Abstract for: Perception is more than time delays
Many system dynamics models include perception as a factor translating actual into observed conditions. It is assumed that true conditions are not available to decision makers and thus, should not be used directly in decision policies in the model. Instead, perceived conditions are used. Typically, perception is modeled by delaying changes in underlying variables. That is, provided that the underlying variable does not change anymore, the perceived value may eventually reach the true condition. In decision sciences and psychology, however, errors in perception are described that lead to systematic differences between actual and perceived conditions. Some of these biases and heuristics prevent a decision maker from ever arriving at the true condition, even if the latter remains constant over time. In this paper, we attempt to broaden the definition and use of perception in system dynamics models. To this end, we present a general model on perception and compare this to a belief updating model formulated in cognitive psychology. We give examples of particular biases and heuristics and how these may be captured in model structures. These model structures may be used in addition to information delays to capture a broader range of perception errors in human decision making.