Table 5 The directions of the cross-cultural differences.

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

Predictors

Results

Formal perceptual attributes

finebrushstroke_roughbrushstroke

While this feature was important for Japanese people, it was not for German people: Japanese evaluate more fine strokes as more beautiful.

darkcolorworld_warmcolorworld

While this feature was important for Japanese people, it was not for German people: Japanese evaluate warm color as more beautiful.

softpastell_intensivestrong

While this feature was important for Japanese people, it was not for German people: Japanese evaluate pastel color as more beautiful.

Content represnetational attributes

negativevalence_positivevalence

This feature was important for both cultures, but more important for German speaking people: same direction, always the positive was appraised as more beautiful.

emotionless_emorionallyloaded

While this feature was important for German, it was not for Japanese people: German evaluated more emotionally loaded paintings as more beautiful.

realisticcontent_imaginativecontent

While this feature was important for Japanese people, it was not for German people: Japanese evaluate realistic painting as more beautiful.

photorealistic_painterly

While this feature was more important for Japanese people than for German people: The Japanese people evaluate photorealistic painting as more beautiful. Note this feature was not significant in the Step 2.1.

loadofcontext_focus

While this feature was more important for Japanese people than for German people: The Japanese people evaluate paintings with loads of contents as more beautiful. Note this feature was not significant in the Step 2.1.