Relationship between language and popularity of social media messages
Lejla Džanko (SWPS Warsaw University, University of Padova)
Social media offers an open platform where anyone can share their thoughts with a wide audience—yet only a fraction of messages go viral while most remain unnoticed. Whether we want to help propagate ideas to see a positive change in the world or to stop false narratives from catching on, it is crucial to study the mechanisms by which the diffusion of information operates in online social networks. Drawing on the insights from the theory of diffusion of innovations (Rogers, 1962) researchers tried to build models of information diffusion in online social networks by studying properties of five key elements: message, users, communication channel, and temporal dynamics. This talk focuses on one of those components—the message itself—and specifically investigates how linguistic features influence popularity, as measured by shares. Drawing on relevant theory from psychology and psycholinguistics as well as results of a systematic review conducted in the field (Džanko et al., 2025) we explore how message framing affects virality. Recognizing the complexity of this relationship, we also consider the roles of message source, context, topic, and mood. The talk will present preliminary findings from a mixed-methods study and outline directions for future research.
English 
Italiano