Abstract
Designers of training-simulators have traditionally depended on task-analysis as the main informational resource for design. Although traditional task-analysis provides information about the necessary details of a task and the required steps in performance, it is deficient in representing a skill and its qualitative dimensions. This deficiency has led to lingering confusions for designers and resulted in using challenging constructs such as fidelity. To provide richer information about skills, this article proposes a novel design method. The method gathers information of skills directly from studying multiple experts by videorecording their performance in target environments. The recordings are then used to create a model of average expert behaviour that can directly guide the design of simulators. An experimental example shows the practicality of the method in providing specific information for human factors/ergonomics practitioners in the design and improvement of training-simulators. By focussing on experts and videorecording their performance, the method provides multiple benefits to the design such as increasing the validity of the performance model, creating a qualitative model of skills, and proposing a generic solution to the design problem.