Abstract
Previous research has shown that the P3b ERP component is associated in some ways with visual awareness, but researchers have debated whether the P3b is a neural correlate of visual consciousness, whether it can occur outside of awareness, and whether it is associated with post-perceptual processes that are distinct from consciousness (e.g., decision processes, task relevance, metacognition). To address these issues, we manipulated stimulus visibility via backward masking during an oddball task. For each participant, target luminance was adjusted during pretesting to yield 75% correct performance with a mask SOA of 50 ms. SOAs of 0, 17, 33, 50, and 200 ms were used in the main experiment. Participants reported the direction of the stimulus (right/left arrows) and rated the confidence of each response. One stimulus was rare (.10) and the other was frequent (.90). We found that P3b amplitude was modulated by SOA, response accuracy, decision processes, and confidence: P3b amplitude was greater for longer than shorter SOAs, greater for correct than for incorrect responses, greater for ‘hits’ than for ‘false alarms’, and larger on high-confidence than low-confidence trials. P3b was absent for missed oddballs, indicating that participants were unable to produce a P3b response unless they were aware of the stimulus, arguing against the idea that P3b can occur outside of visual awareness.