Abstract

By focusing the major functional operations of association cortex, we propose a three-level neural model to represent the mental activity. The model consists of the basic neuron level, the functional module level, and the module organization level. Notice that the basic neuron performs the physiological functions; the functional module is constructed with certain types of neuron as stimulus transformation; and the module organization operates the intelligent functions by interactions between the modules. Since the sensible information is encoded as the density of action potentials by the sensory, the mental information processes can be represented by potential-density distributions over the association cortex. Hence the mental operations can be analyzed through the potential-density transformation. A new concept, called sensation, is also introduced to indicate the distributed density of action potentials. With this concept, the mental activity, such as long-term memory, short-term memory, abstraction, comparison, and language, can be described as the sensation transformation and the sensation group transformation in the module organization level. The mathematical algorithms to analyze the mental activity within this model are given. The discrete algorithm can be used for analyzing the operations of individual modules, while the continuous algorithm describes the sensation transformation and sensation group transformation.

© 1988 Optical Society of America