Abstract

Epileptic pathology is commonly interpreted as a state of increased neural disorder or high entropy. We challenge this interpretation and propose that such conclusions arise from incomplete or unidimensional metric approaches. Using a Unified Neurotopological Framework, we characterize brain dynamics by integrating both informational content and network geometric structure. We quantify neural coherence through a multimodal metric set including Permutation Entropy, Persistent Homology, and Ricci Curvature. The framework is validated using EEG data by comparing status epilepticus with healthy resting-state activity (isostasis). Our results (p < 0.001) show that epileptic seizures do not reflect heightened disorder but instead constitute a catastrophic collapse of network geometry, resulting in disordered low-complexity coherence. Furthermore, we identify a paradox in which the resting healthy brain demonstrates low-ordered complexity, yet remains functionally coherent. These findings suggest that the critical biomarker differentiating health from pathology is not entropy magnitude but the dynamic geometry of neural coherence.