A Multi‐Stage Model of Decision Bias: Implications for Expert Systems

The study of the bias processes that affect decision making is crucial in designing expert systems. This study proposes a multi‐stage model for decision biases which reconceptualizes cognitive styles and decision heuristics within a framework that borrows heavily from research by Posner and McLeod (1982), Tversky and Kahneman (1973) and Ramaprasad (1987). The framework is tested within the context‐bound area of loan making so that the biasing effects of prior experience on decision making can be examined. The results are analyzed with a path‐modeling technique (i.e. covariance structural modeling) that allows testing for indirect as well as direct effects. The results are discussed in terms of the implications for expert systems development