Traditional villages are the common historical and cultural heritage of all mankind. With the intensification of urbanization, the continuation of traditional villages and the inheritance of historical and cultural heritage are facing risks. The research on the driving factors affecting the spatial distribution characteristics, heterogeneity and human land interaction of traditional villages provides a new idea for the protection of traditional villages. This study takes 137 traditional villages in Puxian area, a typical cultural area in the southeast coast, as the research object, analyzes the spatial distribution pattern of traditional villages by using spatial analysis method, and selects 13 factors to analyze the main driving forces and interaction mechanisms through geographical detectors. The results show that (1) Puxian traditional villages are clustered and distributed, and the distribution among counties is uneven, mainly in the state of "one cluster and many scattered points" with more coastal areas and less mountainous areas. (2) Puxian traditional villages are mainly affected by many factors such as nature, space, society and culture. They are more densely distributed in areas with rich cultural heritage, fertile land, flat terrain, suitable climate, close to water systems, developed transportation, backward economy and dense population. (3) Cultural factors are the primary factors affecting the spatial distribution of traditional villages, the order of driving factors' explanatory power is intangible cultural heritage (0.5160) > protected cultural relic units (0.3591) > distance from railway (0.3255) > night light remote sensing (0.3179) > elevation (0.3012) > population density (0.2671) > slope (0.2032) > soil type (0.1804) > precipitation (0.1750) > temperature (0.1744) > land use (0.1492) > distance from river (0.0691)>distance from highway (0.0530). The interaction of intangible cultural heritage, protected cultural relic units and distance from the railway is the dominant factor for the spatial differentiation of traditional villages. Among them, the interaction of intangible cultural heritage∩distance from the railway is the strongest, and the q-value is 0.79, which proves that the interpretation ability of the two factor model is much higher than that of the single factor model. The results of this study reflect that traditional villages and nature, space, society and culture are interdependent, so the protection of traditional villages should be adapted to local conditions.