Human decision-making theories and formal models are increasingly used for developing advanced ICT-based intelligent systems and services. Decision filed theory (DFT) is one of the decision-making theories that has significant potential for practical applications in real-world decision-making situations. Successful empirical studied have shown that DFT theory is able to explain human decision-making behavior in real situations, and the model can be applied as a basis for ICT system and service design. In this chapter, the authors present the results of a systematic literature review conducted for analyzing and synthesizing the evidence of DFT development and its validated usage in different application areas. The results show that the interest in DFT and its applications has grown strongly during the last years. The basic model has been extended to cover more complex decision-making situations and its applications have been widening.
TopBackground
Decision Field Theory, initially proposed by Busemeyer and Townsend (1993), is a dynamic cognitive approach to describe and explain human decision making under risk and uncertainty. It provides a formal theory and a mathematical model of human decision behavior, which takes into account two of the basic phenomena of human decision making that are not covered by most of the other theories: variability of preference of choices and systematic relation between preferences and deliberation time (Busemeyer & Townsend, 1993). Before the theory has been applied to solve multiple options problems of decision making, it has experienced several stages that are the foundation for present theory. The early deterministic subject expected utility (SEU) theory is restricted to the choices between two options (Savage, 1954). This theory can be applied to predict the trend of preferences on options, but it can’t explain the dynamic change of preference. It was followed by the Sequential SEU Theory (DeGroot, 1970), which take into account the sequential order of consecutive decisions. The main problem of the Sequential SEU theory is that the knowledge and experience in the past has no influence on current decision making, which goes against with real human deliberative process. The Random Walk SEU Theory (Busemeyer, 1985) solves this problem by setting an initial preference state. The deterministic decision making theories originate from the basic proposal that the action is either true or false between two conflict actions, which is referred to binary preference relation (Fishburn, 1988). In subsequent, it was used for two or more attributes decision making under dynamic environment (Diederich, 1997). After that, DFT was gradually mature and widely used.
DFT is a method that provides insights into relationship between preference evolution and dynamic decision making. The following original and classic mathematical equation is used to explain the dynamic evolution of preferences during the deliberation time.
In this equation, P is the Preference vector, S is the Growth-decay matrix and V the Valence vector, which are explained below. The equation can be decomposed further as