Abstract: It remains unclear whether higher-level aspects of visual stimuli can be represented without awareness. The global workspace theory suggested that basic stimulus properties such as orientation, spatial frequency, color, and translational motion could still be encoded when rendered invisible. However, higher-level aspects of visual stimuli, for example, face perception, the meaning of words, could not be processed without awareness. Meanwhile, the unconscious binding hypothesis insisted that object-related representation could be formed by early and late perceptual binding even under invisible situation. In this study, we used incomplete pictures to investigated whether gestalt could facilitate perceptual binding during binocular rivalry. Specificly, we scrambled the complete and incomplete pictures to investigate: (1) whether there was a scrambled effect for both kinds of invisible pictures; (2) whether familiarity of the simple line pictures affected the time for stimuli breaking suppression. Continuous flash suppression was used in our experiment. Three factors were included: completeness of the construction (complete vs. incomplete), destruction of the meaning (scrambled vs. non-scrambled) and familiarity (upright vs. inverted). Subjects were instructed to respond as accurately and quickly as possible the appearance of any part of the test image as soon as possible and regardless of the specific content of the image. Eighteen volunteers participated in the experiment (8 males and 10 females). Results showed that non-scrambled pictures occupied less time than scrambled pictures to gain dominance against the identical suppression noise; The scrambled effect was observed both for incomplete and complete pictures; Upright pictures were not faster to enter consciousness than inverted ones. These results suggested that the features in incomplete pictures, even suppressed and invisible, could be bound together to form the object representation by gestalt. Apparently, high-level information (gestalt principles and meanings) of a stimulus did contribute to the strength of breaking suppression during its suppressed phase. Substantial information in the suppression phase of binocular rivalry could be processed to the extent that object-related representations could be achieved by gestalt. Our findings provided direct evidence for the unconscious binding hypothesis.

Key words: unconscious processing, binocular rivalry, gestalt, continuous flashing paradigm