NBL Module 8

In this lecture, I'll be talking about the conduction of the action potential. So in a previous lecture, we talked about how the action potential is first generated at the initial segment of the axon hillock. But to go back just to talk about the function of the action potential. So when there is a significant amount of incoming excitatory activity into a neuron where there's less inhibitory activity, the idea is that this neuron wants to communicate with its target cell that this activity is present so it has been activated, and it wants to be able to communicate this response to the target cell. So the problem is that the axon is often quite long and because these incoming synaptic potentials are graded potentials and they are not regenerated. Typically, this change in the membrane potential, the depolarization, would dissipate or decay before it ever reached the end of the axon. And that's where the action potential comes in. It allows for the communication of this increased excitatory input all the way down the length of the axon to the presynaptic region, where then this neuron will be able to communicate this information to the target cell through synaptic transmission. And therefore, once the action potential has been generated at the initial segment, it needs to be propagated or conducted along the axon to the presynaptic terminus. So action potentials are initiated at the axon initial segment, and then they are propagated down to the presynaptic terminus. And again, we have these incoming graded signals, which are passive signals, and they summate everywhere in the neuron. But if the depolarization is above -55 millivolts, which is threshold, it is sensed by the voltage gated sodium channels, which are present at the initial segment and it will fire an action potential. Now, the voltage gated sodium and potassium channels are localized all along the length of the axon and therefore, once the action potential is produced at the initial segment, that depolarization will be sensed by the nearby voltage gated sodium and potassium channels. Those channels will open, allow for the membrane depolarization and then this depolarization will be propagated along the length of the axon to the presynaptic terminus. So the voltage gated sodium channels then will initially open here and then they will depolarize the membrane. The sodium will flow in, that leads to depolarization of the membrane, and the action potential will travel down to the presynaptic terminus. So it isn't just the production of an action potential, that is the critical function for the axon, but it's the conduction of this action potential as well. So first, I'm going to talk about how the action potential is conducted along the length of an unmyelinated axon. You remember that about half of the axons in the central nervous system are unmyelinated. And typically, an unmyelinated axon is present in a neuron that works locally. So a local circuit neuron is one that is going to work within the same region and communicate information to nearby neurons. These axons are typically shorter than myelinated axons, and they're usually smaller. And continuous conduction occurs in the unmyelinated axon. And in this type of conduction, at every region along the length of the axon, there will be a transmembrane current as well as an axial current. So I want to define these two terms and talk about how the membrane depolarization will depend on the movement of ions across the membrane and the movement of ions inside the axon, within the axoplasm. Now, the critical feature of the unmyelinated axon conduction is that the voltage gated sodium and potassium channels are localized at every region along the membrane of the axon and therefore, the action potential will be regenerated at every specific region along the axon membrane. So going back to our cartoon drawing of the sodium current at the resting membrane potential -67 millivolts, the voltage gated sodium channels are closed. And there is very little sodium current across the membrane, there is the movement of ions by the leak sodium channels and of course the sodium potassium ATPase will be moving sodium ions from the inside to the outside. The membrane potential is at rest about -67 millivolts. And this shows the initial segment of the axon hillock and this is where our voltage gated sodium channels are first localized in terms of their expression. When there are some degraded potentials in the cell body, so you have incoming excitatory activity or less inhibitory activity, the membrane will become depolarized and this will passively spread throughout all of the membranes and it will also spread to the initial segment of the axon hillock. If that depolarization is above -55 millivolts then, that will lead to the activation and opening of the voltage gated sodium channels. And we can see here the membrane potential has reached -55 millivolts, and that's going to activate and open the voltage gated channels. Now the driving force, or the electrochemical potential for sodium is very large at the resting membrane potential and so the sodium ions will move down their electrochemical gradient inside the neuron and that will depolarize the membrane potential. Now very rapidly, these voltage gated sodium channels will inactivate, and therefore they will be inactivated and closed. But the important thing is the sodium ions, which are the sodium current, and then the movement of these ions is going to allow for the movement of this current along the membrane and into the cytoplasm and this is what will produce the depolarization of the membrane potential. So the underlying sodium currents are going to be produced and the sodium ions will be moving and that will depolarize this adjacent membrane along the axon. Now, there is no directionality or specificity to the movement of these sodium ions, they're just going to be moving everywhere. They can also move back along the membrane as well as forward along the membrane and then they can also move by chemical diffusion into the cytoplasm. So the movement of these ions is affected by both the chemical diffusion and the electrical gradient or the coulomb forces and so they're going to be rapidly moving away from the place where they were initially moving across the membrane. >> Movement of these ions and the current that it produces then will depolarize these regions of the membrane. Now, once these depolarization can be sensed by the adjacent voltage gated sodium channels channels. These sodium channels are closed, but they're in an activatable state. So now, these channels will open up and this will allow the sodium ions to flow into the adjacent region, which is going to depolarize the membrane potential. Then the sodium ions will flow in, depolarize the membrane, and the adjacent voltage gated sodium channels can be activated as well. As this process occurs then you have the transmembrane current. You have an axial current within the axoplasm, along the membrane, and within the cytosol, you have transmembrane current, axial current activation of voltage gated sodium channels, which will produce a transmembrane current and an axial current and then those currents will continue to move along the length of the axon, depolarizing the membrane and that depolarization will activate the voltage gated sodium channels. Here we have the activation of the adjacent region. The ions will move in, activate the subsequent voltage gated channels, and this will allow for the movement of the action potential. And the subsequent depolarization of the downstream regions of the membrane. Now, in addition, of course, we have the voltage gated potassium channels, which are also present at the initial segment and they become activated by the depolarization above threshold as well. And once they are activated, they will allow for potassium to move down its electrochemical gradient and because there's more potassium on the inside of the cell than the outside and the membrane potential has become more positive on the inside of the cell, that's going to allow the potassium ions to move across the membrane and this will repolarize the membrane potential behind the depolarization phase of the rising phase of the action potential. Now, just like in the case of the voltage gated sodium channels, there are voltage gated potassium channels all along the length of the axon and their job will be to repolarize the action potential after the depolarization phase, all the way down to the presynaptic region. It turns out that there are these two different currents. The movement of the ions across the membrane is what we call the transmembrane current, and then the movement of the ions within the axoplasm. The axonal cytoplasm is the axial current and these two currents