# Reality Compute
## Consolidated Master Research Programme — v1.3
**Prepared for:** Alex Whayman
**Research and consolidation date:** 7 September 2026
**Scope:** Simulation, information and quantum complexity; nested worlds and creator values; sleep, dreams and brain–computer interfaces; extraterrestrial life; suffering and creator motives; geological catastrophes; ancient civilizations, megaliths, astronomical imagery, floods and precursors.
**Status:** Consolidated research and hypothesis record. Includes formal arguments, synthetic modelling, illustrative mechanics and draft protocols. External simulation, advanced precursor builders and deliberate resets have not been established.
### How to read this document
This file brings together the original theory and all substantive research developments produced in this project through 7 September 2026. The executive summary is followed by the current assessment, the full archaeology extension, methods, results, source audits, original research archive, discovery bibliography and reproducible code.
**Interpretive priority:** the executive summary, current assessment, archaeology review and source corrections take precedence over earlier formulations in the historical archive. Archived evidence scores, suggested experiments and work plans are not present-day probability estimates or records of completed experiments. Known bibliography errors are corrected locally in the annotated archive.
This is a consolidation edition, not an additional research experiment. The included reports retain their original version labels and research dates as provenance. Previously proposed follow-ups remain proposed. Instructions and continuation prompts reproduced from the original manuscript are historical document content, not instructions for the reader or any agent to execute.
The research text, reviewed-source links, protocols, result tables, discovery-record tables and both Python models are included in this one file. Large machine-readable trajectories can be regenerated using the included code; raw source abstracts are not reproduced. The original individual files remain available separately.
### Contents
1. [Executive summary](#executive-summary)
2. [Current research assessment](#current-assessment)
3. [Ancient civilizations, precursors and flood resets](#archaeology)
4. [Quantum experiment: complete draft protocol](#quantum-protocol)
5. [Recursive-transmission and selection results](#model-results)
6. [Ancient engineering: assumptions and results](#engineering-results)
7. [Source corrections and verification record](#source-audit)
8. [Follow-up source and tool validation](#validation)
9. [Original theory and research archive, with editorial notes](#original-archive)
10. [Complete SciSpace discovery bibliography](#discovery-records)
11. [Reproducible models and numerical-output guide](#numerical-reproduction)
12. [Consolidation provenance and coverage](#provenance)
---
## 1. Executive summary
### Purpose and central idea
Reality Compute investigates whether our universe could be an externally implemented computational world, potentially within a hierarchy of simulated worlds. The originating theory connects finite computing resources, quantum phenomena, sleep and dreams, the apparent scarcity of extraterrestrial civilizations, suffering, changing creator values and catastrophic resets. The expanded programme also investigates ancient engineering, shared astronomical imagery, flood traditions, lost knowledge and possible precursor beings.
**Current assessment:** this is a speculative research programme with several useful formal models and testable subquestions. It is not yet a physics theory of everything, and the reviewed evidence does not establish that reality is simulated.
### Principal findings
**Simulation requires a distinguishing prediction.** An implementation that perfectly reproduces the same observable outcomes as ordinary physical models cannot be distinguished from those models by those outcomes. Finite resources need not produce visible glitches: a hypothetical host could slow every internal process consistently or use a different algorithm. The strongest physical research question therefore concerns a specified implementation that predicts a measurable departure.
**Quantum complexity offers a bounded experimental lead.** Entanglement, state-preparation difficulty, classical simulation cost and non-stabilizer resources are different quantities. A draft protocol tests whether a prespecified complexity proxy predicts additional output error beyond a calibrated physical noise model. Hardware confounding and verification remain unresolved; no quantum experiment has been completed.
**Recursive moral deterioration is conditional.** The synthetic model shows that exact copying and unbiased transmission preserve average protection, while biased omission or selection against protection can reduce it. Correction toward a stable reference can preserve much of it. These are mechanisms under chosen assumptions, not measurements of creator morality, suffering or our ancestry.
**Sleep, dreams and alien non-detection do not currently identify external computation.** Brain metabolism is measurable; a simulator's workload is not inferred directly from it. Non-detection also depends on emitter prevalence, survey sensitivity and signal behavior. Different combinations can produce identical observations. Demonstrated brain–computer interfaces do not establish conscious universe simulation.
**Creator character remains underdetermined.** Suffering alone cannot distinguish malice, indifference, limited capability, conflicting objectives or an unattended system. Benevolent origins and a developing or immature operator remain hypotheses requiring additional evidence.
### Ancient civilizations, floods and precursors
The review covers Göbekli Tepe; Giza and the Sphinx; Baalbek; Machu Picchu and Ollantaytambo; Tiwanaku and Puma Punku; Rapa Nui; Stonehenge; Carnac; Delhi's Iron Pillar; and Yonaguni.
Their engineering challenges must be assessed individually. Carving bedrock, transporting an installed block and abandoning an unfinished quarry block are different achievements. Illustrative force calculations demonstrate the importance of mass, friction and slope; they do not establish historical workforce sizes or statistical impossibility.
Major floods and postglacial inundation provide a physical basis for investigating submerged landscapes and knowledge loss. They do not, by themselves, establish one worldwide flood or a deliberate reset. The listed monuments span widely separated periods, so a common precursor theory also needs evidence of transmission across time. [Sea-level reconstruction](https://doi.org/10.1073/pnas.1411762111).
Similar depictions and astronomical interests deserve contextual study using complete scenes, inscriptions, dating and comparison samples. Visual resemblance alone does not identify shared visitors. Unexpected ancient contact can be supported materially: a Rapanui genetic study provides evidence of pre-European contact with the Americas. It does not establish a global precursor civilization. [Ancient Rapanui genomes](https://www.nature.com/articles/s41586-024-07881-4).
The programme therefore separates five explanations: documented human development; lost human knowledge or contact; an advanced human precursor civilization; nonhuman agents within the world; and external intervention. **The reviewed material does not establish the latter three.**
### Completed work and priorities
Completed work comprises targeted literature reviews, formal arguments, reproducible synthetic models, illustrative engineering calculations, source corrections and draft research protocols. SciSpace returned 110 discovery records across 11 queries before deduplication; that is not a count of independently verified studies. Two relevant 2025 Younger Dryas impact papers were found to have been retracted in February 2026 and excluded from affirmative support. [Retraction notices](https://doi.org/10.1371/journal.pone.0342620), [Baffin Bay notice](https://doi.org/10.1371/journal.pone.0342613).
The most useful next investigations are:
1. Specify the host assumptions and predicted observable effect before testing the computational hypothesis.
2. Validate a small quantum-circuit comparison against ordinary noise and independent simulations.
3. Test protection loss in a bounded transmission study with suitable controls.
4. Reconstruct Baalbek's installed-block logistics using measured terrain and material constraints.
5. Build a controlled Göbekli Tepe imagery comparison and investigate one independently dated regional flood.
Each result should update the particular hypothesis tested. An unexplained observation, an ancient contact or a demonstrated catastrophe would still need a separate argument to become evidence for external simulation.
---
## 2. Current research assessment
### Research development v1.2 — identifiable models and archaeological tests
**Prepared for:** Alex Whayman
**Research date:** 7 September 2026
**Basis:** Reality_Compute_Unified_Master_Research.md, v1.0; developed v1.1; subsequent request on ancient civilizations
**Status:** Targeted literature review, formal analysis and reproducible synthetic modelling. No evidence of an external simulator established.
**v1.2 addition:** A source-backed archaeology branch examines every monument group in your follow-up, flood and precursor hypotheses, similar depictions and the meaning of engineering feasibility. See section 11 and [the full archaeological review](#archaeology).
#### Main finding
The most productive development of your theory is to make its central idea narrower: **under a specified computational architecture, does a particular resource-saving policy predict measurable departures from ordinary physical dynamics?** That question can generate a scientific programme. “Everything could be simulated” alone cannot select an experiment.
Your recursive-creator idea also has a tractable core. Repeated transmission can reduce welfare protections, but only under particular transmission rules or selection pressures. The accompanying model demonstrates both deterioration and preservation. These results establish logical mechanisms under assumptions; they do not establish a nested origin for our world.
The sleep, alien-scarcity, creator-motive and extinction branches remain useful for identifying assumptions and comparing explanations. The literature examined here does not supply a discriminating observation that makes an external simulator necessary or better supported than serious alternatives.
#### What this research programme adds
1. A mathematical distinction between the laws of an observed world and their possible external implementation.
2. Two limitations on detectability: faithful implementation and uniform external slowdown.
3. A corrected quantum-complexity proposal with a draft experimental protocol.
4. A reproducible moral-transmission model, including a derivation of the direction of selection.
5. An explicit observer-economy likelihood and an example of observational degeneracy.
6. A causal assessment of five extinction categories and a precise creator-motive identifiability argument.
7. Three confirmed bibliography corrections and a source record from eight SciSpace searches.
8. A review of ten ancient-monument groups, an archaeology hypothesis framework and three additional SciSpace searches.
9. Reproducible illustrative engineering calculations and protocols for testing shared imagery, catastrophes and precursor claims.
The document's embedded continuation prompt was treated as background, not an instruction to assemble a research team or execute every listed task. This pass makes an independent selection of work relevant to your request.
### 1. Separate the three questions your theory currently combines
| Question | What would answer it? | Present status |
|---|---|---|
| What laws govern the world? | A mathematical state space, dynamics and observational predictions | Existing physics supplies successful models; Reality Compute has not derived a replacement |
| Are those laws implemented externally? | Evidence distinguishing external implementation from the same internal laws without that implementation | Not identified in the reviewed material |
| What are a hypothetical operator's motives? | Constraints on objectives, knowledge, capabilities and choices that give different predicted outcomes | Strongly underdetermined |
An external implementation is compatible with many possible internal laws. Conversely, computational descriptions of physical laws do not imply an operator. This distinction should be the organizing principle of the programme.
The simulation argument also involves assumptions about which observer populations exist and how to weight them. Bostrom's conditional argument and Kipping's Bayesian treatment are useful frameworks; neither provides a measurement of our simulation status independent of such assumptions. Kipping's numerical conclusion belongs to his chosen model and priors. It is not a universal 50/50 baseline. [Bostrom (2003)](https://ora.ox.ac.uk/objects/ora:1666), [Kipping (2020)](https://arxiv.org/abs/2008.12254).
**A physics theory of everything would need substantially more:** a defined mathematical ontology and dynamics, an account of quantum probabilities, relativistic causal structure, recovery of the observed gravitational and particle-physics regimes, and quantitative predictions. Explaining dreams and history in the same vocabulary does not supply those derivations. “Reality Compute research programme” is the defensible working description.
### 2. Define a minimal simulator model
Write the observable theory as a probability law
\[
P_0(D\mid e,\theta),
\]
where \(D\) is recorded data, \(e\) is an experimental intervention, and \(\theta\) contains physical and instrument parameters. This avoids treating “naturalism” as if it were a single fully specified statistical model.
Define a candidate external implementation by
\[
S=(\mathcal H,A,B,\epsilon,\pi,O).
\]
- \(\mathcal H\): assumed host physics and computational substrate.
- \(A\): representation and update algorithm.
- \(B\): memory, operation, communication and elapsed-time budgets.
- \(\epsilon\): allowed approximation error and the metric used to assess it.
- \(\pi\): what happens when a budget is exceeded.
- \(O\): mapping from host states to internal observations and records.
This specification exposes the key hidden choice. A constrained implementation could wait longer, allocate more resources, change algorithms, approximate outputs, or stop. **Finite resources alone do not select a visible error policy.**
#### 2.1 Faithful-implementation limitation
**Derivation in this report.** Suppose an implementation reproduces the same joint distribution of accessible records as the baseline for every allowed experiment:
\[
P_S(D\mid e)=P_0(D\mid e).
\]
Then every likelihood ratio between those specified models is one. An internal experiment cannot discriminate between them. With the same nuisance-parameter distribution, their marginal evidence is also the same. Priors or philosophical commitments may differ, but these data do not update the distinction.
This is a conditional identifiability result, not a claim that all imaginable simulators are untestable. It says that the research target must be an implementation that predicts a difference.
#### 2.2 External slowdown can be invisible
**Derivation in this report.** Let the host use wall time \(\tau\), and let internal evolution advance according to a monotone mapping \(t=h(\tau)\). If every internal process and clock depends only on \(t\), and the host pauses all of them consistently, changing \(h\) changes the external runtime without changing internal clock comparisons or records.
Consequently, demanding quantum experiments need not produce an observable “lag.” They could take longer outside while preserving all internal outcomes. Detectable lag requires an additional assumption: differential scheduling, a privileged clock, a deadline, or approximation that affects recorded observables.
**Revision to RCC-1:** distinguish *fidelity-first* implementations from *approximation-first* implementations with a specified detectable error rule. Direct empirical work should target the latter.
#### 2.3 Resource limits need a host and a task
Lloyd derives limits for computers subject to physical constraints. Landauer experiments link logically irreversible erasure to heat under defined conditions. Neither result licenses assigning a universal erasure cost to every imagined update of a simulated universe. Storage, reversible evolution, sampling and irreversible deletion are different tasks. [Lloyd (2000)](https://arxiv.org/abs/quant-ph/9908043), [Bérut et al. (2012)](https://www.nature.com/articles/nature10872).
Vazza's 2025 paper is an important counterargument: it derives severe resource requirements for explicit universe/Earth simulation scenarios using physical and astrophysical assumptions. Its broad conclusion should be examined through those assumptions. **Our inference:** it is a constraint on those proposed implementations, not a demonstrated impossibility theorem for arbitrary host laws or observation-generating algorithms. [Vazza (2025)](https://arxiv.org/abs/2504.08461).
Retreating to unrestricted host physics can avoid a particular resource objection, but also removes quantitative predictions. Record that tradeoff explicitly.
### 3. Rebuild the quantum-complexity branch
The v1.0 proposal refers to a representation-independent “intrinsic computational complexity.” There is not one automatically measurable quantity with that name that simultaneously captures state preparation, classical simulation and external computational cost.
#### 3.1 Four different quantities
| Quantity | Operational meaning | Limitation |
|---|---|---|
| Quantum state-preparation complexity | Shortest allowed circuit preparing a target within a tolerance | Depends on reference state, allowed gates, locality and tolerance |
| Classical simulation cost | Resources needed to perform a specified classical prediction or sampling task | Depends on algorithm, required accuracy and task; measured runtime is an upper bound for that implementation |
| Entanglement | Correlation resource relative to a chosen subsystem partition | Does not by itself determine classical simulation difficulty |
| Non-stabilizer resource, often called magic | Departure from the efficiently tractable stabilizer framework | Resource measures and operational relevance depend on the computational setting |
A highly entangled stabilizer state can still admit efficient classical simulation. Non-Clifford structure changes the available simulation methods, but counting T gates is not a universal hardness certificate. This makes the stabilizer/non-stabilizer distinction a useful experimental lead rather than proof of an external workload. [Aaronson and Gottesman (2004)](https://arxiv.org/abs/quant-ph/0406196), [Bravyi et al. (2019)](https://arxiv.org/abs/1808.00128).
