Micdroscopic computation in human brain evolution

Abstract

When human psychological performance is viewed in terms of cognitive modules, our species displays remarkable differences in computational power. Algorithmically simple computations are generally difficult to perform, whereas optimal routing or „Traveling Salesman”︁ Problems (TSP) of far greater complexity are solved on an everyday basis. It is argued that even „simple”︁ instances of TSP are not purely Euclidian problems in human computations, but involve emotional, autonomic, and cognitive constraints. They therefore require a level of parallel processing not possible in a macroscopic system to complete the algorithm within a brief period of time. A microscopic neurobiological model emphasizing the computational power of excited atoms within the neuronal membrane is presented as an alternative to classical connectionist approaches. The evolution of the system is viewed in terms of specific natural selection pressures driving satisfying computations toward global optimization. The relationship of microscopic computation to the nature of consciousness is examined, and possible mathematical models as a basis for simulation studies are briefly discussed.