Proof of Concept for a Data-Driven Vr Spatial Disorientation Training Tool for Navy Pilots

Abstract

Spatial disorientation (SD) contributes to at least 30% of aviation mishaps, with most being fatal. These mishaps involve not only complex interactions between human, machine, mission, and environment, but also multiple internal neurophysiological systems. Given the complexity of these dynamics, SD competency (i.e., the ability to recognize, confirm, and recover from an SD incident) is hard to predict and teach in the classroom. The Aptima team has developed a data-driven, practical, and realistic virtual reality (VR) training platform for SD called MAVIS, which includes a set of subjective, objective (performance), and physiological measures to rigorously evaluate SD competencies. Preliminary results of this proof of concept confirm the validity of our theoretical framework with a small sample, although rigorous experimentation with a larger sample will be required to validate the approach and justify its operational use. These preliminary results indicated that (1) the MAVIS protocol induces SD; and (2) self-report, performancebased, and neurophysiological measures seem to associate with SD induction and its sequelae.