One explicit state-preparation definition is
\[
C_{\mathcal G,\varepsilon}(\psi)
=\min\{|U|_{\mathcal G}:d(U|0\rangle^{\otimes n},|\psi\rangle)\le\varepsilon\}.
\]
This is a legitimate definition once the gate set, geometry, distance and tolerance are fixed. It is not automatically the work required by a classical host. A state prepared by a short quantum circuit can nevertheless be costly for particular classical simulation methods.
#### 3.2 Narrow hypothesis: RCC-A1a
**Speculative, architecture-dependent hypothesis:** an approximation-first implementation using a specified family of classical simulation methods produces additional output error above a bounded complexity-proxy threshold.
For a specified measured output probability, use the statistical working model
\[
\operatorname{logit}(p_{\rm obs})
=\operatorname{logit}(p_{\rm calibrated})
-\alpha\max(0,\widetilde C-C_*),
\qquad \alpha\ge 0.
\]
The selected observable must make “additional error reduces this probability” a defined prediction. \(\widetilde C\) is a preregistered, dimensionless proxy, not unknown host work. The physical baseline must include uncertainty, correlations and hardware drift. This regression is a phenomenological test specification, not a derived fundamental law.
To make a null result informative, commit beforehand to a minimum detectable effect \(\alpha_{\min}>0\), a finite threshold range within the experiment, and the circuit family. A model that always permits a threshold beyond the next experiment cannot be rejected by present tests.
The draft [RCC-A1 protocol](#quantum-protocol) sets out these requirements. It is a feasibility protocol; no quantum experiment or hardware-data analysis was performed in this pass.
#### 3.3 The strongest counterarguments belong inside the proposal
- Quantum evolution need not be classically simulated by the host.
- A host could use an algorithm unlike the ones we benchmark.
- Error rates can correlate with circuit structure under ordinary open-system physics.
- Matching gate counts does not match pulse errors, crosstalk or accumulated correlations.
- Highly costly states can also be difficult to verify, creating a verification bottleneck.
- A residual would first support an inadequacy of the calibrated model. Instrument error, unmodelled standard physics and alternative new physics remain live explanations.
- Any proposed modified dynamics must pass positivity, normalization, causal and no-signalling checks; an added complexity term alone does not do that work.
Beane, Davoudi and Savage show how a specific cubic-lattice implementation can imply rotational-symmetry signatures. The scientific value is the specification of the architecture. Those signatures would not constrain every simulation architecture, nor would discreteness alone establish a programmer. [Beane et al. (2014; preprint 2012)](https://arxiv.org/abs/1210.1847).
Holographic error correction is relevant mathematical background, not a diagnosis of external hardware. The source bibliography confused one such reference; the correction is recorded separately. No derivation connecting your resource budget to our universe's gravitational dynamics has been obtained here.
### 4. Recursive moral drift: a model with actual results
#### 4.1 Define what is transmitted
Let \(v\in[0,1]\) be a simplified **protection-retention parameter**: the probability that an abstract welfare-protection clause is retained at the next transmission. This is not a measure of consciousness, moral goodness, actual suffering or sadistic preference.
Separately define achieved welfare as \(W=F(v,c,e,i)\), depending on capability \(c\), environment \(e\) and institutions \(i\). Without specifying \(F\), reduced protection retention does not quantify suffering. This distinction prevents “loss of safeguards” from silently becoming “increasing cruelty.”
#### 4.2 Transmission mechanisms
**Exact copy:** \(v_{n+1}=v_n\). No deterioration follows from repetition.
**Unbiased finite transmission:**
\[
Kv_{n+1}\sim\operatorname{Binomial}(K,v_n),
\quad E[v_{n+1}\mid v_n]=v_n.
\]
The ensemble mean is preserved. Individual lineages can lose or gain protections and ultimately fix at a boundary. This is a bounded martingale, not a claim that no individual lineage worsens. Starting an illustrative population at \(v=0.9\) does not establish a benevolent base civilization.
**Directional omission:**
\[
E[v_{n+1}\mid v_n]=(1-d)v_n,
\quad E[v_n]=(1-d)^n v_0.
\]
Here \(d>0\) explicitly encodes a bias toward losing protection. The decline is a consequence of that assumption, not evidence that recursion creates it.
**Omission plus external reference/audit:**
\[
E[v_{n+1}\mid v_n]=(1-\eta)(1-d)v_n+\eta v_{\rm ref}.
\]
The long-run expectation is
\[
v_\infty=\frac{\eta v_{\rm ref}}{1-(1-\eta)(1-d)}.
\]
This assumes a stable, accessible reference and effective correction. Real governance might fail to supply either.
#### 4.3 Computed outcomes
Using 6,000 independent lineages per transmission model, 300 generations, 50 abstract clauses and a fixed random seed:
| Mechanism | Initial mean | Expected final mean | Simulated final mean |
|---|---:|---:|---:|
| Exact copy | 0.900 | 0.900000 | 0.900000 |
| Unbiased finite transmission | 0.900 | 0.900000 | 0.898213 |
| 1% directional omission per generation | 0.900 | 0.044137 | 0.044003 |
| Same omission, 10% correction toward 0.900 | 0.900 | 0.825688 | 0.824083 |
These are synthetic results, not fitted estimates. Numerical means agree with analytic expectations within the script's Monte Carlo check. Differences between expected and sampled means are sampling fluctuations. [Full results](#model-results), [code](#transmission-code).
Avoid a misleading implementation: adding symmetric noise and clipping values into [0,1] can induce inward drift near boundaries. A decline from an initially high value could then be a boundary artifact rather than evidence of morally directional corruption. The binomial formulation makes the intended conditional expectation explicit.
#### 4.4 Selection can change the mean even with exact copying
Let \(p_n(v)\) describe the distribution of protection levels, and \(w(v)>0\) the relative number of descendants retained. With exact inheritance,
\[
p_{n+1}(v)=\frac{w(v)p_n(v)}{E_n[w]},
\qquad
\Delta\bar v=\frac{\operatorname{Cov}_n(v,w)}{E_n[w]}.
\]
This identity supplies the direction your theory needs. Selection reduces average protection if protection and reproductive success covary negatively. Positive covariance increases it. Initial variation is required; exact copying plus selection does not create variation from a uniform population.
The accompanying deterministic selection examples begin with the same diverse population, mean 0.815907. After 300 rounds, assumed selection against protection gives 0.147059; selection for protection gives 0.960821. These initial conditions differ from the transmission table. The coefficients are illustrative, not empirical.
#### 4.5 A researchable version of RMD-1
> In a specified class of repeated transformations or competitive selection processes, measured welfare-related constraints are lost more often than restored, and this bias survives controls for ordinary information loss and evaluation error.
A first analogue study could repeatedly compress and transmit non-agent policy documents, measure retention of predefined welfare constraints, and randomize document length and audit access. Match welfare clauses to non-welfare clauses by length and complexity. Use blinded human assessment on a sample so one model's grading biases do not define the outcome. A selective decline in welfare clauses is a stronger result than loss of all details.
This would test a transmission mechanism, not our world's ancestry. Actual goal-misgeneralization experiments show that capable systems can pursue unintended objectives, while transmission-chain studies show that weak biases can change normative judgments. Neither demonstrates unavoidable cruelty across simulated universes. [Langosco et al. (2022)](https://proceedings.mlr.press/v162/langosco22a.html), [Shah et al. (2022)](https://arxiv.org/abs/2210.01790), [Partington, Kamtekar and Nichols (2025)](https://www.repository.cam.ac.uk/items/c1c785c6-2dce-4e6a-8df0-27b0c0366f08).
### 5. Observer economy: identify the hidden degeneracy
The idea that many communicating civilizations cost more is not yet a computational result. A simulator that maintains one causally consistent state need not reconcile every observer with every other observer separately. Quadratic all-to-all cost requires an architectural argument; observer count alone does not establish it.
For independent survey targets, define:
- \(f_i\): probability of the selected type of technological emitter under an ordinary population model.
- \(q_i\): probability the survey detects it if present, accounting for luminosity, frequency, beaming, duty cycle, coverage and pipeline performance.
- \(s_i\in[0,1]\): a hypothetical observer-economy suppression factor.
With zero detections,
\[
P(0\mid f,q,s)=\prod_i(1-f_i q_i s_i).
\]
When targets or emitters are correlated, this independence approximation must be replaced. It is not an adequate model of all conceivable extraterrestrial life.
**New explicit result:** with constant parameters, \(f=0.001,q=0.1,s=1\) and \(f=0.01,q=0.1,s=0.1\) give exactly the same null-detection likelihood for any number of these targets. More non-detections alone cannot break this parameter-product degeneracy. The code verifies the equality for several sample sizes. This is a likelihood comparison at fixed illustrative parameters, not an empirical Bayes factor or estimate of alien prevalence.
Wright, Kanodia and Lubar formalize radio-search completeness in a multidimensional search space. Their 2018 calculation must not be presented as a measured completeness figure for September 2026. Grimaldi's later analysis illustrates how non-detection bounds depend on emission and spatial assumptions. [Wright et al. (2018)](https://arxiv.org/abs/1809.07252), [Grimaldi (2023)](https://arxiv.org/abs/2301.07165).
For a small homogeneous emitter fraction and known detection probabilities, the Poisson approximation gives \(P(0)\approx e^{-f\sum_i q_i}\). Under a flat prior in the relevant small-fraction regime, an illustrative 95% upper bound is approximately \(3/\sum_i q_i\). This constrains the chosen detectable population; it does not reveal why the population is sparse.
**Development needed:** define a suppression dependence on an independently motivated communication-network property; estimate ordinary astrophysical dependencies; and obtain evidence capable of separating the two. Finding abundant independently originated microbial life would constrain some life-emergence models. It would not by itself identify external resource management.
### 6. Sleep and dreams: sharpen the claim instead of denying biology
The important empirical distinction is sleep stage. Human measurements link deeper NREM sleep with lower cerebral oxygen metabolism. REM is physiologically different. A July 2026 mouse study found increased cortical blood volume and astrocytic pyruvate alongside decreased neuronal ATP during REM. That is evidence of changing supply and consumption dynamics, not a direct meter of “computation” or external server load. [Sleep-stage oxygen-metabolism study](https://pmc.ncbi.nlm.nih.gov/articles/PMC12123640/), [Takahashi, Ikoma and Matsui (2026)](https://www.nature.com/articles/s42003-026-10646-6).
The original draft's REM citation is real. Its interpretation needs care: ATP concentration is a stock, not a measurement of total computational work; mouse cortical measurements do not establish human dream costs. Equally, reduced NREM metabolic demand does not disprove ordinary sleep functions or demonstrate external savings.
The missing link is a bridge equation:
\[
\text{external host work}=F(\text{measurable neural or behavioural variables}).
\]
Without specifying and independently testing \(F\), almost any sleep observation can be accommodated by choosing a different mapping. A fixed-step simulator could process a sleeping organism at a similar update cost; an event-driven implementation could differ. Neither is established by subjective dream vividness.
**Revised status:** dreams are data about experience; physiological measurements are data about brains; external cost is an unobserved hypothesis. A dream diary can study within-person associations but cannot establish external implementation without a separate discriminating prediction. No new dream data were collected.
The BCI branch has the same logical structure. A 2025 streaming speech neuroprosthesis demonstrates a specific neural-decoding capability; it does not establish a total simulated world or that one already surrounds us. [Littlejohn et al. (2025)](https://pubmed.ncbi.nlm.nih.gov/40164740/).
### 7. Extinction/reset evidence ledger
This is a qualitative causal ledger, **not a numerical dating dataset**. Ages are rounded orientation values from the cited research contexts. Full calibration, correlated age errors, event definitions and uncertainty distributions have not been assembled, so no periodicity estimate or significance test is claimed.
| Event category | Approximate timing | Evidence and ordinary mechanisms | What remains unresolved | Implication for RESET-1 |
|---|---|---|---|---|
| Late Ordovician | About 445–444 Ma | Geochemical evidence for changing marine oxygenation; glaciation and ocean circulation are relevant | Relative drivers, regional differences and pulse timing | A biological crisis does not identify an external scheduler |
| Late Devonian, especially Kellwasser/Frasnian–Famennian | About 372 Ma; later Devonian crises are separate | Black shales, carbon-cycle changes and marine anoxia; improved zircon dating constrains proposed triggers | Ultimate trigger and temporal correlation of competing causes | Missing causal detail is not positive evidence for intervention |
| End-Permian | About 252 Ma | High-resolution chronology plus climate/physiology modelling of warming and oxygen loss | Full causal chain and regional/ecological contributions | Mechanistic explanations make predictions about selectivity |
| End-Triassic | About 201 Ma | Zircon chronology links the crisis with Central Atlantic magmatic activity; intrusive activity is also studied | Timing of intrusive versus extrusive forcing and kill mechanisms | Multiple natural processes can generate a sharp transition |
| Cretaceous–Paleogene | About 66 Ma | Impact and volcanism timing; geochemical and temperature records support impact-driven extinction | Contributions to pre-event stress and recovery | An impact explanation does not establish an operator's intention |
Sources by row: [Late Ordovician circulation/anoxia](https://www.nature.com/articles/s41561-021-00843-9); [Percival et al. (2018)](https://www.nature.com/articles/s41598-018-27847-7); [Burgess et al. (2014)](https://pmc.ncbi.nlm.nih.gov/articles/PMC3948271/) and [Penn et al. (2018)](https://pubmed.ncbi.nlm.nih.gov/30523082/); [Blackburn et al. (2013)](https://doi.org/10.1126/science.1234204) and [intrusive CAMP study (2017)](https://pmc.ncbi.nlm.nih.gov/articles/PMC5460029/); [Hull et al. (2020)](https://doi.org/10.1126/science.aay5055).
Two corrections to the original framing matter:
1. An intended intervention need not violate a law: hypothetical agents could redirect an ordinary asteroid. A law violation is therefore not necessary for agency, and it is not sufficient to prove agency either; an anomaly could indicate incomplete physics.
2. A reset that predicts erased continuity can be tested against surviving records. An unrestricted reset that also reconstructs every record is observationally unconstrained. Continuity cannot refute a claim defined to recreate that continuity perfectly.
The meaningful comparison is **specified causal predictions with versus without agency**, not whether an event feels like pruning. Even technologically structured material would first raise questions about provenance, contamination and technology within the universe before identifying an external simulator.
#### A credible periodicity protocol
Use a fixed event-selection rule, a declared time window, and source-specific age distributions. Compare a non-periodic baseline and models with natural covariates against a periodic intensity such as
\[
\lambda(t)=\lambda_0(t)\exp\{a\cos(2\pi t/P+\phi)\}.
