contrast required to evoke a response). These cells in turn get their input from the diffuse bipolar cells, which recieve their input from L and M cones. They disply center surround inhibition, where an ON center receptive field has L+M+ in its center and an L-M- in its surround, and vice versa for the OFF center receptive field. It is clearly achromatic. M cells respond only transiently to a maintained stimulus (they respond best to modulations at 20 Hz or greater, and continue to respond up to 60-80 Hz), and amacrine cells are suspected to be the major source of this temporal differentiation.
The P cells, in layers 3 to 6, are much smaller and slower, but come with the added advantage that they can do complex calculations such as colour detection (lower temporal resolution, higher spatial resolution). They recieve their input from the Midget ganglion cells, which have higher spatial frequency selectivity, and lower contrast sensitivity. These in turn get their input from the midget bipolar cells which receive their input from individual L or M cones. Most data indicates that the center and the surround are composed of different cones. These P cells generally give sustained responses to a maintained stimulus (they respond best to patterns modulated around 10 Hz, and can't respond above 20-30 Hz).
Between each of the M and P layers lies a zone of very small cells, the K cells. S cones project to special 'blue cone' bipolar cells, which in turn provide input to small bistratified ganglion cells, which project to the K layers. The ganglion cells also recieve input from the diffuse bipolar cells (L+M+) which forms the chromatically opponent S-(L+M+) cells. The L+M signal is collected from the DB2 and DB3 diffuse bipolar cell to the bistratified ganglion cell
It has been pointed out by Dong&Atick that the retina only removes spatial correlation from the visual input, and delivers an encoded signal which is temporaly redundant. They propose that the LGN's role may be to reduce temporal correlation and complete the spatial-temporal decorrelation process
Lagged and Non-Lagged Cells
The geniculate cells can be classified, based on their responses, into two groups, and, based on their input, further into two groups, for a total of four groups. The input arriving from the geniculate cells can be designated one of two spatial categories, 'on' or 'off'. The output can be designated one of two temporal categories, 'lagged' or 'non-lagged'. Generally geniculate input is dominated by one category, therefore, we need not concern ourselves with a third case where both on and off input compete. The on cells recieve their input from on-center cells, and likewise for the off cells. The Lagged cells represent the temporal group which exhibits a decrease in luminosity over time, and the non-lagged cells represent the temporal group which exhibits an increase in luminosity over time. We therefore get the following four groups:
- On-Lagged: a decrement of light and on-center
- On-Non-Lagged: an increment of light and on-center
- Off-Lagged: a decrement of light and off-center
- Off-Non-Lagged: an increment of light and off-center
Burst vs Tonic spikes
Burst spikes have shorter latencies between stimulus and response (they are activated more quickly). They have smaller receptive field centers.
X vs Y cells
X cells respond to contrast reversal of fine sine gratings at the fundamental temporal modulation frequency with the null positions one quarter cycle away from positions for peak response?!?!? Y cells on the other hand are at twice the modulation frequency away and approximately independent of spatial phase?!?!? X cells are more linear and Y cells are more non linear. I think Y cells are transient, whereas X cells are sustained.
- What role does separating the K, M and P layers present? Is it to optimize temporal reduction?
- Are the X and Y cells just the P and M cells?
- What's white noise?
- what does it mean to whiten something?
- what's spatial/temporal noise?
- Why is it efficient to have things as spread out as possible (decorrelated)
- how are neurons in the LGN perfectly alligned?
- What role does cortical feedback play?
- How can a Type II cell have a center?
