Comparing aesthetic preferences and agreement ratings in depictions of natural and urban environments

Studies have shown that individuals usually prefer natural over built environments, assuming a biological preparedness for nature. Consensus in preference ratings has been reported, though with inconsistencies, with the largest effects in the comparison of faces and abstract art. The present study aimed to assess agreement in a set of environmental images displaying natural as well as built environments. We hypothesized that individuals would show greater agreement for the natural environment than for the built environment.

Ninety-eight participants (69.4 % of whom were female) rated 200 images from four environmental categories (50 each): very scenic natural, mid-scenic natural, mid-scenic built, and un-scenic built, with respect to valence and arousal using self-assessment-manikin rating-scales. Agreement ratings were assessed applying the mean-minus-1 method. Analyses were conducted with linear mixed-effects models. The results exhibit a linear increase from relaxing to stressful and most preferred to least preferred for very scenic natural images over mid-scenic natural and mid-scenic built to very un-scenic built environment. Significant differences in the agreement ratings between the groups emerged for valence, but only for the mid-scenic built category in comparison to the other categories. For arousal, agreement was significantly different only in the very scenic natural images with lowest consensus ratings. No intuitive agreement on natural environments emerged, contradicting existing findings.