Modelling the emergence of design descriptions across schemata

An exploration of how connectionist systems can be used to model important aspects of design reasoning is presented: the way design appears to involve memory; and a 'holistic' kind of reasoning by which designs appear to emerge from that memory. A simple connectionist system is used to demonstrate how information about schemata (in this case, subfloor construction types) is stored implicitly after the system is exposed to a number of specific examples. New schemata can emerge from this information. The paper includes a discussion of design, a comparison between a connectionist and a frame-based approach to representing design knowledge, a technical explanation of a connectionist system, and the demonstration of an implemented example.