Understanding and modeling experiences in urban environments using a computational model of emotion: a conceptual framework: Urban Transitions 2022

Creating high-quality environments which can provide positive experiences for citizens is one core task for urban developments. Momentary experiences in the city directly influence people's affective states and eventually contribute to their long-term subjective wellbeing. The influences from urban environments on people’s momentary affective experiences have been widely explored. However, most of previous studies focus on describing the nature of the intermediate emotional state without trying to unpack the generation and effects of emotion mechanisms. There is a paucity of psychological theory-driven modeling of emotional experiences which can support goal-directed emotion computation and serve simulation and prediction of experiential-aspect of the environment.In the field of artificial intelligence, the general purpose computational models of emotion (CMEs) are flourishing with the aim of endowing artificial agents with human-like emotion mechanisms. The CMEs provide interesting insights for understanding the emotion system but have been rarely applied in the context of experiencing the built environment.Within the scoping review, we build on literatures from both fields of affective experiences in built environments and generic CMEs to reach a combined understanding and a computational framework for modeling affective experiences in the built environment. In particular, we focus on appraisal theory as the emotion theory base and agent logic as the model architecture to formulate affective experiences in environments. Primary and secondary emotions are elicited through different appraising rules on environmental stimuli and function in conjunction with goal-oriented rational judgement to produce adaptive behaviors during the experience. A working prototype of the model is developed taking train station areas, an urban place serving citizens with diverse daily purposes (e.g., transferring, shopping, wandering), as an application scenario to further demonstrate the affective experiences differentiation.The well-structured CME tailored for experiencing the city can be embedded into the digital urban information system (e.g., a digital twin of city) to maximize its efficiency in monitoring and predicting, ultimately supporting better urban design decisions recognized by citizens.