will both contribute then to the action potential conduction. Now, for an unmyelinated axon, there will always be a transmembrane current followed by an axial current, and then there will be depolarization of the membrane potential and then that will activate voltage gated sodium channels and so you'll have a transmembrane current that will be followed by an axial current and this will happen in a continuous fashion, all the way down the length of the axon. You remember that the current is equal to the voltage times the conductance. So when there is the conductance of the opening of the voltage gated sodium channels, then that's going to be responsible for the transmembrane current. For the axial current, you just have the conductance of the axoplasm itself, which allows for these ions to move throughout the axoplasm. In addition to the local circuit neurons, there are also the principle or projection neurons. These neurons have axons that are myelinated and one of the major functions of myelination is to help increase or enhance the rate or the speed of the conduction of the action potential. This shows that some axons are unmyelinated and other axons are myelinated and these are types of axons that are found in the peripheral nervous system, where the majority of axons are in fact myelinated. What this demonstrates is that there are different types of sensory neuron axons and these sensory axons can differ with respect to their diameter and also to whether or not they are myelinated. It turns out that there are two ways to enhance the speed of action potential conduction. One of them is to increase the diameter of the axon, and the other is to myelinate that axon. First, I'll talk about the effect of the diameter on the rate of action potential conduction. For this, it's very useful to think about the water hose analogy. Here we think about the water flowing through a water hose and we can compare the flow of water to the current flowing through a wire or the current flowing through an axon. In this case, the voltage would be analogous to the water pressure the current, will be how much of the water is flowing through the water hose and the resistance would be anything that might slow down the flow of the water, such as a blockage of the water hose or a kink in the water hose. So you can imagine that one way to increase the rate of flow of the water through the hose is to make the water hose larger. And by making the water hose larger, that's going to decrease the resistance and that will increase the rate of flow of the water through the water hose and in an analogous manner, by increasing the diameter of the axon, that will increase the flow of current through the axon. Remember that the current is equal to the voltage times the conductance. Again, current is the movement of ions, and we can think of this as the rate of movement of ions per second. The conductance of ions through the axoplasm depends on the diameter of the axon. In this drawing here, you can see that a smaller diameter axon would have a greater resistance, whereas the larger diameter axon would have a smaller resistance or a greater conductance. That is one mechanism that neurons use to increase the rate of conduction of the action potential along an axon is by increasing the diameter of the axon. But the problem for neurons is that as the diameter of the axon increases, it takes up more space within either the nerve or the tract. In addition, as the axon becomes larger and larger, it becomes more vulnerable to shear. There is a theoretical limit to how large an axon can be before it just takes up too much room and it becomes more susceptible to damage. These types of axons that are shown here. The diameter of these axons are a little over 13 micro meters in diameter, and that's the largest diameter axon and this would be for a specific type of sensory neuron and you can see that the smallest diameter axons are on the order of a fraction to about one micron in diameter. So this is almost a 10 fold increase in the diameter of the axon but axons don't typically get much larger than this again, because of the space and the susceptibility. Now, the other way that the neurons can increase the rate of action potential is conduction by myelination and what myelination does is that enhances both the speed and the efficiency of action potential conduction. In the central nervous system, the majority of axons that originate from the principal or projection neurons, in which the axons are going to be traveling from one region of the brain to another region of the brain, these are typically the myelinated axons in the CNS and you remember that the oligodendrocytes are the myelinating cells in the central nervous system, while the Schwan cells are the myelinating cells in the peripheral nervous system. In the PNS, the majority of axons are in fact myelinated. There are just a few of these fibers here, which are the C fibers, which are typically unmyelinated. Interesting these are the axons from the neurons that are involved in pain, sensation, and transmission into the CNS. >> So myelination is a mechanism then that enhances the speed and efficiency of action potential conduction. And because of the myelination, it really changes the way that the action potential is propagated along the axon. So this depicts what a myelinated axon would look like. And in a myelinated axon, there are two specific regions: We have the myelinated region that has got the myelin wrapped around it, and usually, there are many layers to the myelin. So it's not just a single membrane, but this membrane wraps around and around to form many layers. And so what this will do is it's going to completely insulate and isolate this part of the axon from the extracellular fluid. And then there are gaps between the myelinated regions and these gaps are called the nodes of Ranvier. And at the node region, this is where the plasma membrane of the axon does have access to the extracellular fluid. So this is the place where the ion channels and the active transporters are localized and this is where the movement of ions across the membrane will occur. So these regions that are myelinated, not only is the myelin membrane completely covering the plasma membrane and so preventing access to the extracellular fluid, but it turns out that in these myelinated segments, there are no channels or transporters localized in this region either. So the major differences then are that the voltage-gated channels and the active transporters that are involved in establishing the ionic gradients for the resting membrane potential, these are all selectively localized at the nodes of Ranvier. Now, because the voltage-gated sodium and potassium channels are only localized at the nodes, this means that this is the only place where the transmembrane current can occur. So this is the only place where there will be the influx of sodium and the efflux of potassium. And this is the region where the depolarization of the membrane will occur and this is the region where the action potential will be regenerated in a myelinated axon. So this means that the action potentials are regenerated only at the nodes of Ranvier. So we can see here that the voltage-gated sodium channels will be activated, that's going to lead to the rising phase of the action potential, and then the presence of the voltage-gated potassium channel will lead to the repolarization, the falling phase of the action potential. So the action potential will be generated at this node and then what happens is the sodium ions are going to flow through the axoplasm along the length of the myelinated segment. So between the nodes of Ranvier, there is only passive movement of current and a passive change in the membrane potential. But because the myelinated segment is so well insulated, that means that those sodium ions can travel very efficiently because no sodium ions are going to be lost across the membrane to the leak channels or to the active transporters and so that sodium current then will then move along the length of the myelinated segment, and this will produce a very rapid movement of those ions and those sodium ions will be able to depolarize the membrane of the downstream node of Ranvier and this will activate the voltage-gated sodium and potassium channels found at this subsequent node. And what this means is that if we were to look at this response, we would see a large depolarization and repolarization of the membrane that occurs here and then there would just be this passive movement of current, and then there would be another depolarization that would occur at this node. So this type of conduction is called saltatory conduction. And saltatory means proceeding by leaps or discontinuous. And so we see there's a large