Abstract
Although estimating travel distance is essential to our ability to move through the world, our distance estimates can be inaccurate. These odometric errors occur because people tend to perceive that they have moved further than they had. Many of the studies investigating the perception of travel distance have primarily used forward translational movements, and postulate that perceived travel distance results from integration over distance and is independent of travel speed. Speed effects would imply integration over time as well as space. To examine travel distance perception with different directions and speeds, we used virtual reality (VR) to elicit visually induced self-motion. Participants (n = 15) were physically stationary while being visually "moved" through a virtual corridor, either judging distances by stopping at a previously seen target (Move-To-Target Task) or adjusting a target to the previous movement made (Adjust-Target Task). We measured participants' perceived travel distance over a range of speeds (1-5 m/s) and distances in four directions (up, down, forward, backward). We show that the simulated speed and direction of motion differentially affect the gain (perceived travel distance / actual travel distance). For the Adjust-Target task, forwards motion was associated with smaller gains than either backward, up, or down motion. For the Move-To-Target task, backward motion was associated with smaller gains than either forward, up or down motion. For both tasks, motion at the slower speed was associated with higher gains than the faster speeds. These results show that transforming visual motion into travel distance differs depending on the speed and direction of optic flow being perceived. We also found that a common model used to study the perception of travel distance was a better fit for the forward direction compared to the others. This implies that the model should be modified for these different non-forward motion directions.