Behavioural and neural correlates of performance and expertise in an immersive virtual reality motion simulator

点击翻译按钮获取中文摘要

Powerful vehicles like airplanes and automobiles can be dangerous when operated without great skill. That skill can be acquired through simulation training - an approach that is especially important in complex and potentially dangerous situations where there is too much risk for real practice. The use of simulated environments for training is a growing and critically important area of research. Advances in technology have made training simulators more available, but we still do not know which aspects of a simulated experience lead to effective learning. Without this knowledge simulator training can be ineffective, resulting in costly mistakes and even loss of life. One important question is whether training simulators should incorporate physical motion. When we navigate through space, vision is our dominant sense, and visual cues provide a strong perception of self-motion. Although we often underestimate the contribution of kinesthetic perception, vestibular and proprioceptive cues are highly integral to our experience. These motion cues play a critical role when learning to operate moving vehicles. Transport Canada Civil Aviation licensing requires that pilots train in a motion simulator, but outside of aviation, most simulators have limited or no motion, in part because it is expensive and difficult. We need to know what is (and is not) necessary to create effective training experiences in simulated environments. Conventional wisdom in simulator design has been to strive for higher realism and fidelity in reproducing the environment, but some sensations may not contribute to training and can present problems for multisensory integration. Mismatches between visual, vestibular, and other motion cues cause discomfort and illness, which disrupts learning. Rather than trying to fully recreate physical motion, we need to learn how to simulate the perceptual experience of physical motion. To this end, we must understand multisensory integration in simulated environments. This project will use behavioural and brain measures to reveal the dynamic range of visual and motion perception within which successful integration occurs, and to learn how to build effective training simulators. Most of what we know about the neural correlates of vestibular motion perception comes from studies that isolate specific vestibular components. While these tightly controlled, laboratory experiments are crucial, our interest is to measure perception and integration while people are immersed in learning to drive or fly. Our perception of self-motion is highly sensitive to information presented to multiple senses. We combine engineering and neuroscience expertise, developing technology and tools to study learning and performance in a motion simulator under different training scenarios. We use electroencephalography to measure brain responses, we monitor head and eye movements, and we collect physiological measures of stress responses. Using these measures, together with a range of behavioural observations, we will build a taxonomy of behavioural and brain correlates of motion perception and multisensory integration. We will use these tools to advance understanding about what aspects of an immersive environment lead most effectively to skill acquisition, and to improve transfer of training from simulation to real life. We aim to apply this new understanding to a new generation of design for training simulators. The work will lead to technological advances and safer and more effective training. As our population ages, so do our drivers; there is evidence that older adults rely on multisensory integration to a greater extent than young adults. This work will lead to important new methods of evaluation and re-training for elderly drivers.