The Predictive Power of Trajectory Motion
When the central region of an obliquely oriented line is bisected by a wide, vertical opaque occluder, observers misperceive the two line segments as being misaligned (the Poggendorff illusion). If the oblique line segment is replaced with a spot moving on an oblique trajectory, little if any misalignment is perceived. This accurate alignment of oblique segments depends upon the consistent motion of the dot along the oblique trajectory and not other temporal or spatial characteristics of the motion-defined segments since random plotting of the dot along each oblique segment resulted in robust misalignment. The nullification of the Poggendorff illusion was also obtained if only one of the segments was defined by a moving spot so long as the spot moved in a direction that ‘pointed’ to the static segment. Moreover, if the occluder boundary was defined by rows of vertically moving dots, was filled with vertically moving dots or was a real (cardboard) occluder, the motion-defined oblique segments were still perceived to be aligned with little error, consistent with the unimpaired detection of a trajectory dot in noise interrupted by similar occluders [Watamaniuk, S. N. J. & McKee, S. P. (1995). ‘Seeing’ motion behind occluders. Nature, 377, 729–730]. The results are interpreted as evidence that trajectory motion produces a cascade of activity in appropriately aligned motion detectors, in the direction of motion, that continues after the moving object has been occluded to produce a prediction of where the moving object should reappear.
Watamaniuk, S. N.
(2005). The Predictive Power of Trajectory Motion. Vision Research, 45 (24), 2993-3003.