Abstract
Purpose: Artists specializing in two-dimensional media choose the most aesthetically appealing spatial composition and scene architecture for the subject of their artwork. Palmer et al (2008) discovered a strong inward and center facing aesthetic bias. Barghout et al (2010, 2008) found that a model of scene architecture composed of hierarchically nested spatial-taxons was consistent with the "entry level" object labels proposed by Jolicour et al (1984). This model followed a rank-frequency distribution that was independent of image content and image subject numerosity, and was consistent with a law of least effort. In this study, we explored if these established findings were invariant under aesthetic and non-aesthetic spatial composition condtitions. Methods: Paper surveys consisting of black and white drawings of inward facing and outward facing figures were distributed. Participants were asked to draw an ‘x’ on the center of the subject of the image. Participants were asked to rank the aesthetic of the image using a Likert scale and write down a few words to describe the image. An N by M factorial design was used, where N represented single vs. dual subject composition and M represented aesthetic vs. non-aesthetic configuration. K-means clustering analysis determined spatial-taxon designation of the images as defined by Barghout et al (2010,2008). Rank-frequency distributions for spatial-taxons and corresponding word labels were fit using linear regression analysis. Results: Our dataset of annotated images with empirically derived spatial-taxons and their relative frequencies provides insight into the interaction effects between aesthetic configuration, spatial-taxon organization of scene architecture, and word labels. The frequency distribution of spatial-taxons for both aesthetic and non-aesthetic dual-subject figures followed a Gaussian distribution while single-subject figures with a non-aesthetic configuration followed an inverted Gaussian frequency distribution. Keywords: aesthetic preference, spatial composition; rectangular frame, spatial-taxon, scene perception, entry-level categories, law of least effort, object recognition
Meeting abstract presented at VSS 2013