Decoding the Public's Real-Time Emotional and Cognitive Responses to the Changing Climate on Social Media: Computational Analysis Using Weibo and Meteorological Data.

Background: Climate change poses a significant threat to mental health and well-being worldwide. Existing research on the associations between climate change-related events and mental well-being primarily focuses on clinical outcomes and often measures associations at single time points. The long-term effects and variability of the changing climate on more subtle nonclinical but widespread mental well-being remain relatively unexplored. Additionally, the underlying mechanisms that link changing climate events to real-time emotional well-being and pro-environmental actions have rarely been studied. Revealing real-time nonclinical mental well-being and its underlying mechanism is crucial for the early detection of at-risk individuals. This knowledge can also inform future interventions aimed at improving the public's risk perception and empowering communities to manage related challenges effectively.

Objective: This study aimed to understand the association between the changing climate and expressed emotional well-being by integrating multiple data sources, including social media posts about climate change on Weibo (N=76,514), 20 years of regional meteorological data (N=216,476 records), and regional vulnerability data in China.

Methods: This study proposed and tested a new mechanism that connects meteorological factors with expressed emotional well-being through three cognitive responses identified from social media posts: thinking styles, social affiliations, and somatosensory experiences. Psycholinguistic analysis, structural equation modeling (SEM), and multiple regression models were used to examine the mediation of these three conceptual factors, as well as the moderating effects of regional vulnerability and seasonal changes on the influence of climate change on the public's expressed emotional well-being and downstream pro-environmental tendencies.

Results: The SEM results revealed that extreme hot days are associated with decreased emotional well-being when talking about climate change (total effect=-0.712, 95% CI -0.894 to -0.531, P<.001), and these effects were mediated by three proposed mediators: social affiliations (indirect effect=-0.445, 95% CI -0.537 to -0.347, P<.001), analytical-intuitive thinking style (indirect effect=-0.100, 95% CI -0.126 to -0.073, P<.001), and somatosensory experiences (indirect effect=0.022, 95% CI 0.005-0.041, P=.02). Additionally, regression analysis indicated that the association between increased temperatures and expressed emotional well-being is moderated by seasonal changes (β=-.091, 95% CI -0.159 to -0.023, P=.009) and regional population density (β=-.068, 95% CI -0.118 to -0.018, P=.007). In the crude model examining associations between weather indices and expressed pro-environmental tendencies, the results showed that extreme hot days are associated with reduced pro-environmental tendencies (odds ratio [OR]=0.802, 95% CI 0.747-0.861, P<.001). However, after controlling for expressed emotional well-being and cognitive responses, such associations were less pronounced.

Conclusions: The findings highlight the need for interventions that promote mental well-being in response to climate change and the importance of cognitive responses in developing positive coping strategies and enhancing emotional resilience. This approach could empower individuals to create a positive self-reinforcing cycle that encourages pro-environmental behaviors.