The fundamental principles of neuronal encoding

deludedgod's picture

The fundamental principle behind all signals between neurons in the brain is that of ion channels. These channels are gated. The cell membrane does not normally admit ions. The channels are shut unless they receive the proper signal. There are a variety of signals which will open the channels. Certain voltage differences across the membrane will open channels. Such channels are called voltage gated ion channels. VGIC are the most common types of channels in the logic gates in the brain, but they are not the only ones. There are, in addition, channels which open in response to the presence of ligands called neurotransmitters. These are called transmitter gated ion channels (TGIC). The other extremely important feature of these channels is that they are nearly always ion selective. Thus, we can separate the ion concentration gradients and voltage gradients that result from the charges of ions and the concentration of individual ions. This ability to differentiate between the concentration gradient and electric field is extremely important for the ability of neurons to transmit signals.

In neurons, this principle allows the propagation of an electrical signal from a transiently induced change in voltage to propagate along the neuron. Neurons are cells with long and spindly fibers that reach out from them. These long portions can be very long indeed, some of them extend almost 1.5m in length, from head to foot. These are called axons and they always propagate electrical signals away from the main body of a neuron. The neuron also has shorter outreaching arms that receive signals from other axons, called dendrites. Between an axon and a dendrite, a junction called a synapse forms and there is a gap, called a synaptic cleft, between the axon of one neuron and the dendrite of another. In this way, the two are electrically isolated.

If the electrical signal is conducted passively, it will get weaker at a rate directly proportional to distance travelled from its transiently induced signal. The signal needs to be propagated, or amplified at nodes along the axon in order for the maintenance of an appreciable signal.

Suppose that a transiently induced signal causes a slight depolarization of the membrane, raising the charge slightly (because the resting potential is negative). Because there is such a large induced change in voltage per cubic centimeter, this change will open channels that respond to it, in this case, Na+ channels. Because Na+ is maintained with a very high concentration outside the cell membrane, the influx of more Na+ down its electrochemical gradient will move naturally towards the Na+ equilibrium potential, since the cell is electrically excitable. This potential is roughly 50mV. The transient signal that allows for the movement of more Sodium into the membrane causes a greater depolarization which causes more sodium channels to open at an exponentially amplifying rate. That is, the flow rate of sodium ions into the neuron is a positive feedback loop which abruptly halts when the electric field induced blocks any more from entering. When the level of Sodium reaches a critical threshold, the nueron is said to fire and an electrical signal is propagated down the axon. However, this system is not enough to generate a proper action potential. If the exponential depolarization of the membrane simply occurred by the high concentration of VGNaC, then the membrane would repeatedly be subject to induced spasms, since the membrane would depolarize, and repeated adduction of Na+ into the cell membrane would cause the cell to stop at the Na+ equilibrium potential, and all of the Na+ channels would remain open. So, the neuron would simply sit there as the Na+ reached its equilibrium, freely moving through the open channels.

The cell is saved from these electrical spasms by two mechanisms. The first is Voltage gated Na+ channel inactivation and the second is Voltage gated K+ channels. These help to rapidly depolarize the membrane to bring it back to its normal potential in preparation for the membrane to fire again, which helps to rapidly close the Na+ channels.

The VGIC can exist in three states, open, closed and inactive. In the inactive state, the Na+ VGIC enters a refractory period where it cannot admit Na+, but also cannot close again for a small period of time, until the membrane is fully repolarized. There exists a mechanism on the Na+ VGIC that allows the blocking of the influx of Na+ while the channel is still open. This block is sensitive to membrane depolarization. There exists a similar mechanism on K+ VGIC which allows for their inactivation. The inactivation of Na+ VGIC also stops the action potential from spreading backwards, and ensures the signal is propagated unidirectionally along the axon.

If the inactivation of Na+ channels was the only mechanism to ensure that the membrane was fully repolarized quickly, then it would be a long time before the inactive Na+ channel was restored to its closed state, since the membrane will decay towards its resting potential only slowly. Another mechanism ensures that the membrane is rapidly restored to its polarized state. These are the K+ VGIC. Like Na+ VGIC, the K+ VGIC can exist in three states, and a short peptide "leash" on the VGIC will respond to membrane depolarization and blocks the efflux of K+. This is necessary to maintain the K+ gradient for rapid restoration by the Na/K pump (potassium tends to move out of the cell due to its concentration gradient. The pump must maintain this or the gradients will be destroyed as the ions come to equilibrium across the membrane). The rapid efflux of K+ overwhelms the transient influx of Na+ and thus brings the cell back to resting potential more quickly.

The efflux of K+ through the membrane allows for the rapid repolarization (charge becomes more negative) of the membrane, and brings the cell back to its resting potential, which allows for the inactive Na+ channels to rapidly close in response to the firing of action potential. Now that we know how the signals are propagated, we should learn how they are translated once they reach the end of a neuron.

