Author: cxdig

Energy and Information

Klaus Jaffe

The literature contains many contradictory conceptual descriptions of therelation between Energy and Information. Here I argue that Information is not energy.Information is a representation or description of spatiotemporal arrangements (order)of matter and energy, encoded onto a physical substrate. Known substrates includeelectromagnetic waves, material structures, chemical molecules, and neural networks—whether in brains or computers. In quantum mechanics, when information pertainsto elementary particles, the object of study and the substrate encoding the informationcoincide. This leads to view energy and information as the same phenomenon, leadingto counterintuitive and often perplexing interpretations of reality. The proposeddefinition distinguishes between thermodynamic entropy and information entropy,enabling a consilient bridge between quantum and classical mechanics, geneticinformation, human knowledge, personal models of the world, consciousness andempathy. It facilitates the study of different kinds of information—especially the kindthat generates free energy and enables useful work, as studied by infodynamics. Thecentral insight is that energy and information are distinct, irreducible properties of theuniverse. Understanding their interplay requires considering four foundationalelements: spacetime, matter, energy, and information. These clarifications arefundamental for research in natural and artificial intelligence

Read the full article at: SSRN

Diffusion of complex contagions is shaped by a trade-off between reach and reinforcement

Allison Wan, Christoph Riedl, and David Lazer
PNAS 122 (28) e2422892122
How does social network structure amplify or stifle behavior diffusion? Existing theory suggests that when social reinforcement makes the adoption of behavior more likely, it should spread more—both farther and faster—on clustered networks with redundant ties. Conversely, if adoption does not benefit from social reinforcement, it should spread more on random networks which avoid such redundancies. We develop a model of behavior diffusion with tunable probabilistic adoption and social reinforcement parameters to systematically evaluate the conditions under which clustered networks spread behavior better than random networks. Using simulations and analytical methods, we identify precise boundaries in the parameter space where one network type outperforms the other or they perform equally. We find that, in most cases, random networks spread behavior as far or farther than clustered networks, even when social reinforcement increases adoption. Although we find that probabilistic, socially reinforced behaviors can spread farther on clustered networks in some cases, this is not the dominant pattern. Clustered networks are even less advantageous when individuals remain influential for longer after adopting, have more neighbors, or need more neighbors before social reinforcement takes effect. Under such conditions, clustering tends to help only when adoption is nearly deterministic, which is not representative of socially reinforced behaviors more generally. Clustered networks outperform random networks by a 5% margin in only 22% of the parameter space under its most favorable conditions. This pattern reflects a fundamental trade-off: Random ties enhance reach, while clustered ties enhance social reinforcement.

https://www.pnas.org/doi/abs/10.1073/pnas.2422892122

Participatory Evolution of Artificial Life Systems via Semantic Feedback

Shuowen Li, Kexin Wang, Minglu Fang, Danqi Huang, Ali Asadipour, Haipeng Mi, Yitong Sun

We present a semantic feedback framework that enables natural language to guide the evolution of artificial life systems. Integrating a prompt-to-parameter encoder, a CMA-ES optimizer, and CLIP-based evaluation, the system allows user intent to modulate both visual outcomes and underlying behavioral rules. Implemented in an interactive ecosystem simulation, the framework supports prompt refinement, multi-agent interaction, and emergent rule synthesis. User studies show improved semantic alignment over manual tuning and demonstrate the system’s potential as a platform for participatory generative design and open-ended evolution.

Read the full article at: arxiv.org

A critical phase transition in bee movement dynamics can be modeled using a 2D cellular automata

Ivan Shpurov, Tom Froese

The collective behavior of numerous animal species, including insects, exhibits scale-free behavior indicative of the critical (second-order) phase transition. Previous research uncovered such phenomena in the behavior of honeybees, most notably the long-range correlations in space and time. Furthermore, it was demonstrated that the bee activity in the hive manifests the hallmarks of the jamming process. We follow up by presenting a discrete model of the system that faithfully replicates some of the key features found in the data – such as the divergence of correlation length and scale-free distribution of jammed clusters. The dependence of the correlation length on the control parameter – density is demonstrated for both the real data and the model. We conclude with a brief discussion on the contribution of the insights provided by the model to our understanding of the insects’ collective behavior.

Read the full article at: arxiv.org

An AI tool for scafolding complex thinking: challenges and solutions in developing an LLM prompt protocol suite

This paper reports an exploratory study examining the interaction between a theoretical framework for Complex Thinking and AI (LLMs), in terms of its potentialities and constraints. The aim was to develop and conduct a preliminary pilot evaluation of a tool comprising a prompt protocol suite for use with an LLM, to scafold Complex Thinking. The tool is designed for use by an individual or group in relation to a given Target System of Interest (i.e., a real-world system, a problem, or a concern), supporting the development of more complex understandings of such systems that can guide more efective and positive actions and decisions. We describe the process of developing a suite of prompt protocols for scafolding particular properties of Complex Thinking and report on the outcomes of a pilot test evaluation with a set of users across diferent domains.

Melo, A. T., Renault, L., Caves, L., Garnett, P., Lopes, P. D., Ribeiro, R., & Santos, F. (2025). An AI tool for scaffolding Complex Thinking: Challenges and solutions in developing an LLM prompt protocol suite. Cognition, Technology & Work. https://doi.org/10.1007/s10111-025-00817-6