Heated conversations in a warming world: affective polarization in online climate change discourse follows real-world climate anomalies

Aman Tyagi, Joshua Uyheng, Kathleen M. Carley

Abstract: Climate change research describes online discourse as sharply polarized, echoing real-world divides in society. Yet while polarization in online climate change discourse has been extensively studied in terms of isolated communities and echo chambers, less is known about the extent of affective polarization that characterizes the hostile nature of intergroup interactions. Utilizing a combination of machine learning and network science tools, we design a methodological pipeline that quantifies the extent to which stance groups interact with more negative sentiment toward out-group members than in-group members. We apply this framework to 100 weeks of Twitter discourse about climate change. We find that deniers of climate change (Disbelievers) are more hostile towards people who believe (Believers) in the anthropogenic cause of climate change than vice versa. We also observe that Disbelievers use more words and hashtags related to natural disasters during more affectively polarized weeks as compared to Believers. Finally, using vector autoregression analysis, we find that climate anomalies in terms of both severe temperature and storms predict asymmetric shifts in online climate change discourse: Disbelievers grow more hostile toward out-groups, while Believers become less affectively polarized. These findings resonate with prior work on the asymmetric nature of polarization in contentious discourse, both around climate change and beyond. Our work also extends existing findings around temporal associations between climate anomalies, divided media representations, and real-world conflicts. We conclude with implications for science communication and studying affective polarization in online discourse, especially concerning the subject of climate change.