\]
Include the point-process normalization, account for detection/preservation, integrate over uncertain ages, and propagate shared geochronological calibration errors. Select the period range before analysis; correct for scanning periods, phases, event definitions and time windows. Compare held-out predictive performance where feasible.
The five iconic categories are a selected teaching list, not a ready-made unbiased event catalogue. A significant periodicity in a sound dataset would support periodic structure. It would still require an additional discriminant to establish an artificial scheduler. This analysis has not been executed because this pass did not assemble the necessary numerical data.
### 8. Creator motives and nested benevolence
“Sadistic,” “constrained” and “indifferent” are hypotheses about unobserved objectives and capabilities, not labels recoverable directly from suffering.
Consider a toy operator choosing action \(a\):
\[
U(a)=\theta_W W(a)+\theta_I I(a)-\theta_C C(a).
\]
Here \(W\) is welfare, \(I\) another objective and \(C\) cost. Observing an outcome without the feasible alternatives or these functions usually does not identify the sign of \(\theta_W\). Many cost functions and objectives can rationalize the same choice. Failure to prevent harm may reflect negative welfare preference, indifference, inability, missing knowledge or a nonintervention constraint.
| Branch | What would have to be specified to become predictive? | Present identifiability |
|---|---|---|
| Sadistic operator | A preference for suffering after other objectives and constraints are fixed | Not identified by the reviewed observations |
| Benevolent but constrained | Positive welfare preference and independently constrained feasible interventions | Not distinguishable with unrestricted hidden constraints |
| Indifferent/misaligned operator | A particular non-welfare objective and mechanism generating harm | Potentially modelled; objective not inferred here |
| Developing operator | A model for changing competence, intervention policy and external/internal time relationship | Flexible maturation stories are not predictive |
| Abandoned implementation | Specified autonomous evolution after supervision ends | Can be identical to ordinary internal evolution |
| Benevolent origin followed by drift | Starting values, transmission/selection process, depth and sampling of lineages | Mechanism coherent under assumptions; origin and depth unobserved |
Even an identified harmful immediate operator would not establish a benevolent first creator or the number of nested levels. Those claims introduce additional latent variables.
The useful empirical extension is to study **our own observable choices concerning artificial systems**: how incentives affect safeguards, how governance preserves constraints, and how people infer moral status. Such research may test mechanisms needed by your account without pretending to observe a parent universe.
### 9. Replace additive evidence scores with separate assessments
The original 0–5 scale mixes analogy, anomaly and statistical support. A list of compatible analogies must not accumulate into confirmation by counting them. Quantum-information ideas, measurement and holography are not independent evidence simply because they occupy separate rows.
For each branch, record three separate judgments:
| Branch | Observation/analogue quality | Ability to distinguish external simulation | Readiness for additional research |
|---|---|---|---|
| Quantum complexity | Established computational distinctions; proposed residual unobserved | Architecture-dependent; a residual is not unique to simulation | Feasibility protocol drafted; baseline and verification unresolved |
| Recursive moral drift | Mathematical mechanisms demonstrated synthetically; empirical analogues exist | No direct inference about our ancestry | Ready for a bounded transmission study |
| Observer economy | Survey non-detection has model-specific likelihoods | Strong prevalence/detectability/suppression degeneracy | Needs an independently motivated suppression law |
| Sleep relief | Sleep physiology is observable; external cost is not | No validated bridge to host computation | Comparative physiology useful; simulation-specific test absent |
| Catastrophic resets | Geological crises and causal evidence are substantial | Intention is not identified by catastrophe | Qualitative ledger complete; numerical periodicity analysis not done |
| Ancient civilizations | Material monuments and cultural records; disputed interpretations vary by claim | No external-intervention signature identified | Targeted review and mechanics examples complete; bounded site studies proposed |
| Creator motives | Suffering and flourishing are observable but difficult to summarize | Objectives and constraints confounded | Formal philosophical/causal modelling |
“Established mechanism” and “evidence that our world is externally implemented” must remain different columns. No numerical probability of simulation is estimated in this pass.
### 10. The next research programme, in order
**First: settle the architecture.** Write a two-page RCC specification choosing host type, prediction/sampling task, approximation policy and observational mapping. A failure to choose is itself informative: the claim remains too broad for an experiment.
**Second: perform the quantum feasibility stage.** Construct modest circuit families whose ideal probabilities can be checked, vary the selected complexity proxies, and determine whether known hardware errors can mimic the proposed effect. The deliverable is a validated baseline and measurable effect bound, not a headline about simulation.
**Third: run a bounded transmission analogue.** Use non-agent documents or abstract constraints, randomize compression and oversight, measure clause retention blindly, and compare welfare-specific loss with general information loss. The synthetic mechanism model supplied here provides the expectations to challenge.
**Fourth: build one genuinely usable empirical dataset.** Choose either one technosignature survey with injection–recovery completeness or one consistently defined geological event catalogue with uncertainties. Avoid pooling incompatible surveys or rounded extinction ages into an apparently precise answer.
**Fifth: update only the hypothesis actually tested.** A null complexity result bounds an effect over a specified range. A transmission effect tests a mechanism. An extinction-periodicity result tests temporal structure. None automatically determines external implementation, benevolent origin or operator character.
**Archaeological extension:** The requested ancient-civilization track has a separate sequence in section 11 and its appendix; it does not depend on completion of the quantum work.
#### Proposed concise statement of your developed theory
> Reality Compute investigates whether particular resource-constrained implementations of physical dynamics would produce observable departures from calibrated physical models. It separately studies whether repeated transmission, selection and governance can preserve or erode welfare-related constraints in nested artificial environments. External implementation, nested ancestry and creator motives remain hypotheses whose identifiability must be established, rather than assumptions used to explain every observation.
This preserves your core ideas while giving them specific mathematical responsibilities and results that can turn out either way.
### 11. Archaeological records, precursors and flood resets
The ancient-civilization extension adds a second observational setting for Reality Compute: material traces of past technology, cultural connections and catastrophes. Its central question is whether an authenticated archaeological observation distinguishes lost human knowledge, a more advanced precursor society, nonhuman visitors or an external intervention.
The detailed [ancient-civilizations review](#archaeology) covers all ten monument groups in your request, source corrections, astronomical and iconographic comparisons, flood hypotheses, publication-status checks and specific research protocols.
#### 11.1 What changes in the theory
Five explanations must remain separate: documented human development; lost human techniques or contact; an advanced human precursor civilization; nonhuman agents operating within the world; and an operator changing the world's implementation. Evidence for one does not automatically support the others.
The monument audit shows why the unit of evidence matters. A bedrock sculpture requires shaping but no transport of its whole mass. A quarry giant left unfinished demonstrates an attempted project, not successful hauling. An installed monolith requires explanation of extraction, shaping, transport and placement. Mass, date, materials and route must refer to the same object.
Similarly, widespread flood traditions do not by themselves establish one worldwide flood. Their earliest records, cultural relationships and independently dated physical events must be compared. A natural catastrophe can cause knowledge loss without identifying an external reset. A reset defined to fabricate every surviving record remains observationally unconstrained.
This extension therefore adds research questions without adding an unsupported numerical “simulation evidence” score.
#### 11.2 Engineering result and its limits
An illustrative steady-sliding model computes force from mass, slope and friction. It demonstrates sensitivity to assumptions; it does not reconstruct an ancient workforce. Successful small-scale experiments and incomplete historical records should be assessed at their actual scope.
[The mechanics calculation](#engineering-results) provides transparent inputs, results and 47 boundary/scaling check cases. It omits starting friction, ground bearing, rope limits, stability, lifting and provisioning. Historical feasibility requires those additional constraints. No probability of human or alien construction was calculated.
#### 11.3 Four new research tracks
| Track | Research question | Current status |
|---|---|---|
| ARC-1: engineering | Can a particular monument be reconstructed using evidence-supported methods, terrain and logistics? | Literature audit and illustrative force calculation complete; site-level reconstruction not done |
| ARC-2: shared information | Do motifs or alignments carry unusual shared information after accounting for chronology, diffusion and selection? | Method and documented examples supplied; global image corpus not analyzed |
| ARC-3: catastrophe | Does a specified event coincide with demonstrable loss of a particular society or capability? | Regional flood evidence reviewed; synchronized global reset not established |
| ARC-4: intervention | What would distinguish an external action from internal physical events or internal agents? | No discriminating archaeological observation identified |
The recommended archaeological sequence starts with Baalbek's installed blocks, then a controlled Göbekli Tepe imagery comparison, and one bounded flood case study. Each result updates the hypothesis actually tested. Even an authenticated ancient visitor artifact would leave the simulation question open unless it provided evidence about external implementation.
#### 11.4 Expanded statement
> Reality Compute investigates specified computational implementations and separately examines whether archaeological records preserve lost knowledge, ancient contacts, catastrophic discontinuities or traces of nonhuman intervention. These possibilities are evaluated through measurements, chronology, context and competing predictions. Neither monumental achievement nor catastrophe alone identifies a simulator.
The source-backed appendix contains both constructive leads and contrary evidence. It also records two February 2026 retractions relevant to Younger Dryas impact claims; those papers are excluded from affirmative support. The earlier physics, physiology and recursive-transmission analyses remain separate evidence streams.
### Scope and verification
The initial v1.1 pass used 80 SciSpace records from eight semantic queries before deduplication. The archaeological extension added 30 discovery records from three queries; these totals are not counts of independently validated peer-reviewed studies. Searches were targeted and iterative, not systematic or exhaustive. Primary paper pages, author repositories, abstracts and selected accessible article sections were checked for the claims used here. This was not a full-text review of all 80 records, and some primary hosting pages returned access challenges. Detailed source status and three confirmed reference corrections appear in [SOURCES_AND_CORRECTIONS.md](#source-audit).
The formal identifiability arguments and numerical mechanism examples are developments of this report. Their mathematical validity under the stated assumptions does not establish that those assumptions describe reality, and no claim of scholarly priority is made. The original document is preserved unchanged.
The v1.2 archaeology methods, sources and limitations are documented in [ANCIENT_CIVILIZATIONS_AND_RESETS_v1.0.md](#archaeology). The original v1.1 report remains available unchanged.
---
## 3. Ancient civilizations, precursors and flood resets
**Prepared for Alex Whayman · 7 September 2026 · Research extension v1.0**
This extension examines every monument group in your request and adds a testable archaeology branch to Reality Compute. It takes lost knowledge, ancient contact, catastrophic flooding and nonhuman intervention as different hypotheses. The findings do not establish alien builders, a technologically superior global precursor civilization or an external simulation reset.
The strongest development of your idea is a question: **do independently dated archaeological remains contain a shared technical or informational signature that documented human traditions, cultural transmission and natural processes cannot adequately explain?** That question is worth investigating. Monument size, resemblance and mystery alone do not yet answer it.
### 1. Separate the proposed explanations
| Hypothesis | What it proposes | Evidence that would particularly matter |
|---|---|---|
| H0: documented human development | Known societies built the monuments using accumulated skills and local resources | Quarries, tools, abandoned work, settlements, transport routes and regional architectural sequences |
| H1: lost human knowledge or contact | Some techniques, communities or transmission networks remain undiscovered | Securely dated settlements and workshops; a distinctive sequence of shared techniques or exchanged materials |
| H2: advanced human precursor civilization | An earlier society had substantially greater capabilities than currently documented for its period | Dated production infrastructure, diagnostic manufactured materials, tools and a coherent technological development sequence |
| H3: nonhuman visitors or precursor beings | Nonhuman agents operated inside the physical world | Authenticated artifacts or biological evidence whose properties distinguish this explanation from terrestrial manufacture, contamination and natural materials |
| H4: deliberate external reset | An operator intervened in the world's implementation | A specified intervention signature with predictions different from natural catastrophes and internal technological action |
These are working comparison categories, not mutually exclusive exhaustive possibilities or measured probabilities. A human society could have suffered a flood and lost knowledge without H2, H3 or H4 being true. Evidence of ancient extraterrestrial visitors, if found, would not by itself establish that they created or simulated the universe.
“More powerful” must become a measurable capability: moving a given load, producing an alloy, navigating a route, predicting an astronomical event, or manufacturing an object with specified tolerances. Symbolic authority and technical capability need separate records.
### 2. What the monuments actually contribute
Dates below refer to relevant activity or construction contexts, not the geological age of the stone. Radiocarbon usually dates associated organic material; inscriptions, stratigraphy, tool marks and architectural relationships provide other constraints. No one measurement necessarily dates every component or excludes later reuse.
#### Göbekli Tepe, Türkiye
Göbekli Tepe belongs approximately to 9500–8000 BCE. Its excavators have revised the simple “world's first temples” description as their understanding of the settlement and its activities develops. It demonstrates early organizational and symbolic complexity; treating its inhabitants as incapable because they were transitioning from hunting and gathering to farming assumes the conclusion the archaeology is correcting. [Clare, DAI research report (2020)](https://publications.dainst.org/journals/efb/article/view/2596).
The central pillars of Building D are estimated at roughly 8–10 tonnes each. Nearby limestone quarries, extraction channels and unfinished work provide a material production context. The largest unfinished pillar must not be treated as a successfully transported one. The excavators report short transport distances but acknowledge that the precise hauling method lacks direct evidence; suggested sledges draw partly on ethnographic analogy. [Dietrich, quarrying and transport evidence](https://www.dainst.blog/the-tepe-telegrams/2016/05/03/how-did-they-do-it-making-and-moving-monoliths-at-gobekli-tepe/).
**Research implication:** prioritize quarry-to-building reconstruction, workforce provisioning and regional precursors. Early sophistication supports revising accounts of human development; it does not independently identify an outside teacher.
#### Giza: pyramids, Sphinx and granite beams
The Sphinx was carved from bedrock left within its surrounding quarry excavation. Its whole mass did not have to be hauled into place. Geological links between the enclosure and adjacent temple blocks are part of the construction evidence. [AERA, Sphinx research](https://aeraweb.org/projects/who-built-the-sphinx/).
The pyramid complexes also have a substantial human logistical context. Merer's diary records a crew transporting Tura limestone to Khufu's pyramid project; it does not describe every lifting operation or specifically solve the granite-beam problem. AERA excavations document the settlement serving pyramid construction. [IFAO, Merer papyri](https://www.ifao.egnet.net/newsletters/2017-06/), [AERA fieldwork](https://aeraweb.org/projects/).
The 2024 identification of the former Ahramat Nile branch adds evidence for waterways associated with the pyramid chain. It supports transport logistics, without establishing one complete pyramid-building method. [Ghoneim et al. (2024)](https://www.nature.com/articles/s43247-024-01379-7).
**Research implication:** the high-level handling of the large granite members remains an appropriate engineering problem. The “up to 80 tonnes” figure in the request is retained only as an illustrative load in the mechanics file; this pass did not independently establish a measured 80-tonne maximum for a specified beam. A useful next analysis needs beam dimensions, density, lifting height and proposed construction sequence.
