Abstract: The technology of connected vehicles based on human machine interface (HMI) plays an important role in improving the overall safety level of fog highway, however, its influence on the driver's attention cannot be ignored. This paper analyzed the driver's visual characteristics, evaluated the degree of distracted influence of the HMI, which is based on connected vehicle system, built the freeway connected vehicle experiment platform based on the driving simulation technology, designed four experimental scenarios for two visibility levels of light fog and heavy fog, and two technical levels of with and without HMI prompts, and quantified the AttenD area, the AttenD area per unit time and the proportion of distracted to evaluate the visual load of the self-designed HMI on the driver under different conditions. Results show that the driver's attend data will fluctuate more frequently and more violently under the condition of HMI prompt. The technical level mainly affects the attend area and the attend area per unit time, that is, the visibility level has a significant impact on the driver's visual distraction with HMI. With HMI prompt, the proportion of the driver's visual distraction will increase significantly, which will have a certain negative impact on road traffic safety. To improve the safety of using HMI, all HMI should be evaluated and optimized according to different weather conditions. At the same time, the research results provide a general evaluation method for evaluating the effectiveness of HMI of connected vehicle system.