depolarization followed by a short period of time, and then a large depolarization, et cetera. So why does this speed up then, the rate of action potential conduction? So we talked about the transmembrane currents and the axial currents. What happens is that by having the presence of the myelin membrane, that means that there is no access of these voltage-gated channels to sodium because the extracellular fluid is completely covered, and in addition, there are no voltage-gated channels that are present within the myelinated membrane segment. And what that will do then is that's going to increase the membrane resistance because it prevents the movement of ions through channels within these myelinated segments and that will enhance the axial current. So once the sodium flows across the membrane at the node of Ranvier, then that sodium ion is going to be able to zip down the axoplasm and will depolarize the membrane at the downstream node. Again, the current is equal to the voltage times the conductance. If we have no conductance at this membrane, then there will be no transmembrane current and there will be no leakage of ions across the membrane so it has a very good resistance and that will enhance the axial current, which will then predominate. Now, in addition to the movement of the sodium ions down their chemical gradient, the other part of the driving force is the electrical gradient. And you remember that the downstream node is going to have a resting membrane potential of about minus 67 millivolts because it hasn't been depolarized yet. And because this downstream node of Ranvier is negative, it's going to have a very high attractive electrical force on those sodium ions. So not only will the sodium ions be moving along the length of the membrane by chemical diffusion, but now it's going to have a large electrical force on it as well. And the movement of sodium down the axon can be very rapid and it can be much more rapid than depending on the opening of channels, a transmembrane current, the movement of sodium ion, the opening of channels, transmembrane current, and so on. Completely, we're getting around the need to have any transmembrane current and that will force the axial current to take charge of the movement of the ions and that will then lead to a much faster rate. So we see that the action potential is generated at the node of Ranvier. So here's our action potential that's depending on the activation of voltage-gated sodium channels and voltage-gated potassium channels. But because there are no voltage-gated channels here, then this current is just going to be moving along the axoplasm here and you can see that the current will be great enough to depolarize the membrane and activate the voltage-gated channels at the downstream nodes. So we see that the voltage-gated sodium and potassium channels are concentrated at the nodes and between the nodes, the current is traveling by only the axoplasm. But this current traveling is much faster than having two open voltage-gated channels and depend on the transmembrane current, and so we see that the depolarization, it appears to jump from one node to the other node, though, of course, it doesn't really jump, it's just that the internode movement of the current is much faster than it would be in continuous conduction. >> So interestingly, in addition to the potential across the membrane, which is the transmembrane potential. It turns out by having this situation the axon has added an additional potential, and this is what we call the transnodal potential. So as I mentioned, as those sodium ions move across the membrane they're going to have their positive charge, but they are going to be attracted to the negative charge that would be present as the resting membrane potential in the downstream node. So in addition to the potential across the membrane then we also have we have an additional potential that has built up between the nodes, and that's going to add to the electrical force on those sodium ions, and that's one of the reasons why the movement of these ions is enhanced in the myelinated axon. So just a quick review then we have these two different types of conduction. We have continuous conduction, and we have saltatory conduction. The continuous conduction is for unmyelinated axons and the saltatory conduction occurs in the myelinated axons. For the myelinated axons we have the concentration and localization of the ion active transporters and the voltage gated ion channel at the nodes of Ron VA. The presence of these will enhance the speed of action potential propagation because it's going to favor the axial current, which is much faster than the transmembrane current and the dependence on the opening of voltage gated channels to underlie that current. In addition, the myelination also enhances the efficiency of action potential propagation, and the reason for this is because these channels are localized only at the nodes. You need fewer channels in order to do exactly the same job. So these channels are localized only here, they're not expressed or localized along the length of the axon. So if you look at the total number of voltage gated channels and transporters it is much less than it would be in continuous conduction, because they only need to be concentrated at the nodes. In addition, the ion transporters that are producing the ionic gradients and the resting membrane potential are also only localized at the nodes of Ron VA. So basically there is no membrane potential in the mylinated segments, and so we talk about the sodium potassium ATPAs. It uses a lot of ATP hydrolysis in order to maintain the gradients and the resting membrane potential, and so by having fewer of these active transporters as well it requires much less energy to maintain the resting membrane potential only at the nodes of Ron VA. So to review then the voltage gated sodium and potassium channels are localized and concentrated only at the nodes of Ron VA, and the action potential is also only regenerated at the nodes of Ron VA. And between the nodes of Ron VA in the myelinated segment the current flows by only the axoplasm, which is much faster than the transmembrane current. Because the myelination prevents any loss of current across the membrane, it is also more efficient. And in addition to the transmembrane potential, which is going to be the driving force for the movement of ions across the membrane in the myelinated axon, there's also a transnodal potential which is going to provide an additional component to the driving force. You need fewer channels and transporters, and that's going to use less energy, and hence it will be more efficient. So if we look at the comparison of the speed of conduction of the action potential in an unmyelinated and myelinated axon, you can see that it can be very dramatic. So in an unmyelinated axon depending on what the diameter of the axon is, we have the rate of propagation can be between about 1-10 millimeters per millisecond. Of course, the larger the axon diameter the faster the rate of conduction will be. However, in a myelinated axon the rate of conduction depending on the diameter and other properties of that axon, usually varies between about 20 and 100 millimeters per second. So there can be about a 5-10 fold increase in the speed of action potential conduction by having a myelinated axon. So this begs the question then, why aren't all axons myelinated within the nervous system? And one of the ideas about why this is is because myelination is expensive. It's expensive because it requires this additional cell, the Glial cells, which provide the myelination. And these cells require energy for all of their cellular functions, so all of the oligodendrocytes in the CNS and the Schwan cells in the peripheral nervous system then they are going to be necessary, they're going to be required to provide the myelination for these axons. And in addition, the myelination itself also takes up space. We think about the evolution of the nervous system across all the different species, and one thing that has occurred during evolution is that the number of neurons within the brain and the peripheral nervous system has increased as well as the density. So humans have a greater density of neurons within a specific volume and so one idea is that it's a trade off, so that you have to have a trade off between increasing the number of neurons those axons take up space. And so one of the ideas is that if neurons are working locally they are short enough that they don't need to increase their rate. Whereas if a neuron is working at a distance then it will need to have myelination in order to produce a rate of action potential conduction, that will be sufficient to signal to the target cell within a specific period of time.

