Motion noise removes the effect of visual path information on human heading estimation

Abstract

We have shown that visual path information helps heading estimation when the display does not contain sufficient optic flow cues (see Li et al, 2007 Perception 36 Supplement, this issue). Here, we investigate whether adding motion noise, which disrupts the reliability of path information, to environment points similarly affects heading perception in displays with or without path information. The display (110 deg694 deg) simulated motion on a circular path and observers used a joystick to rotate their line of sight until deemed aligned with true heading. Motion noise, as assigned random velocities of environmental points, was introduced to the display at levels of 0%, 25%, 33%, and 50%. Heading precision decreased monotonically with increasing motion noise in a similar manner for displays with and without visual path information (mean heading precision ± SE across six observers: 2:9 ± 0:4 deg vs 3:0 ± 0:2 deg at 0%, 4:1 ± 0 deg vs 5:5 ± 0:6 deg at 25%, 6:4 ± 0:4 deg vs 7:3 ± 0:8 deg at 33%, and 12:8 ± 0:8 deg vs 13:4 ± 0:5 deg at 50%). Motion noise is effective in removing the influence of path information on human heading estimation.