The visual search efficiency is affected by the color of the interferer. When the interferer is close to the target color, the visual search efficiency will decrease, and the same effect can be found in the relationship between background and the target. Prior studies mainly focus on the influence of existing background and interference factors on visual search, without considering special visual factors such as negative afterimages. A negative afterimage is a visual phenomenon in which adaptation to a chromatic stimulus induces color perception of the opposite chromaticity after the removal of the stimulus. When appears on the information search interface, the negative afterimage will have to be seen as the background or a distractor of visual search. This study is to explore the effect of visual image on visual search efficiency. In experiment 1, we used a two-factored internal design of 2 (with afterimage/no afterimage) ×3 (color relationship of afterimage color and search information: similar color/complementary/neutral color).we combined the van Lier afterimage with visual search paradigm--using the red "cross- shaped " four-pointed star and the green " x-shaped " four-pointed star overlapping as stimulus material, two types of contours as inducing material with 5 English letters. Firstly, the color stimulating material of 1200ms is presented, and then the contour of the four-pointed star with visual search information is presented. The subject found the target letter and reacted. After a1200ms visual mask, the correct answer was input, and the next trial was performed at intervals of 3000ms. A complete procedure was comprised of 4 blocks: 2 block A (with afterimage) and 2 block B(no afterimage), and each block contained 30 trials(5 information places × 3 conditions×2 contours). Reaction times and error rates were recorded. However, it had a lower afterimage perception accuracy than the prior studies and maybe father influence the result. To find out the effect of the adding information on afterimage perception, in experiment 2, we used a 2(with visual information/no visual information) ×2(afterimage color: red and green) ×3(information color: red, green and gray) design and same procedure as experiment 1. Repeated measurement analysis of variance for the obtained data, and we found that: 1. The main effect of the afterimage is significant（F=38.372，P<0.0001）. In the case of a afterimage present, the subject's RTs is significantly slower than in the case of no afterimage;2. The main effect of color is not significant, and the relationship between the color of the afterimage and the color of the search information does not significantly interfere with the visual search efficiency; 3.The color-filled portion of the afterimage also has an interference effect as the visual search for the original afterimage;4. The adding visual information does not weaken the perception of the afterimage color. These results show that the afterimage itself will reduce the efficiency of visual search, and whether the color of the afterimage or other factors interfere with the visual search remains to be explored. Therefore, in the work of high visual search efficiency, the visual information display interface design should minimize the triggering of the rear image to improve the visual search efficiency, thereby improving the task success rate. In addition, the results of this study also provide an objective assessment method for afterimage research. In addition to the subjective report of the subject, the experimenter can also visually search for the reaction to determine whether the afterimage occurs.