In this lecture, I'll wrap up talking about the action potential by answering some important questions. So we talk about the action potential as traveling along the axon, as the action potential is conducted along the axon. So it starts out with a depolarization in this region. And then as the action potential moves down the axon, this leads to depolarization of the adjacent and downstream regions. But I just wanted to point out that what is really traveling when the action potential moves along the axon are the ions. So the ions are moving across the membrane and through and down the axoplasm. You remember that the movement of ions is current. And so while we talk about the action potential, as a change in the membrane potential, as the process that is moving along the axon. It's actually the current that's underlying this change in the membrane potential that is leading to the movement of the action potential. So an interesting question is, why does the action potential move in only one direction down the axon? Why doesn't it spread to the upstream regions as well in either continuous or saltatory action potential propagation then we have the movement from the initial segment in one direction down to the presynaptic region. And it turns out that it is the inactivation of the voltage gated sodium channels. Which makes them refractory for about a millisecond or so. Meaning that even when the membrane is depolarized, once those voltage gated channels have been inactivated, they can't open up again until the membrane potential returns to the resting membrane potential. Because the voltage gated sodium channels require the repolarization of the membrane potential, that occurs mainly by the activation of the voltage gated potassium channels and the movement of potassium ions to reset these channels. They have to go back from the closed inactive state back to the closed but active state. And this requires the repolarization of the membrane potential. So once these voltage gated sodium channels have been activated, they will remain in the closed inactive state until the membrane potential has been repolarized and will reset the membrane potential back to the resting membrane potential. So behind the action potential then, these are the voltage gated sodium channels that are opening here. But behind the action potential, these voltage gated sodium channels are inactive and closed. And because of this then, even though the membrane is depolarizing in this region here, these channels cannot be opened again until they go back to the resting membrane potential. So this prevents the backward movement of the action potential. Once the action potential has been produced. So only the channels that are upstream are sensitive in the closed, but active and activatable state. The ones behind are closed, and in the inactive state. And this inactivation of the voltage gated sodium channels leads to what we call the absolute refractory period. And this again, ensures the directionality of the action potential. This period of time is about one millisecond. You can see that during this period, this is the time that is required for those voltage gated potassium channels and the potassium current to repolarize the membrane potential back to the resting membrane potential. Once the resting membrane potential has been reached, then those inactivated voltage gated channels will undergo a conformational change that will set them back into the closed but active state. In addition, this time period, this absolute refractory period, which is defined as the period of time in which no matter how much depolarization occurred within the neuron, another action potential could not be produced. This also sets the maximum theoretical rate or frequency of action potential generation. And because this is one millisecond here, if it takes one millisecond to reset those voltage gated channels back from the inactive closed state to the active closed state, and then this means that it requires a millisecond in order to do this. And so the maximum frequency of action potentials is one action potential per millisecond, which would be 1,000 action potentials per second, which would be 1,000 Hertz. Now, a typical neuron never fires at this high frequency. Action potentials usually fire in the order of about one every two to three to five milliseconds or so when a neuron is firing very rapidly. So even though this is the theoretical max, neurons never fire at such a high frequency. So you can see here that when the membrane is depolarized, it produces an action potential. During this millisecond time frame, even if you depolarized the membrane potential to a greater and greater level. And a subsequent action potential could not be fired because these voltage gated sodium channels are still in the inactive state. Now if we move a little bit further along here to about 1.5 milliseconds or so, then if there is a depolarization that is sensed at the initial segment that is greater than threshold, then a subsequent action potential could be produced. So the interesting difference between the graded potentials and the action potentials is not only their magnitude, but the fact that the action potential is what we say, all or none. So we have the incoming synaptic signals that are generated in the dendrites and the cell body. And these are the graded signals are electrotonic signals, and the current that is produced just will passively spread throughout the cell body and throughout the dendrites. And these passive signals, again, a neuron will receive thousands of these incoming signals and they're going to summate with each other and there will be a passive spread of the current and also a passive spread of the membrane potential. Once the threshold is produced here at the initial segment, then the action potential will be produced, and then it will be conducted along the length of the axon. But the action potential itself does not actually contain any information except to say that threshold was produced at the initial segment in order to generate the action potential. So as long as threshold is reached then, the action potential has a very characteristic magnitude and time course. And the time course and magnitude are independent of the magnitude of the initial signal that was produced at the cell body at the initial segment that triggered it. And so what that means is that you can see here the graded potentials, we have some very small graded potentials. If you had a number of small graded potentials that all came into the neuron at the same time, then that would passively depolarize the membrane potential. And if that is reached at the initial segment, if threshold is reached, then that will produce an action potential. So no matter how large the stimulus is that produces the action potential, the action potential always has the same characteristic, magnitude, and duration. So this then will. >> This type of information processing then will allow us to ask the question then, how does a neuron communicate? How much activity it actually receives within the cell body? If this response just needs threshold to be reached, then how is information about the activity that is occurring within the neuron, how is that information communicated to the downstream target neuron? And the answer to that is that, one of the main ways that neurons will encode information is by the frequency of action potentials that are generated. So it turns out that when you have a sub threshold stimulus, you get no response. And if you have a small stimulus here, in which threshold is reached, so you have a depolarization of the membrane potential, which will reach threshold and give -55 millivolts. Then what you can see is that a small stimulus will lead to the production of low frequency action potentials. In contrast, if you have a large depolarizing stimulus, say that stimulus depolarized the membrane potential to about -40 or -30 millivots, what you can see is that's going to lead to a greater frequency of the action potentials that are generated at the initial segment. So the number of action potentials that are produced then, so we talk about the duration of the action potential train depends on how long the membrane potential is depolarized above threshold, and the frequency of the action potentials will depend on the magnitude of the summed depolarization that is being sensed at the initial segment. So even though a single action potential itself only contains the information that there was some activity within the neuron, it's the frequency of these action potentials then, that will encode the information about the magnitude and the duration of the depolarization that is occurring within the neuron. So this brings us to the relative refractory period. An important question is, how soon or how quickly after the initial action potential is produced, can a neuron fire the next or subsequent action potential? And this firing rate depends on the relative refractory period. The relative refractory period is dependent on the activation of the voltage gated potassium channels, which will repolarize the membrane potential. Then the additional potassium channels that will be involved in the after hyperpolarization or the undershoot phase. So what this means is that when there is an action potential that is produced by an incoming depolarization at the initial segment, it has to reach threshold here. You can see that if during this time you had the same type of small stimulus, it would not be able to induce a subsequent action potential because you've got this constant efflux of potassium which is then constantly bringing that membrane potential in through the hyperpolarization of the membrane. However, if at the initial segment there is a larger depolarization stimulus, then during this period of time there would be enough depolarization to overcome that hyperpolarization that was induced by the voltage gated potassium channels and therefore, during the relative refractory period, a subsequent action potential could be produced. But you have to have a larger stimulus in order to produce that. So this period depends on the potassium current which is repolarizing the membrane potential and the potassium currents that are involved in hyperpolarizing the membrane potential. And basically, what this means is, is that you need a larger depolarization to meet threshold during the relative refractory period. Now neurons differ in their relative refractory period to a great extent. So this can be very short. If there are many voltage gated potassium channels and other potassium channels, this will repolarize the membrane potential much more quickly and that will be important because then those voltage gated sodium channels will be able to be reset and can be activated again. So that is going to shorten this period of time, but when you have many voltage gated potassium channels, you're going to require a larger depolarization at the initial segment in order to generate a subsequent action potential. Now after this time, after this relative refractory period, then the membrane potential is back to the resting membrane potential. All the potassium channels and sodium channels are closed. After this five milliseconds or so, then it is much easier for an action potential to be generated just by meeting threshold at the initial segment. So we can see here that the frequency of the action potential generation depends on the magnitude and the duration of the summed responses that are present in the cell soma initial segment. So if we have a stimulus here, if we have a depolarization of the membrane at the initial segment that only lasts for a millisecond or so, you can see that that will generate a single action potential. But if we have a depolarization of the membrane that lasts for many milliseconds or seconds, what you can see is that as soon as the membrane potential has repolarized back to the resting membrane potential, if there is still a depolarization of the