Abstract
Can semantic activation before test displays affect figure assignment? Five experiments show the answer is yes, although context matters. Participants viewed black and white displays divided into two equal-area regions by a central border and reported which side they perceived as figure. One side depicted a familiar object (upright or inverted); the complementary side depicted a novel object. A word denoting either the familiar object at a basic level (BL) or an unrelated (UNR) different-category object preceded each display. In Experiment 1, where words were unmasked, Semantic Priming effects were observed: For both upright (p < .001) and inverted (p < .03) displays, reports of perceiving the figure on the familiar side of the border (Fam = Fig) were significantly greater following BL than UNR primes. Semantic Context effects were also observed: Independent of prime type, Fam = Fig reports were substantially greater for inverted displays than had previously been observed without primes. Thus, processing word meaning induces upweighting of familiarity as a figural prior. In Experiments 2–5 words were masked. Effects observed varied with both prime-display SOA (100ms or 170ms) and whether an induction task preceding the masked word emphasized meaning or perceptual features. For 100ms SOAs, both Semantic Context effects and Semantic Priming effects were observed when the induction task emphasized meaning (p < .005; Experiment 2), but not when it emphasized perceptual features (p >.55; Experiment 3). For 170ms SOAs, which allowed more time for semantic activation, Semantic Context effects were observed following both induction tasks, suggesting that the semantic induction task in Experiment 3 speeded semantic processing of the prime. These results suggest that (a) expectations regarding specific objects can increase the probability of perceiving a region suggesting that object as figure, and (b) the familiarity prior is upweighted in contexts where meaning is emphasized.
Acknowledgement: ONR 00014-14-1-067