Abstract
Decades of research on visual perception has uncovered many phenomena, such as binocular rivalry, backward masking, and the attentional blink, that reflect ‘failures of consciousness’. Although stimuli do not reach awareness in these paradigms, there is evidence that they nevertheless undergo semantic processing. Object substitution masking (OSM), however, appears to be the exception to this rule. In OSM, a temporally-trailing four-dot mask interferes with target perception, even though it has different contours from and does not spatially overlap with the target. Previous research suggests that OSM has an early locus, blocking the extraction of semantic information. Here, we refute this claim, showing implicit semantic perception in OSM using a target-mask priming paradigm. Across two experiments, we manipulated the semantic congruency between target words and the color of the mask, and observers made a speeded identification response to the mask color, followed by a target identification (Experiment 1) or detection judgement (Experiment 2). In Experiment 1, we obtained a strong, systematic effect of the semantic congruency of the target word on response time to the mask (priming), such that responses were faster for compatible compared with incompatible target-mask trials. This was the case both when observers correctly identified the target, and, critically, when they did not. In Experiment 2, we also obtained a priming effect both when the target was correctly detected, and when it was missed. Strikingly, however, a pattern of negative priming was observed (faster responses to compatible trials) when the target was missed, whereas the opposite pattern was found when the target was detected. This result converges with previous findings that unconscious and conscious processing can lead qualitatively different patterns of priming. Most importantly, the significant effect of semantic congruency from masked targets in both experiments reveals that semantic information suppressed via OSM can nevertheless guide behavior.
Australian Research Council.