Abstract
In metacontrast masking, perception of a brief stimulus (the target) is reduced or eliminated under some conditions by a second, non-overlapping brief stimulus (the mask). In Type-A masking, target visibility is low when mask and target onset simultaneously (stimulus onset asynchrony [SOA] is 0) and improves as SOA increases. Type-B masking is a u-shaped function of SOA. Target visibility is poor at intermediate SOAs (50–150 ms) but good at shorter and longer SOAs. Mask-to-target energy ratio (M/T) can often determine which type of masking will occur. Type-A masking occurs when the mask is stronger (i.e. larger, greater contrast, or longer duration) than the target (M/T[[gt]]1). Type-B masking occurs when M/T≤1. Type-B metacontrast has been reported in monocular, binocular, and dichoptic viewing conditions. In cyclopean conditions only Type-A metacontrast has been observed until recently (Krueger, Dobelbower & Phinney, Society for Neuroscience, 2006). We varied mask duration, holding target duration constant, to determine the effect of M/T on metacontrast with cyclopean stimuli. Stimuli were dynamic random dot, red-green anaglyphs (320×240 pixels, 50% density) viewed with red-green glasses. Observers viewed a three-target array (3×9° horizontal bars) and judged which display quadrant lacked a target (4 AFC) with SOAs of 0 to 800 ms for mask onset (8 bars vertically flanking the 4 target positions). Type-B masking functions were observed when the masks were shorter than or equal to target duration (16, 50, and 83 ms mask duration [MT≤1]), or slightly longer (150 and 183 ms). Type-A masking functions were observed when the masks were of much longer duration (316 ms). These results are similar to those in masking with luminance stimuli where Type-B masking occurs with M/T near 1, and Type-A masking with M/T greater than 1. They may also explain past failures to demonstrate Type-B masking with cyclopean stimuli, which typically involved M/T[[gt]]1.