Thermodynamic Time and the Architecture of Cognition
Year: 2026
Type: article
Abstract: This paper develops a unified framework in which cognition is understood as a thermodynamic process operating across biological, cultural, and planetary scales. Rather than treating time as a symbolic or psychological construct, the study argues that time is a physical gradient that regulates the stability, evolution, and decay of cognitive systems. Drawing on archaeoastronomy, nonlinear dynamics, ancient metrology, and contemporary machine learning, the work demonstrates how early civilisations encoded astronomical cycles into practical instruments such as the wheel‑cross. These devices anchored memory, navigation, and social coordination to the physical sky, providing a stable cognitive architecture grounded in thermodynamic reality. In contrast, modern AI systems reveal the fragility of symbolic cognition when detached from physical grounding, exhibiting catastrophic forgetting analogous to cultural loss in human societies. By comparing ancient sky‑regulated knowledge systems with contemporary machine intelligence, the paper proposes a cross‑scale model of cognition in which thermodynamic constraints shape both ancient scientific instruments and modern computational architectures. The result is a synthesis linking prehistoric measurement systems, planetary regulation, and the future trajectory of artificial intelligence—showing that cognition, whether human or machine, ultimately emerges from the same underlying thermodynamic arrow of time.
Language: English
Subfield: Archeology (arts and humanities)
Field: Arts and Humanities
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7204032942