Unveiling political influence through social media: network and causal dynamics in the 2022 French presidential election

Introduction During the 2022 French presidential election, we collected daily Twitter messages on key topics posted by political candidates and their close networks. Our objective was to investigate how political communities influence one another and how central topics shape the dynamics of online political debate. Methods Using a data-driven approach, we analyzed interactions among political parties and applied Convergent Cross Mapping (CCM), a causal inference technique, to move beyond traditional correlation analyses. This allowed us to uncover directional influences among political communities and distinguish genuine causal effects from spurious correlations. Results We found that some parties are more likely to initiate changes in activity while others tend to respond, highlighting asymmetric relationships and hidden dynamics within the social media political network. Specific issues, such as health and foreign policy, emerged as catalysts for cross-party influence, particularly during critical phases of the election campaign. Discussion Our findings demonstrate that CCM provides a powerful tool for capturing causal dynamics in political discourse. The results offer a novel framework for understanding how influence propagates within online political networks and have practical implications for campaign strategists and media analysts seeking to monitor and respond to shifts in political influence over time.