MODELLING NATURALISTIC DECISION MAKING USING AN ARTIFICIAL NEURAL NETWORK: PILOTS' RESPONSES TO A DISRUPTIVE PASSENGER INCIDENT

This paper describes a study conducted within a naturalistic decisionmaking paradigm in which a disruptive passenger threatens the safety of a hypothetical air flight. 65 professional members of flight crews participated in semi-structured interviews during which they described their decisionmaking process for dealing with this situation. An artificial neural network was used to model the decisions made on the basis of the situation assessment activities undertaken to produce an empirically verifiable model of the participants' decisionmaking process. Cross-validation of results showed that decision outcomes could be very accurately predicted on the basis of this model. It is suggested that neural networks may be a viable way of modeling naturalistic decisions.