Abstract
Weisstein, Maguire, and Brannan (1992) proposed antagonistic interactions of dorsal-M and ventral-P streams both within a region and between regions play a central role in figure-ground perception. Based on this model, a region with a stronger ventral-P biased “Figure Signal” is perceived as figure, whereas a region with a stronger dorsal-M biased “Ground Signal” is perceived as ground. Last year we reported results supporting their model using an artificial scotoma paradigm where a small, peripherally viewed target figure fades into the background. Using red and blue light to reduce dorsal-M activity, fade times for color combinations involving red or blue were longest whether those colors were in the target, strengthening its Figure Signal, or in the background, weakening its Ground Signal (Plummer, Brown, & Song, VSS, 2018). In the present study, we tested their model further using figure-ground reversible Maltese crosses. The crosses were either figure-ground ambiguous consisting of left- and right-tilting sectors of equal area (Exp 1) or figure-ground biased consisting of small sectors oriented vertical/horizontal, biased to be seen as figure, compared to larger, obliquely oriented sectors biased to be seen as ground (Exp 2). We measured the duration the alternate crosses were perceived as figure, testing every possible color combination of red, blue, green, and gray. Red and blue were expected to reduce dorsal-M activity thereby strengthening the Figure Signal/weakening the Ground Signal in those sectors resulting in them being seen more often as figure compared to green and gray. Red sectors were perceived as figure more often compared to the other colors in both experiments. Blue sectors were not. The present results provide further support for Weisstein et al’s model while also indicating red was more effective than blue at reducing dorsal-M activity for our centrally viewed figure-ground stimuli.