Artificial Neural Systems and Decision Behavior in A Dynamic Environment

This paper explores the role of neural networks for decision making in dynamic environments which are characterized by risks and uncertainties, and also provides experimental evidence from a simulated data. Theoretical support is derived from theories of affective balance, and self‐organized criticality. The simulation is conducted for a two‐person‐constant sum game. The findings of the experiment are helpful in extending to managerial decision making which involves varying degrees of uncertainties. Such decisions are affected by forces both internal and external to the company, and making judgments in such a fuzzy future is highly probabilistic. It is suggested, therefore that neural networks are better able to capture the interactive dynamics of variables operating in a managerial decision environment. In sum, the findings indicate that decisions in general and business decisions in particular can greatly benefit from the parallel computational capabilities of neural networks.