Abstract
The vast majority of studies on motion in depth perception used horizontally moving dots as stimuli. Here we measure perceived azimuth and elevation angle of an oriented line moving in 3D. Although motion direction of an oriented line is highly ambiguous three observers gave similar and reliable estimates of 3D motion direction.
In two experiments we presented moving line stimuli to the left and right eye on calibrated flat CRT monitors with a refresh rate of 120 Hz in a two-screen Wheatstone configuration. On each trial observers verged on a fixation cross flanked by nonius lines at a viewing distance of 55 cm. An oriented line travelled back and forth on a 3D trajectory inside a circular aperture. Line orientation (±45 and 90 deg) and orientation disparity (−6 to +6 deg) was independently varied in randomly intermixed trials. In each open-loop trial the observer repeatedly viewed the line moving in depth before a string of dots was superimposed inside the aperture. The observer adjusted orientation and horizontal disparities of the string of dots until they matched perceived 3D motion direction of the oriented line.
The results indicate systematic effects of 3D line orientation on perceived azimuth and elevation angle. We discuss whether our findings are compatible with intrinsic/extrinsic borders and predictions from generalized motion-in-depth models (Lages & Heron, VSS 2009).