Recently, we are faced with serious concerns related to a rapidly aging population, declining birthrate and decreasing population in Japan. In addition, the structure of big cities versus rural regions in national land and that of urban areas versus rural areas in rural regions are observed. These structures trigger remarkable disparities among regions. In this situation, how to revitalize the regions is a huge problem. In this study, we focus on interactions between regions as one of the most fundamental and important factors for regional revitalization. The aim of this study is to clarify the interaction mechanisms. In order to achieve this aim, interaction models based on the interaction purpose are constructed, because people make a move based on a different behavioral theory for each purpose of interaction. These models are applied to an analysis of interactions in Japan. An interaction model for sightseeing was based on the utility maximization theory. When the utility function was maximized, time constraints were used as a constrained condition. The estimation result of the model was satisfactory and showed that variables related to travel time between regions have a major effect on interactions. In addition, interactions are affected by variables representing the regional attraction of a destination region and the convenience of traffic between the regions. On the other hand, we assumed behavior related to business interactions was based on the profit maximization theory and a production function was used as a constrained condition. In addition, a business trip generation model and an interaction assignment model for business were constructed. Results of the profit maximization model and the business trip generation model clarified that the volume of production and business trip generation influence each another. In addition, business trip generation is affected by the convenience of traffic between regions. It can be said that the interaction model for business reflects the current situation. These results enable us to clarify interaction mechanisms in more detail. Therefore, we can use these models to clarify the effect of infrastructure development on an interaction. JEL Classification: C0, D03, R15