Abstract: Feature integration theory, the classic visual attention model, provides a theoretical foundation for optimizing visual communication. This paper proposes a set of glyphs named Rose Shape that aim at improving users' visual information search in visualization. Rose Shapes integrate rich and easy-to-catch attributes for mapping data variables, enabling automatic and unconscious information processing in the pre-attentive stage. A case study illustrating the levels of education funding in the United States in 2008 validates the feasibility of this approach. This study also conducts empirical experiments using eye tracking technologies to detect users' first fixation and fixation time during the first 5 seconds when they read RoseShapes. Based on the results of experiments, we propose methods to optimize users' visual experiences by controlling the similarities between Rose Shapes.