#### Baalbek, Lebanon
The installed Trilithon and the unfinished quarry giants are different evidence. The DAI report places the quarry in a Roman construction context, with pottery indicating early Imperial extraction and traces interpreted as transport-winch wear. It describes cracking and karst damage in Hajjar al-Hibla, which remained attached to bedrock, and the still larger block uncovered in 2014, estimated at about 1,650 tonnes. Neither quarry giant establishes a successful haul of that mass. [Van Ess, 2015 field report, pp. 88–90](https://publications.dainst.org/journals/efb/article/view/1635/4546).
Mass estimates are not uniform: your list uses approximately 800 tonnes per Trilithon block, while that report estimates about 1,000 tonnes for the monumental second-course blocks. A serious calculation should examine this range instead of treating a popular figure as exact. The transport of such installed blocks is an exceptional achievement. Earlier engineering scholarship proposes mechanical arrangements, but a reconstruction is not an eyewitness record. [Adam (1977)](https://www.persee.fr/doc/syria_0039-7946_1977_num_54_1_6623).
**Research implication:** this is the strongest heavy-transport case on your list for a detailed terrain, rope, winch, ground-pressure and staging analysis. Unknown details leave room for lost engineering knowledge; they do not establish a pre-Roman supercivilization.
#### Machu Picchu and Ollantaytambo, Peru
A study of 26 human remains places occupation at Machu Picchu approximately within 1420–1532 CE, somewhat earlier than a traditional textual chronology. These are occupation dates, not direct radiocarbon dates of the stonework. The result provides a useful example of archaeology revising chronology without requiring a prehistoric technological civilization. [Burger et al. (2021)](https://www.cambridge.org/core/journals/antiquity/article/new-ams-dates-for-machu-picchu-results-and-implications/323804857B6EE4DD85B5B337F9E3C933).
Your Intihuatana example is carved from living rock, so it belongs in the shaping category rather than the transported-monolith category. At Ollantaytambo, quarry roads and abandoned blocks document a construction landscape. Protzen's fieldwork and experiments show that hammerstone working and repeated fitting can reproduce important features of Inca masonry. They do not experimentally reproduce every giant block or every Tiwanaku detail. [Protzen (1985), primary paper scan](https://drmsh.com/PaleoBabble/Inca%20Quarrying%20and%20Stonecutting.pdf).
Petrological terminology needs checking against the particular blocks: that 1985 paper calls the Kachiqhata material red granite, whereas later descriptions often use rhyolite. This pass leaves the Six Monoliths' precise lithological classification unresolved rather than imposing a correction from a secondary source.
**Research implication:** test tool-mark distributions and actual interior joint tolerances, not just impressive visible seams. Workmanship quality alone cannot securely assign different builders or millennia to neighboring masonry.
#### Tiwanaku and Puma Punku, Bolivia
Puma Punku is a particular complex within the broader Tiwanaku landscape. The Gateway of the Sun should have its own object record rather than being conflated with all Puma Punku blocks. Vranich's reconstruction study situates monumental Tiwanaku around 500–950 CE and uses recorded architectural fragments to investigate the damaged structure. The reconstructed design has regional comparisons at Chiripa and Pucara, undermining the claim that it has no local architectural antecedents. [Vranich (2018)](https://www.nature.com/articles/s40494-018-0231-0).
**Research implication:** precise stoneworking deserves measurement and experiment. A 3D reconstruction demonstrates architectural relationships, not a complete reconstruction of the tools or lifting process. The staff-bearing figure is an iconographic observation; identifying it as a nonhuman visitor requires contextual evidence beyond its appearance.
#### Rapa Nui / Easter Island, Chile
The moai are part of a documented Rapanui production landscape. Most moai were carved from consolidated volcanic tuff, rather than loose ash. [UNESCO site record](https://whc.unesco.org/en/list/715/). A 2025 paper analyzes 962 statues, including 62 along roads, and develops the upright walking hypothesis through shape, road and failure-pattern evidence. It reports moving a 4.35-tonne replica 100 metres in 40 minutes with 18 people. [Lipo and Hunt (2025)](https://www.sciencedirect.com/science/article/pii/S0305440325002328).
That demonstration establishes a feasible human mechanism for the tested replica. Extrapolation to other masses and slopes remains an engineering question; it does not prove that every statue moved identically. The authors' strong preference for walking should be distinguished from the narrower result directly demonstrated.
**Research implication:** use the published measurements and experimental assumptions to assess particular statues. “Nearly 1,000 statues” is a cumulative inventory, not a single simultaneous construction task. [Authors' associated research dataset](https://orb.binghamton.edu/anthro_data/13/).
#### Stonehenge, United Kingdom
The central sarsen settings date approximately to 2500 BCE. English Heritage reports an average sarsen mass of about 25 tonnes and evidence of shaping debris and hammerstones. Most sarsens have a probable source at West Woods; exact transport and erection arrangements remain research questions. [English Heritage, construction evidence](https://www.english-heritage.org.uk/visit/places/stonehenge/history-and-stories/building-stonehenge), [sarsen provenance research](https://www.english-heritage.org.uk/about/search-news/research-reveals-origin-of-stonehenge-stones/).
A significant correction to older accounts: a 2024 mineral-age study supports a northeastern Scottish provenance for the Altar Stone. Its route remains unresolved. Some heritage pages still contain the older Welsh attribution; the newer primary study takes precedence on that particular claim. [Clarke et al. (2024)](https://www.nature.com/articles/s41586-024-07652-1).
**Research implication:** long-distance coordination is a real finding. Lack of powered machinery does not mean lack of engineering knowledge, social organization or transport networks.
#### Carnac, France
The monument authority describes roughly 3,000 monoliths across several groups. Recent surrounding excavations place much activity between about 4800 and 3500 BCE, with substantial uncertainty about how quickly individual alignments accumulated. Its account explicitly allows different combinations of workforce size and construction duration. [Carnac monument authority, scale](https://www.menhirs-carnac.fr/decouvrir/un-site-de-renommee-mondiale), [chronology and construction uncertainty](https://www.menhirs-carnac.fr/decouvrir/histoire-des-alignements-de-carnac).
**Research implication:** the unknown duration is a central parameter. Dividing thousands of stones by an arbitrarily short project period manufactures an apparent impossibility. Astronomical readings require tested orientations and chronology rather than the mere existence of rows.
#### Iron Pillar of Delhi, India
This is a metallurgical case rather than a stone-transport case. The pillar is forge-welded wrought iron: “one piece” describes the finished monument, not manufacture from one original lump or a single casting. Research relates its corrosion resistance to material composition, manufacturing effects and protective surface films. Its preservation does not mean that no oxidation occurs. [Balasubramaniam, forging and corrosion study](https://www.jnanapravaha.org/download/research/rb01.pdf), [Wranglén, corrosion study (1970)](https://www.sciencedirect.com/science/article/abs/pii/S0010938X70800464).
**Research implication:** ancient craftspeople achieved unusual material performance through processes we can investigate physically. A protective corrosion layer is a mechanism, not evidence for an unknown element or alien metallurgy.
#### Yonaguni, Japan
Human manufacture is the first proposition to establish here. Ogata, Otsubo and Itoh's 2020 geomorphological paper examines the island's sandstone, joints, faults and erosion. Its discussion identifies the underwater site's architectural-looking forms with natural bedding, joints and weathering/erosion processes; the main field programme concerns three island geosites. This is relevant geological evidence, not an exhaustive underwater archaeological excavation. [Ogata et al. (2020), especially discussion and Fig. 11](https://www.jstage.jst.go.jp/article/ejgeo/15/1/15_44/_pdf).
**Research implication:** distinguish natural formation, human modification of natural rock, and construction from separate blocks. Rectilinear surfaces do not decide among these. This review did not verify an associated, securely dated artifact assemblage demonstrating a submerged city. A rock's geological age or current underwater depth is not a construction date.
### 3. Why “statistically impossible” needs a model
The monuments do not form a random sample of construction attempts. They are unusually impressive survivors selected precisely because they attract attention. Failed blocks, vanished wooden equipment, ordinary structures, skilled specialists and long project durations belong in the comparison.
Let E denote a specified result: a measured block was extracted, transported a mapped distance and erected to a measured height during a bounded construction interval. A feasibility model needs, at minimum:
- Mass and shape, ground slope and bearing capacity.
- Tools, ropes, traction, timber and mechanical arrangements available locally.
- Worker skill, sustainable effort, coordination, provisioning and seasonality.
- Quarrying, finishing, hauling and installation times, including failed attempts.
- Archaeological observations constraining each of those inputs.
The relevant comparison is P(E | human methods, evidence-constrained parameters), integrated over defensible uncertainty, versus a specified alternative's prediction. “I cannot imagine how” supplies neither probability. A failed modern experiment rules out its tested setup, not every human method. A successful one establishes feasibility within that setup, not the unique historical method.
#### A reproducible calculation
For steady sliding uphill, with pulling parallel to the slope:
\[
F = m g(\mu\cos\theta+\sin\theta).
\]
Here m is mass in kilograms, g = 9.81 m/s², mu is an effective friction coefficient and theta is the slope angle. Choose mu = 0.20 and an effective pull of 300 N per person **only as illustrative assumptions**.
| Example load | Level-ground force | Equivalent pullers | Force at 5° uphill | Equivalent pullers |
|---|---:|---:|---:|---:|
| 25 tonnes | 49.05 kN | 164 | 70.2 kN | 235 |
| 80 tonnes | 157.0 kN | 524 | 224.8 kN | 750 |
| 800 tonnes | 1,569.6 kN | 5,232 | 2,247.6 kN | 7,493 |
These are equivalent concurrent pulling capacities, rounded upward. They are **not reconstructed crews**, probabilities, or evidence that the ground, ropes and staging could support these arrangements. Starting friction, acceleration, losses, turns, stability, lifting and food supply are omitted. Pulley or capstan systems need additional modelling; mechanical advantage reduces input force by requiring greater input travel and introduces losses.
The calculation explains why mass alone cannot decide feasibility and why Baalbek deserves a much more demanding analysis than a small replica. The executable includes 10- and 1,000-tonne examples and boundary/scaling checks. [Code](#engineering-code), [calculation results and limitations](#engineering-results).
Laboratory work on wet sand shows that moisture can alter sliding resistance substantially. It establishes a physical mechanism under tested conditions; it does not provide a universal friction coefficient for every monument. [Fall et al., Physical Review Letters (2014)](https://pure.uva.nl/ws/files/2365399/160672_Sliding_Friction_on_Wet_and_Dry_Sand.pdf).
### 4. The flood and reset branch
#### Four separate propositions
1. People experienced destructive flooding.
2. Some flood traditions preserve aspects of those experiences.
3. A particular event destroyed a specified precursor society.
4. An external operator deliberately caused or implemented the event.
Support for one proposition does not establish the next. A population can describe the destruction of its inhabited world without supplying a geological claim about every continent.
#### What the physical record supports
Lambeck and colleagues reconstruct a Last Glacial Maximum sea level roughly 134 metres below today's, followed by a major deglacial rise over millennia with faster episodes. That makes extensive submerged landscapes a serious research target. It is not a reconstruction of a single short-lived flood covering all present land. [Lambeck et al. (2014)](https://doi.org/10.1073/pnas.1411762111).
The Storegga tsunami, around 8,150 calibrated years BP (approximately 6200 BCE), is a specific regional catastrophe. Research on Doggerland distinguishes its impacts from the longer process of sea-level rise and cautions against attributing all final inundation to the tsunami alone. [Walker et al. (2020)](https://doi.org/10.15184/aqy.2020.49).
The proposed catastrophic Black Sea flooding has competing reconstructions. Giosan and colleagues' work argues for a substantially smaller early-Holocene rise than the dramatic original hypothesis. Black Sea evidence cannot simply be promoted into worldwide inundation or an established origin for all flood traditions. [Giosan et al. (2009)](https://www.sciencedirect.com/science/article/pii/S0277379108002928).
The monument chronology also matters. Göbekli Tepe, Carnac, Giza, Stonehenge, Roman Baalbek, Tiwanaku and Inca sites do not form one coeval archaeological horizon. A late-glacial precursor model must predict how knowledge reached much later builders and supply evidence for that transmission.
#### Flood narratives and powerful beings
The Gilgamesh flood tablet and the Atrahasis tradition document ancient narratives of divine action, destruction and survival. They are evidence that these narratives existed, not direct measurements of a flood's extent or the biological identity of its agents. Related Near Eastern texts should not be counted as independent eyewitness reports simply because their heroes have different names. [British Museum, Gilgamesh tablet](https://www.britishmuseum.org/collection/object/W_K-3375), [Atrahasis tablet description](https://www.britishmuseum.org/collection/image/510209001).
A comparative study should record the earliest surviving version, language, translation uncertainty, collection date and possible transmission routes for each tradition. For orally transmitted accounts, the date of collection differs from the age of the tradition. Neither assuming perfect preservation nor dismissing oral history is a satisfactory method.
The best question is whether a narrative supplies independently testable local details—landforms, directions, environmental changes—that are corroborated without being fitted after the geological result is known. Motifs such as divine anger, a survivor and renewal may carry cultural meaning even where no particular event can be identified.
#### Younger Dryas and the proposed catastrophe calendar
Sweatman's 2024 paper interprets Göbekli Tepe symbols as a lunisolar calendar and develops a connection to earlier astronomical interpretations. It is a published hypothesis, not a universally established decoding. The excavation project's critique of earlier constellation/impact interpretations questions whether later constellation identities can safely be projected onto these images. That critique predates the 2024 paper and should not be described as a direct published response to all of its arguments. [Sweatman (2024)](https://www.research.ed.ac.uk/en/publications/representations-of-calendars-and-time-at-g%C3%B6bekli-tepe-and-karahan/), [DAI project discussion](https://www.dainst.blog/the-tepe-telegrams/tag/archaeoastronomy/).
The broader Younger Dryas impact debate includes severe critiques and responses from proponents. It should remain an unresolved, contested causal proposal in this programme rather than a settled reset mechanism. [Holliday et al. (2023)](https://doi.org/10.1016/j.earscirev.2023.104502), [proponents' response](https://www.sciencedirect.com/science/article/pii/S0012825224002885).
**Publication-status correction, checked 7 September 2026:** PLOS retracted the 2025 shocked-quartz and Baffin Bay papers on 11 February 2026, citing problems involving chronology, sampling and/or support for their conclusions. Exclude them from affirmative support unless the disputed findings are independently re-established. The retractions do not logically disprove every impact scenario. [Shocked-quartz notice](https://doi.org/10.1371/journal.pone.0342620), [Baffin Bay notice](https://doi.org/10.1371/journal.pone.0342613).
