Tracking art’s evolution with entropy and complexity

Mar 13, 2019, 7:56 AM EDT
(Source: Art Gallery ErgsArt - by ErgSap/flickr)
(Source: Art Gallery ErgsArt - by ErgSap/flickr)

Entropy is too thermodynamic a term to bear any proximity to art. Well, a trio of physicists is upending that notion with an intriguing method of tracking art’s evolution.

Physicists Higor Y.D. Sigaki, Matjaž Perc, and Haroldo V. Ribeiro scanned nearly 140,000 paintings, created between 1031 and 2016, by some 2,000 artists in around 100 styles, and mapped out the degrees of complexity and entropy in these artworks, writes

The trio found a characteristic pattern in the paintings from different eras when analyzed for their entropy (the disorder or randomness) and complexity (the variability of patterns) within an image. The works dating back to the 17th century and prior showed relatively low levels of entropy and high complexity than found in Modern Art.

This entropy-complexity-based analysis yielded a timeline for art’s evolution that perfectly meshes with the established trajectory of art history. This novel probe classifies art in three categories, namely, Renaissance, Neoclassicism and Romanticism; Modern Art; and Contemporary/Postmodern Art — all these groups occupying different positions on the entropy-complexity spectrum.

The researchers suggest that these metrics can be instrumental in studying the interactions between different artistic periods and in attributing specific artistic styles to lesser-known works, notes Scientific American.