Differential effects of objective and perceived environments on place attachment with emotion mediation: a case study in Nanjing, China

Research increasingly links the built environment to place attachment, with expected benefits for perceived restorativeness and resident wellbeing. Yet prior work has concentrated on macro scale attributes and provides limited evidence on how objective and perceived environments shape attachment at the micro scale of the streetscape. Few studies separate their effects or test whether residents’ emotion mediates these links. This study tests mediation between objective and perceived streetscape characteristics and place attachment and clarifies their roles in theory. We assess the distinct contributions of physical features and subjective appraisals at a unified human scale. We further examine heterogeneity by residence duration and age. The empirical design combines semantic segmentation of street level imagery with survey responses from 797 residents in Nanjing. We estimate a partial least squares structural equation model and conduct multigroup analysis to evaluate mediation and differences across groups. Results reveal that natural environment is positively related to place attachment only through emotion, with significant standardized indirect effects of 0.025** and 0.019**. The artificial environment is negatively related through both pathways, with a direct effect on place dependence of −0.203*** and an indirect effect of −0.044***. The perceived built environment shows the strongest association with attachment, with total standardized effects of 0.595*** for dependence and 0.623*** for identity. Heterogeneity analysis indicates that the attachment preferences of medium-term residents gradually shift towards the natural environment, while older adults exhibit reduced sensitivity to artificial environmental factors. Together, the findings advance micro scale attachment research and inform design that fosters belonging and wellbeing.