<p>Social networks have profoundly influenced human thinking and behaviour patterns and have effectively promoted the development and evolution of the urban tourism economy. However, the front-end impact of social network information differences on travel intention and the back-end effect of travel intention on actual behaviour have not been considered from the perspective of behavioural psychology. Based on the three-stage division of SOR theory, namely, “Stimulus‒Organism‒Response”, this study takes the tourist destination image, personal characteristics and information quality of key opinion leaders (KOLs) as external stimulus variables, combines basic theories such as the theory of planned behaviour and the technology acceptance model, and builds a theoretical framework to analyse the impact of social network information on travel intention. It also incorporates variables such as individual socioeconomic attributes and tourism attributes and establishes a tourist destination selection behaviour model (ICLV) that considers latent psychology variables, analysing the mechanism whereby multiple factors influence travel behaviour intention and behaviour in social network interaction scenarios. Specifically, we analysed 522 valid questionnaires from two internet celebrity cities (ICCs), Chongqing and Chengdu. The results show that the personal characteristics and information quality of KOLs can effectively stimulate tourist travel intention by influencing travel motivation, travel attitude and perceived value but have no significant effect on tourist destination decisions. Furthermore, tourist destination decision-making behaviour is affected mainly by factors such as the travel attitude, destination image, individual attributes of tourists and travel attributes. This study enriches the research and empirical investigation of tourist travel behaviour in the digital environment and provides a reference for the creation of ICCs and the marketing of tourist destinations.</p>