The influence of social interaction and emotional states on residents’ decision-making on energy-efficiency retrofit: an improved multi-agent simulation approach

Purpose – This study examines how social interaction and emotional states shape residents’ behavioral decision-making regarding energy-efficiency retrofit (EER) of existing residential buildings, aiming to promote EER acceptance and mitigate greenhouse gas emissions. Design/methodology/approach – A multi-agent simulation model grounded in social interaction theory and overreaction theory was developed to analyze residents’ EER decision-making. The model integrates residents’ self-assessment, social interaction and emotional states, capturing their dynamic evolution over time. Behavioral evolution was further illustrated through the analysis of fluctuation patterns in simulation curves. Findings – Social interaction significantly enhances EER acceptance by strengthening observational learning and demonstration effects. However, high network connectivity and strong emotional sensitivity can amplify short-term fluctuations, exhibiting a “double-edged sword” effect. Costs of EER implementation and maintenance exert the strongest influence on residents’ decisions, whereas visible and immediate government subsidies effectively trigger short-term participation. In contrast, indirect benefits, such as house appreciation and environmental benefits, exert comparatively limited impact under the model’s baseline parameterization. Originality/value – This study provides a novel analytical framework by integrating social interaction and emotional states into the analysis of residents’ EER behaviors. It addresses a gap in existing research, which often neglects the psychological and social dimensions of behavioral decision-making in energy efficiency contexts. The findings offer actionable insights for designing effective policies to foster EER acceptance, emphasizing the importance of social networks and emotional dynamics in shaping collective behaviors.