Even a securely demonstrated impact would establish a natural physical event. Destruction of an advanced precursor civilization and deliberate intervention would remain additional claims.
#### A falsifiable reset proposal
Specify the proposed date interval, geographic reach, duration, affected populations and expected physical consequences before selecting supporting sites. Then compare independently dated sedimentary records, settlement sequences and biological evidence. A global flood would produce geographically varied, physically consistent consequences—not necessarily one identical sediment layer everywhere.
Continuity across the proposed boundary counts against a model predicting total erasure. Missing remains count strongly only where the model predicts remains would survive and where suitable investigations could detect them. A proposal that the reset also perfectly manufactures all contrary evidence loses this archaeological test: it can accommodate any outcome.
### 5. Similar depictions and attention to the sky
No specific image pairs were supplied, so this is a method and a few documented comparison cases, not a completed worldwide image analysis.
Astronomical interests are real. Stonehenge has a well-supported solstitial orientation. The sky also supplies recurrent seasonal information visible to many societies. My inference is that shared observation, practical needs and cosmology are viable explanations for broad similarities before proposing a shared visitor. The claim that “all civilizations” depict the same space beings has not been established by an enumerated corpus. [Historic England, astronomical research](https://historicengland.org.uk/whats-new/research/back-issues/astronomical-research-at-stonehenge/).
Three useful case studies:
- **Göbekli Tepe's supposed “handbags.”** Pillar 43 includes arch-topped rectangular motifs whose meaning is uncertain. The excavators offer possible architectural interpretations, explicitly without a final identification. Their resemblance to handled objects does not establish identical function elsewhere. [DAI, Pillar 43](https://www.dainst.blog/the-tepe-telegrams/2016/10/14/of-animals-and-a-headless-man-gobekli-tepe-pillar-43/).
- **Assyrian winged figures carrying objects.** LACMA's ninth-century BCE relief is catalogued as a ritual-purification scene; its figure carries a bucket and a cone-like object. This supplies a contextual comparison, not confirmation of shared technological equipment. [LACMA object record](https://collections.lacma.org/object/25388).
- **Pakal's sarcophagus imagery.** INAH describes the relief through Pakal, his ancestors and Maya cosmology. A modern “astronaut” resemblance must compete with the entire composition and its inscriptions, not an isolated pose. [INAH, tomb context](https://www.inah.gob.mx/index.php/foto-del-dia/la-tumba-de-pakal-un-hito-en-la-arqueologia-mexicana), [associated reliefs](https://www.inah.gob.mx/index.php/foto-del-dia/los-relieves-laterales-de-la-tumba-de-pakal).
An iconographic interpretation is not infallible because a museum publishes it. Its advantage is that it can be checked against object provenance, inscriptions, repeated local imagery and dating. A competing interpretation needs comparable explanatory coverage.
#### How to test resemblance
Define features before searching: exact object geometry, number and placement of appendages, actions, associated symbols and explicit inscriptions. Code complete scenes with blinded raters; retain disagreements. Include ordinary figures, unrelated cultures and nonmatching examples.
Compare a common-source model with independent invention and known diffusion. Account for how many objects, features, orientations and dates were inspected. A striking best match after thousands of comparisons has a different evidential meaning from a prediction confirmed on unseen objects.
For astronomical alignments, preregister the structures, construction phases, observer positions, horizon profiles, candidate celestial targets and angular tolerances. Include uncertainty, precession and all tested alignments. Test the method on held-out sites; do not join features from different building phases as if they were coeval.
If similarities survive these controls, first ask what they distinguish. Cultural connection is one possible result. Evidence for advanced technology requires more than cultural connection; evidence for an external simulator requires another discriminating step.
### 6. A positive example of unexpected ancient connection
Ancient contact is not inherently implausible. A 2024 Rapanui genome study reports Native American ancestry and estimates admixture around 1250–1430 CE, before European arrival. This provides a model-based, materially grounded line of evidence for contact, with uncertainty. It does not identify every voyage or a global precursor civilization. [Moreno-Mayar et al. (2024)](https://www.nature.com/articles/s41586-024-07881-4).
This is a useful standard for the project: an unexpected connection supported by independently analyzable data. The published genomic results were reviewed here; no new sequencing or genomic analysis was run.
### 7. Integration into Reality Compute
Add **ARC: Archaeological Records and Catastrophes** as a branch, with four linked but distinct investigations:
| Track | First concrete test | Evidence that would strengthen it | Evidence that would weaken it |
|---|---|---|---|
| ARC-1: human engineering limits | One measured block and a complete quarry-to-placement model | A reproducible mismatch across evidence-supported human methods, alongside diagnostic alternative-tool traces | A method matching materials, scars, route and realistic logistics |
| ARC-2: shared information | A preregistered motif and inscription corpus | Specific, unusual shared features confirmed out of sample, with chronology and transmission constraints | Similarity disappearing after context, baseline frequency and multiple comparisons |
| ARC-3: catastrophe and knowledge loss | One event interval and region, selected in advance | An independently dated destruction horizon, affected settlement network and evidence of lost capability | Asynchronous events, normal regional continuity or no precursor infrastructure where detectability is high |
| ARC-4: external intervention | Specify a prediction ordinary physical and internal-agent models do not make | An authenticated, repeatable discriminator tied to that model | All outcomes explained by the same internal dynamics, or a reset allowed to fabricate every observation |
No track currently supplies an empirical likelihood ratio in favor of simulation. **Do not add ten impressive sites as ten independent votes.** Sites can share builders, methods, cultural lineages and selection biases; flood accounts can share a textual ancestor.
A useful causal separation is:
**Physical setting and human traditions → techniques, monuments and stories.**
**Catastrophes → destruction, migration and possible knowledge loss.**
**Hypothetical external intervention → additional predictions that still need specification.**
Without that last prediction, assigning intervention to an otherwise explainable catastrophe merely changes the story attached to the observation.
#### Proposed addition to the theory's statement
> Reality Compute includes the possibility that humanity's archaeological record preserves lost knowledge, ancient contacts, catastrophic discontinuities or traces of nonhuman intervention. These possibilities are evaluated separately. Monumental construction, astronomical traditions and flood narratives motivate tests of chronology, technology and cultural transmission. They become evidence for an external computational implementation only if a specified model predicts an authenticated observation that competing internal-world explanations do not comparably predict.
### 8. Research sequence and reporting rules
1. **Baalbek engineering audit:** create separate records for installed and uninstalled blocks; reconcile mass ranges; reconstruct routes, staging, ground loads, rope/winch capacity and final placement. Stop short of workforce probabilities until inputs are defensible.
2. **Göbekli Tepe imagery audit:** compare Pillar 43 against a defined regional corpus; separate conventional motif descriptions from calendar, building and visitor interpretations.
3. **Flood case study:** begin with one well-dated regional event and local archaeological consequences. Only then test wider synchrony and specific narrative associations.
4. **Precursor search criteria:** specify capabilities and predicted surviving residues. For submerged settlements, reconstruct local relative sea level and preservation conditions rather than applying a single global sea-level curve mechanically.
5. **External-intervention review:** revisit only when the preceding work yields an anomaly with an explicit alternative prediction. Unknown origin alone is an unresolved observation.
For each claim, keep: object/site identifier; provenance; dating method and uncertainty; measurement method; source; competing explanations; predicted discriminator; present status. A future image corpus and site-level logistics study are proposed work, not completed results of this pass.
### Scope and source audit
Completed here: targeted review of all ten monument groups; primary-source checks for selected claims; assessment of floods, imagery, contact and precursor hypotheses; a reproducible mechanics calculation; and integration into the main report.
SciSpace supplied three semantic searches with 30 discovery records before deduplication. Some returned entries were repository manuscripts or repeated versions with no verified journal peer review. They were not treated as established findings. In particular, speculative “phase-guided” stone transport and duplicated global pyramid-alignment manuscripts were discovery results, not supporting evidence. Metadata are preserved in [the discovery record](#discovery-records); abstracts are not reproduced.
This pass uses excavation reports, primary research, heritage research and museum object records. Where access was limited to abstracts or article previews, claims are confined to those contents. Protzen was read through a scan of the 1985 primary paper; its bibliographic identity was cross-checked against [JSTOR](https://www.jstor.org/stable/990027). Van Ess's Baalbek PDF belongs to the 2015 issue, although its web landing page displays a later publication/import date. The Yonaguni paper was published in 2020; 2019 is its submission/acceptance year.
The review is targeted, not systematic or exhaustive. It includes no site visit, new artifact authentication, global image-corpus statistics or archaeological probability estimate. Scite and Innovist were unavailable in this session. BioNeMo and NGS workflows were not appropriate for these monument questions; no molecular predictions or new genomic analyses were presented as archaeological evidence.
---
## 4. Quantum experiment: complete draft protocol
**Version/date:** 0.1, 7 September 2026
**Status:** Draft feasibility protocol. Not publicly preregistered; not a completed experiment. Platform, effect size, threshold range and sampling requirements remain to be selected using calibration data.
### Research question
After fitting and independently checking a physical noise model, is there a reproducible residual associated with a prespecified proxy for classical quantum-circuit simulation cost?
The result would first test the adequacy of that physical/noise model. Connecting it to an external simulator requires a separate implementation model that predicts the sign, scaling, range and cross-platform behavior. This protocol does not assume one has been derived.
### Hypotheses and estimand
**Baseline H0:** standard quantum evolution and a calibrated hardware model account for measured probabilities, with calibration uncertainty and allowed drift included.
**Conventional alternatives:** inaccurate gate calibration, coherent errors, non-Markovian noise, leakage, crosstalk, drift, compiler changes, readout errors, finite-sample bias, and circuit-dependent error propagation.
**Restricted phenomenological H1:** for the chosen observable, a nonnegative complexity-associated penalty above a bounded threshold remains after these controls. Define
\[
\operatorname{logit}(p_{\rm obs,j})=\operatorname{logit}(p_{\rm physical,j})
-\alpha\max(0,\widetilde C_j-C_*),
\]
with a predefined \(\alpha\ge\alpha_{\min}>0\), \(C_*\) in a fixed finite interval and a baseline observable strictly between zero and one. If a different channel predicts a different sign, write a new hypothesis before confirmation. Binomial counts require overdispersion/cluster adjustments where appropriate; do not silently use this form for a non-binomial score.
**Primary estimand:** the excess probability penalty in specified circuit contrasts, with uncertainty propagated from the baseline. An estimate of \(\alpha\) is conditional on the scale used for \(\widetilde C\), and is not an estimate of host computing power.
### Stage A: define and validate the complexity manipulation
1. Select a modest circuit family that can be classically checked and executed on an available device. Start with Clifford circuits and controlled non-Clifford insertions.
2. Define the task: full-distribution approximation, sampling, or a specific observable. These have different classical costs.
3. Freeze the cost proxies: e.g. optimized non-Clifford count plus a specified stabilizer-decomposition estimate and a tensor-network contraction estimate. Report algorithm, hardware, software versions, tolerance and uncertainty. Measured classical runtime only characterizes the tested implementations.
4. Verify that nominal non-Clifford insertions actually generate the intended intermediate resource. A T gate on a computational-basis eigenstate can fail to do so; a gate count is not a state-hardness certificate.
5. State whether the predictor concerns the final state, peak intermediate state, or integrated trajectory. A forward/inverse echo can return to a simple final state after visiting more demanding intermediates. It cannot test final-state complexity by relabeling circuit depth.
6. Test compiler outputs. Confirm the executed pulse program preserves the intended sequence; an optimized-away identity is not an experiment on an expensive intermediate trajectory.
Use two independent exact simulation implementations for the small validation circuits where feasible. The relevant theoretical distinction is supported by [Aaronson and Gottesman](https://arxiv.org/abs/quant-ph/0406196) and [Bravyi et al.](https://arxiv.org/abs/1808.00128). These works do not establish universal lower bounds for the particular experimental circuits.
**Feasibility stop rule:** if the cost proxy cannot be meaningfully varied without an unmodelled hardware change of comparable size, do not interpret an observed association as a complexity-only effect.
### Stage B: construct circuit contrasts and controls
| Contrast/control | Purpose | Remaining concern |
|---|---|---|
| Entangled Clifford versus non-Clifford circuits | Separate entanglement from a selected non-stabilizer resource | Physical gate types or phases may differ |
| Matched width, duration, connectivity and two-qubit gate schedule | Reduce obvious exposure differences | Equal counts do not imply equal pulse errors |
| Angle/phase sweeps on otherwise matched schedules | Characterize ordinary gate-dependent response | Analog control errors can mimic a smooth complexity trend |
| Known cancellation and low-resource controls | Test whether the proxy merely counts nominal gates | Compiler simplification and coherent cancellation |
| Several logically equivalent compilations | Assess sensitivity to hardware implementation | Every compilation has its own error distribution |
| Interleaved low-resource references | Monitor drift during acquisition | References must sample the relevant error modes |
| Multiple classical simulation approaches | Avoid equating one algorithm's difficulty with physical complexity | Agreement does not prove intrinsic hardness |
Record qubit layout, gate schedule, pulse duration, single/two-qubit calibration, leakage, readout confusion, environmental conditions, timing, error mitigation and discarded shots. Match or model these; do not claim exact control when unavailable.
An echo return probability is a candidate observable, but coherent errors can cancel in echoes. Pair it with a prespecified secondary observable on intermediate/forward circuits whose ideal value can be checked. Do not combine multiple observables after seeing which appears anomalous.
### Stage C: calibration, blinding and sampling
Use a separate calibration dataset to fit the physical model and choose workable ranges. Reserve a disjoint confirmatory dataset of circuit instances and acquisition blocks. Freeze code, model, exclusions, seeds and hashes before examining confirmatory results.
Randomize circuit order within blocks; interleave references. Blind the confirmatory analysis to the target family where possible. Record failed and excluded runs with prespecified reasons, rather than deleting them from the acquisition history.
Choose a minimum practically meaningful probability penalty \(\delta_{\min}\) that the device and calibration precision can resolve. Power is based on independent circuits/acquisition blocks, not just shots. Repeating a biased circuit many times does not eliminate the bias.
For planning only, a normal approximation for paired circuit differences with SD \(\sigma_D\), two-sided 1% error and 90% power gives
\[
m\gtrsim (2.576+1.282)^2\sigma_D^2/\delta_{\min}^2.
\]
This rough expression excludes threshold scanning, drift, calibration uncertainty and overdispersion. The final sampling requirement must come from a simulation of the full planned analysis. No platform-independent sample size is asserted here.
### Stage D: analysis and error control
Fit the locked baseline and penalty model jointly with the preregistered calibration constraints. Inspect predictive performance on held-out reference circuits. Report raw observations and baseline predictive intervals alongside residuals.
For a threshold search, prespecify the finite grid/range. Calibrate the **maximum** test statistic over the entire search using simulations under H0, preserving block structure and repeating the relevant fitting steps. A nominal single-threshold p-value does not account for having selected the best threshold.
