Abstract
Purpose: People easily and accurately identify the movement of points of light that correspond to the joints of a walking person. Although it is known that motion of additional moving points of light can mask the motion of a point light walker, little is known about what aspects of the point light walker influence its detectability within a mask. Method: Twenty-nine naive participants viewed artificially created point light walker sequences with masks for approximately one second. Three properties of point light walkers within masks were examined: 1) translation (translation versus no translation), 2) location (static versus random point light walker location across trials), and 3) orientation (upright versus inverted). Each participant was randomly assigned to a given combination of the translation and location variables. For the particular translation-location combination assigned, each participant completed 100 trials at each orientation. Results: Statistical analyses showed that all variables influenced the detectability of point light walkers (as measured by d'). Participants were better at detecting the point light walker when it was not translating (that is, walking in place), its location was static, and it was upright. There were also statistically significant interactions (all two-way and the three-way) between translation, location, and orientation. Conclusions: The detectability of point light walkers within a mask depends on their orientation, translation, and location. Although a person walking through our field of view is a common occurrence, it is apparently easier to mask that motion (a point light walker with translation) than the presumably less common occurrence of someone walking in place (a point light walker without translation).