A mathematical method borrowed from topology can reveal structural properties of visual art that correspond to how people perceive and respond to them, according to a new study published this week in the open-access journal PLOS Computational Biology by Jacek Rogala of the University of Warsaw, Poland, Shabnam Kadir of the University of Hertfordshire, UK, and colleagues.
For generations, researchers have tried to understand why certain works of art move us more than others. But a direct link between image properties and viewer response has remained elusive.
In the new study, researchers applied a technique called persistent homology—a method from computational topology that captures the structure of an image at multiple scales—to two sets of abstract paintings: works by recognized abstract artists, and "pseudo-art" generated by AI programs designed to produce art-like images. The topological method, the researchers found, was able to clearly distinguish real art from pseudo-art.
The researchers further analyzed a broader range of paintings by eminent abstract artists such as Wassily Kandinsky, Mark Rothko, and Jackson Pollock. They found that these artists' pieces converged on a specific rate of violation of a mathematical relationship known as Alexander duality. This value captures how artists balance visual structures at the edges of a composition versus its interior. The authors propose, based on this finding, that abstract artists may intuitively follow a mathematical "golden rule" in structuring their compositions
Then, the team studied people's eye movements and recorded their brain activity as they viewed the sets of images, both in a laboratory setting and in a gallery. People interacted differently with the two types of images: viewing real art was associated with more stable, integrative brain processing, while pseudo-art prompted more exploratory eye movements and greater perceptual uncertainty.
When the researchers mapped the eye movements with the topological features they had previously pinpointed, they found that where people looked at art corresponded to the structural features identified by the topological analysis.
"An important part of our study was to explore the relationship between topologically derived image features, eye movement, and aesthetic experience," the authors say. "Our research showed that our newly developed method not only clearly distinguished between two sets of images but also allowed us to map topological features onto gaze fixation heat maps."
Senior author Jacek Rogala adds: "What struck me most is that we could actually see the gallery environment doing something measurable. It wasn't just a backdrop — it changed which images held attention and for how long. That's a result you can put numbers on, and it still feels surprising."
In your coverage please use this URL to provide access to the freely available article in PLOS Computational Biology: https://plos.io/4tUFhPR
Citation: Dmitruk E, Bajno B, Kot L, Dreszer J, Bałaj B, Ratajczak E, et al. (2026) Art's hidden topology: A window into human perception. PLoS Comput Biol 22(5): e1014156. https://doi.org/10.1371/journal.pcbi.1014156
Author Countries: Poland, United Kingdom
Funding: This research was supported by the University of Warsaw under the Priority Research Area V of the "Excellence Initiative – Research University" programme. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
