The default mode network (DMN) has been reported to be involved in a variety of important cognitive functions and received increasing attention in neuroscience recently. Its dysfunction is also reported to be associated with multiple psychiatric disorders. However, the causal information flow (effective connectivity) within the default mode network remains poorly understood. In this study, we explored the effective connectivity pattern between 4 key DMN brain areas based on a high resolution 7T resting state fMRI dataset using a cutting-edge spectral dynamic causal modelling technique. Results showed that there was a distinct effective connectivity pattern among the DMN nodes. We found medial prefrontal cortex(MPFC) and bilateral inferior parietal cortex(IPC) sent information to the posterior cingulate cortex(PCC), which suggested that the PCC might be a hub region that collected information from other DMN areas. Besides, a causal influence was found from bilateral IPC to MPFC, and from left IPC to right IPC. This work was the first 7T fMRI study that investigated effective connectivity pattern among DMN nodes, which may promote our understandings about the functions of DMN and benefit future research in DMN-related psychiatric disorders.