Rumor is a widespread social phenomenon, while the psychological black box of rumor spreading still remains underexplored. The existing research largely examined the issue of rumor from the aspect of the phenomenon, but the in-depth exploration of individual’s psychological process and characteristics during rumor spreading is limited. This paper attempts to investigate the influence of unfairness appealing rumor on people’s cognitive appraisal and emotional response in a context of Genetic Modified (GM) food risk. Drawing on previous research, we hypothesize that the emotions could be classified as consequence-related vs. morality-related. Further, rumor which provokes a feeling of unfairness would result in more intense ethical emotions, while maybe not the case for consequentialist emotions and perceived risks. Furthermore, we infer that rumor refuting by negate the credibility of rumor would offset the influence of it on emotional responses. To verify these assumptions, an experiment was taken on 150 undergraduates, which were assigned into 3 groups randomly. The group X read the basic message of GM bean oil. For the group Y, we inserted a faked message into the basic message to provoke a feeling of unfairness, while a clarification of that rumor was added for the group Z. Manipulations tests show that rumor indeed provoke a feeling of unfairness, while the reported unfairness decreased after rumor refuting. We analyzed the structure of the emotion ratings by applying a principal component analysis with varimax rotation, which supported the distinction between morality-based and consequence-based emotions. The results indicted that the feeling of unfairness not only directly led to stronger ethical emotions (such as angry, indignant), but also increased the consequentialist emotions (such as fear and worry) as well as the perceived risks indirectly. We argue that the ethical emotions caused by unfairness in the context of GM risk produced a casual effect on risk perception, where the participants intend to overestimate the risks of GM foods to justify their increased ethical emotions. Many psychological experiments have confirmed the spillover effect of incidental emotions on judgment, e.g. angry people inclined to underestimate risks (see Lerner et al., 2015). However, the induced ethical emotions in our experiment were integral to the following judgments and it resulted in higher perceived risks. The contrary effects imply a difference between incidental emotion and integral emotions on judgment. What’s more, we found that negating the rumor immediately could control the influence of rumor on perceived risks and emotions to some extent. These findings suggest that rumors governance must pay special attention to the emotional response of the public. To verify these assumptions, an experiment was taken on 150 undergraduates, which were assigned into 3 groups randomly. The group X read the basic message of GM bean oil. For the group Y, we inserted a faked message into the basic message to provoke a feeling of unfairness, while a clarification of that rumor was added for the group Z. Manipulations tests show that rumor indeed provoke a feeling of unfairness, while the reported unfairness decreased after rumor refuting. The results indicted that the feeling of unfairness not only directly led to stronger ethical emotions (such as angry, indignant), but also increased the consequentialist emotions (such as fear and worry) as well as the perceived risks indirectly. We argue that the ethical emotions caused by unfairness in the context of GM risk produced a casual effect on risk perception, where the participants intend to overestimate the risks of GM foods to justify their increased ethical emotions. Many psychological experiments have confirmed the spillover effect of incidental emotions on judgment, e.g. angry people inclined to underestimate risks (see Lerner et al., 2015). However, the induced ethical emotions in our experiment were integral to the following judgments and it resulted in higher perceived risks. The contrary effects imply a difference between incidental emotion and integral emotions on judgment. What’s more, we found that negating the rumor immediately could control the influence of rumor on perceived risks and emotions to some extent. These findings suggest that rumors governance must pay special attention to the emotional response of the public.