Mechanisms Linking Urban Plant Landscape Spatial Characteristics to Stress Regulation and Attention Recovery

ABSTRACT As an important component of landscape design, the impact of planted landscape space on human health has received considerable attention in modern urban landscape gardening research. However, most existing studies focus on the overall benefits of urban green space and lack quantitative analyses of planted landscape spatial features (e.g., spatial scale, slope, and plant community structure level) about psychological recovery and physiological regulation. This study investigated the effects of different scale groups, slope groups, and plant composition groups on stress reduction and attention recovery. Thirteen spatial models of planted landscape space were constructed using virtual reality technology, covering five scale groups, five slope groups, and three plant composition groups. Fifty‐nine participants were randomly assigned to three experimental groups, and their physiological (HR, SpO 2 , EEG) and subjective (stress, attention, preference) responses were assessed and analyzed using ANOVA with post hoc tests to determine significant differences among spatial groups. The results showed that the largest scale space (12,000 m 2 ) was the most effective for stress reduction (HR: p ≤ 0.001; β/α: p = 0.016; PMRelaxation: p = 0.038; PMStress reduction questionnaire: p ≤ 0.001); the smallest scale (400 m 2 ) was more beneficial for attention restoration (θ/β: p ≤ 0.001; PMAttention: p = 0.002; PMEngagement: p ≤ 0.001; PMExcitement: p = 0.036; PMInterest: p = 0.037; PMFocus: p = 0.029); and the moderate scale (6000 m 2 ) received the highest subjective preference (Preference questionnaire: p = 0.047). Gently sloped space (5%) demonstrated the best performance across stress reduction, attention restoration, and subjective preference (HR: p ≤ 0.001; β/α: p = 0.007; PMStress: p = 0.005; PMRelaxation: p = 0.002; PMStress reduction questionnaire: p ≤ 0.001; θ/β: p = 0.039; PMAttention: p = 0.043; PMEngagement: p ≤ 0.003; PMExcitement: p = 0.037; PMInterest: p = 0.015; PMFocus: p = 0.019; Preference questionnaire: p ≤ 0.001). Mid‐level planting was most significant for attention restoration and preference (HR: p ≤ 0.001; θ/β: p = 0.041; PMAttention: p = 0.068; PMEngagement: p ≤ 0.020; PMExcitement: p = 0.022; PMInterest: p = 0.008; PMFocus: p = 0.036), while lower‐level planting was more conducive to stress regulation (β/α: p = 0.039; PMStress: p = 0.011; PMRelaxation: p = 0.035; Stress reduction questionnaire: p ≤ 0.001; Preference questionnaire: p ≤ 0.001). This study provides a quantitative basis for the spatial design of planted landscape space, elucidates the mechanisms through which different spatial features influence psychological and physiological health, and offers theoretical support for urban green space planning.