Title:   Are simulators just for fun?

Simulators are complex machines that incorporate many different electrical and mechanical sub-systems controlled by an array of computers in hard real time. The purpose of the simulator seems obvious: to allow a human an as close as possible experience of the "real thing", to train, study, or entertain that person. Simulators were built since many years and have become very sophisticated, to the point that a pilot can log flight hours in a simulator. The general question is: how can we make them more effective in training people, how can we avoid that we train something wrong? Do we really need self-motion to train people? Interestingly this is not clear and not easy to answer. Due to the limited access to full motion simulators much research was focused on the specific aspects for training to fly or operate machines. Many fundamental questions are open, some of the design decisions seem ad hoc. In this talk I will explain how our simulator designed to allow us to conduct experiments by presenting stimuli including vision, motion, and sound; and recording responses including tactile responses, EEG and eye tracking with precision timing. We cover about experiment design, data collection, and evaluation and some results. A simulator challenges it's researchers in many areas, from mechanical vibration, over control and signal processing to the knapsack problem are encountered. This work is truly interdisciplinary, a close collaboration between Neuroscience and Engineering needed to be developed.