membrane at the initial segment, a subsequent action potential will fire. So here you can see that the time course, the duration of the summed graded potentials will then be indicating the frequency of the action potentials that will be generated by that particular neuron. So the frequency of action potentials depends on the time course of the summed responses at the initial segment. The frequency of action potentials also depends on the magnitude of the summed degraded potential. Change you can see here. So this is just a small depolarization will lead to a single action potential. If we depolarize the membrane even more, that's going to lead to a series of action potentials. And again, if we have an even greater depolarization, that will lead to a higher frequency. >> So this means that the duration of action potentials and the frequency of action potentials depends on how long the depolarization occurs at the initial segment and also how large that depolarization is present at the initial segment. So even though the individual action potential loses its ability to encode information, we now know that this is the frequency of action potentials. That is one of the mechanisms that neurons use to encode information. Now, information is also likely to be encoded in the pattern of action potential generation as well. And even though we know a little bit about the frequency code, the pattern of action potentials also is thought to be very important for encoding information. Now, we talk about the information encoding as part of what we call the neural code. And the question in the neural code is if the neurons are just using action potentials, then how is that information encoded? So we now know that the neural code involves the frequency of action potential, and we suspect that it involves the pattern of action potential generation. Now, at the presynaptic region, this is where the action potential is going to be decoded into synaptic transmission. So when we talk about synaptic transmission mechanisms, in the next few modules, we'll then talk about this decoding mechanism, whereby the action potential information will then be decoded into synaptic transmission. So this is a nice illustration of how the frequency of action potential depends on the magnitude and the duration of the summed graded responses in the cell soma. So this is a type of pyramidal neuron that might be present in the cerebral cortex and excitatory glutamatergic neuron. And you can see that it's receiving information into one of its dendrites. And this axon is firing action potentials at a frequency of about five per second, five action potentials per second. So this would be a frequency of five hertz. And through the summation of these membrane potentials then at the initial segment an action potential will be generated with some low frequency, by the axon on the other side of this neuron. If we increase the incoming activity, so the input synaptic activity, if we increase that 20 fold to, say, 100 hertz. So that's 100 action potentials per second, so that's an action potential every 10 milliseconds. So this is a really rapidly firing, very active input here at this synapse. The graded responses will be much greater and they will passively spread throughout the cell body here and it's going to lead to a larger and a more persistent depolarization of the membrane potential at the initial segment. And the output of that is going to be a higher production in the rate of action potentials by this particular neuron. Now, it's not an exact match, so it's not like you get five hertz input, and then you get a five hertz output. It's not a direct computation, but it's a reflection of the amount of incoming activity. And of course, this is an artificial situation, because this spiny neuron has got thousands and thousands of spines, which all have synaptic inputs, some of which are excitatory, some of the inputs might be inhibitory, and so there will be integration of all of these incoming synaptic inputs, which will then determine the magnitude and the duration of the summed graded responses here at the initial segment, which will then lead to an output in terms of the action potential frequency. So I wanted to go onto another question which has to do with the myelinated axon. So an important question is the fact that we have these nodes of Ranvier, which are the unmyelinated segments that separate the myelinated segments. And these are critical because these are the places where the action potential is regenerated by the voltage gated sodium and potassium channels within the myelinated neurons. So a question is, these nodes of Ranvier, why are they a specific distance apart within a myelinated neuron? Why aren't they a shorter distance or why are they not longer? And the idea to answer this is that they have to be close enough together so that as the membrane potential is decreasing within the myelinated segments through the axial movement of the sodium ions, these two regions have to be close enough such that as the membrane potential will be decaying between the nodes, that by the time the current reaches that downstream node, that current will still be significant enough and substantial enough in order to activate the voltage gated sodium channels and lead to the generation of an action potential at the downstream node. So they have to be close enough together then, so that as the membrane potential decays, that membrane potential change will still be great enough in this region to depolarize the next node. >> And this leads to the next question of what happens if there is a loss of myelin within a myelinated axon. And this is exactly what happens in demyelinating disorders like multiple sclerosis. In multiple sclerosis, there is demyelination of axons in the central nervous system. And the onset of multiple sclerosis usually occurs in individuals in their late teens or their early 20s. And it usually presents with specific signs and symptoms, including problems with vision, with movement, and also with some cognitive effects as well. And in these demyelinating disorders, it affects the myelin that is present within the central nervous system. And in this cartoon here, we have damage to the myelin. And this damage to the myelin then will often prevent the action potential signal from being conducted from the initial segment down to the presynaptic terminus. Now, this depicts a type of motor axon. And there are a number of motor effects in multiple sclerosis usually affects the ability to walk at first, as well as coordination, but there are also sensory effects as well. And this is a magnetic resonance image showing an individual at different times in different regions. And the sclerosis part of multiple sclerosis, it shows that there are these plaques or sclera here in which there is a loss of myelin, and you can see this within the brain. Now, this demyelinating disorder is thought to be one of the types of autoimmune disorders. And in autoimmune disorders, the body will make antibodies to what are normally healthy and self-types of proteins. And in the case of multiple sclerosis, the immune cells will make antibodies to the myelin, and then this will lead to an attack by the immune cells on the myelin, which leads to a damage and destruction of the myelin membranes. Now, in addition, one of the other effects in multiple sclerosis is there is a breakdown of the blood-brain barrier. And the normal vascular endothelial cells, which are keeping those tight junctions, these vascular endothelial cells, there will be the production of gaps. The immune cells will be able to cross across the blood-brain barrier. The antibodies to myelin will be able to cross the blood-brain barrier. And this is going to lead to the destruction of the myelin in the axons that are in these regions around the blood vessels as well as many other regions within the central nervous system. You also have activation of microglial cells, infiltration of T cells and other immune cells which will then lead to this feed-forward destruction of the myelin membrane. Now, we don't know what the cause of multiple sclerosis is. It's thought that there is a combination of some type of genetic modifications as well as some type of environmental exposure. It's been speculated that contributing factors include things like smoking, decreases in vitamin D, pathogens like bacterial or viral infections have also been implicated as well. So multiple sclerosis is a demyelinating disorder. And what happens is that you have the loss of the myelin membrane in specific regions. Now, going back to the discussion of why the nodes of Ranvier are a specific distance apart, you can see here that when there is a loss of the myelin membrane because there are no voltage-gated sodium and potassium channels within this segment of the axon that was myelinated previously, there is no mechanism to boost and regenerate the action potential within this region. When the myelin is lost, now the sodium and potassium currents that were produced at the nodes of Ranvier, they're going to move passively along the axon as it would in a healthy axon. But without the myelin membrane to prevent the loss of that current, these sodium ions are able to leak back across the membrane, they can be transported across the membrane, and so that sodium current is lost. So during this passive decay, there's a much more rapid loss of this current, and so there is a much faster change in the membrane potential. And basically, what this means is that even though there was this large sodium current and this large depolarization at this membrane segment by the time it reaches this part of the axon, then the change in the membrane potential will not be great enough to activate the voltage-gated sodium and potassium channels that are in this downstream node. And so what that means is that the action potentials will not be regenerated and they will eventually decrease and the rate can even stop. And if there are no action potentials then that are being conducted and arriving at the presynaptic region, then that's going to mean that there will be no synaptic transmission. So that means that, for example, the lower motor neurons, which are conducting signals out to the skeletal muscles, there will be a loss of those signals, which will then prevent muscle contraction because there will be no information that will be transmitted to the neuromuscular junction. Now, multiple sclerosis was initially thought to be just a demyelinating disorder which leads to the loss of synaptic transmission, action potential conduction, and synaptic transmission. But we now know that during multiple sclerosis there is also degeneration of axons. And it's sought that the reason for this is that oligodendrocytes provide myelin that allows for the fast and efficient conduction of action potentials by saltatory conduction. But in addition, oligodendrocytes also provide trophic support. So they release trophic factors like nerve growth factors and brain-derived neurotrophic factors that bind to axons and keep them healthy and surviving. And once the demyelination of the membrane occurs in addition to this, the oligodendrocytes themselves are not functioning properly, and so you have the loss of oligodendrocytes as well as their ability to myelinate. And so this can lead to a loss of that trophic factor, survival support. And so in some cases, in some regions of the brain, these axons will degenerate. Now, you may remember from our discussion of glial cells that once axons degenerate in the central nervous system, they are unable to regenerate. And the reason for this is that the presence of the oligodendrocytes and other activated astrocytes and microglia, they actually prevent the regeneration of these axons. So once these axons have degenerated, then this leads to an irreversible damage to the central nervous system and is one of the reasons that in multiple sclerosis there's a degenerating loss of function over a number of years.

