Melvin, Chen and Erik, Cambria and Cecilia, Laschi and Gianmarco, Mengaldo (2025) Intelligence is Physical: Energy, Information, & a Thermodynamically Grounded Theory of Minds & Machines. [Preprint]

Text matryoshka_updated_with_names.pdf Download (421kB)

Abstract

Despite the prominence of formalist theories of intelligence within the artificial intelligence research tradition, intelligence cannot be fully understood as a purely formal or symbolic process. Instead, it emerges from the physical interactions of energy, matter, and information across nested levels of complexity, from basic physical structures to life, neural systems, and adaptive behaviour. In this paper, we introduce a hierarchical, thermodynamically grounded theory of intelligence that emphasizes the coupling of sensory, cognitive, and motor processes with metabolic and informational constraints. By explicitly considering the energy costs and physical instantiation of computation, our theory provides a principled framework for understanding natural intelligence and guiding the design of sustainable, efficient, and adaptive artificial systems.

---

Export/Citation:

EndNote | BibTeX | Dublin Core | ASCII/Text Citation (Chicago) | HTML Citation | OpenURL

Social Networking:

Share |

---

Item Type:	Preprint
Creators:	Creators Email ORCID Melvin, Chen melvinchen@ntu.edu.sg 0000-0003-1756-8666 Erik, Cambria cambria@ntu.edu.sg 0000-0002-3030-1280 Cecilia, Laschi cecilia.laschi@nus.edu.sg 0000-0001-5248-1043 Gianmarco, Mengaldo mpegim@nus.edu.sg 0000-0002-0157-5477
Keywords:	Thermodynamic; intelligence; nested ontology; coupling logic; scaling laws; information engine; entropy
Subjects:	Specific Sciences > Complex Systems Specific Sciences > Cognitive Science Specific Sciences > Artificial Intelligence
Depositing User:	Dr. Melvin Chen
Date Deposited:	05 Jan 2026 13:56
Last Modified:	05 Jan 2026 13:56
Item ID:	27741
Subjects:	Specific Sciences > Complex Systems Specific Sciences > Cognitive Science Specific Sciences > Artificial Intelligence
Date:	30 December 2025
URI:	https://philsci-archive.pitt.edu/id/eprint/27741

Monthly Views for the past 3 years

Monthly Downloads for the past 3 years

Plum Analytics

Actions (login required)

View Item