This study was carried out to investigate whether the context effects of facial emotions of surrounding people on the judgment of a person’s facial emotions vary with information on the group memberships of the target person and surrounding people. The context conditions were 2 (surrounding people’s facial emotions: happiness and sadness) x 2(group membership: the same group and different groups) and there was a no context conditions for calculations of context effects. Both context conditions and no context conditions were within-participant designs. The facial emotion conditions of the surrounding people were processed so that the happiness or sadness of four persons would appear around the target person in a circle with a visual angle of 13° per emotion and the group membership conditions were processed so that the uniform colors of the target person and the surrounding people would be the same or different. The no context condition was to judge the emotions of the target person without surrounding people. The tasks to be performed by the participants under the context conditions were first, seeing the facial emotions of four surrounding people expressing the emotion of happiness or sadness to judge whether the target person won or lost the game. After judging the victory or defeat, the target person on the center was presented. Now, the second task to be performed by the participants was a two-alternatives forced choice task to judge the facial emotion of the target person as happiness or sadness. At this time, one target face was randomly presented at a time among faces morphed into happy faces or sad faces at six levels with differences of 20% while the facial emotions of the surrounding people were presented with happiness or sadness without being morphed. The PSE was estimated under each condition based on the data of the participants'' happiness response rates by the level of the target face morphing. The differences between the PSE values under the context conditions and the PSE values under the no context condition were calculated to calculate the context effects. According to the results of the experiments, the interaction effects were shown to be significant. Under the same group / happiness context condition, the effect to reduce the PSE was significant compared to the no context condition and under the same group / sadness context condition, the effect to enhance the PSE was significant compared to the no context condition. This means that when the target person is in the same group as the surrounding people and the surrounding people show happy emotions, the participants judge that the target face shows happy emotions even if it shows relatively less smiles. In addition, when the target person is in the same group as the surrounding people and the surrounding people show sad emotions, the participants judge that the target face shows happy emotions only when it shows relatively more smiles. On the contrary, in the case of different group conditions, there was no significant difference from the no context condition whether the facial emotion of the surrounding people was happiness or sadness. The results of this study suggest that the effects of the facial emotions of surrounding people can vary with information on the group memberships of the target person and surrounding people.