#542: Exploring Unstructured Collaboration While Navigating Misinformation in Computer-Supported Collaborative Argumentation: Implications for Design

The proliferation of socio-scientific misinformation is an inherently social challenge. While Computer-Supported Collaborative Argumentation (CSCA) environments offer a promising pedagogical approach for collective knowledge construction, their effectiveness relies heavily on productive group dynamics. To avoid over-scripting future interventions, this formative Design-Based Research (DBR) study explores the baseline collaborative breakdowns that occur when students navigate misinformation without explicit social guidance. The study analyzed the audio transcripts of two middle school triads collaboratively mapping arguments about a controversial topic, utilizing established interaction analysis frameworks to evaluate peer talk and social modes of co-construction. Findings indicate that, within these groups, students defaulted to two distinct dysfunctional patterns: an epistemic conflict avoidance pattern, characterized by cumulative talk and quick consensus-building, and a social dominance pattern, characterized by disputational, competitive fact-checking. Structural scaffolds alone appeared insufficient to foster the critical exploratory talk required to evaluate deceptive claims, absent social interaction protocols. The study concludes by offering targeted design implications, suggesting that future CSCA environments might benefit from embedded collaboration scripts, such as structured "Devil's Advocate" and "Lead Fact-Checker" roles, to disrupt these natural tendencies and better support apt epistemic performance.