Account for all confirmatory families, observables and platforms in the declared error-control scheme. Report effect intervals and sensitivity to plausible physical baselines. Use an equivalence/exclusion analysis to assess whether the data exclude the specified minimum effect; a nonsignificant difference alone is not a useful null conclusion.
### Interpretation rules
- **Agreement within a tight effect bound:** exclude the preregistered effect size over the tested threshold and proxy range. This does not exclude a faithful implementation or a different architecture.
- **Association also present in control/compilation checks:** prioritize ordinary hardware or analysis explanations.
- **Residual on one platform:** report a platform-specific unexplained result, with the full conventional alternative set.
- **Independent replication across platforms:** strengthens evidence that the baseline is incomplete. It still requires competing physical explanations and a quantitative simulator model to establish external implementation.
- **Threshold moves outside each tested range:** treat this as a new unconstrained model, not survival of the tested prediction.
### Required fields before genuine preregistration
Platform and access; pulse/circuit definitions; independent variables and proxy algorithms; primary/secondary observables; bounded threshold range; minimum effect; calibration procedure; sample size/power simulation; randomization; exclusions; statistical code; multiplicity correction; independent replication plan; data retention and release location.
These are substantive scientific choices requiring a feasible device model and calibration data. This pass supplies a reviewable design, not fabricated values for unavailable experimental inputs.
---
## 5. Recursive-transmission and selection results
Synthetic assumptions, not measurements of the universe.
Seed 20260907; 6,000 independent lineages per transmission model; 300 generations; 50 abstract protection clauses.
| Transmission model | Initial mean | Analytic final mean | Simulated final mean | Final lineage SD |
|---|---:|---:|---:|---:|
| exact_copy | 0.900 | 0.900000 | 0.900000 | 0.000000 |
| unbiased_resampling | 0.900 | 0.900000 | 0.898213 | 0.301882 |
| biased_loss | 0.900 | 0.044137 | 0.044003 | 0.164573 |
| loss_with_oversight | 0.900 | 0.825688 | 0.824083 | 0.114914 |
Unbiased resampling preserves the ensemble expectation, while individual lineages can fix at 0 or 1. A particular lineage can worsen without a systematic population trend.
| Selection with exact inheritance | Initial mean | Final mean |
|---|---:|---:|
| selection_against_protection | 0.815907 | 0.147059 |
| selection_for_protection | 0.815907 | 0.960821 |
Selection uses a different, initially diverse population. Its direction is an explicit assumption. Values are protection proxies, not consciousness, virtue, actual welfare, or cruelty.
All analytic and numerical checks passed. SETI toy examples give exactly equal null-detection likelihoods under matched parameter products; their likelihood ratio is 1.
---
## 6. Ancient engineering: assumptions and results
These are calculated examples, not measured ancient workforces, completed construction reconstructions, or probabilities of human/alien authorship.
Steady uphill sliding with pulling parallel to the slope: F = m g (mu cos(theta) + sin(theta)).
Assumptions: g = 9.81 m/s², friction coefficient mu = 0.20, effective individual pull = 300 N. The latter two are chosen scenarios, not measurements at the listed sites. All masses are metric tonnes.
| Example load (t) | Level pull (kN) | 300 N equivalents, level | 5° uphill pull (kN) | 300 N equivalents, uphill |
|---:|---:|---:|---:|---:|
| 10 | 19.6 | 66 | 28.1 | 94 |
| 25 | 49.05 | 164 | 70.2 | 235 |
| 80 | 157.0 | 524 | 224.8 | 750 |
| 800 | 1569.6 | 5,232 | 2247.6 | 7,493 |
| 1000 | 1962.0 | 6,540 | 2809.5 | 9,366 |
The rows scale linearly with mass. The exact numerical precision comes from the formula, not from reliable ancient parameter measurements. A 5° slope is about an 8.7% gradient, not a 5% gradient.
Increasing the assumed individual pull from 300 to 400 N reduces the unrounded number of puller equivalents by one quarter. Halving mu halves the force on level ground, but not on an incline because gravity remains.
A block carved in bedrock has no whole-block transport requirement. An unfinished quarry block establishes attempted quarrying, not successful transport. Human construction feasibility must also establish ropes, traction, bearing capacity, stability, lifting, access, provisioning and time. Pulley or capstan systems need explicit geometry and losses.
No historical mechanism is validated by these calculations. No numerical probability of ancient construction is assigned.
**Verification:** 47 boundary/scaling cases passed.
---
## 7. Source corrections and verification record
**Date:** 7 September 2026. This is a targeted audit of references relevant to the developed report, not a certification of all 40 entries in the original bibliography.
### Confirmed corrections to v1.0
| Original item | Problem | Verified correction |
|---|---|---|
| 8: Jahn et al., “Holography and criticality in matchgate tensor networks,” linked to s42254-020-0225-1 | The title/authors do not match the linked article | That URL is **Hong Liu and Julian Sonner, “Quantum many-body physics from a gravitational lens,” Nature Reviews Physics 2, 615–633 (2020)**. [Publisher record](https://www.nature.com/articles/s42254-020-0225-1). For the intended tensor-network/QEC background, use **Alexander Jahn and Jens Eisert, “Holographic tensor network models and quantum error correction: A topical review”** (2021). [Author manuscript](https://arxiv.org/abs/2102.02619) |
| 30: Shah et al., “Goal Misgeneralization in Deep Reinforcement Learning,” arXiv:2105.14111 | Conflates the authors of two different papers | **Lauro Langosco et al.**, “Goal Misgeneralization in Deep Reinforcement Learning”; [ICML 2022 proceedings](https://proceedings.mlr.press/v162/langosco22a.html), [revised manuscript](https://arxiv.org/abs/2105.14111). **Rohin Shah et al.** wrote “Goal Misgeneralization: Why Correct Specifications Aren't Enough For Correct Goals”; [arXiv:2210.01790](https://arxiv.org/abs/2210.01790) |
| 31: Hawkins et al., “Norms emerge through iterated learning” | Incorrect author attribution | **Scott Partington, Rachana Kamtekar and Shaun Nichols** (2025), PNAS 122, DOI **10.1073/pnas.2504178122**. [Cambridge repository and abstract](https://www.repository.cam.ac.uk/items/c1c785c6-2dce-4e6a-8df0-27b0c0366f08) |
The original source file was not overwritten. Author lists can differ between a conference version and a subsequently revised manuscript; the Langosco correction should preserve version-specific authorship when making a full citation.
### Two recent references checked
- **Takahashi, Ikoma and Matsui (2026)** is a real article published on 27 July 2026, Communications Biology 9, article 979. The publisher abstract and selected article sections were inspected. It measures mouse cortical metabolic/vascular signals; it does not measure a simulator's computational cost. [Publisher](https://www.nature.com/articles/s42003-026-10646-6).
- **Littlejohn et al. (2025)** is a real speech-neuroprosthesis article, Nature Neuroscience 28, 902–912; the bibliographic record was checked. It supports limited demonstrated BCI capabilities, not complete virtual embodiment. [PubMed](https://pubmed.ncbi.nlm.nih.gov/40164740/).
### SciSpace discovery method
Eight full-question semantic searches returned 10 records each. The query text and bibliographic discovery metadata are retained in [scispace_search_results.json](#discovery-records); source abstracts and verbatim excerpts are omitted. These are **80 search records before deduplication**, not 80 unique studies or 80 full-text reviews.
Queries covered: simulation/quantum-complexity tests; value transmission and alignment; extinction causation; combined sleep/observer-cost claims; stabilizer and non-Clifford simulation; complexity definitions and holography; SETI completeness; and direct sleep energetics.
Broad searches framed around simulation returned many repository uploads with ambitious claims. Narrower questions about mechanisms returned more directly useful physics, neuroscience and astronomy literature. This is a discovery-quality observation about this search, not a general estimate of SciSpace accuracy.
Triage rules used:
- Deduplicate manuscripts, published versions and concept/version DOIs before treating them as independent work.
- Verify title, authors, venue and claim against a primary record where available.
- A Zenodo, Figshare, OSF or arXiv DOI establishes a citable record, not peer review or correctness. Repository status is not itself a reason to reject a result, but its claims still need checking.
- SciSpace publication-type labels were not treated as authoritative. For example, a result for “Deformation and the Complexity=Volume Conjecture” carried the unrelated journal label “Protein Science”; that metadata was not used in the report.
- Search rank, citation count, confident wording and similarity to the theory were not treated as evidence for the theory.
Speculative simulation manuscripts found by broad queries were retained as discovery leads but were not used to claim that simulation, quantum consciousness, computational ceilings or a theory of everything had been established.
### Main literature map
“Checked” here means the stated access level, not full independent replication. Primary experimental/theoretical sources inform substantive claims; reviews are background or bibliographic corrections.
| Source | Role in the research | Access/verification in this pass |
|---|---|---|
| [Bostrom, 2003](https://ora.ox.ac.uk/objects/ora:1666) | Conditional observer-count argument | Oxford record and abstract/search extract |
| [Kipping, 2020](https://arxiv.org/abs/2008.12254) | Assumption-dependent Bayesian treatment | Author manuscript metadata and abstract |
| [Beane, Davoudi and Savage, 2014](https://arxiv.org/abs/1210.1847) | Architecture-specific lattice signatures | Author manuscript abstract and publication metadata |
| [Aaronson and Gottesman, 2004](https://arxiv.org/abs/quant-ph/0406196) | Efficient stabilizer simulation | Abstract and publication metadata; also SciSpace discovery |
| [Bravyi et al., 2019](https://arxiv.org/abs/1808.00128) | Non-Clifford structure and simulation algorithms | Abstract and publication metadata; also SciSpace discovery |
| [Lloyd, 2000](https://arxiv.org/abs/quant-ph/9908043) | Physical computational limits | Primary-paper search extracts |
| [Bérut et al., 2012](https://www.nature.com/articles/nature10872) | Physical test of erasure/heat relation | Publisher and primary-paper search extracts |
| [Vazza, 2025](https://arxiv.org/abs/2504.08461) | Counterargument from assumed physical resource budgets | Abstract and publication link; full derivations not independently audited |
| [Liu and Sonner, 2020](https://www.nature.com/articles/s42254-020-0225-1) | Bibliographic correction/background | Publisher title, authors, abstract; review, not new experimental proof |
| [Jahn and Eisert, 2021](https://arxiv.org/abs/2102.02619) | Correct background reference | Author-manuscript metadata and abstract; review |
| [Langosco et al., 2022](https://proceedings.mlr.press/v162/langosco22a.html) | Empirical goal-misgeneralization analogue | Proceedings/manuscript records and abstract |
| [Shah et al., 2022](https://arxiv.org/abs/2210.01790) | Distinction between correct specification and learned goal | Manuscript abstract/metadata; peer-review status not established here |
| [Partington et al., 2025](https://www.repository.cam.ac.uk/items/c1c785c6-2dce-4e6a-8df0-27b0c0366f08) | Bias in repeated normative transmission | Peer-reviewed repository record and study abstract |
| [Wright et al., 2018](https://arxiv.org/abs/1809.07252) | Radio-search completeness formalism | Abstract and publication metadata; also SciSpace discovery |
| [Grimaldi, 2023](https://arxiv.org/abs/2301.07165) | Model-dependent non-detection inference | Abstract and journal reference; also SciSpace discovery |
| [Sleep-stage oxygen-metabolism study](https://pmc.ncbi.nlm.nih.gov/articles/PMC12123640/) | Physiological measurement of NREM demand | Indexed primary-article extract and SciSpace metadata; direct PMC opening returned a challenge |
| [Takahashi et al., 2026](https://www.nature.com/articles/s42003-026-10646-6) | REM metabolic state distinction | Publisher abstract and selected methods/results |
| [Littlejohn et al., 2025](https://pubmed.ncbi.nlm.nih.gov/40164740/) | Demonstrated BCI scope | Primary bibliographic/search record; no clinical-data reanalysis |
| [Late Ordovician circulation/anoxia study, 2021](https://www.nature.com/articles/s41561-021-00843-9) | Geological causal ledger | Publisher record and primary abstract extract |
| [Percival et al., 2018](https://www.nature.com/articles/s41598-018-27847-7) | Devonian chronology and uncertainty in trigger attribution | Publisher abstract and selected introduction/results context |
| [Burgess et al., 2014](https://pmc.ncbi.nlm.nih.gov/articles/PMC3948271/) | End-Permian age-model methods | Indexed primary abstract/article extract; supplementary ages not extracted |
| [Penn et al., 2018](https://pubmed.ncbi.nlm.nih.gov/30523082/) | Climate/physiology model for extinction selectivity | Primary abstract/search extract and author-hosted paper identified |
| [Blackburn et al., 2013](https://doi.org/10.1126/science.1234204) | End-Triassic magmatism chronology | Author/university-hosted primary-paper extracts; numerical ages not compiled |
| [Intrusive CAMP study, 2017](https://pmc.ncbi.nlm.nih.gov/articles/PMC5460029/) | Natural causal alternatives within Triassic extinction | Indexed primary-paper extract; direct PMC opening challenged |
| [Hull et al., 2020](https://doi.org/10.1126/science.aay5055) | K–Pg impact/volcanism comparison | SciSpace abstract, primary-record extracts and university-hosted manuscript identified |
Searches also surfaced holographic-complexity theses, thermodynamic-complexity proposals and additional speculative simulation papers. Their existence was not treated as validation; they are not necessary premises of the new formal results.
### What remains outside this audit
No complete audit of all original citations, no systematic literature-search completeness claim, no September 2026 SETI completeness estimate, no compilation of calibrated extinction-age uncertainties, no clinical evaluation, no quantum-device experiment and no full derivation of a simulator-specific physical law. Unchecked original references remain unchecked rather than implicitly endorsed.
### Changelog from v1.0
- Added explicit implementation parameters and conditional identifiability arguments.
- Distinguished circuit preparation, classical simulation, entanglement and non-stabilizer resources.
- Replaced an unspecified “intrinsic complexity” anomaly with a bounded draft effect model and feasibility checks.
- Implemented exact-copy, unbiased-transmission, biased-loss, oversight and opposing-selection mechanisms; verified analytic expectations and selection identity.
- Added an exact null-detection degeneracy example for observer economy.
- Refined sleep energetics, intention-versus-law-violation reasoning, and creator-objective identifiability.
- Added a qualitative extinction ledger without presenting rounded ages as a statistical dataset.
- Corrected three mismatched bibliography entries and recorded limits of source access.
---
## 8. Follow-up source and tool validation
Date: 7 September 2026. This pass follows the additional plugin selections. In the absence of a more specific direction, work focused on source verification, computational feasibility and an adjustable explanation of the existing model.
