Table 4 The directions of the results in the model in Step 2.2.

From: Cross-cultural comparison of beauty judgments in visual art using machine learning analysis of art attribute predictors among Japanese and German speakers

Significant predictors

The direction of the results

Formal perceptual attributes

visualharmony_disturbingforms

Paintings with visual harmony were evaluated as more beautiful, and with disturbing forms as less beautiful.

simple_complex

Complex paintings were evaluated as more beautiful, and simple ones as less beautiful

lesscolor_colorful

Colorful paintings were evaluated as more beautiful, and paintings with less colors as less beautiful.

Content representational attributes

negativevalence_positivevalence

Paintings with positive valence were evaluated as more beautiful, and with negative valence as less beautiful.

emotionless_emotionloaded

Emotion loaded paintings were evaluated as more beautiful, and emotion less ones as less beautiful.

Back to article page