Purchase this article with an account.
Ross Goutcher, Gunter Loffler; Motion transparency in combined first and second order stimuli. Journal of Vision 2005;5(8):142. doi: 10.1167/5.8.142.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Two superimposed luminance gratings of identical orientation and opposite direction of motion are seen as moving across one another (i.e. moving transparently) only if they are of different spatial frequencies (SF). Identical SF gratings produce counter-phase flicker. This suggests that opposite motions cancel each other within SF channels. Here we show that transparent motion is perceived with two superimposed gratings of identical orientation and SF, when one component grating is a first-order stimulus, whilst the other is a second-order stimulus. More precisely, this stimulus is the sum of two identical dynamic 2D noise carriers, one subject to luminance modulation (LM) the other subject to contrast modulation (CM) by a sinusoidal envelope (SF = 4.6cpd, temporal frequency = 1.6Hz). The perception of this stimulus is examined in a 2AFC task. Participants were presented with a LM+CM stimulus, where the motions of the component gratings were either in the same or opposite directions. The amplitudes of the CM and LM gratings were varied (10% – 22.5% and 5% – 25% contrast, respectively) and participants were asked to choose the interval containing transparent motion. Results show that the amplitude of the LM component affects the CM amplitude required for the perception of transparency: higher LM amplitudes require higher CM amplitudes before transparency is seen. In a second experiment, participants adapted to a transparent LM+CM stimulus. Following adaptation, participants showed elevated contrast detection thresholds when the direction of motion of a translating CM or LM test grating was identical to that of the corresponding component of the adaptor. These findings support the idea of separate pathways for the detection of first and second-order motion. However, the observed contrast dependency suggests that these pathways are not wholly independent. Our results further suggest that second-order information may aid the segmentation of multiple motion signals.
This PDF is available to Subscribers Only