The ridges left by a brush survive as a microscopic landscape. Researchers at Case Western Reserve University mapped that surface and used AI to compare patches in two El Greco-associated works, finding a relationship that could change how parts of Baptism of Christ are attributed.[1][2]
The team studied works associated with El Greco, including Christ on the Cross at the Cleveland Museum of Art and Baptism of Christ in Toledo. Their method combined fine-scale topographic scanning of the painted surface with AI analysis of relationships among patches. One question was whether portions long understood as workshop completion might have more in common with another area of a painting than art history had assumed.[1][2]
Looking at the hills and valleys of paint
Evidence at a glance
| Measure | Value | Context |
|---|---|---|
| Works examined | Two paintings | Christ on the Cross and Baptism of Christ |
| Measured evidence | Microscopic surface topography | Brushstroke texture analysed as spatially related patches |
| Attribution result | Suggestive relationship | Not a definitive finding that one historical artist painted every region |
A conventional photograph records colour and reflected light. A topographic scan records relief: a tiny raised ridge, a groove or a variation in the surface. At the scale of a brush bristle, texture may preserve information about technique, handling and material. The researchers divided scanned areas into small patches and used a computational approach to examine similarities and relationships across the surface rather than judging a single dramatic stroke.[1][2]
The distinction is important. Two areas can look different in colour and iconography while sharing similar underlying handling. Equally, two visually similar faces may have been painted with different mixtures, different tools or by different people. Surface texture gives investigators another layer of evidence, one that does not require attributing every patch solely from visual style.
The team found strong consistency in Christ on the Cross. In Baptism of Christ, regions previously attributed to different hands showed underlying connections. The researchers say the pattern could reflect a common set of materials or possibly a single hand—alternatives that further art-historical evidence must distinguish.[1][2]
Separating a painter’s hand from a workshop’s materials
A workshop often shares brushes, pigments, preparation methods and training. The master and assistants may deliberately imitate one another. A restorer, centuries later, may add paint or modify surface texture. A detector that finds similar patterns must therefore distinguish individual style from common materials, common methods and later conservation history. The researchers explicitly treat their results as suggestive and open to confirmation from broader evidence.[1][2]
Securely attributed reference works and documented restoration history provide ways to test the interpretation. If similar texture follows a shared material or tool rather than an individual, that is still useful information about the workshop, but it answers a different question.
The science is a second conversation with the painting
The surface measurements add evidence that an ordinary photograph does not capture. They can guide the next material analysis or historical comparison, returning the attribution debate to marks that remain on the canvas.[1][2]
Sources & further reading
- Case Western Reserve researchers using AI at hair-width scale to reveal Renaissance master’s hidden hand Case Western Reserve University · 21 Apr 2026 · Accessed 24 Oct 2026
- PATCH: A deep learning method to assess heterogeneity of artistic practice in historical paintings Science Advances / PMC · 17 Apr 2026 · Accessed 24 Oct 2026