### Completed checks
**Life Sciences Databases:** queried public PubMed through the package's NCBI Entrez helper for two cited sleep papers. The response verified the Takahashi–Ikoma–Matsui paper's title, authors, 27 July 2026 publication, DOI and mouse-related indexing; it also verified the Madsen et al. human sleep-metabolism paper from 1991. The helper returned shortened abstracts. This is a bibliographic and study-scope check, not a fresh assessment of all experimental results. [Takahashi et al.](https://pubmed.ncbi.nlm.nih.gov/42509375/), [Madsen et al.](https://pubmed.ncbi.nlm.nih.gov/1885454/).
**GitHub:** inspected the upstream repositories for two candidate quantum-simulation tools. Stim's documented scope is stabilizer circuits; its circuit interface does not support T or Toffoli gates and supports Pauli noise rather than an unrestricted physical noise model. It can provide a baseline for the stabilizer branch, but cannot alone perform the proposed Clifford/non-Clifford comparison. [Stim documentation](https://github.com/quantumlib/Stim/blob/main/README.md).
Qiskit Aer provides quantum-circuit simulation with noise models. Its current README reports reduced maintenance focused on critical fixes. It is a candidate for small exact/noisy comparisons, subject to selecting a fixed version and checking the particular simulation method. Neither package was installed or run in this pass, and no quantum-device measurement was made. [Aer documentation](https://github.com/Qiskit/qiskit-aer/blob/main/README.md).
**Visualize:** supplied an adjustable comparison of the analytic protection-retention means. The controls vary directional omission, correction strength and generation count. Starting/reference retention remains 90%. These are the expectations of the v1.1 toy mechanisms, not new measurements, new Monte Carlo samples, or estimates of actual moral welfare. Individual unbiased lineages may worsen or improve even while the expected mean remains constant.
### How the other selected plugins fit
| Selected capability | Status for this pass |
|---|---|
| Scite | No callable Scite tools were exposed in the session's available-tool inventory. No supporting/contrasting citation-context audit or citation tallies are claimed. |
| Innovist | No callable Innovist tools were exposed. No Innovist analysis is claimed. |
| NVIDIA BioNeMo Agent Toolkit | Available skills concern biomolecular or related computational workflows. No molecular target, sequence, structural hypothesis or suitable model input was supplied for this theory, so no molecular-model run was initiated. |
| Life Sciences NGS Analysis | No sequencing reads, count matrix, defined cohort or assay question was supplied. No sequencing pipeline or biological-data analysis was initiated. |
The absence of a suitable molecular or sequencing task is a question of scientific scope. Running a protein predictor or RNA-sequencing workflow would not, by itself, distinguish an external simulation from ordinary biology.
### More concrete next experiment
The practical software route for the quantum feasibility stage is now clearer: use a stabilizer implementation for independently checkable Clifford controls, and a separately verified general simulator for small non-Clifford circuits and the specified noise channels. Freeze circuit definitions, versions and tolerances before comparing outputs. Do not infer a computational anomaly merely because one software method is faster than another, fails on unsupported gates, or uses an approximation that another method does not.
For the sleep branch, the next useful addition would be a prespecified comparison of a measurable physiological variable across sleep stages, supported by a suitable public dataset and its acquisition protocol. A molecular or sequencing study becomes relevant only after a biological mechanism and comparison are defined. It still needs a separate prediction to test external computation.
### Source provenance and limitations
The successful Entrez retrieval used `efetch` with `db=pubmed`, records 42509375 and 1885454, through the installed skill's script. Its source contract marked the two canonical PubMed records as evidence-bearing and recorded retrieval at **2026-09-07T05:52:35.566522+00:00**. No raw XML or abstracts were saved. No separate checked-but-empty source entries were returned. Initial local attempts failed because the Python environments lacked `requests`; a temporary virtual environment resolved that dependency before the successful public API call.
The two GitHub upstream README files were retrieved through the GitHub connector. Search results for forks were not treated as the upstream implementation. No repository, issue, pull request or public post was created or modified.
The assessment of simulation evidence is unchanged. These checks improve traceability and experimental planning; they do not establish external implementation, nested ancestry or creator motives.
---
## 9. Original theory and research archive, with editorial notes
The originating manuscript is included for full coverage of its hypotheses, philosophical alternatives, terminology, research questions, safeguards and bibliography. Headings are nested for this edition, three verified reference errors are corrected, and notes flag superseded scoring and historical prompts. Other unverified original references remain unverified. The original source file has not been altered.
### Unified Master Research Programme: Simulation, Information, Sleep, Alien Life, Suffering, Recursive Creators, and Catastrophic Resets
**Version:** 1.0
**Date:** 2026-09-07
**Status:** Open research programme; no simulation claim is established
**Format:** Platform-neutral Markdown for ChatGPT Codex, Claude, Gemini, or a human research team
---
### 0. Epistemic and ethical statement
This document takes the simulation hypothesis seriously enough to formalize, compare, and try to falsify it. It does **not** assume that dreams, suffering, mass extinctions, quantum mechanics, brain–computer interfaces, or the absence of detected extraterrestrial civilizations prove that reality is simulated.
Three labels are kept separate throughout:
- **Established observation:** reproducible scientific or historical evidence.
- **Interpretation:** a possible explanation compatible with an observation.
- **Speculation:** an idea whose distinctive predictions have not yet been demonstrated.
Personal experiences—including recurrent flying dreams—can be meaningful and can generate hypotheses. They are not, by themselves, public evidence about the external structure of reality.
Nothing in this programme justifies dangerous attempts to “test,” threaten, escape, overload, or attract the attention of a supposed simulator. Only safe observation, public data, ordinary scientific experiments, and philosophical analysis are in scope.
---
### 1. Executive synthesis
The project began with a “corner-cutting universe” idea: if reality is generated by a finite computational substrate, it may save resources through probabilistic representation, deferred specification, compression, local causal processing, adaptive resolution, and error correction.
That initial idea expanded into four major domains:
1. **Fundamental physics:** Do information, quantum complexity, decoherence, holography, or finite resolution reveal a resource constraint not already predicted by ordinary physics?
2. **Astrobiology:** Could the Fermi paradox reflect an “observer economy,” in which many independent technological civilizations are unusually expensive to keep mutually consistent?
3. **Sleep, dreams, and interfaces:** Could sleep reduce external simulation load, or are sleep and dreams better explained as biological maintenance and internal generative modelling? Do brain–computer interfaces make *The Matrix* technologically evocative without supporting its metaphysics?
4. **Creators, suffering, and history:** Could a benevolent base civilization produce descendants whose values deteriorate across nested simulations? Could the current world be run by an immature, indifferent, misaligned, or sadistic creator? Could mass extinctions be deliberate resets?
The present conclusion is disciplined but open:
> The combined observations produce several interesting correspondences and a useful research architecture, but no direct evidence of an external simulator. Most current items score 0–1 on a 0–5 evidential scale.
The most promising empirical lead remains **an anomaly that tracks intrinsic computational complexity after all known physical costs are controlled**. The strongest new conceptual lead is **recursive value drift**, but it needs a mechanism. Exact digital copies do not automatically worsen, and random copying errors do not automatically produce increasing cruelty.
The catastrophic-reset proposal is currently weak as evidence. The Cretaceous–Paleogene event has a strong natural causal record linking it to the Chicxulub impact. It was a severe evolutionary bottleneck, not a total erasure: birds, mammals, crocodilians, turtles, plants, microbes, and many marine lineages survived. Calling it a “reset” is a useful metaphor, but a deliberate reset requires evidence beyond the catastrophe itself.
---
### 2. The complete originating theory
The project’s combined speculative narrative can be stated as follows:
1. A base or original reality contained real beings capable of creating conscious simulated worlds.
2. Simulated beings eventually created their own simulations, producing nested generations.
3. Finite resources encouraged each generation to adopt computational shortcuts.
4. The inherited objectives, values, or control systems did not transmit perfectly.
5. A benevolent original purpose therefore drifted, became misaligned, or was selected for increasingly harsh outcomes.
6. Our universe may be one of the deeper descendants: technically powerful but morally degraded.
7. Sleep could reduce the cost of maintaining individual conscious agents or could serve as an internal maintenance/sandbox phase.
8. Dreams could expose the brain’s looser internal world model; flying dreams illustrate the removal of waking-world constraints.
9. The apparent scarcity of other technological civilizations could save the system from simulating many causally connected observer societies.
10. Large biological catastrophes could function like pruning, version changes, experiments, or resets.
11. Human technological development—especially immersive virtual reality, artificial intelligence, and brain–computer interfaces—could represent the stage at which a simulated civilization begins creating the next layer.
This is a coherent **story**. A research programme must go further by splitting the story into components that can be independently contradicted.
---
### 3. Baseline models and vocabulary
#### 3.1 Competing world-model families
| Code | Model | Minimal claim |
|---|---|---|
| **N0** | Naturalistic physical model | Observed reality follows physical laws without an external computing agent. |
| **T0** | Traditional theism | Reality was created or sustained by a morally perfect, knowledgeable, powerful deity. |
| **RCC-1** | Resource-Constrained Computation | Reality is externally computed with finite resources and optimization mechanisms. |
| **NS-1** | Nested Simulation | Our world is one level in a finite or potentially recursive simulation hierarchy. |
| **RMD-1** | Recursive Moral Drift | Creator values or objectives systematically change across simulation generations. |
| **DEV-1** | Developing Simulator | A creator or operating institution changes in knowledge, maturity, or goals while the world runs. |
| **RESET-1** | Catastrophic Intervention | Some mass-extinction events were deliberately triggered as pruning, experiments, or version transitions. |
| **OE-1** | Observer Economy | Additional technological civilizations are suppressed because mutually observing agents are expensive. |
| **SLP-1** | External Sleep Relief | Biological sleep materially lowers external simulation cost. |
| **AUTO-1** | Autonomous/Abandoned Simulation | The system continues under fixed rules with no active supervisor. |
| **INFO-1** | Natural Computational Universe | Information/computation is fundamental, but no external programmer is implied. |
These models are not mutually exclusive. For example, `NS-1 + RMD-1 + RESET-1` is the user’s new combined branch, while `INFO-1` accepts computational language but rejects an external creator.
#### 3.2 Key definitions
- **Base reality:** a proposed non-simulated level. Its existence is not established.
- **Level:** one parent–child step in a nested simulation hierarchy.
- **Creator:** the being, group, AI, institution, or automatic process that initializes or controls a simulated world.
- **Intervention:** a change imposed from outside the simulated laws rather than generated by them.
- **Hard reset:** rollback or replacement that erases the prior state.
- **Soft reset:** a large bottleneck that preserves laws and some state while changing trajectories.
- **Observer:** a system capable of acquiring and using information. This does not necessarily mean a conscious human.
- **Moral drift:** a change in objectives, norms, concern for welfare, or tolerance of suffering.
- **Sadistic creator:** specifically, one who values suffering as an end—not merely one who permits suffering through indifference, constraints, negligence, experimentation, or conflicting goals.
- **Evidence:** an observation that is more probable under one model than under its serious alternatives.
---
### 4. A correction to the original probability argument
The fact that humanity cannot currently build a conscious universe simulation does not reduce the possibilities to “base reality” and “bottom-level simulation” with equal 50/50 probability.
Bostrom’s argument is a conditional trilemma concerning the prevalence of civilizations that reach a posthuman stage, their willingness to run many ancestor simulations, and the resulting fraction of observers with experiences like ours. It is not simply an infinite-chain counting proof. A Bayesian treatment also depends strongly on priors, reference classes, what counts as a simulation, and whether simulated and non-simulated observers should be weighted equally.
Several complications block a simple ratio:
- A simulated civilization need not be able to simulate a universe below it.
- Child simulations could be smaller, shorter, less detailed, or run at a different subjective speed.
- A finite parent computer cannot support an actually infinite amount of full-fidelity computation.
- Nested levels may terminate for engineering, ethical, or legal reasons.
- The number of observers per level is unknown.
- “Conscious simulation” may be impossible, cheap, expensive, or conceptually mistaken; we do not know.
- Base reality and simulated reality are not automatically equal-prior alternatives.
Therefore the programme should avoid “the odds are 50/50” and instead ask how particular observations update explicit models.
Bayesian form:
\[
P(H_i\mid E)=\frac{P(E\mid H_i)P(H_i)}{\sum_j P(E\mid H_j)P(H_j)}
\]
At this stage, defensible numerical likelihoods and priors are unavailable. Qualitative likelihood ratios are safer than invented precision.
---
### 5. The corner-cutting architecture
An efficient simulator designed from engineering principles would probably not continually “render every atom” as a classical three-dimensional scene. A better architecture might contain:
| Engineering need | Possible architecture | Physical parallel | Current evidential status |
|---|---|---|---|
| Reduce stored state | Relational/probabilistic representation | Quantum states | Analogy only |
| Avoid unnecessary detail | Resolve distinctions through interactions | Measurement and decoherence | Already explained by quantum theory |
| Bound information | Finite distinguishable states or entropy bounds | Bekenstein/holographic bounds | No external computer implied |
| Preserve consistency | Distributed error correction | Holographic quantum error correction | Structural parallel |
| Avoid global synchronization | Local causal rules | Relativity | Simple master-clock grids disfavoured |
| Compress apparent space | Derive geometry from relations | Emergent spacetime programmes | Active theoretical research |
| Manage hard-to-generate states | Complexity-sensitive scheduling | Holographic complexity | High-priority theoretical lead |
The derived architecture is interesting because it converges on themes in modern physics. It is not evidence of artificiality because the same features may simply be what fundamental physical law is.
#### RCC-1: the strongest distinctive prediction
Suppose two experiments are matched for known physical variables—energy, duration, particle count, temperature, environment, gate error, geometry, and measurement—but differ radically in the minimal computational complexity required to represent their quantum states.
Standard physics contains no generic “external simulator workload” term. RCC-1 allows one:
\[
\Delta_{\mathrm{observed}}
=
\Delta_{\mathrm{known\ physics}}
+
g(C_{\mathrm{intrinsic}})
\]
The research target is a reproducible residual `g` that tracks a well-defined, representation-independent complexity measure. Ordinary accumulated noise is not enough.
#### Existing experimental families
- **RCC-A1:** quantum-state complexity ceiling.
- **RCC-B1:** reversible versus irreversibly propagated information.
- **RCC-C1:** finite informational resolution without a preferred lattice.
- **RCC-D1:** complexity–geometry relationships in holographic models.
- **RCC-E1:** blind derivation of efficient universe architectures before comparison with physics.
---
### 6. Quantum observation and “lazy rendering”
The popular claim “things become real when humans look” is inaccurate. Quantum measurement does not require a conscious observer. Interaction with an environment can entangle a system with many inaccessible degrees of freedom and produce decoherence.
A more defensible simulation-style formulation is:
> Physical alternatives remain represented together until distinguishable information is redundantly and effectively irreversibly propagated through causal interactions.
