ABSTRACT
The problem of decision making is a matter of directing and maintaining the continuous flow of behavior towards some set of goals rather than as a set of discrete episodes involving choice dilemmas. This chapter considers Bayesian, regression, and heuristic approaches to decision making and decision boundaries in dynamic contexts in order to explore some of these parallels. In the context of manual control, the primary use of Bayesian models is in the Kalman filter, which is a Bayesian estimator. One analytic tool for understanding human decision making is multiple regression. Multiple regression is a mathematical technique for predicting one variable from a weighted sum of observations. Linear describing functions of human tracking behavior are analogous to linear regression models of decision making. The chapter focuses on just a few parallels between control theory and decision making.
