Generating Flight Illusions Using Galvanic Vestibular Stimulation in Virtual Reality Flight Simulations.

Abstract

Vestibular flight illusions remain a significant source of concern for aviation training. Most fixed-based simulation training environments, including new virtual reality (VR) technology, lack the ability to recreate vestibular flight illusions as vestibular cues cannot be provided without stimulating the vestibular end organs. Galvanic vestibular stimulation (GVS) has long been used to create vestibular perception. The purpose of this study is to evaluate the ability of GVS to simulate common flight illusions by intentionally providing mismatched GVS during flight simulation scenarios in VR. Nineteen participants performed two flight simulation tasks-take off and sustained turn-during two separate VR flight simulation sessions, with and without GVS (control). In the GVS session, specific multi-axis GVS stimulation (i.e., electric currents) was provided to induce approximate somatogravic and Coriolis illusions during the take-off and sustained turn tasks, respectively. The participants used the joystick to self-report their subjective motion perception. The angular joystick movement along the roll, yaw, and pitch axes was used to measure cumulative angular distance and peak angular velocity as continuous variables of motion perception across corresponding axes. Presence and Simulator Sickness Questionnaires were administered at the end of each session. The magnitude and variability of perceived somatogravic illusion during take-off task in the form of cumulative angular distance (p < 0.001) and peak velocity (p < 0.001) along the pitch-up axis among participants were significantly larger in the GVS session than in the NO GVS session. Similarly, during the sustained turn task, perceived Coriolis illusion in the form of cumulative angular distances (roll: p = 0.005, yaw: p = 0.015, pitch: p = 0.007) and peak velocities (roll: p = 0.003, yaw: p = 0.01, pitch: p = 0.007) across all three axes were significantly larger in the GVS session than in the NO GVS session. Subjective nausea was low overall, but significantly higher in the GVS session than in the NO GVS session (p = 0.026). Our findings demonstrated that intentionally mismatched GVS can significantly affect motion perception and create flight illusion perceptions during fixed-based VR flight simulation. This has the potential to enhance future training paradigms, providing pilots the ability to safely experience, identify, and learn to appropriately respond to flight illusions during ground training.