At the synapse there is a gap between the axon and the dendrite called the synaptic cleft which separates the two cells and keeps them electrically isolated. When the signal reaches the end of an axon, it signals the influx of Ca2+ through the VGIC that allow for it to enter. This in turn signals the release of neurotransmitters through synaptic vesicles into the synaptic cleft. These transmitters then bind to transmitter-gated ion channels and allow for the influx of Sodium into the post-synaptic neuron, which allows for the continued propagation of the signal. There are two ways, in general, that transmitters work. They can either be excitatory or inhibitory. The former will bind to ion channels that allow for the influx of Na+ hence depolarizing the membrane. The latter will bind to ion channels that allow for Cl- influx of K+ efflux for the repolarization of the membrane hence inhibiting the sending of signals throughout the membrane. The neurotransmitters are varied and can serve to do either, but they are usually confined to one role. Acetylcholine, serotonin and adrenaline are the principle excitatory neurotransmitters, while GABA, dopamine and glycine are usually inhibitors.

So, once a signal is propagated through the gap between two neurons, how does the other neuron respond? Additionally, how does the neuron integrate all the signals it receives from all the other neurons it is synapsed with? Neurons ultimately act as integrating computational devices by virtue of two principles: temporal summation and spatial summation. The cell body of a neuron and the dendrites are coated with the terminal of axons. There are numerous axon terminals that coat the cell body. There are two types of membrane potentials that the axons can transmit. They can transmit inhibitory posy synaptic potentials and excitatory post-synaptic potentials. Each potential will either help to depolarize or hyperpolarize the membrane therefore determining if it will fire or not. But the important principles is that firing is all or nothing. Below a certain threshold, the cell cannot start the positive feedback loop. Once the threshold is reached, the rate of depolarization is self-accelerating and so it acts like a very abrupt switch has been thrown. Overall, the sum of the potentials that are received by the cell body will determine whether or not the post-synaptic action potential is generated. Overall, the membrane gradient is uniform, but the overall potentials will determine the membrane potential. This is called spatial summation, and it is a principle method of computation by neurons.

Another key principle relates to the frequency of signals received by a single pre-synaptic membrane that forms a cleft with a pos-synaptic cell . This is called temporal summation. The frequency of action potentials will determine the signal that a pre-synaptic cell sends. If action potentials rapidly follow each other, they can superimpose to form larger potentials which have a greater influence on the signal it sends. This translates frequency into magnitude of PSP.


Yet this appears to be problematic since the firing of Action potentials is all or nothing. It would appear that a continuously graded variable is integrated over the surface of the neuron to produce an on/off response. If this were the case, then tremendous amounts of information would be lost in the signalling process and the whole exercise would be a huge waste. There must be some way that the neuron encodes the on-off response (fire or don't fire) in the form of a continuously graded variable. It does such a thing, in the form of firing frequency. 

Together, temporal and spatial summation provide the post-synaptic membrane with a compuational integration to form the total post-synaptic potential, the magnitude which is a continuously graded variable. This then needs to be encoded in a method that can be integrated for computation. The frequency of firing of an action potential by a post-synaptic membrane is directly proportional to the post-synaptic potential. This encoding is done at the base of the junction between the axon and the main cell body called the axon hillock. This is the basis for every logical operation performed in the brain.

This provides you with a sound basis of the universal principles of logic gates in neurons, but it doesn’t fully address the higher functions of these processors. One of particular interest is memory. A crucial function of the nervous systems in higher vertebrae is the ability to learn and remember. A key effect of this is Long term potentiation which occurs in neurons which mediates memory. A principle effect of this occurs in cells inside the hippocampus, and if the cells inside are destroyed, the organism seems to lose the ability to form long term memories. Although their ability to recall preexisting memories seems unhindered.

LTP, or long term potentiating is a remarkable ability of neurons in the hippocampus. It refers to a strongly enhanced response of a post-synaptic membrane to a pre-synaptic action potential that results from repeated rapid firing from the pre-synaptic membrane. This can last for days, weeks, etc. depending on intensity. When I say a strongly enhanced response I mean that the magnitude of the post-synaptic potential increases. LTP will occur on a post-synaptic neuron which is already strongly depolarized and that receives a signal from pre-synaptic neuron. If any other synapses are contacting the Post-synaptic membrane that are firing at the same time, those particular synapses will also undergo LTP at the surface of the post-synaptic membrane, even if those pre-synaptic membranes were only firing single action potentials. LTP works by the following steps:

The post-synaptic membrane has glutamate gated Na+ channels and NMDA gated channels which are Ca2+ permeable. The first are only transmitter gated, and therefore are opened when a pre-synaptic potential causes the release of glutamate into the cleft. The latter are a rare TVGIC (transmitter and voltage gated ion channel) that therefore integrate two signals. First, NMDA must be bound, or the channel won't open. Second, the channel has a plug in the form of an Mg2+ ion which can only be removed when the NMDA is bound and the membrane is depolarized. The influx of Ca2+ induces a signal whereby more glutamate TGIC are inserted into the membrane. As such, whenever that synapse fires again, the response of the Post-synaptic membrane is greatly enhanced. The loss or destruction of these cells blocks the formation of long term memories but does not impede recollection of pre-existing memories.



"Physical reality” isn’t some arbitrary demarcation. It is defined in terms of what we can systematically investigate, directly or not, by means of our senses. It is preposterous to assert that the process of systematic scientific reasoning arbitrarily excludes “non-physical explanations” because the very notion of “non-physical explanation” is contradictory.


Books about atheism