Neurons as decision-making devices

Separate synaptic potentials are too small to activate the voltage-dependent sodium channels therefore they do not trigger action potentials. Rather than they are passively conducted over the nerve cell membrane, getting smaller both with time and distance as they extend. This decay of little potentials is determined solely by the physics of the neuron. In general, the smaller the diameter of neuron, axon, or dendrite along with a potential is spreading, the minimum is the distance over which it will decay, and faster this will occur, and the slower the potential is conducted. This is very critical in determining how neurons integrate their inputs and how information is processed in the nervous system. Additionally it accounts for why action potentials that do not decay in time and distance are required for long-distance transmission. The Synaptic potentials decay to zero in a few millimeters in most of neurites, and hence cannot carry information to great distance. Though, some short interneurons (example, those in the retina) do not fire the nerve impulses, but rely on the synaptic potentials for transmission along with their neurites.

Numerous thousands of synapses are created on each neuron, both excitatory and inhibitory. At any certain time a subset of these will be activated to produce epsps and ipsps. The special property of these graded potentials is that they summate, or add altogether. Whenever a sufficient number of epsps are formed, then in summing they will drive the axon hillock membrane potential across the threshold for triggering action potentials and the neuron will fire. The axon hillock is critical as, being the area of a neuron with the highest density of voltage-dependent sodium channels, it has the lowest threshold. When at any instant deficient excitatory synapses are activated, and a high level of excitatory synaptic input is more than the offset by production of ipsps from inhibitory input, then the axon hillock will never be driven across the threshold and the cell will not fire. Therefore, the neurons are decision-making devices. The decision—to fire or not—is really taken by the axon hillock on the basis of whether the total sum of epsps and ipsps causes its membrane potential to become more positive than the firing threshold. It is the operation which constitutes information processing by individual neurons. In engineering words, a synapse converts the digital signals into analog ones (i.e., postsynaptic potentials). The neuron then integrates all its analog signals over a short time and compares the outcome of that integration with a given threshold to decide whether to fire. Whenever it does fire the output is digital.

The experiments on pyramidal cells illustrate that about 100 excitatory synapses, on average, should be activated at similar time to trigger an action potential. Though, the efficacy with which a synapse can influence firing totally depends on its position. As postsynaptic potentials decay as they spread passively towards axon hillock, the synapse far out on distal dendrite will have less effect than one closer to the cell body. In this situation it is noteworthy that on pyramidal cells there are only about 250 inhibitory synapses on the cell body although 10000 or so excitatory axodendritic synapses. Relative strength of the synapse in contributing to the neuron’s output is its weighting. This require not be a fixed property but may change with the time.