Abstract
Recent research has shown that visual search can proceed just as efficiently through randomly moving items as through stationary items (Hulleman, PBR, 2009). Here we tested whether the visual system can efficiently segregate moving and stationary items. We spatially interleaved stationary and randomly moving search sets, and independently varied the moving and stationary set sizes. This allowed us to measure the influence of each set (defined as the reaction time by set size slope) as a function of task relevance. Stimuli were gray disks marked with white lines. We tested three types of search task: Easy Feature search for a vertical line among lines tilted ±60° (N = 12); Difficult Feature search for a vertical line among lines tilted ±30° (N = 12); and Spatial Configuration search for rotated Ts among rotated Ls (N = 8). Observers were asked to search for either moving or stationary targets in separate blocks. In all three experiments, we replicated Hulleman's finding that search through moving and stationary stimuli is equally efficient in terms of the effect of relevant items (all p > .30). The effect of irrelevant items (e.g., moving items when searching for a stationary target), however, varied with search difficulty. For Easy Feature search, relevant items (1.7 ms/item) actually had less effect on reaction time than irrelevant items (3.5 ms/item), with a trend towards stationary items generating greater interference than moving items. For Difficult feature search, search through relevant items was inefficient (23.7 ms/item), but irrelevant items were largely ignored (6.7 ms/item). Spatial Configuration search was much more difficult (39.9 ms/item). Irrelevant stationary items were easy to ignore (2.7 ms/item), but irrelevant moving items caused substantial interference (25.9 ms/item). For very efficient searches, the target may be found before segregation comes into play. More difficult searches allow the visual system time to employ segregation mechanisms.