Relationships among seven aesthetic response scales were studied by requiring subjects to rank fifteen paintings on each scale, using a between-subjects design. Three of the five evaluative scales used, likeability, pleasingness, and preferability were strongly positively intercorrelated. Using these scales to examine painting content (landscape, portrait, still-life) and style (Impressionism, Surrealism, etc.) effects, however, revealed that the scales did not always yield similar results. Although content effects were similar on all three scales, likeability and preferability were relatively insensitive to style effects, while pleasingness was more sensitive. These sensitivity differences appear to be linked to variation in the degree of intersubject agreement on the different scales, leading to the suggestion that some scales, such as pleasingness, are relatively homogeneous while others, such as likeability, are more heterogeneous. Another commonly-used evaluative scale, interestingness, was unrelated to the other four but was relatively sensitive to style effects. Data are also presented on an additional evaluative scale, wish to see again, and on two descriptive scales, complexity and familiarity. Overall, the results suggest that conclusions drawn from studies using aesthetic scales may depend crucially on the particular scale used and in particular that the commonly used likeability and preferability scales, despite their apparent ecological validity, may not be the most informative ones.