Neurons propagate/conduct APs along the axon so that the axon presynaptic region can become depolarized and can participate in/initiate synaptic transmission. Propagation/conduction of the action potential ensures that the presynaptic terminus undergoes a large depolarization that will produce synaptic transmission and communication with the target cells. What really travels/moves are the Na+ and K+ ions inside the axon, which are the Na+ and K+ currents.

The factors that affect the movement of ions are chemical diffusion and coulomb forces, which together underlie the DF-electrochemical gradient. The two currents are the transmembrane current (through ion channels) and the axial current (in the axoplasm). The axial current depends on the diameter of the axon while the transmembrane current depends on the presence of ion channels and whether the membrane is myelinated

In continuous conduction, the AP activates VGNa+ channels in the adjacent membrane region, Na+ flows into the axon (the transmembrane current) and that Na+ diffuses in the axon (the axial current). Since VGNa+ channels are located all along the axon, the Na+ needs to travel only a short distance in the axon, and as it does, it depolarizes the adjacent membrane potential, and thus activates the adjacent VG Na+ channels. There is sequential activation of the VG Na+ channels all along the entire length of the axon. This involves the transmembrane Na+ current followed by a small axial Na+ current, followed by a transmembrane Na+ current and so on. In a similar manner, VG K+ channels are located all along the axonal membrane where they repolarize the AP, producing an outward K+ current at every membrane along the axon. In continuous conduction, the AP is regenerated at every membrane region along the axon.

The two mechanisms that increase the rate of AP conduction are increasing the diameter of an axon, and myelination. The types of neurons that need an increased rate of AP conduction are neurons with long axons. These include both somatic and autonomic lower motor neurons and sensory neurons (whose axons are located in the PNS in nerves), and in the CNS the projection/principal neurons that need to send their signals over long distances. Projection/principal neurons extend their axon outside of one CNS region where its cell body is located, and make synapses with target neurons in another region of the CNS.

Myelinating cells in the CNS are oligodendrocytes. Myelinating cells in the PNS are Schwann cells. In addition to affecting the speed and efficiency of the action potential, myelinating cells also provide trophic (survival) support to the axons they myelinate.

Nodes of Ranvier are small gaps in the myelinated axon that are the regions of theaxonal plasma membrane that are not myelinated. The Na+/K+ ATPase, leak channels, VGNa+ channels and VGK+ channels are all localized to the Nodes of Ranvier. As these are the regions where there is no myelin membrane, this is where Na+ and K+ ions cross the membrane to establish the RMP, and to regenerate the AP (depolarize or repolarize the membrane potential). The Nodes are spaced at the distance they are so that they are far enough to favor the axial current, but close enough so the membrane potential doesn’t decay below threshold before it reaches the adjacent Node of Ranvier

In saltatory conduction, which occurs in a myelinated axon, the AP is regenerated only at the Nodes of Ranvier, because those are the only regions of the axonal membrane where the VG channels are located and where the channels have access to the extracellular fluid. At the Node, an AP is generated by activation of VG Na+ channels, and produces the Na+ current as Na+ flows across the membrane (a transmembrane current). Those Na+ ions then move along the axon (the axial Na+ current), which depolarizes the membrane potential. There are no VG channels in the myelinated segments. There are also no leak channels or pumps in the myelinated segments. In addition, the myelin covers the axonal membrane so there is also no extracellular fluid around the axonal membrane. This also means that as the axial current is moving along the axon, the ions can’t flow back across the membrane (through leak channels or transporters), and so no current is lost across the membrane. The axial current flows passively along the axon and as it does, it depolarizes the adjacent membrane within the myelinated segment. Some Na+ ions do diffuse into the axoplasm, and so some current is lost with distance from the Node and time. However, by the time the current has spread to the adjacent Node, it is still large enough that it will depolarize the membrane potential above threshold (-55 mV) so that it will activate the VG channels at the next Node, which regenerates the AP. Therefore, the AP is regenerated only at the Nodes. Between the Nodes in the myelinated segments, there is only passive movement of current. In a similar manner, VG K+ channels are located only at the Nodes, so the repolarization of the AP occurs only at the Nodes as well.

