Negative Emotional Responses on Social Media During Crisis Events

ABSTRACT Japan's 2023 discharge of nuclear wastewater into the ocean sparked widespread negative emotions on social media, underscoring the mental health impact of a globally salient environmental event. However, empirical evidence remains limited on how controversial environmental risks generate distinct patterns of emotional responses. To address this gap, this study analyzes large‐scale Weibo data from Japan's 2023 wastewater discharge to examine how differentiated negative emotions are shaped by crisis appraisal cues and social media interactions. Posts were classified into three discrete emotions—anxiety, sadness, and anger—using a sentiment lexicon. We then applied Latent Dirichlet Allocation (LDA) topic modeling to identify thematic contexts. We further used regression analysis to assess the effects of these factors across different emotional outcomes. Results show that attribution‐independent emotions (anxiety and sadness) are primarily driven by cognitive appraisals of the event's danger, damage, and urgency; in contrast, attribution‐dependent anger is amplified by discussions of responsibility and social interactions. Notably, rational communications by influential users significantly mitigate anger levels. By integrating the Social‐Mediated Crisis Communication (SMCC) framework with Cognitive Appraisal Theory, this study elucidates how crisis cognition and social interaction jointly shape emotional dynamics. The findings offer guidance for developing differentiated management strategies for negative emotions in environmental crises, enabling more precise and effective regulation of public sentiment.