Conscious percept formation using fuzzy entropy measures of neuronal multiplex signals

摘要

One goal of automation is to mimic the smoothness and efficiency of human performance. Fuzzy logic based soft computing and engineering works towards this goal. The obvious difference between machines and humans is that of life. If one could understand the method by which human perception and motor response takes place one might conceive of a means to automate these capabilities. The anatomy and physiology of the living nervous system are notable in this respect. In the mammalian brain, the pyramidal neuron of the cerebral cortex plays a key role in perception. Pyramidal cell axons exhibit clusters of action potentials that form a multiplex code allowing multiple parallel patterns of information to travel in the same time frame along that axon. Selective decoding at different target locations within the central nervous system then takes place. We propose that the anatomy of the cerebral cortex and pyramidal neuron is uniquely suited to distribute a multiplex signal, and that this property then provides the basis for a common code for perceptual and motor representations. Thinking of the anatomy in terms of geometry, we are able to predict the increase in length of the apical dendrites of pyramidal neurons from cortical layers 2 to 6 on the basis of a property of the doubled fuzzy hypercube (unit square). A decrease in fuzzy entropy of incoming signals takes place as engrams and new signals arrive bband are passed down through these layers to pyramidal cell layer 5. Perceptions and motor action plans are separated out by demultiplexing so that conscious percepts are executed by the thalamoreticular system and corresponding motor plans by the corticospinal tract. This is possible because of differential demultiplexing mediated by inhibitory gabaergic neurons of the same multiplex signal at the relevant target areas. The important feature of this process is its plasticity, due to synaptopoeisis which is a continuous remodeling process of synaptic location.