Both types of conduction are similar in that they both produce an action potential (a large depolarization followed by repolarization and hyperpolarization) where they are produced. They both involve the opening and closing of the voltage gated Na+ and voltage gated K+ channels. They are both initiated at the initial segment and travel along the axon in one direction to the presynaptic terminus. They both are required to depolarize the presynaptic region to initiate synaptic transmission

In an unmyelinated axon, conduction is continuous since the VGNa+ and VGK+ channels are located all along the length of the axon. These channels produce a transmembrane inward Na+ current followed by a transmembrane outward K+ current. Continuous conduction doesn’t require much axial current, just enough to depolarize the adjacent VG channels. So the rate is slower (1-10 m/sec) because it takes time to open the VG channels, for the Na+ ions to flow in and then move to activate the adjacent VG channels. In saltatory conduction, the VG Na+ and VGK+ channels are located only at the Nodes of Ranvier (the unmyelinated regions). Thus the transmembrane Na+ and K+ currents are produced only at the Nodes. In the myelinated segments, more of the Na+ and K+ currents travel along the axon (axial currents). Saltatory conduction is much faster (20-100 m/sec) because it relies more on the axial current, which is must faster than the transmembrane current (since the transmembrane current requires more time to open the VG channels and Na+ to flow in etc). The axial current can involve more movement of charges by Coulomb forces, which is faster than chemical diffusion. In addition, in myelinated axons there is also an internodal potential, which adds to the membrane potential and increases the electrochemical gradient. Saltatory conduction is more efficient because it requires fewer total VG channels (since they are localized at only the Nodes) and less energy is required to produce the resting membrane potential (which occurs at the Nodes and then spreads passively).

Increasing the Rm will decrease the transmembrane current. In myelinated axons, the transmembrane current occurs only at the Nodes. By increasing the membrane resistance in the myelinated segments, it prevents ions from crossing back across the membrane as easily. This means that more ions move within the axoplasm since they are not lost across the membrane, and this enhances the axial current in the axoplasm. The axial current is faster than the transmembrane current since it doesn’t require the opening of ion channels and also, the axial current can involve the movement of charges by Coulomb forces, which is faster than chemical diffusion. This increases the speed of depolarization along the axon. As mentioned above, it also enhances the efficiency since the channels and the active transporters only need to be present at the Nodes (so it requires fewer of these)

The inactivation of the VG Na+ channels ensures the unidirectional conduction of the AP. Once they are activated/opened, the VG Na+ channels close and inactivate. Hence, even though the downstream segment of the axon mediates a Na+ current and is depolarized, the upstream VG Na+ channels can’t open again because they are inactivated. They require repolarization of the AP back to the RMP (through the function of the VG K+ channels) to return to the closed-active state, where they can then be opened again and produce another AP. However, by the time this happens (repolarizing back to RMP and the VG Na+ channels returning to the open-active state), the Na+ current and depolarization of the downstream AP is too far away (and thus will have already passively decayed) to have any effect on the upstream VG Na+ channels

If the myelin membrane is damaged or lost, as in MS, the action potentials could stop completely (AP conduction would be stopped). This could prevent the AP signal from reaching the end of the presynaptic region, and thus prevent synaptic transmission. This is what happens in MS, and some of the most vulnerable axons are long axons in the spinal cord that control motor neurons that innervate muscles in the legs, and the optic nerve involved in transmission in the visual system. (Only CNS axons are demyelinated in MS.) Demyelination can also be detrimental to the health of the axon because oligodendrocytes, which provide myelin, also provide trophic (survival) support to the axons they myelinate. Without trophic support, the axon could degenerate. As we learned earlier, once CNS axons degenerate, they can’t regenerate in the CNS, leading to an irreversible loss of axons

AP conduction is faster in a myelinated axon because myelination favors the axial current over the transmembrane current. The axial current moves so fast because of the Coulombic interactions, in which ions are attracted to or repulsed by nearby ions. This facilitates the movement of an ionic charge rapidly through the axoplasm (which has a theoretical speed equal to the speed of light), which is much faster than the rate of the chemical diffusion of ions in solution.

In the simplest systems, the frequency and duration of the train of APs depends on the magnitude and duration of the summed graded potentials (inputs from synapses) in the neuron. The neural code refers to how information is encoded in APs (frequency, timing, pattern, and/or number- or some other aspect) and then decoded at the synapse in synaptic transmission. From Wikipedia: The link between stimulus and response can be studied from two opposite points of view. Neural encoding refers to the map from stimulus to response. The main focus is to understand how neurons respond to a wide variety of stimuli, and to construct models that attempt to predict responses to other stimuli. Neural decoding refers to the reverse map, from response to stimulus, and the challenge is to reconstruct a stimulus, or certain aspects of that stimulus, from the spike sequences it evokes

A sequence, or 'train', of spikes may contain information based on different coding schemes. In motor neurons, for example, the strength at which an innervated muscle is contracted depends solely on the 'firing rate', the average number of spikes per unit time (a 'rate code'). At the other end, a complex 'temporal code' is based on the precise timing of single spikes. They may be locked to an external stimulus such as in the visual and auditory system or be generated intrinsically by the neural circuitry. Whether neurons use rate coding or temporal coding is a topic of intense debate within the neuroscience community, even though there is no clear definition of what these terms mean.