The LGN eventually projects onto one of the six layers of the primary visual cortex (aka V1), but mostly onto layer 4. From here information is passed onto V2, V4 and MT, and finally Inferotemporal and Parietal 7. It is important to note that both hemispheres of the visual cortex are symmetric, and they simply receive their inputs from the respective LGN hemisphere. As is the case for the other areas of the cortex, the visual cortex has 6 distinct layers. The visual cortex is much more complicated than the LGN or the retina because it has many more cells, many more horizontal connections, and although it has all of its inputs feeding in from the LGN, it is difficult to predict where that information streams out to, therefore it has, as present, an unknown bandwidth. It is known, however, that only twenty percent of the neurons are intrinsic (accept incoming info?), the other eighty percent project to other parts of the brain, possibly within the neocortex. Conservative estimates place the number of different types of neurons in the cortex at around 1000 (Masland)
Simple Cells & Complex Cells
There are three subgroups of cells within the cortex: Simple cells, complex cells, and complex cells with end inhibition. The simple cells respond only to a bar of a certain size, and oriented at a certain angle. The complex cells on the other hand can detect edges of any size and/or orientation. Complex cells can also detect corners, in which case we label them with end inhibition. It may be the case that an array of simple cells could connect to a complex cell to give rise to its behaviour. To illustrate this point, consider the case where a complex cell has a receptive field of that of 100 simple cells (10x10 plane) and for each 'cell' it has four simple cells oriented at 90 degrees to each other. Then with 400 simple cells, the complex cells can react to a whole host of bars, in one of four positions, anywhere along its receptive field. If we consider a complex cell with 3600 simple cells, then it can distinguish bars of any orientation every ten degrees. Throughout the entire mammalian visual stream, one will find that there is a great deal of hierarchies and precise stratification, the visual cortex is no exception.
Ocular Dominance Columns
An interesting formation within the visual cortex is that of Ocular Dominance Columns (ODCs). Horizontally along the layers of the cortex, there exist strips which are entirely controlled by one eye or the other, but not both. It has been suggested that activity dependent competition causes the formation of ocular dominance columns. The columns are 250 to 500 wide, and taking the soma to be from 4 to 100 wide, that means that every column has about 5 to 75 neurons.
Not suprisingly, if one samples the cortex perpindicular to its surface, one we encounter similar orientation preferences amongst the neurons. The columns here are much more percise than ODCs as these are on average only 20 to 50 wide, which means that there are less than a dozen neurons in each column. When the a map of the OCs are superimposed with a map of the ODCs it can be seen that where the ODCs meet, OCs cross perpindicular to the two ODCs. This makes sense since the neurons for both eyes in that area should have the same orientation preference.
Color perception occurs by comparing two (or more) different types of cones. Within the retina the S,M, and L cones combine antagonistically to form red-green and blue-yellow spectral opponent pathways. Within the red-green pathway signals from L and M cones are opposed. Within the blue-yellow pathway signals from S cones oppose combined signals from L and M cones. It has been suggested that color coding could piggyback the antagonistic center surround organization that exists in ganglion cells. One way to do this would be to have the center be excitatory towards one cone (say L) and the surround be inhibitory to another cone (say M); we then get the situation whereby the center responds to red, and the surround responds to anything besides green. Therefore, the signal generated would be greatest to a red spot on a blue (or neutral) background.
- S cones contact either a single type of On (blue) bipolar cell or Horizontal cells, getting more attention from the HII cell.
- L cones contact Off bipolar cells(diffuse bipolar cells) or horizontal cells
- M cones contact Off bipolar cells(diffuse bipolar cells) or horizontal cells
- Rods connect to a single type of ON bipolar cell, and then to the AII amacrine cell, which forms an inhibitory connection with the cone-hyperbolarizing bipolar cell, and an excitatory connection with the cone-depolarizing bipolar cell, thus creating the parallel pathways for the rods. Rods can also connect to one of the horizontal cells, however, their input does not interfere with that of the cones. rodsrod bipolar cellsAII amacrine cells ON or OFF cone bipolar cellON or OFF ganglion cell
- Surround inhibition is the primary function of the retina, and it is the first step in the two step spatio-temporal decorrelation of incoming light. It starts in the bipolar cells, with cone input forming the center, and horizontal cell input forming the surround of the receptive field.
- The Magnocellular layers receive their input from the parasol ganglion cells (M cells). They receive their input from diffuse bipolar cells, which receive synergistic input from L and M cones. On center parasol cells receive excitation from L and M cones in its center, and inhibition in its surround, and vice versa for Off center parasol cells. Since the L cones are not opposing the M cones we observe no spectral opponency.
