The spatial behavior of tourists is an important part of the research on congestion management and sustainable planning of tourism destinations. Combined with user-generated content (UGC) and site-based survey data, this study conducted an overlaying analysis between street network configurations that resulted from space syntax and tourist preferences. Based on space syntax, tourist movement is influenced by the distribution of scenic spots and the structure of tourist trails in scenic mountain areas. The results reveal that the distribution of scenic spots has a significant impact on tourist flow and visitors’ choices of entrance to the mountain; the volume of online sign-ins is highly correlated with landscape attention, axial control values and the local integration value of the trails; and tourists’ attention focuses on the entrance area and the few tourist-sight markers. This study advances the understanding of the spatial patterns of within-destination tourist behavior; this knowledge will be helpful in alleviating congestion in mountain scenic areas and providing effective guidance for tourists to plan an ideal tour route.