Hinrichs et al. apply discrete geometry to decode interbrain synchrony

Traditional methods for measuring interbrain synchrony rely on fixed, correlation-based approaches that only describe surface-level patterns. A new paper by Nicolás Hinrichs, Noah Guzmán, and Melanie Weber introduces a fundamentally different perspective: **discrete geometry**. Inspired by network science, the researchers model interbrain connectivity as a dynamic geometric structure that evolves during social exchanges. Their pipeline computes curvature distributions across neural networks and uses entropy metrics to identify critical transitions—moments when the geometric configuration shifts significantly. This captures richer dynamics than simple correlation scores. The 4-page paper includes one figure and two appendices, and was accepted at the NeurIPS 2025 Workshop on Symmetry and Geometry in Neural Representations (NeurReps), with proceedings published in PMLR (Volume 325, pages 145–152).

The geometric method positions hyperscanning—simultaneous brain recording from multiple people—as a tool for uncovering causal mechanisms in social behavior. By interpreting connectivity changes through evolving curvature, the framework moves beyond descriptive observations toward a more explanatory model of neural interaction. The authors argue that this geometric lens can reveal how brains dynamically coordinate during conversation, cooperation, or conflict. While still in the workshop stage, the approach promises to enhance brain-computer interfaces and social neuroscience studies by providing a principled way to detect when and how neural networks reconfigure during real-time interaction. The paper is available on arXiv (ID: 2509.10650) and has been updated through four versions, with the latest revision in July 2026.