Neuroanatomy 1:00-2:00Scribe: Lauren Paul

Wednesday, January, 27, 2010Proof: Jennifer Grimes

Dr. KraftEarly Visual ProcessingPage1 of 7

LI-Lateral Inhibiton; AP=Action Potential; NT=Neurotransmitter

Retinal Physiology Vocabulary [S1]
<NO AUDIO> On and off bipolar cell—how do we take one sign and turn it into two signs
On board—brighter than its surrounding
On slide see negative contrast—things that are darker than surrounding
Positive and negative contrast are the 2 things vision uses to see things. They are really important
<START AUDIO> How do you encode positive and negative contrast? That’s something I want you to take away from this lecture.
Another thing to take away is that in the retina we will be processing images that pay attention to a direct pathway and a surround. In other words, if I’m looking directly at this person studying hard and I can tell that he’s getting a high score on some game he’s playing, I’m also interested in the all the people around him. So the retina is going to process things in a local and a global sense. In other words, center and surround. cross to a direct process and what’s around. Retina processes in a local vs. global sense. Sensor and surround.
If I were to take the illumination in this room and adjust it up by a factor of 100 over the period of this lecture, you wouldn’t notice. If I were to adjust it up or down by 5% in 10 ms, you would notice right away.
The retina is interested in change. That change could be light gray to dark gray level of one computer next to another in space or it could be a change in time. When you’re looking at one object and the illumination from that object or the surround changes over a period of a time.
We’re going to talk about spatial summation and temporal summation. We’re going to talk about the idea of divergent and convergent wiring—convergent wiring means in order to increase sensitivity we pool information. If I need to know how well I’m teaching, I don’t just listen to the response of my favorite students but of all the students. I will get an average. That will give me a sensitive indication of how well the class is going. In the retina, pooling info gives you a more sensitive indicator of what’s going in terms of light levels.
Divergent wiring—in one point in space (remember we have a mobile camera) I need to examine light, dark, light coming on and going off, whether it’s blue or yellow for color information processing, temporal information to see if the object is moving. All of those signals are being processed for each location in the retina.
That means the output for primary sensors at that location in the retina are going to diverge and go out into circuits that compute many different properties—color, changes over time, light and dark. <GOES TO SLIDE 3>
Lateral Inhibition: What does it do for you? [S2]
Now what it says, if you have primary receptors, those indicate pressure so the same circuit exists in the sensory system in your skin. In the retina, we call them photoreceptors. There’s a light hitting photoreceptors B and D and it generates a signal. But in the second order neurons, there is some comparison going on.
So receptor A generates a signal which feeds through an interneuron and an inverted synapse onto photoreceptor B. That’s called lateral inhibition. And that happens in the retina with just a single interneuron called the horizontal cell.
A practical example of this would be if this young lady here gets a 90 on her test, she’s going to be really happy. But if she could take that 90 and give an inhibitory signal of 20% to all the people around her, that means that 20% would take away 18 points from everyone else’s tests. It would make her very unpopular but it would enhance the appearance of her signal.
So what that lateral inhibition of the neighboring cells does is enhance the appearance of the signal in the center.
What you can see in the white graph below is the stimulus of the primary receptor. A is receiving 10 units of excitation, B-50, C-25, D-50, and E-10. And you can see that there’s a difference but lateral inhibition does two things.It reduces the overall level of excitation. You can see that the gray bars which show you the affect of LI are all depressed. Everything has gone down a bit. If you spread negative signals, it causes general, overall depression.
Important thing is that the ratio of the stimulus to the non-stimulus stimulated cells is greatly enhanced. Before it was 50:25, a ratio of 2:1. After LI, a simultaneous feedback of -20% of every cell to its neighboring two cells, the ratio is now 8:1. We have a signal of 43 in the second order neuron to 5 in the central neuron.
The ratio has been enhanced, or the difference of these 2 excitations has been enhanced, from 2:1 to 8:1. The retina is not interested in absolute levels of illumination but in contrast (changes over time and space). So what this LI has done, it has taken the signal and created a very, very large difference in the adjacent stimulus sum. So we now have an 8:1 ratio. So if in your brain you had to determine if there were 2 points of a light or one larger spot of light, and it required a ratio of 4:1 or greater, without LI you would see this as a large spot covering this area. With LI, the same exact raw stimulus hitting the same number of density photoreceptors would now be perceived as two spots of light in two different locations.
So you can see how the neural processing has enhanced the spatial discrmination of the retina. A similar thing could be done in time but is hard to demonstrate.
