Scene scaling during simulated forward accelerations: Are explicit size cues used?

In many applications for motion simulators, movements which are considerably larger than the motion range of the actual setup are simulated. An important question is therefore how to design inertial movement envelopes and visual scenes so that the simulated movements are perceived as veridically as possible. In the study presented here, we investigated how participants derive an estimate of a visual forward acceleration and how they compare it to an inertially simulated forward acceleration (cross-modal matching). We tested two possibilities: participants could either derive an absolute estimate of visual self-acceleration from objects of known size in the visual scene (images of people), which they compare to an estimate of absolute inertial acceleration; or they could resort to ’range matching’, where they do not use absolute estimates but separate relative estimates of visual and inertial cues in their respective stimulus ranges, which they then compare. We found that in our experiment subjects’ responses were indeed largely consistent with a ’range matching’ strategy. Explicit visual size cues were only used by 4 of the 12 participants, and across all participants, size cues did not have a significant effect on visual-inertial matching. We conclude that when exposed to linear accelerations in a simulator environment, participants do not necessarily use visual size cues to derive an absolute estimate of the visual acceleration, but that most adapt to the range of presented stimuli, and perceive visual and inertial accelerations as ’matching’ if they have similar relative magnitudes. This suggests that one should be careful when interpreting results of cross-modal magnitude matching experiments, since they might be strongly influenced by the ranges of stimuli presented in the experiment