Self-relevance predicts the aesthetic appeal of real and synthetic artworks generated via neural style transfer

What determines the aesthetic appeal of artworks? Recent work suggests that aesthetic appeal can to some extent be predicted from a visual artwork’s image features. Yet, a large fraction of variance in aesthetic ratings remains unexplained and may relate to individual preferences. We hypothesized that an artwork’s aesthetic appeal depends strongly on self-relevance. In a first experiment, observers viewed real artworks and rated them for aesthetic appeal and self-relevance. Aesthetic appeal was positively predicted by self-relevance. In a second experiment, we developed a method to create synthetic, self-relevant artworks, by using deep neural networks that transferred the style of exist- ing artworks to photographs. Style transfer was applied to self-relevant photographs which were identified based on autobiographical memories, self-identity, interests, common activities and pref- erences. Self-relevant, synthetic artworks were rated as more aesthetically appealing than matched control images, at a level similar to real artworks. Thus, self-relevance is a key determinant of aesthetic appeal, independent of artistic skill and image features.