So this is lateral inhibition and very important for center surround organization. <GOES BACK TO SLIDE 3>
Mach Bands[S3]
Mach bands—if you look closely at the edges of these 13 bars, the edge at the right hand side appears a little darker than the mean level of the rest of the bar. And the edge immediately adjacent to it appears a little bit lighter than the rest of that. If I were to take a scanner that tells me exactly what the light intensity is, it would be exactly square steps but your vision system doesn’t process it the same way.
We have something here called edge enhancement—the neural signal takes average information and subtracts it from the adjacent location. What that means is you have something called lateral inhibition. In order to get information out of the retina and into the CNS, you have to do a couple of different things. One, you want to get the information from the photoreceptor; there’s a direct pathway: photoreceptor-bipolar cell-ganglion cell. The ganglion cell has an axon, it goes to the brain. That’s the shortest, quickest pathway to get information out of the retina to the brain.
A lot of computation occurs in the retina. A lot of computation is comparison of lateral or surround effects on the center position. This is a demonstration of lateral inhibition and of retinal wiring which is shown in the previous slide. <GO BACK TO SLIDE 2>
<PICKING UP AGAIN FROM SLIDE 2> So you can imagine that if I have a spot, a cell, which is right here, it’s receiving input form the light bouncing off the slide and comparing it to the light immediately around it. But they are all equal. So subtraction of the surround from the center stimulus gives you essentially a 0 output.
The retina is not interested in that uniformity. At the edge, the cell here is stimulated greater and has more of an inhibitory effect on the cell which is there. So the perception is that it’s darker, darker at the left edge than here because of that shared inhibition. If that doesn’t quite make sense, work it out with math on that slide I showed you before. You can see by placing the receptors across this type of illumination what you would get.
What you should see (goes to board) is that if the light level is changing in a step-wise fashion, the perception is showing an enhancement of the edge. The ratio of the perception is greater than the absolute change of illumination. That’s accomplished by lateral interaction of nerve cells in the retina. So let’s take a look at the retina.
Tartuferi 1887 [S4]
Picture of retina by Italian anatomist about 120 years ago.
The retina has a photoreceptor layer. There are 3 layers of nuclei in the retina. One is made of photoreceptors (outermost).
Sclera, choroid (blood supply) RPE, photoreceptors (outer layer of nuclei).
There’s a plexiform (synaptic region) between the photoreceptors and the 2nd order neurons.
The 2nd motor neurons are bipolar and horizontal cells. The bipolar and horizontal cell nuclei reside in the inner nuclear layer, inner because the inside of the eye is in this direction. The outer nuclear layer is next to the sclera and the inner nuclear layer is close to the inside of the eye, which, of course, is filled with jelly called vitreous. So this is sometimes called the inner retina or the vitreal portion of the retina.
Second synaptic region called inner plexiform layer—exists where bipolar cells make synaptic contact with the ganglion cells and amacrine cells.
5 basic neural cell types exist in the retina.
Photoreceptor, bipolar, and ganglion cells—this is the direct output pathway. They are vertically arranged in this slide.
2 horizontal type cells—make lateral connections. Those that would be responsible for generating the LI we just talked about. Those are horizontal cells in the synapse in the outer plexiform layer and then a cell called amacrine cells which make the same type of connections in the inner plexiform layer.
Key points to remember[S5]
Key concepts about the retina
The visual system covers an incredible range of orders of magnitude of light levels. Approx 1010, that is 10 orders of magnitude of light levels over which the retina can see changes in light and respond to those.
Part of that is covered by the factin our photoreceptor layer, there are 2 types of photoreceptors:
Rod cells—specialized for very dim light conditions. They cover the lowest 4 orders of magnitude.
Cone cells—don’t respond well to dim lighting condition. You can’t use them in a moonlight setting. They take over at middle ranges of light levels and have a tremendous ability to adapt. They cover the top 6 orders of light. We’re all using our cones right now. I know this because I can see in color. They come in 3 different colors. Some sample the red end of the spectrum and some sample the blue end.
We get color vision by comparing the relative excitation in one locale of the retina to the red, blue, and green photoreceptors.
So the direct pathway (as stated previously) is from the photoreceptor to a bipolar cell to a ganglion cell.
Ganglion cells are spiking neurons in the retina. They generate action potentials, which travel the vast distance of about 25mm back to the brain.
Horizontal cells and amarcine cells—make up lateral connections and are responsible for LI. And something I want to talk a lot about and that is the center-surround organization.
