Abstract
Do visual numerosity judgments depend only on the perceptual strength of the difference between targets versus distractors? If so, then the visual property that makes targets different from distractors should not matter. Specifically, if the perceptual strength of the difference between targets of type T1 and distractors of type D1 is equal to the perceptual strength of the difference between targets of type T2 and distractors of type D2, then these two combinations of targets and distractors should support identical performance across various tasks requiring numerosity judgments. To address this question, we assessed the accuracy of four observers who each, in 12 separate runs, made judgments while the magnitude of the difference between targets and non-targets was varied. Each run involved a different one of the possible combinations of four numerosity discrimination tasks and three visual properties. All targets and distractors were equiluminant and brighter than the gray background. Target objects differed from non-targets in their size, orientation, or color along the L-M axis. Three of the numerosity discriminations used a field of 14 objects and required discriminating 0 versus 1 targets, 1 versus 2 targets, and 2 versus 6 targets. The fourth task required discriminating 5 versus 15 targets in a field of 35 objects. All four observers showed clear differences: these judgments depend on more than simply the perceptual strength of the difference between targets versus distractors. These results suggest that more than one mechanism is involved in numerosity judgments (e.g., parallel individuation and analog magnitude estimation) and that these processes have differential access to information conveyed by the three visual properties.