This paper identifies the relative effectiveness of two mechanisms of emotional contagion on shared emotion in teams: explicit mechanism (active spreading of one’s emotion) and implicit mechanism (passive mimicry of others’ emotion). Using social network analysis, this paper analyzes affective communication networks involving or excluding a focal person in the process of emotional contagion by disaggregating team emotional contagion into individual acts of sending or receiving emotion-laden responses. Through an experiment with 38 pre-existing work teams, including undergraduate or MBA project teams and teams of student club or co-op officers, we found that the explicit emotional contagion mechanism was a more stable channel for emotional contagion than the implicit emotional contagion mechanism. Active participation in affective communication, measured by outdegree centrality in affective communication networks, was positively and significantly associated with emotional contagion with other members. In contrast, a team member’s passive observation of humor, measured by ego network density, led to emotional divergence when all other members engaged in humor communication. Our study sheds light on the micro-level process of emotional contagion. The individual-level process of emotional convergence varies with the relational pattern of affective networks, and emotion contagion in teams depends on the interplay of the active expresser and the passive spectator in affective networks.