Attentional Effects of Environmental Scenes After Disruption of Global Spatial Coherence: Evidence from Scrambled Images

Urbanization and increasing cognitive demands have heightened interest in identifying environmental features that support attentional restoration. Based on Attention Restoration Theory, this study investigated whether different environmental categories influence specific attentional processes after global spatial coherence is disrupted by image scrambling. A within-subjects experimental design was conducted with 26 undergraduate participants across four environmental categories: natural, cultural heritage, urban nighttime, and urban daytime environments. An image-scrambling procedure was applied to disrupt global spatial coherence while retaining multiple visual properties of the original images, and attentional performance was assessed via the Attention Network Test to measure the efficiency of alerting, orienting, and executive-control networks using repeated-measures ANOVAs. The results revealed significant environment type × testing phase interactions for alerting and executive control after Greenhouse–Geisser correction and Holm adjustment, whereas the orienting interaction was not significant. Follow-up analyses showed a significant within-condition pre–post improvement in executive control only in the natural condition after Holm correction; these within-condition comparisons do not establish differences in benefit across environments. Neither the alerting nor the executive-control interaction remained significant after RT screening, indicating sensitivity to preprocessing choices. The orienting null finding should be interpreted cautiously because the orienting difference score showed variable reliability across administrations. The executive-control interaction remained significant after adjustment for the specified session-position terms, although these analyses do not rule out practice effects or other temporal influences. After adjustment for pre-test executive-control performance and session position, the natural-versus-urban-daytime post-test contrast was not significant. These findings suggest that different scrambled environmental categories were associated with different patterns of executive-control change in the primary analysis, but do not establish a robust advantage of the natural condition independent of baseline performance and RT-processing choices. However, because the scrambling manipulation retained multiple visual properties simultaneously, the present study cannot determine which specific visual attributes were responsible for these effects. Because no intact-scene condition was included, the present study cannot determine whether the observed changes differ from those elicited by intact environmental scenes or experimentally separate the relative contributions of local visual information, global spatial coherence, and scene-level meaning.