Abstract. Emotional signals from the face and body are normally perceived as an integrated whole in everyday life. Previous studies have revealed an incongruent effect which refers to distinctive behavioral and neural responses to emotionally congruent versus incongruent face-body compounds. However, it remains unknown which kind of the face-body compounds caused the incongruence effect. In the present study, we added neutral face and neutral body stimuli to form new face-body compounds. Forty subjects with normal or corrected-to-normal vision participated in this experiment. By comparing the face-body compounds with emotional conflict and face-body compounds with neutral stimuli, we could investigate the source of the incongruent effect. For both behavioral and event-related potential (ERP) data, a 2 (bodily expression: happiness, fear) × 2 (congruence: congruent, incongruent) repeated-measure analysis of variance (ANOVA) was performed to re-investigate the incongruent effect and a 3 (facial expression: fearful, happy, neutral) × 3 (bodily expression: fearful, happy, neutral) repeated-measure ANOVA was performed to clarify the source of the incongruent effect. As expected, both behavioral and ERP results have successfully repeated the incongruent effect. Specifically, the behavioral data showed that emotionally congruent versus incongruent face-body compounds were recognized more accurately ( p < .05). The ERP component of N2 was modulated by the emotional congruency between the facial and bodily expression showing that the emotionally incongruent compounds elicited greater N2 amplitudes than emotionally congruent compounds ( p < .05). No incongruent effect was found for P1 or P3 component ( p = .079, p = .99, respectively). Furthermore, by comparing the emotionally incongruent pairs with the neutral baseline, the present study suggests that the source of the incongruent effect might be from the happy face-fearful body compounds. We speculate that the emotion expressed by the fearful body was much more intensive than the emotion expressed by the happy body and thus caused a stronger interference in judging the facial expressions.