The relationship between nature exposures and attention restoration, as moderated by exposure duration: A systematic review and meta-analysis
Cameron N. Bell, Rebecca George, Cynthia A. Honan, L. Bell, Alex T.W. Jolly, Allison Matthews
Abstract
Much prior research suggests that exposure to natural environments is associated with improved cognitive performance, in comparison to non-natural exposures. However, the presence and size of such nature benefits are inconsistent. This meta-analysis aimed to determine the influence of nature exposure duration, a likely moderating factor, on cognitive performance change. In total, 273 individual outcomes from 80 studies were included in correlated hierarchical effects (CHE) multilevel meta-analyses. When analysing all cognitive outcomes, the average performance improvement was larger for nature exposures compared to non-natural comparators. Of 11 cognitive domains identified in the literature, Working Memory and Attentional Control were the only domains to show reliable nature benefits. However, subgroup analysis suggested that effect size differences between cognitive domains may not be significant. Meta-regressions indicated exposure duration significantly moderated natural/non-natural differences in restoration in a non-linear relationship, though the shape and significance of this relationship within each cognitive domain was not consistent. Nature benefits were generally larger for participants who underwent cognitive fatigue, but other experimental design variations did not produce meaningful differences in effects. Overall, effects sizes were small and heterogeneity was substantial. The largest difference in cognitive restoration between natural/non-natural exposures appears to occur after approximately 30 min of environmental exposure. • Nature is more restorative than non-nature exposure for overall cognitive capacity. • Exposure durations moderate the restorative effects of nature. • Benefits are largest when nature exposures are approximately 30 min long. • Overall, effects are small and substantially heterogenous.