Key points to remember [S6]
We’ve talked a lot about the retina being interested in light and dark edges. In order to do that, there is actually specialization where one pathway takes increments of light, called the “on pathway” and whenthe light changes from darker to brighter, it generates a burst of APs that says “HEY!” the light just came on in this location.
At that exact same location, there is a duplicate circuit which takes a signal and inverts it. So when the light goes out at that physical space, at that photoreceptor, a burst of APs goes to the brain saying, “A light just went off right there.” So there’s an on and an off pathway. Both of which send APs. They are wired differently though.
Key feature of retinal output is the center-surround receptive field. It exists in the bipolar cells and the ganglion cells.
Some of our perceptual limits are set at the retina itself, and some are higher up.
Retina anatomy—overview [S7]
Cartoon of retina. Again, ganglion cells are up. Light comes through the ganglion cells, through the inner plexiform layer, through the inner nuclear layer, through the outer plexiform layer, through the outer nuclear layer, then finally to that portion of the photoreceptors that actually contains the pigments that absorbs the light.
Excess light which doesn’t get absorbed is generally absorbed by the retinal pigment epithelium, which is why when you look through someone’s pupils, it’s black. That’s the pigment of the back of the eye. The entire retina has to be transparent in order for the photons to get through.Don’t want light scattering around in the back of your eye because it will cause confusion as to where the light came from.
When you are driving at night and come across deer or opossum, and you get that bright reflection from their eyes that’s because in their retinas they have mirror-like crystals that bounce light back to the retina. It’s great for dim-light reception because lights bounce back to the retina but very poor for resolution because you have a lot of scattered light coming back there. But that’s the reflection you get from nocturnal animals.
NO TITLE [S8]
This is histological section of the retina on the left showing you an actual view of primary photoreceptors, you can see cone cells (larger in diameter) outer segments that contain visual pigment, and the retina pigment epithelium behing them.
Outer nuclear layer, outer plexiform layer, inner nuclear layer, inner plexiform layer, ganglion cells, and nerve fibers heading over to the optic nerve.
Flip it around so you can read the labels in the next slide
NO TITLE [S9]
Here we see this is the more physiological type of presentation you’ll see in books.
Here is a single cone photoreceptor, which is sending it’s excitor output to two different neurons. So that’s representative of the primary on and off pathways I discussed earlier.
At the synapse, a photoreceptor signals to light which is a hyperpolarization. Curtis—in your general knowledge of neurophysiology when a nerve cell is excited, what does a membrane potential do? Answer: depolarizes. Right, and that depolarization spreads to the next cell via a synaptic connection.
The excitatory signal for photoreceptor is darkness—it depolarizes photoreceptors.
When the light comes on, the absence of darkness generates a hyperpolarization. It’s that that spreads and changes the NT release.
When a cell hyperpolarizes, what happens to the rate of NT release? It slows down or it goes down. So, the hyperpolarization produces a decrease in the NT release. Perhaps contrary to what you normally would think, when a light comes on in a particular area of the retina, the photoreceptor hyperpolarizes and the rate of NT release goes down. That is the signal that the light has arrived.
If you just think about that darkness deploraizes the cell, it will all make sense from a neurophysiological standpoint.In addition to that, you realize the light going on and off are both important signals in the retina. Retina encodes both light going on and off at every location. That’s accomplished through the on and off bipolar cells.
One bipolar cell, when the light goes on, it depolarizes. The photoreceptor hyperpolarizes, NT release goes down and the on bipolar depolarizes. That sends a depolarizing signal to the ganglion cells and when they depolarize, what happens to their AP frequency? Increases. When a cell that’s going to generate APs is depolarized the spike frequency goes up. So when the light comes on, this on ganglion cell increases its spike. That’s the signal that the light came on in this particular location of the retina. When that light goes off, it generates an opposite signal which leads to a burst of APs from the off ganglon cell reporting to the brain that in that location, the light just went off.
Vertical/direct pathway: output [S10]
Signals that they generate are carried out of the retina by the ganglion cells shown here. There are lots of different morphological cell types.
The retina is not only interested in the fine details. Those of you in the front trying to discern the pattern of my tie and those of you in the middle trying to discern where I am in the room—those are fine details or more coarse details and for that you might need ganglion cells which have small receptive fields as demonstrated by this midget system.
A midget ganglion cell, which has a very narrow dendritic arbor, collect information from the very narrow spatial location and is responsible for our high acuity vision.
We don’t have high acuity vision everywhere because it takes a lot of cells to do that. But where we do have high acuity vision, we have medium acuity vision and low acuity vision.