Abstract
It has previously been shown that a complete image can be integrated from the visual system from incomplete parts when flashed briefly after one another. However, most of these experiments used a single target. Less is known about whether multiple targets can be integrated in the same way. Our series of experiments explored the role of integration versus separation when two target images are flashed. We compared recognition of Chinese characters under scrambled and unscrambled conditions. We also looked at the effect of polarity on recognition. Native speakers were recruited to take part in a two-session experiment. In the first session, participants were shown a random Chinese character in each trial, flashed rapidly across a monitor. The brightness varied in between trials and the results of the first session was used to calculate a specific level of brightness for each participant for the second session. In the second session, participants were randomly assigned to either the scramble vs non-scramble task or the matching vs non-matching polarity task. In both tasks, participants were shown two characters and asked to recognize both. In the first task, subjects were shown either two scrambled (half of one character + half of the second character) or two unscrambled characters in each trial (one full character followed by the second full character). There was no difference in recognition when we flashed scrambled versus non-scrambled. We concluded that participants were able to integrate correct halves in the scrambled condition, despite each display being a nonsense character on its own. In the second task, subjects were shown either two dark characters or a bright and dark character. Subjects were able to recognize both characters with similar accuracy across both these conditions, suggesting that the role of integration favors interpretations in a way that is consistent with prior learning.