Competition between grouping principles

The perceived organization of elements in a display is affected by the strength of grouping rules available (e.g., Ben-Av & Sagi, 1995). A 'winner-take-all' approach predicts that only the dominant organization is represented, and hence, ultimately reaches conscious perception (e.g., Kubovy & van den Berg, 2008).We tested the hypothesis that in situations where more than one organization is possible in the display, multiple representations are constructed, and while one of them is chosen for conscious perception the others are suppressed. We used the primed matching paradigm in which the observers are exposed to a prime stimulus that is followed by a pair of test figures. The test pair can consist of identical or different figures, and the observer's task is to give a "same" or "different" response to that pair. "Same" responses are relatively faster and/or more accurate when the test stimuli are similar to the prime than when they are dissimilar to it (e.g., Beller, 1971). The prime consisted of a matrix of elements organized (e.g., into columns/rows) by two grouping rules (dominant: brightness/color similarity; non-dominant: shape similarity) that lead to either the same organization (no-competition condition), or to two different organizations (competition condition; e.g., columns by brightness and rows by shape). A neutral prime containing no apparent organization served as a control condition. In the competition condition test figures were similar to one or the other of the two possible organizations in the prime. By varying the duration of the prime we were able to examine the time-course of representations construction (e.g., Razpurker-Apfeld & Kimchi, 2007). The results show a reduction of priming effects for the dominant organization in the competition condition compared with the no-competition condition at the longest prime duration, suggesting that although the dominant cue dominates early organization, a competition seems to emerge later in time.