The Pattern Extraction Architecture: A Connectionist Alternative to the Von Neumann Architecture

A detailed connectionist architecture is described which is capable of relating psychological behavior to the functioning of neurons and neurochemicals. The need to be able to build, repair and modify current electronic systems with billions of hardware components has been met through a seldom appreciated aspect of the von Neumann architecture: the hardware architecture is compatible with a simple functional architecture which can support precise translation between functional descriptions at many levels of detail down to the individual hardware components, through the use of a common functional element, the instruction. Existing neural network models have been developed to simulate different aspects of cognition, but do not offer a behavioral architecture analogous with the von Neumann architecture. The brain has experienced intense evolutionary selection pressures analogous with the requirements to build, repair and modify electronic systems. These pressures have resulted in a neural architecture which is compatible with a simple functional architecture based on the use of the common functional element, the pattern extraction. The pattern extraction functional architecture is the basis for intellectual understanding of brain functioning. Physiological structures can be understood as an efficient partitioning of function within the constraints imposed by the properties of neurons. Behavioral phenomena such as declarative memory can be understood in neuron terms, including the reasons for distribution of memory traces. A range of physiological and psychological evidence is discussed. Electronic simulation demonstrates that key psychological functions can be emulated by the architecture.