Purchase this article with an account.
Bochao Zou, Igor Utochkin, Yue Liu, Jeremy Wolfe; Binocularity and Visual Search – Revisited . Journal of Vision 2016;16(12):1283. doi: 10.1167/16.12.1283.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
How does dichoptic information guide attention in visual search? Wolfe and Franzel (1988) reported that binocular rivalry could guide attention only weakly but that binocular luster (shininess) "popped out", producing very shallow RT x set size functions. We have revisited the topic with new and improved stimuli and found that our lustrous stimuli did not always pop-out while search for rivalry can be quite efficient. Stimuli were presented on an imaginary circle. In Experiment 1, ~1.5 deg. gratings were presented to each eye: Orthogonally oriented to produce rivalry, same orientations to produce fusion. Search for rivalrous targets was more efficient than search for fused targets but still inefficient (RT x setsize slopes: 29 vs. 58 msec/item). In Experiment 2, each stimulus was a 2x2 checkerboard of vertical and horizontal gratings so every stimulus contained both orientations. Now search for rivalrous pairs was much more efficient (12 msec/item). The use of checkerboard stimuli may eliminate distracting orientation salience signals that masked the salience of rivalry in Experiment 1. In Experiment 3, 9 observers searched for shiny disks defined by binocular luster among matte disks with variable grey levels (eliminating reliable contrast and luminance cues). Luster is produced by presenting darker disks to one eye; lighter to the other. Search for targets defined by luster among matte distractors is fairly efficient (Target Present: 16 msec/item, average error rate 2.8%). In Experiment 4, observers searched for two types of targets (white shiny or black shiny) among two types of distractors (white matte or black matte) or vice versa. Search for lustrous targets among matte distractors was markedly more efficient than search for matte among luster (Target Present: 15 ms/item vs. 47 ms/item, average error rate 7% vs. 21%). This clear asymmetry suggests that luster may serve as a basic feature in visual search.
Meeting abstract presented at VSS 2016
This PDF is available to Subscribers Only