Psychophysics of natural scenes: A comparison of measurement methods for useful integration

This contribution compares two different methods to measure some global properties (GPs) of natural images/scenes, in relation to the perceived restorativeness (PR). According to the scene-centered approach, GPs are basic characteristics of a scene (e.g., openness, expansion, depth) that guide us in identifying other higher-level characteristics (e.g., whether a scene represents a natural or built environment). Indeed, the literature has shown, on the one hand, that images of natural environments, as opposed to built ones, generate a greater PR (i.e., the sense of recovery of emotional and cognitive resources; Berto, 2014), and on the other hand, that GPs play a role in this process (Celikors, 2022). We investigated the topic by comparing two different methods of measuring GPs. We conducted an experiment consisting of two phases. Eight photos depicting natural scenes were shown to seventy-five participants. In the first phase, we considered six GPs (Greene & Oliva, 2009), openness, expansion, depth, concealment, navigability, and transience, measured by means of six unidimensional scalings, and PR dimensions of fascination, coherence, being away and scope, measured by PRS scale (Pasini et al., 2014). In this phase, we performed a pairwise comparison of the six GPs among the eight images. Correlation and cluster analysis were conducted. In the second phase, the same eight images were rated on the six GPs on a 7-point Likert scale, and correlation and regression analysis were conducted. The results, strengths, and weaknesses of the two measurement methods are discussed