Pilot Task Management: Testing an Attentional Expected Value Model of Visual Scanning

ARL 1 ABSTRACT Two models of information acquisition in visual scanning are described. A descriptive model identifies the role of event salience, effort, expectancy and value in influencing where and when people look at different channels to sample information in dynamic environments. An optimal prescriptive model accounts for the role of expectancy and value, as these characterize the properties of channels necessary to serve tasks that may differ in their importance. The prescriptive model is validated against the data from four experiments in which skilled pilots flew a high fidelity visual flight simulator, while engaged in traffic detection with (Experiments 1, 3, and 4) and without (Experiment 2) a cockpit traffic display, and with different forms of data link displays (Experiment 4). All four experiments provided a good fit between model predictions and the percentage of time that pilots spent viewing different areas of interest in the airplane environment. Implications of the model to training optimal strategies, and to possible refinement of training policies are described.