- The parvocellular layers recieve their input from the midget ganglion cells (P cells) which are the main (if not only) cells that form the red-green opponent pathway (L cones oppose M cones). The receptive field center is formed by either the L or the M cone, but not both, and the surround formed by the other (Type 1 cell), and rod input is always synergistic to the center, and inhibitory to the surround. This provides the basis for red-green spectral opponency. The input comes from midget bipolar cell which connects to no other cells, and recieves its input from one L or M cone. As one moves from the fovea to the periphery, the opponency degrades into an additive input from both L and M cones, and they become similar to the parasol cells. Therefore, from the periphery, the parasol does relay achromatic information to the parvocellular layer.
- The midget ganglion cells also could relay Blue-OFF/Yellow-ON signals since themidget bipolar cells which they receive input from in turn recieve an off signal from S cones, and surround inhibition from L and M cones via HII.
- The koniocellular layers receive their input from the bistratified ganlgion cells which form the Blue-ON/Yellow-OFF pathway (excitation from S cones combined with inhibition from L and M cones). These cells recieve input from the blue cone bipolar cells which synapse exclusively with S-cones, and also the diffuse bipolar cells carrying the synergistic LM signals. There is no classical center surround chromatic inhibition, only an overlapping signal (Type 2 cell). This is incredibly strange since the S cones formed long before the L or M cones.
- Since horizontal cells deal only with center surround inhibition, it has been suggested that the amacrine cells provide spectral opponency.
- Burst Mode vs Tonic Mode ?
- Lagged vs Non lagged?
Alitto, Henry J; TG Weyand, WM Usrey (2005). "Distinct Properties of Stimulus-Evoked Bursts in the Lateral Geniculate NUcleus". Journal of Neuroscience 25 (2): 514–523.
Bailey, Julian (2007). 9 Month Report.
Blitz, Dawn M; WG Regehr (2005). "Timing and Specificity of Feed-Forward Inhibition witih the LGN". Neuron 45: 917–928.
Bloomfield, Stewart A; RF Dacheux (2001). "Rod Vision: Pathways and Processing in the Mammalian Retina". Processes in Retinal and Eye Research 20 (3): 351–384.
Boycott, Brian; W Heinz (June 1999). "Parallel Proccessing in the Mammalian Retina". Ivestigative Opthalmology & Visual Science 40: 1313–1327.
Cai, Daqing; GC DeAngelis, RD Freeman (1997). "Spatiotemporal Receptive Field Organization in the Lateral Geniculate Nucleus of Cats and Kittens". American Physiological Society.
Chapman, Barbara; KR Zahs, MP Stryker (1991). "Relation of Cortical Cell Orientation Selectivity to Alignment of Receptive Fields of the Geniculocortical Afferents that Arborize within a Single Orientation Column in Ferret Visual Cortex". Journal of Neuroscience 11 (5): 1347–1358.
Connolly, Michael; DV Essen (1984). "The Representation of the Viual Field in Parvicellular and Magnocellular Layers of the Lateral Geniculate Nucleus in the Macaque Monkey". Journal of Comparative Neurology 226: 544–564.
Dacey, Dennis M (1996). "Circuitry for Color Coding in the Primate Retina". Proc. Natl. Acad. Sci 93: 582–588.
Dacey, Dennis M (2000). "Parallel Pathways for Spectral Coding in Primate Retina". Annu. Rev. Neurosci. 23: 743–775.
Dacey, Dennis M (1999). "Primate Retina: Cell Types, Circuits and Color Opponency". Progress in Retinal and Eye Research 18: 737–763.
Dan, Yang; JJ Atick, RC Reid (1996). "Efficient Coding of Natural Scenes in the Lateral Geniculate Nucleus: Experimental Test of a Computational Theory". Journal of Neuroscience 16 (10): 3351–3362.
De Valois, Russell L; NP Cottaris, SD Elfar, LE Mahon, JA Wilson (2000). "Some Transformations of Color Information from Lateral Geniculate Nucleus to Striate Cortex". PNAS 97 (9): 4997–5002.
DeVries, Steven H (2000). "Bipolar Cells Use Kainate and AMPA Receptors to Filter Visual Information into Separate Channels". Neuron 28: 847–856.
