When Expertise Fails: Designing for High Uncertainty Decision Making in Virtual Worlds
(Room S320E)
27 Nov 18
5:00 PM
-
5:30 PM
High risk, high volatility task environments require extensive hands-on training, detailed protocols and the experience of expert decision makers who can anticipate, adapt and respond effectively to complex situations. But what happens when even expert-level decision making reaches its limits? How do people behave when protocols and best practices fail to account for unanticipated sources of risk? How can we develop better learning paradigms to address training needs in environments where the best course of action is unknown? Operators in these environments know that experience is key: exposure to repeated, high risk events helps them develop sophisticated response strategies. But what happens when novel challenges render expert decision models obsolete? Immersive simulations and wargames can create ideal learning environments to address this issue, by rapidly accelerating exposure to a variety of rare, nonroutine, or hazardous events. However, two primary issues constrain their ability to deliver effectively: First, many of
these programs are focused on replicating events that can assure proficiency among operators. Using the Dreyfus five stage model of expertise, we argue that immersive simulation is most effective when it pushes decision makers beyond proficiency assurance and is able to destabilize their mental models, challenging them toward innovative approaches to chaotic situations. Second, training needs in industry, military and government organizations are often highly constrained by time, money, capacity, schedule, and staffing requirements. Immersive, multi-player Virtual Worlds allow us to relieve some of this training burden by providing a platform to sandbox novel techniques for expert decision making in extraordinary conditions. However, many limitations and challenges still remain. Drawing from our own implementation efforts, we will discuss some of the successes, failures, constraints and opportunities virtual environments offer as a platform to d