The research of this Presidential Young Investigator is devoted to advancing the theories of effective information display and applying these theories to new areas. The first area to be studied is user interfaces for expert systems. Presently, maintaining the rules of an expert system as its database is modified and updated is a difficult and highly resource intensive activity. With new techniques, the user interface could be improved and maintenance time decreased. An ultimate goal of this research is to develop techniques to translate the rules of expert systems into a higher-level language that is more compatible with the way that people think. Prior research indicates that humans do not normally manipulate rules when solving a problem. Instead, humans seem to have well-practiced abilities in the manipulation of spatial information. These manipulations can often be summarized in terms of simple rules. Thus, the rules are a byproduct of the information processing instead of a direct method of mental manipulation. So, an expert system interface which is compatible with human information processing might translate the rules into spatial entities to be manipulated directly by the user.
