Integrated Predictive Workspace Theory (IPWT)

Academic Preprint：

PDF | 中文版本

What is IPWT?

TL;DR: Consciousness is a dynamic process that emerges in the workspace to minimize free energy, with maximizing synergistic information as the optimal computational strategy.

Project Timeline

• June 5, 2025: IPWT proposed.
• June 7, 2025: Formalization completed.
• June 8, 2025: Manuscript moved to this standalone repository.
• June 18, 2025: IPWT has completed its "PoPI", aligning with the solid experimental evidence from Dr. Luppi's work "A synergistic workspace for human consciousness revealed by Integrated Information Decomposition.".
• June 25, 2025: Manuscript underwent a thorough rewrite and workflow changes. Completed the two core computation packages: ΩID and ΣPI.
• August 4, 2025: Friston's A beautiful loop confirmed our trajectory. So long, and thanks for all the fish.
• January 11, 2026: Urbina-Rodriguez et al. demonstrated that LLMs spontaneously develop "Synergistic Cores" similar to the human brain, providing definitive evidence for the emergence of synergistic information processing as a fundamental property of intelligence.

Acknowledgements

The nascent form of this theoretical framework was initially developed as a foundational setting for the science fiction world-building project "Chain://". This project is dedicated to exploring the profound impacts that the core concept of Formalized Realism might trigger at social, technological, and philosophical levels. In the process of constructing the inherent logic and operating rules for the "Chain://" universe, we gradually refined and formalized the core ideas of this framework. Therefore, the birth of this research is inseparable from the vast imaginative space and rigorous logical deduction requirements provided by science fiction creation.

We extend our sincere gratitude to all artificial intelligence systems involved in this cross-disciplinary exploration, which made it possible to transform fictional world settings into a serious scientific theory. The resulting technical applications and sociological unit tests are documented in the Chain:// project.

Citation

If you wish to cite this work, please use the following BibTeX entry:

@misc{ipwt2025,
  author       = {Rui, L.},
  title        = {{Integrated Predictive Workspace Theory: Towards a Unified Framework for the Science of Consciousness}},
  year         = {2025},
  publisher    = {Zenodo},
  doi          = {10.5281/zenodo.15676304},
  url          = {https://doi.org/10.5281/zenodo.15676304}
}

License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.