This research tested the visual rightness theory of pictorial composition's assertion that the induced organizational structure of a visually right (i.e., "good") design is perceptually salient and judged superior by anyone viewing it regardless of his or her training in the visual arts. Stimuli for Experiments 1 and 2 consisted of 16 reproductions of paintings by renowned artists and an experimentally reconstructed less-well-organized version of each art stimulus. It was found that design professionals (Experiment 2) were significantly more successful at detecting the original versions than were participants untrained in the visual arts (Experiment 1) (hit RATES=64% and 55%, respectively). In Experiment 3 participants replaced a major structural element removed from each of six pictures of the stimulus set at the location where they thought it appears in the original. A significant number of untrained participants and those with training in design theory were in agreement as to the location of each element within its pictorial field; the location chosen conformed to its compositional structure but not its actual location in the original. Findings demonstrate that the ability to detect the induced structural skeleton of a painting resulting from a visually right design does not require expert knowledge of design principles whereas the ability to discriminate between several articulation possibilities of the same composition does require formal training.
Locher, P. J., Stappers, P. J., & Overbeeke, C. J. (1999). An empirical evaluation of the visual rightness theory of pictorial composition. Acta psychologica, 103(3), 261-280. https://doi.org/10.1016/S0001-6918(99)00044-X