After decades of rapid economic growth, environmental issues are threatening people’s health and societies around the world are facing severe challenges with environmental pollution. And environmental problems are usually recognized as social dilemmas which imply conflicts between individual and collective benefits. Empirical studies explored the effect of social observation on pro-environmental behavior. However, the effect of social observation on pro-environmental behavior in previous studies are not completely consistent. Some researchers found that social observation increased pro-environmental behavior; however, some studies found that social observation did not affect pro-environmental behavior. In order to find out the possible causes of contradictions in previous research conclusions and understand the mechanisms of pro-environmental behavior, two experiments were conducted to investigate the effect of social observation on pro-environmental behavior and the effect of social distance. Experiment 1 explored the effects of social observation on pro-environmental behavior under the presence or absence of observers. A total of 34 participants were recruited in Experiment 1a to investigate the effect of social observation without reciprocity on pro-environmental behavior. In Experiment 1b, 37 participants were recruited to further explore the effect of social observation with reciprocity on pro-environmental behavior. The results of Experiment 1a showed that social observation without reciprocity had no effect on pro-environmental behavior. But the results of Experiment 1b showed the participant's pro-environmental behavior under social observation without reciprocity condition was greater than that without observation (control condition). But the results of Experiment 1b showed the participant's pro-environmental behavior under social observation without reciprocity condition was greater than under control condition. Experiment 2 recruited 40 participants to explored the effect of social observation and social distance on pro-environmental behavior. The results of experiment 2 showed that social distance had a moderate effect on the relationship between social observation and pro-environmental behavior. The participant's pro-environmental behavior is not affected by social observation when environmental toward friends, while the pro-environmental behavior is greater under social observation relative to the control condition when the environmental toward acquaintances and strangers. These results show that people will regard pro-environmental behavior as a strategy to pursue fame and benefit in the future under social observation, and social distance had a moderate effect on the relationship between social observation and pro-environmental behavior. When the social distance between the deciders and the potential victims or beneficiaries of the environment is close, the pro-environmental behavior may be a kind of intuitive instinctive behavior, which is not affected by social observation, while as social distance increased, pro-environmental behavior is based on external motivation, which is affected by social observation. This enlightens us that when the social distance between the deciders and the potential victims or beneficiaries of the environment is close, the government should emphasize the possible consequences of environmental decision-making. With the increase of social distance, the government can improve pro-environmental behavior through external incentives, such as increasing the openness and benefit of pro-environmental behavior.