A Structural Model of Professionals’ Perceptions of Contextual Information Processing in Human–AI Collaborative Design of Rural Public Spaces

Artificial intelligence and spatial data technologies have expanded information-processing capabilities in public space design, but greater access to information does not necessarily improve design judgment.This study proposes a structural model linking spatial environmental cognition, behavior and needs identification, sociocultural understanding, multi-source information integration, design problem construction, human-AI collaborative judgment, and use feedback and dynamic optimization in rural public space design.A cross-sectional survey using purposive recruitment was conducted among professionals engaged in rural public space design and related development activities across eastern, central, western, and northeastern China.Of 500 questionnaires distributed, 468 valid responses were obtained.Confirmatory factor analysis, structural equation modeling, and 5,000 bootstrap resamples were used to test the proposed relationships.The results showed that contextual cognition was significantly associated with multi-source information integration, which was further associated with design problem construction and human-AI collaborative judgment.Human-AI collaborative judgment was positively associated with use feedback and dynamic optimization.Significant indirect associations were also identified through multi-source information integration and design problem construction.These findings support a structural pathway from contextual cognition to information integration, task specification, collaborative judgment, and feedback updating in human-AI collaborative rural public space design.