Rate coding: The rate coding model of neuronal firing communication states that as the intensity of a stimulus increases, the frequency or rate of action potentials, or "spike firing", increases. Rate coding is sometimes called frequency coding. Rate coding is a traditional coding scheme, assuming that most, if not all, information about the stimulus is contained in the firing rate of the neuron. In most sensory systems, the firing rate increases, generally non-linearly, with increasing stimulus intensity. Any information possibly encoded in the temporal structure of the spike train is ignored. Consequently, rate coding is inefficient but highly robust with respect to the ISI 'noise'. In rate coding, learning is based on activity-dependent synaptic weight modifications. Rate coding was originally shown by ED Adrian and Y Zotterman in 1926. In this simple experiment different weights were hung from a muscle. As the weight of the stimulus increased, the number of spikes recorded from sensory nerves innervating the muscle also increased.

Spike-count rate: The spike-count rate, also referred to as temporal average, is obtained by counting the number of spikes that appear during a trial and dividing by the duration of trial. The length T of the time window is set by the experimenter and depends on the type of neuron recorded from and to the stimulus. In practice, to get sensible averages, several spikes should occur within the time window. Typical values are T = 100 ms or T = 500 ms, but the duration may also be longer or shorter. Despite its shortcomings, the concept of a spike-count rate code is widely used not only in experiments, but also in models of neural networks. It has led to the idea that a neuron transforms information about a single input variable (the stimulus strength) into a single continuous output variable (the firing rate). There is a growing body of evidence that in Purkinje neurons, at least, information is not simply encoded in firing but also in the timing and duration of non-firing, quiescent periods.

Time-dependent firing rate: The time-dependent firing rate is defined as the average number of spikes (averaged over trials) appearing during a short interval between times t and t+Δt, divided by the duration of the interval. It works for stationary as well as for time-dependent stimuli. To experimentally measure the time-dependent firing rate, the experimenter records from a neuron while stimulating with some input sequence. The number of occurrences of spikes nK(t;t+Δt) summed over all repetitions of the experiment divided by the number K of repetitions is a measure of the typical activity of the neuron between time t and t+Δt. A further division by the interval length Δt yields time-dependent firing rate r(t) of the neuron, which is equivalent to the spike density of PSTH. The time-dependent firing rate coding relies on the implicit assumption that there are always populations of neurons.

Temporal coding: When precise spike timing or high-frequency firing-rate fluctuations are found to carry information, the neural code is often identified as a temporal code. A number of studies have found that the temporal resolution of the neural code is on a millisecond time scale, indicating that precise spike timing is a significant element in neural coding. Such codes, that communicate via the time between spikes are referred to as interpulse interval codes, and have been supported by recent studies. Neurons exhibit high-frequency fluctuations of firing-rates, which could be noise or could carry information. Rate coding models suggest that these irregularities are noise, while temporal coding models suggest that they encode information. If the nervous system only used rate codes to convey information, a more consistent, regular firing rate would have been evolutionarily advantageous, and neurons would have utilized this code over other less robust options. Temporal coding supplies an alternate explanation for the “noise," suggesting that it actually encodes information and affects neural processing. Temporal codes employ those features of the spiking activity that cannot be described by the firing rate. For example, time to first spike after the stimulus onset, characteristics based on the second and higher statistical moments of the ISI probability distribution, spike randomness, or precisely timed groups of spikes (temporal patterns) are candidates for temporal codes

Phase-of-firing code: Phase-of-firing code is a neural coding scheme that combines the spike count code with a time reference based on oscillations. This type of code takes into account a time label for each spike according to a time reference based on phase of local ongoing oscillations at low or high frequencies. It has been shown that neurons in some cortical sensory areas encode rich naturalistic stimuli in terms of their spike times relative to the phase of ongoing network oscillatory fluctuations, rather than only in terms of their spike count. The local field potential signals reflect population (network) oscillations. The phase-of-firing code is often categorized as a temporal code although the time label used for spikes (i.e. the network oscillation phase) is a low-resolution (coarse-grained) reference for time. As a result, often only four discrete values for the phase are enough to represent all the information content in this kind of code with respect to the phase of oscillations in low frequencies. Phase-of-firing code is loosely based on the phase precession phenomena observed in place cells of the hippocampus. Another feature of this code is that neurons adhere to a preferred order of spiking between a group of sensory neurons, resulting in firing sequence. Phase code has been shown in visual cortex to involve also high-frequency oscillations. Within a cycle of gamma oscillation, each neuron has its own preferred relative firing time. As a result, an entire population of neurons generates a firing sequence that has a duration of up to about 15 ms.

Population coding: Population coding is a method to represent stimuli by using the joint activities of a number of neurons. In population coding, each neuron has a distribution of responses over some set of inputs, and the responses of many neurons may be combined to determine some value about the inputs. From the theoretical point of view, population coding is one of a few mathematically well- formulated problems in neuroscience. It grasps the essential features of neural coding and yet is simple enough for theoretic analysis. Experimental studies have revealed that this coding paradigm is widely used in the sensor and motor areas of the brain.

Correlation coding: The correlation coding model of neuronal firing claims that correlations between action potentials, or "spikes", within a spike train may carry additional information above and beyond the simple timing of the spikes. Early work suggested that correlation between spike trains can only reduce, and never increase, the total mutual information present in the two spike trains about a stimulus feature However, this was later demonstrated to be incorrect. Correlation structure can increase information content if noise and signal correlations are of opposite sign. Correlations can also carry information not present in the average firing rate of two pairs of neurons. A good example of this exists in the pentobarbital-anesthetized marmoset auditory cortex, in which a pure tone causes an increase in the number of correlated spikes, but not an increase in the mean firing rate, of pairs of neurons.

Independent-spike coding: The independent-spike coding model of neuronal firing claims that each individual action potential, or "spike", is independent of each other spike within the spike train.

Position coding: A typical population code involves neurons with a Gaussian tuning curve whose means vary linearly with the stimulus intensity, meaning that the neuron responds most strongly (in terms of spikes per second) to a stimulus near the mean. The actual intensity could be recovered as the stimulus level corresponding to the mean of the neuron with the greatest response. However, the noise inherent in neural responses means that a maximum likelihood estimation function is more accurate.

Sparse coding: The sparse code is when each item is encoded by the strong activation of a relatively small set of neurons. For each item to be encoded, this is a different subset of all available neurons. In contrast to sensor-sparse coding, sensor-dense coding implies that all information from possible sensor locations is known. As a consequence, sparseness may be focused on temporal sparseness ("a relatively small number of time periods are active") or on the sparseness in an activated population of neurons. In this latter case, this may be defined in one time period as the number of activated neurons relative to the total number of neurons in the population. This seems to be a hallmark of neural computations since compared to traditional computers, information is massively distributed across neurons