Testing day: The effects of processing bias induced by Navon stimuli on the strength of the Muller-Lyer illusion

Explanations for the cognitive basis of the Muller-Lyer illusion are still frustratingly mixed. To date, Day s (1989) theory of perceptual compromise has received little empirical attention. In this study, we examine the merit of Day s hypothesis for the Muller-Lyer illusion by biasing participants toward global or local visual processing through exposure to Navon (1977) stimuli, which are known to alter processing level preference for a short time. Participants (N = 306) were randomly allocated to global, local, or control conditions. Those in global or local conditions were exposed to Navon stimuli for 5 min and participants were required to report on the global or local stimulus features, respectively. Subsequently, participants completed a computerized Muller-Lyer experiment where they adjusted the length of a line to match an illusory-figure. The illusion was significantly stronger for participants with a global bias, and significantly weaker for those with a local bias, compared with the control condition. These findings provide empirical support for Day s conflicting cues theory of perceptual compromise in the Muller-Lyer illusion