Physical Computation and Finite Information: Hypercomputation, reliable access, and neural simulation
Hypercomputation requires an unbounded supply of accessible algorithmic information when reliable answers are obtained by effective processing at recognizable finite stages. Confining all the contributing resources to one fixed bounded spacetime region would require unbounded information capacity there. Algorithmic information theory makes this requirement precise: a noncomputable task has finite answer tables whose shortest descriptions have unbounded length, even when the table size is given. We bound this complexity by the information needed to describe resources that reproduce the answers. The bound separates information in an exact physical state from information that can be reliably extracted, including from quantum states. Temporal proposals assume access to a limiting answer after infinitely many steps. A computable history can have a noncomputable limit, but Lloyd's operation bound excludes such infinite completion at finite energy and duration within its counting model. Relativistic proposals require a separate accounting of the entire contributing history. Gandy-type conditions connect effective simulation to the structure and locality of physical mechanisms. The remaining physics problem is to bound accessible information and computational activity for the complete resource, including its preparation and readout. This account specifies what hypercomputation claims about brains would require of physics, while keeping computability distinct from the realization claims of biological naturalism.
Authors
- Giulio Ruffini (ORCID: https://orcid.org/0000-0003-3084-0177)
Institutions
- Neuroelectrics Barcelona SLU (Spain) (ES)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-16
- DOI
- https://doi.org/10.5281/zenodo.24281423
- Primary Topic
- Computability, Logic, AI Algorithms
- Type
- preprint