Neural Basis of Action Simulation in Architectural Perception: A Multi-voxel Pattern Analysis Study

A growing body of neuroscience evidence indicates that perceiving everyday objects and environments triggers mental simulations of potential actions, preparing the brain for movement even before any motion occurs. Although previous research on navigation has demonstrated how the spatial configuration of corridors and doorways guide locomotion, it remains unclear whether architectural elements (e.g., doors, windows, stairs) that require direct interaction evoke distinct, action-specific representations in the brain. Thus, the aim of this study is to investigate whether the passive perception of such elements (viewed without any intention to act) spontaneously recruits motor-related brain networks and whether these activations are specific to the actions implied by each element. fMRI data were collected from 31 participants during passive viewing of each architectural element followed by imagery of four actions: one reflecting the element's primary function, one representing manipulation, and two corresponding to peripheral actions. Using multi-voxel pattern analysis, we created classification ratio maps to identify action-architectural element associations. Our analyses revealed that passive perception significantly elicits neural patterns aligned with the elements' primary function while also encoding sensorimotor representations of the required interactions. Furthermore, our results show that this anticipatory process activates motor-related regions, including premotor cortex and the dorsal visual pathway. These results underscore the crucial role of action simulation in architectural perception and demonstrate the brain's readiness to engage with environmental affordances at the sensorimotor level. Consequently, we propose possibilities for a neuroscience-informed design, suggesting that architecture can be optimized to align with the embodied nature of human cognition.