Purchase this article with an account.
Gene R. Stoner, Thomas D. Albright, Jay Hegde'; Depth order perception in first- and second-order motion stimuli. Journal of Vision 2003;3(9):799. doi: https://doi.org/10.1167/3.9.799.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Dynamic cues for depth ordering of surfaces include (1) progressive uncovering and/or covering of the background texture by the foreground surface (‘accretion-deletion cue’) and (2) the fact that texture elements of a foreground surface should move at the same velocity as the boundary between foreground and background (‘boundary flow cue’).
We studied depth order perception using three first-order and three second-order motion stimuli. One first-order stimulus offered both depth cues and the other two first-order stimuli offered only the boundary flow cue. Two second-order stimuli (defined by either contrast modulation or ‘flicker’) offered only the accretion-deletion cue. The other second-order stimulus offered neither depth cue and was defined by both contrast modulation and flicker. Na?ve subjects (n=10) viewed stimuli and indicated perceived depth order by a key press. For stimuli with the accretion-deletion cue, all subjects reported the depth order that was consistent with that cue (p < 0.01), regardless of the presence of the boundary flow cue and whether the stimulus was first- or second-order. Accretion-deletion thus appears sufficient for depth order perception.
The efficacy of the boundary flow cue varied considerably across subjects, with some subjects showing no effect. No consistent depth order bias was evident when neither depth cue was available. In addition to highlighting the importance of the accretion-deletion cue relative to the boundary flow cue, our results show that second-order motion defined by either contrast modulation or flicker provides depth order information. Previous neurophysiological studies have found that some visual neurons respond to these types of second-order stimuli. This study suggests that such responses may encode information about depth order as well as motion.
This PDF is available to Subscribers Only