Relating language, logic, and imagery

Abstract

The world is a continuum, but words are discrete. Sensory organs map the continuous world to continuous mental models of sights, sounds, actions, and feelings. Those mental models, which represent a moving 3-D virtual reality (VR) are the semantic foundation for all versions of language and logic. A common model for cognition must be able to process and relate all modalities. Kyndi technology represents all information in graphs. They include conceptual graphs for symbolic information and arbitrary graphs for 2-D icons or 3-D VR. All graphs are stored in Cognitive Memory, which can find approximate mappings for arbitrary graphs in logarithmic time. Those mappings include formal unification for logics, and informal analogies for case-based reasoning. The analogies can even map conceptual graphs to the graphs derived from imagery or VR simulations. For reasoning, Peirce’s rules of inference for existential graphs can support operations on arbitrary icons, such as the diagrams in Euclid’s geometry. Those rules, when adapted to conceptual graphs, can map symbolic languages to and from Euclidean style geometrical reasoning. With two new rules of inference, called observation and imagination, the Standard Model of Cognition can support mental models without making any current software obsolete.