Authors:
Walter Crist, Eric Piette, Karen Jeneson, Dennis J. N. J. Soemers, Matthew Stephenson, Luk van Goor, Cameron Browne

Venue:
Antiquity, 2026

Topics:
archaeogaming, general game playing, AI simulation, cultural heritage, use-wear analysis

Links: PDF · Publisher

This paper investigates a Roman-era stone artefact discovered in Coriovallum (modern Heerlen, Netherlands), whose incised lines suggest the presence of a board game but do not match any known game.

The study combines use-wear analysis and AI-driven simulations to evaluate candidate rule sets. By analysing abrasion patterns on the stone and comparing them with simulated gameplay trajectories, the authors identify plausible game mechanics consistent with the observed wear.

Results indicate that the artefact was most likely used as a game board for a blocking-type game, extending the historical evidence for this class of games and demonstrating the effectiveness of combining archaeological methods with general game AI.

This work is part of the broader Digital Ludeme Project and the GameTable COST Action, which aims to reconstruct and analyse historical games using computational methods.

The approach integrates Ludii-based simulations with physical evidence from archaeology, enabling the reconstruction of rules for games that were never formally recorded.

More broadly, this research demonstrates how artificial intelligence can contribute to cultural heritage studies, offering new ways to interpret artefacts and understand past human behaviour through gameplay reconstruction.

Crist, W., Piette, E., Jeneson, K., Soemers, D. J. N. J., Stephenson, M., van Goor, L., Browne, C. (2026). Ludus Coriovalli: using artificial intelligence-driven simulations to identify rules for an ancient board game. Antiquity.

@article{crist2026ludus_coriovalli,
  author  = {Crist, Walter and Piette, Eric and Jeneson, Karen and Soemers, Dennis J. N. J. and Stephenson, Matthew and van Goor, Luk and Browne, Cameron},
  title   = {Ludus Coriovalli: using artificial intelligence-driven simulations to identify rules for an ancient board game},
  journal = {Antiquity},
  year    = {2026},
  doi     = {10.15184/aqy.2025.10264}
}