Abstract
What distinguishes conscious from non-conscious visual perception? We investigated this question from an information-processing perspective by exploring which spatial frequencies (SFs) are correlated with observers’ responses during conscious vs. non-conscious face perception. Specifically, we used a face-gender repetition priming paradigm and the SF Bubbles technique (Willenbockel et al., 2010) to precisely map the SFs that prime as a function of awareness. A "visible prime" condition was set up by presenting the stimulus sequence mask-blank-prime-blank-mask-target (prime, blank, and mask durations ≤ 50 ms); an "invisible prime" condition was created by reversing the order of the masks and the blanks (see also Dehaene et al., 2001). Twenty grayscale face photographs (10 males; visual angle ~3º) served as primes and as targets, whereby the prime faces were randomly SF filtered trial-by-trial. Results show facilitatory priming effects in response times for both visibility conditions, albeit smaller for the invisible prime condition. A multiple linear regression on the SF filters from each trial and the transformed response times revealed that fast responses were linked to specific SFs (~12 cycles per face width) in the visible prime condition, but not to any specific SFs in the invisible prime condition. Interestingly, the SFs that led to faster responses in the visible prime condition led to slower responses in the invisible prime condition. The results imply that different visual information primes as a function of awareness and therefore provide strong support for a qualitative conscious/non-conscious dichotomy.
Meeting abstract presented at VSS 2012