Dong, Dawei W; JJ Atick (1995). "Temporal Decorrelation: a Theory of Lagged and Nonlagged Responses in the Lateral Geniculate Nucleus". Computation in Neural Systems 6: 159–178.
Essen, David C Van; CH Anderson (1995). Information Processing Strategies and Pathways in the Primate Visual System.
Gegenfurtner, Karl R; DC Kiper (2003). "Color Vision". Annu. Rev. Neurosci 26: 181–206.
Grunert, Ulrike (1997). "Anatomical evidence for Rod Input to the Parvocellular Pathway in the Visual System of the Primate". European Journal of Neuroscience 9: 617–621.
Hammond, P (1972). "Chromatic Sensitivity and Spatial Organization of LGN Neuron Receptive Fields in Cat: Cone-Rod Interaction". Journal of Physiology 225: 391–413.
Ichida, Jennifer M; VA Casagrande (2002). "Organization of the Feedback Pathway from Striate Cortex (V1) to the Lateral Geniculate Nucleus (LGN) in the Owl Monkey". Comparative Neurology 454: 272–283.
Jones, Helen E; IM Andolina, NM Oakley, PC Murphy, AM Stillito (2000). Spatial Summation in the Lateral Geniculate Nucleus and Visual Cortex.
Kaplan, E; RM Shapley (1982). "X and Y Cells in the Lateral Geniculate Nucleus of Macaque Monkeys". Journal of Physiology 330: 125–143.
Lee, Barry B; VC Smith, J Pokorny, J Kremers (1997). "Rod Inputs to Macaque Ganglion Cells". Vision Res. 37 (20): 2813–2828.
Lee, JW; SP Chae, MN Kim, SY Kim, JH Cho (2001). "A Moving Detectable Retina Model Considering the Mechanism of an Amacrine Cell for Vision". ISIE: 106–109.
MacNeil, Margaret A; RH Masland (1998). "Extreme Diversity among Amacrine Cells: Implications for Function". Neuron 20: 971–982.
Masland, Richard H (1996). "Processing and Encoding of Visual Information in the Retina". Current Oppinion in Neurobiology 6: 467–474.
Masland, Richard H (2001). "The Fundamental Plan of the Retina". Nature Neuroscience 4 (9).
Masland, Richard H (23 May 2003). "The Retina's Fancy Tricks". Nature 423: 387–388.
Masland, Richard H (2004). "Neuronal Cell Types". Current Biology 14: R497–R500.
Masland, Richard H (2001). "Neuronal Diversity in the Retina". Current Oppinion in Neurobiology 11: 431–436.
Merigan, W H; JHR Maunsell (1993). "How Parallel are the Primate Visual Pathways?". Annu. Rev. Neurosci. 16: 369–402.
Reid, R Clay (2002). "Space and Time Maps of Cone Photoreceptor Signals in Mcaque Lateral Geniculate Nucleus". Journal of Neuroscience 22: 6158–6175.
Reinagel, Pamela (1999). "Encoding of Visual Information by LGN Bursts". Neurophysiology 81: 2558–2569.
Shah, Samir; MD Levine (1993). Information Processing in Primate Retinal Cone Pathways : A Model.
Soloman, Samuel G; AJR White, PR Martin (2002). "Extraclassical Receptive Field Properties of Parvocellular, Magnocellular, and Koniocellular Cells in the Primate Lateral Geniculate Nucleus". Journal of Neuroscience 22 (1): 338–349.
Soucy, Ed; Y wang, S Nirenberg, J Nathans, M Meister (1998). "A Novel Signaling Pathway from Rod Photoreceptors to Ganglion Cells in Mammalian Retina". Neuron 21: 481–493.
Srinivasan, MV; SB Laughlin, A Dubs (1982). "Predictive Coding: a Fresh View of Inhibition in the Retina". Proc. R. Soc. Lond. B216: 427–495.
Xu, Xiangmin; JM Ichida, JD Allison, JD Boyd, AB Bonds, VA Casagrande (2001). "A comparison of Koniocellular, Magnocellular, and Parvocellular Receptive Field Properties in the Lateral Geniculate Nucleus of the owl Monkey (Aotus Trivirgatus)". Journal of Physiology 531 (1): 203–218.