Perceptual fluency and eye movements when viewing urban and natural scenes

Abstract

A number of eye-tracking studies have shown that viewing natural environments is associated with reduced eye movement activity compared with viewing built environments. This has been linked to the cognitive benefits of viewing nature and explained in terms of Kaplan and Kaplan's Attention Restoration Theory. However, the theory has recently been criticized for the lack of empirical evidence supporting its framework. The first aim was to replicate the results of previous eye movement studies using different visual stimuli. In addition, we investigated whether reduced eye movements when viewing natural versus urban images could be explained by greater perceptual fluency and fractal complexity of the images. The participants (N=66) viewed images of forests with and without foliage and images of urban apartment buildings while their eye movements were recorded. The self-reported perceptual fluency and fractal complexity of the presented images were measured. Analysis of eye movements revealed significantly less eye movement activity (a reduced number of fixations that are longer) when viewing natural images than urban ones, consistent with previous findings. There was no significant difference between images containing foliage and those without for any of the measured variables. However, mediation analysis did not show significant effects of perceptual fluency on eye fixations. Moreover, while previous research suggests that fractal structure may be one of the mechanisms underlying perceptual fluency, mediation analysis did not reveal any significant effects of fractal complexity on eye fixations. This raises the question of what causes the differences in eye movement patterns and the restorative effects of nature versus urban scenes. Further research should address the specific spatio-cognitive dimensions of natural images and the individual differences that may affect how people move their eyes when processing different types of scenes.