Tailoring decision support to individual users

Abstract

Computer-based systems have been developed to help people make decisions. Only some of these systems try to tailor support to the individual user; even fewer try to assess a user's needs within the framework of a psychological theory. Behavioural decision theory (BDT) provides a framework for assessing individual needs. Models of individual users can be created using existing methodologies developed in studies of single-process BDTs (e.g., the linear model). We have developed a new methodology for creating such models within the framework of constructive process BDTs. Our methodology estimates an individual's decision strategy. Tests of this methodology show it is very successful in classifying participants as using one of several strategies. We discuss the implications of these findings for the design of future decision support systems.