neuron

A typical vertebrate neuron responds to the many sources of electrical impulses that impinge on its cell body and on its branches -- its "dendrites" -- in three ways. Some inputs excite the neuron, some inhibit it, and some modulate its behavior. If the neuron becomes sufficiently excited, it respondes ("fires") by sending an electrical pulse or "spike" down its output cable--its axon.
Neurons are rather different from most cells. Mature neurons do not move about, nor do they divide. If a mature neuron dies, it is rarely replaced by a new one. Neurons have a more spikey shape than most cells, and the axon of a neuron can be very long -- as much as several feet. Neuroscientists today believe that the brain records an event by strengthening the connections between groups of neurons that participate in encoding the experience. (see engram)

Warren McCulloch and Walter Pitts developed a theorem for the activities of the neuron that abstracted the pinkish-gray tissue of the brain and its known electrical, chemical, and anatomical properties to the construction of formal neural nets isomorphic to the relations of propositional logic and showed how neurons might represent logical assertions. For instance, if pulses travelling along either of two excitatory fibers suffice to cross the synaptic gap, then this physical situation corresponds to the formal statement that "If either A or B is true, then C is true." A different logical statement corresponds to the situation where fiber A excites the neuron at regular intervals, but fiber B is inhibitory: "Assuming A is always true, C is true whenever B is not true."

This "all or nothing" model of excitition lends itself to digital rather than analog interpretation.

The essential result of the Pitts-McCulloh analysis has been summarized as follows: "That anything that can be completely and unambiguously described, anything that can be completely and unambiguously put into words, is ipso facto realizable by a suitable neural network."

Still, how do the neurons of the brain work in a coordinated and time-sensitive manner? While many activities of the brain are functionally segregated and localized, activity is synchronized and coordinated despite the lack of a central coordinatory area. Gerald Edelman calls this capacity reentry. Edelman claims to have "solved" consciousness as a process of neural Darwinism, in which groups of neurons compete to form an effective representation of the world. He calls this the theory of neuronal group selection. (TNGS)