Abstract
Scene context has been shown to influence object recognition; it is not clear what level of visual processing is required for this effect to manifest. Specifically, it is unclear if such object/context interactions may exist in the absence of conscious awareness. By conducting experiments with and without the use of continuous flash suppression (CFS), we examined how context (background) congruency affects target recognition and response time. We used animal and vehicle images in natural or man-made scenes, which formed congruent/non-congruent image groups (100 images each). By comparing among three experimental designs (b-CFS, plain 2AFC, and 2AFC-CFS), we found the response time in the congruent scenes was significantly faster than in the incongruent scenes in plain 2AFC (without suppression). This congruency effect persisted only in the vehicle group when under b-CFS suppression. When combining the two paradigms (2AFC-CFS), the results replicated the congruency effect from the plain 2AFC condition. This indicates that the congruency effect does not emerge at the lowest levels of perception, but requires additional processing, necessitating a degree of conscious access.