Search

Motion simulators are widely employed in basic and applied research to study the neural mechanisms of perception and action under inertial stimulations. In these studies, uncontrolled simulator-introduced noise inevitably leads to a mismatch between the reproduced motion and the trajectories meticulously designed by the experimenter, possibly resulting in undesired motion cues to the investigated system. An understanding of the simulator response to different motion commands is therefore a crucial yet often underestimated step towards the interpretation of experimental results. In this work, we developed analysis methods based on signal processing techniques to quantify the noise in the actual motion, and its deterministic and stochastic components. Our methods allow comparisons between commanded and actual motion as well as between different actual motion profiles. A specific practical example from one of our studies is used to illustrate the methodologies and their relevance, but this does not detract from its general applicability. Analyses of the simulator inertial recordings show direction-dependent noise and nonlinearity related to the command amplitude. The Signal-to-Noise Ratio is one order of magnitude higher for the larger motion amplitudes we tested, compared to the smaller motion amplitudes. Deterministic and stochastic noise components are of similar magnitude for the weaker motions, whereas for stronger motions the deterministic component dominates the stochastic component. The effect of simulator noise on animal/human motion sensitivity is discussed. We conclude that accurate analyses of a simulator motion are a crucial prerequisite for the investigation of uncertainty in self-motion perception., Comment: 24 pages, 6 figures, 1 table