According to linguistic relevance hypothesis(LRH), people have the ability to categorize the world that they have experienced, influenced by language and culture. Thus, researchers who agree with LRH argued that people with different language organizing experiences should have different world schemas.Some relevant arguments came from research on the relationship between color word and color cognition: color perception should be influenced by the physical properties of the light wave and the biological characteristics of the human eye, and also by language and culture.Although there is no clear boundary among the various visible light–waves from red to violet at the perceptual level, the continuous spectrum is divided into different color regions. In the literature, a "color category perception effect" was proposed that people were more likely to distinguish colors from different colors than those that landed in the same area. However, it has still lacked of discussions on the essential mechanism of this effect. Namely, it is still not clear whether this effect is a perceptual phenomenon or cultural phenomenon(i.e., effects from language application and language labels). Using perceptual task(Experiment 1) and classification tasks containing memory(Experiment 2 and 3), assumption that language application and language labels affect color categorizing was tested in the present study. In Chinese, there are clear and distinct language labels for colors RED, PURPLE, BLUE, and GREEN in Chinese, but intensities of relevance between RED and PURPLE and between BLUE and GREEN are different in everyday expressions(language application): connection of the mental conception(conceptual connection) between RED and PURPLE is much closer than those between BLUE and GREEN. With the boundary colors of the "red–purple" color pair(RGB: 255, 0, 255) and the "blue–green" color pair(RGB: 0, 255, 255) as base points, a vertical demarcation line was drawn on the RGB chromatography. Four color blocks of different lightness saturation levels were randomly selected, upon which two colors on both the left and right sides of the boundary were selected respectively. The distance between two neighboring color blocks(including two neighboring colors that are on either side of the color boundary) is equal on the chromatography. In Experiment 1, three colors that have equal optical range constitute one set of experimental material. Participants were asked to judge as quickly and as accurately as possible whether the left or the right color block looked more similar to the middle one, and to press the corresponding button on a response box. 30 college students from the Han nationality participated in the experiment. In Experiment 2, materials were identical to Experiment 1 and 44 college students from the Han nationality were instructed to remember the colors and to identify as quickly and as accurately as possible whether the following colors belong to the left or to the right of the color pair, and to press the corresponding button on a response box.In Experiment 3, using identical materials, 44 participants were asked to judge as quickly and as accurately as possible whether the left or the right color looked more similar to the standard one, and to press the corresponding button on a response box. Results showed that intensity of conceptual connection effect was not involved in perception task, but in classification tasks and recognition tasks. Conceptual connection, rather than language labels, which might be triggered by comparison, was the main reason that affected performances of classification. According to the present findings, we believe that language labels and conceptual connection are both intermediate in color processing, and coding of colors in memory has a direct function in this process.