Social transmission refers to a process in which an observer (OB) acquires new information about the environment including threat situations, through the action of familiar conspecifics. Recently, a number of studies employing observational threat conditioning (OTC) in which OB mice expressed defensive responses following indirect exposure to pair-housed partner mice (demonstrator: DE) which were receiving repeated footshocks, have produced interesting insights into the social mechanisms of emotional transfer. However, the nature of the transmitted information or the critical cognitive processes involved in OTC is not clear. In a series of experiments, we investigated the key elements involves in triggering socially-induced defensive responses. In Exp.1, we compared the effectiveness in conveying a threat of two different types of defensive reactions of DEs: the circa-strike activity burst (CSAB) vs. freezing. The results show that the CSAB is more effective than freezing in inducing defensive freezing in an OB. In Exp. 2, we investigated different types of the OBs’ defensive responses by measuring their change in head orientation or their “gazing” at the DEs, and their temporal synchrony with the DEs’ defensive reactions in the form of their CSAB. The results show that OBs’ gazing was significantly correlated with the DEs’ CSAB, especially the DEs’ jumping behavior, but not with the freezing of the DEs, indicating that jumping is a more effective trigger stimulus in inducing attentional capture in conspecific partner animals. In Exp. 3, the role of visual information was tested. The result shows that the OBs’ level of freezing was significantly reduced when visual information was blocked by an opaque partition. In Exp. 4, to confirm the critical role of visual attention, we introduced distracting flashing lights, which were switched on and off at random intervals during the conditioning process. With all other conditions being maintained unaltered, the OBs in the distractor condition displayed a significantly decreased level of freezing, indicating that the visual attention paid to the DEs by the OBs during the conditioning process was critical for the social transmission of threat. Taken together, the results of the current study strongly suggest that socially transmitted defensive behavior is dependent on the specific behavioral elements of a DE’s defensive behavior, and moreover, that a visual attentional process is required during the OTC.