Purchase this article with an account.
Pascal Mamassian, Pawan Sinha; Apparent motion from apparent lightness. Journal of Vision 2001;1(3):276. doi: 10.1167/1.3.276.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Apparent motion is perceived when two spatially separated tokens are presented sequentially and if the observer matches the token in the first frame with the token in the second frame. The present study investigates whether this correspondence depends upon the physical brightness of the tokens or upon their perceived lightness. Apparent motion perception was assessed with a display in which two discs placed in opposite corners of a checkerboard were replaced by two discs in the other two corners of the checkerboard. Observers were asked to report the perceived direction of motion (horizontal or vertical) in this bistable display. Dynamic noise was introduced between each trial to reduce the strong hysteresis effect in this type of stimuli. The main independent variable was the aspect ratio of the checkerboard. In addition, the lightness of the discs was manipulated to favour either a horizontal or a vertical motion. While all discs had the same mid-grey luminance, perceived lightness was varied by introducing a shading gradient to the checkerboard background. Therefore, if motion correspondence is based on perceived lightness, the background luminance manipulation should significantly bias the perceived direction of motion. The frequency of vertical motion perception increased with the horizontal-vertical ratio of the checkerboard. Two psychometric curves were derived for the two background luminance manipulations that changed the perceived lightness of the discs. Even though the physical luminance of the discs was kept constant, these two curves were shifted one relative to the other. This result indicates that the lightness of the discs was taken into account during the resolution of the correspondence problem. We have demonstrated that apparent motion is influenced by the perceived lightness of objects. This result suggests that motion correspondence operates on a mid-level representation in which lightness has already been computed. This interpretation is consistent with recent evidence that indicates that apparent motion acts on an intermediate surface-based representation.
This PDF is available to Subscribers Only