The next speaker in this session at the 2026 International Communication Association conference in Cape Town is the great Yangliu Fan, whose focus is on cross-positional agreement and disagreement in social media contexts – with a particular focus on climate discourse on Reddit.
The approach here explores discourse features to encode their complexity, structures, and dynamics within a higher-dimensional space. This extracts discourse features from each post and subsequent comments, and examines comments’ agreement or disagreement with the previous post of comments; from the embedding of these features in a higher-dimensional spaces it is then possible to construct several similarity networks.
Such networks are mostly topic- and format-driven; posts with the same topical categories are closely related to each other, for instance. This can also be used to track discourse flows within the debate over time. What this results in are several observations about the impacts of agreement and disagreement on debate cohesion; threads driven by disagreement are often more cohesive in their debate focus.