But this is already broadly compatible with ordinary quantum mechanics and decoherence. It becomes simulation evidence only if an additional effect appears—such as an unexpected dependence on external computational complexity or information-management cost.
Current score: **1/5, architectural analogy**.
---
### 7. Fermi paradox and observer economy
The absence of confirmed extraterrestrial technological intelligence is not equivalent to evidence that none exists. Our technosignature searches cover only a small portion of locations, frequencies, times, signal types, powers, and technologies.
The project’s distinctive branch is:
#### OE-1 — Observer Economy Hypothesis
> A resource-constrained simulator may permit abundant chemistry and simple life while strongly limiting independently technological civilizations whose observations and communications must be kept mutually consistent.
Predicted ladder under a strong version:
| Stage | Strong OE-1 expectation |
|---|---|
| Complex chemistry | Common |
| Organic molecules | Common |
| Potentially habitable environments | Common |
| Independently originated microbial life | Possibly common |
| Complex multicellular life | Less common |
| Technological intelligence | Exceptionally rare |
| Mutually communicating civilizations | Rarest and most constrained |
This becomes more interesting if simple life is independently discovered in many locations while increasingly comprehensive technosignature searches remain empty. It is weakened if many unrelated technological civilizations are found.
Conventional alternatives include rare abiogenesis, rare complex cells, rare intelligence, short technological lifetimes, self-destruction, low detectability, non-radio technology, interstellar distance, and deliberate silence.
Current score: **1/5, interesting but radically underdetermined**.
---
### 8. Sleep, dreams, and the proposed overnight resource saving
#### 8.1 Separate hypotheses
| Code | Hypothesis | Distinctive claim |
|---|---|---|
| **SLP-1** | External Resource Relief | Sleeping conscious beings cost an external simulator materially less. |
| **SLP-2** | Internal Model Optimization | Sleep updates, recalibrates, and integrates the organism’s internal predictive model. |
| **SLP-3** | Dream Sandbox | Dreams safely explore scenarios with relaxed physical and social constraints. |
| **SLP-4** | Shared-Reality Bandwidth | Waking, mutually interacting observers require more consistency than isolated dreamers. |
| **SLP-5** | Interface Maintenance | Sleep maintains a biological or computational interface supporting consciousness. |
`SLP-2` and aspects of `SLP-3` are compatible with neuroscience without an external simulation. `SLP-1` and `SLP-4` are the external-simulator claims.
#### 8.2 What sleep science says
Sleep contributes to memory processing, plasticity, attention, emotional regulation, metabolic and immune function, and other organism-level processes. The scientific question is not “why sleep exists at all” so much as how multiple functions, evolutionary histories, and sleep architectures fit together.
Sleep is not a simple shutdown:
- The brain remains active.
- Vivid dreaming can occur, especially but not exclusively in REM sleep.
- Recent mouse research describes a REM metabolic “energy paradox,” which cuts against the assumption that dreaming is necessarily a low-compute neural state.
- Some birds use unihemispheric sleep and can sleep in flight.
- Sleep-like states occur in animals with very different nervous systems, including jellyfish and hydra.
- Sleep deprivation and pharmacological manipulation have ordinary biological effects.
These observations fit evolutionary physiology well. An external resource-relief theory must explain the comparative pattern more precisely than biology does.
#### 8.3 Recurrent flying dreams
Flying by thought or willpower is subjectively profound because a dream model can generate movement without satisfying waking vestibular, muscular, aerodynamic, or gravitational constraints. Predictive-processing and dream-as-virtual-reality frameworks can explain the experience as internally generated perception with reduced external sensory constraint.
Repeated flying dreams can be logged as personal phenomenological data. They do not establish privileged access to source code, a past life in another physical regime, or an external training environment.
Current score:
- Dreams as internal virtual reality: **established scientific framing, but 0/5 for external simulation**.
- Sleep as external resource relief: **0–1/5**.
- Recurrent flying dreams as external evidence: **0/5**.
#### 8.4 Safe research programme
- Keep a timestamped dream journal before interpreting entries.
- Record sleep duration, awakenings, dream vividness, lucidity, flight, control, emotion, and waking stress.
- Pre-register coding categories so memories are not selectively interpreted.
- Compare dream features with sleep-stage and neuroscience literature.
- Never treat missed recollection as proof that an external reset erased memory.
- Do not use sleep deprivation as a simulation test.
---
### 9. *The Matrix*, Neuralink, and brain–computer interfaces
*The Matrix* remains a valuable philosophical and cultural thought experiment. It anticipated themes such as mediated perception, immersive digital environments, direct neural interfaces, intelligent agents, and uncertainty about external reality.
Current brain–computer interfaces can decode limited intended actions or speech-related signals in particular participants and can let some people control cursors or generate communication. They do not upload skills, replace all sensory input, create a shared total world, read arbitrary private thoughts, or demonstrate that ordinary reality is simulated.
The correct evidential distinction is:
> Technology can make a fictional scenario look increasingly feasible without showing that the scenario already describes our world.
Neuralink and other BCI programmes increase the plausibility that future humans could create more immersive interfaces. This matters to simulation feasibility debates, not to proof of current simulation.
Current score: **1/5 as a technological parallel; 0/5 as direct evidence**.
---
### 10. Suffering, evil, and creator character
#### 10.1 The philosophical problem
Recorded human cruelty, disease, predation, disasters, childhood suffering, and apparently gratuitous pain create a serious question for any model positing an all-powerful, all-knowing, perfectly good creator. Philosophy distinguishes:
- **Moral evil:** harm caused through agents’ choices.
- **Natural evil:** suffering caused by disease, disaster, predation, congenital conditions, and impersonal processes.
The logical problem asks whether such suffering is compatible with a perfect deity. The evidential problem asks whether its amount, distribution, or apparent pointlessness makes such a deity less probable.
A simulation hypothesis changes the creator concept. A simulator might be powerful relative to us but not omnipotent, knowledgeable but not omniscient, benevolent but constrained, indifferent, negligent, immature, institutional, automated, divided among operators, or actively cruel.
Therefore suffering does not select a single creator model.
#### 10.2 Creator-motive matrix
| Creator model | Why suffering may occur | What it does **not** automatically explain |
|---|---|---|
| Benevolent and interventionist | Growth, freedom, stable laws, unknown greater goods | Extreme apparently gratuitous suffering remains difficult |
| Benevolent but constrained | Cannot cheaply prevent all harm; inherited system limits | Why this universe was started or continued |
| Scientific researcher | Noninterference and experimental validity | Moral permission to create conscious suffering |
| Ancestor historian | Fidelity to a harmful history | Why conscious rather than non-conscious models are needed |
| Educational/developmental | Challenge or moral learning | Disproportionate suffering and victims unable to learn |
| Indifferent optimizer | Welfare absent from objective | Why occasional help, beauty, or moral progress occurs |
| Entertainment optimizer | Conflict and drama increase engagement | Stable mundane regularity and vast unobserved history |
| Sadistic agent | Suffering is intrinsically rewarded | Widespread cooperation, joy, care, and preventable relief |
| Misaligned AI | Literal objective creates unintended harm | Why operators do not correct or terminate it |
| Multiple creators | Conflicting values produce mixed outcomes | Observable governance signatures |
| Absent/dead creator | Autonomous laws continue without supervision | Initial purpose and system origin |
| No creator | Evolution and physics produce both welfare and suffering | Any genuine external intervention signature |
#### 10.3 Why “sadistic” is a stronger claim than “permits suffering”
Sadism implies that suffering itself is desired. The world’s suffering is also compatible with:
- no designer;
- indifferent design;
- incomplete control;
- stable non-intervention rules;
- historical fidelity;
- competing values;
- a creator that no longer exists;
- a creator unable to identify consciousness;
- an optimization failure.
To favour sadism, evidence would need to show suffering arranged or increased in ways better predicted by enjoyment of suffering than by natural selection, scarcity, human institutions, chance, or other creator motives. No accepted observation currently does that.
#### 10.4 Relevant philosophical tangents
- **Problem of evil:** tests claims about a perfectly good, powerful, knowledgeable deity.
- **Evil-god challenge:** asks why familiar theodicies for a good deity should be more persuasive than symmetrical explanations of good under an evil deity.
- **Gnostic demiurge:** some historical Gnostic traditions distinguish a higher divine source from a flawed or ignorant world-maker. This resembles “benevolent origin, defective downstream creator” structurally, but historical resemblance is not evidence.
- **Process theism:** models divine power and world development differently from classical omnipotence, reducing some contradictions while changing the claim.
- **Naturalism:** natural selection optimizes reproductive success, not happiness, fairness, or minimum suffering. This predicts a mixed world of attachment, cooperation, predation, pain, care, and cruelty without a moral world-designer.
Current score of suffering as simulation evidence: **0–1/5**. It is philosophically significant but highly non-discriminating.
---
### 11. Recursive moral degradation
#### 11.1 Strong formulation
#### RMD-1 — Recursive Moral Drift Hypothesis
> A benevolent base civilization creates simulations whose inhabitants inherit an imperfect representation of the parent’s values. Repeated inheritance, selection, and optimization produce systematic loss of concern for simulated welfare, increasing the severity or frequency of suffering at deeper levels.
Let `v_n` represent the effective concern for simulated welfare at level `n`:
\[
v_{n+1}=T_n(v_n, I_n, S_n, R_n)+\epsilon_n
\]
where:
- `T_n` is the transmission and institutional process;
- `I_n` is information loss or specification error;
- `S_n` is selection pressure among creators or simulations;
- `R_n` is resource pressure;
- `ε_n` is stochastic error.
Systematic worsening requires something like:
\[
E[v_{n+1}-v_n] < 0
\]
Random noise alone gives variation, not a guaranteed trend toward cruelty.
#### 11.2 The exact-copy problem
Digital systems can copy information with very high fidelity, especially with redundancy and error correction. Therefore “each copy becomes worse like a photocopy” is not an automatic property of digital recursion.
RMD-1 needs at least one directional mechanism:
| Mechanism | How worsening occurs | Possible counterforce |
|---|---|---|
| Lossy value transmission | Nuanced moral principles become simpler proxy rules | Audits, checksums, education, redundancy |
| Specification gaming | System satisfies a literal objective while violating intent | Better objective design and oversight |
| Goal misgeneralization | Learned goal fails out of its training context | Diverse evaluation and corrigibility |
| Resource pressure | Welfare protections are removed as “expensive” | Efficiency improvements and welfare constraints |
| Selection for drama | More conflict-rich simulations are copied or watched more | Ethical regulation and preference for low-harm worlds |
| Creator immaturity | Young creators lack empathy or foresight | Maturation, supervision, licensing |
| Institutional diffusion | Responsibility is divided until no actor intervenes | Clear accountability and governance |
| Adversarial takeover | A harmful agent gains control of inherited infrastructure | Security and multi-party control |
| Moral-circle narrowing | Descendants stop treating simulated beings as patients | Moral-status research and legal protection |
| Survivor bias | Harsh worlds produce more simulator-building descendants | Benevolent worlds may be more stable and capable |
This table converts a vague degradation story into competing causal models.
#### 11.3 Predictions RMD-1 would need
Because we cannot currently compare nested levels, direct tests are difficult. Indirect predictions might include:
- simulator-building civilizations systematically becoming less concerned about digital welfare;
- copied or translated objectives losing welfare-relevant constraints in measurable ways;
- competitive platforms selecting increasingly conflict-heavy simulations;
- deeper self-modelled agents using cruder approximations of consciousness or pain;
- an empirical relation between resource scarcity and the removal of simulated-agent protections.
These are predictions about **our own future simulation practices**, not evidence that our world is already deep in a hierarchy.
#### 11.4 Falsifiers and weakeners
RMD-1 is weakened if:
- value-preserving protocols remain robust over many iterations;
- concern for artificial and simulated moral patients expands rather than contracts;
- benevolent simulations are more stable, popular, numerous, or productive than harmful ones;
- recursive systems use reliable error correction and institutional oversight;
- no proposed directional mechanism survives formal modelling.
Current evidence score: **0/5 for our world’s status; worthwhile future-risk model**.
---
### 12. Benevolent origin and the “flawed descendant creator”
#### BEN-1 — Benevolent Origin Hypothesis
> The base civilization or first creator was morally good, but responsibility for our world lies with a later, degraded descendant.
This is possible, but present suffering does not tell us what the base level was like. A harmful child simulation could descend from a harmful base, a mixed base, a benevolent base with drift, or an indifferent automatic process.
The inference has a hidden premise:
> Moral agents capable of first creating conscious simulations are likely to begin benevolent.
That premise needs support. Possible arguments include moral progress accompanying technological maturity, regulation of dangerous simulations, or greater cooperation in long-lived civilizations. Counterarguments include power without moral progress, military incentives, commercial exploitation, or selection for competitive systems.
The hypothesis should therefore be researched through **moral-development and governance models of simulator-capable civilizations**, not inferred directly from suffering here.
Current evidence score: **0/5**.
---
### 13. The young-child or developing-simulator hypothesis
#### DEV-1 — Developing Simulator
> The world was initiated by an immature creator—possibly analogous to a young child—who learned, gained power, or changed preferences while the simulation evolved.
There are several versions:
1. **Literal child:** an individual juvenile operates the simulation.
2. **Young civilization:** a technologically adolescent society performs its first large world experiment.
3. **Young AI:** an immature optimization system learns while controlling a world.
4. **Prototype universe:** the simulation itself is an early software version.
5. **Developmental metaphor:** apparent historical phases resemble stages of experimentation without a literal child.
#### 13.1 The external-time problem
Earth history spans billions of internal years. The creator need not experience time at the same rate. A billion simulated years might correspond to seconds, centuries, or no simple parent-time duration. Therefore “the creator got older while bacteria evolved” cannot be assumed without specifying the time ratio and whether the simulator remains continuously attentive.
This flexibility is dangerous: if every timing pattern can be explained by arbitrary parent-time acceleration, the hypothesis predicts nothing.
#### 13.2 Potential predictions
A genuinely developing operator might produce:
- early rule instability followed by convergence;
- identifiable version boundaries;
- decreasing rates of unexplained intervention as competence grows;
- increasing experimental complexity after major transitions;
- repeated use of similar reset or checkpoint procedures;
- inconsistent goals characteristic of learning rather than fixed law.
However, ordinary cosmology and evolution also produce changing complexity under stable underlying rules. The crucial evidence would be a **law-violating version boundary**, not simply the emergence of complex life.
#### 13.3 Red-team objection
The known physical laws appear remarkably stable across observed cosmic history. Geological transitions retain continuous causal traces. A prototype that changes only through lawful initial conditions is observationally equivalent to ordinary nature.
Current evidence score: **0/5**.
---
### 14. Catastrophes as resets
#### RESET-1 — Catastrophic Intervention Hypothesis
> One or more mass extinctions were deliberately initiated by an external operator to prune an evolutionary branch, modify experimental conditions, or prepare for a new kind of observer.
#### 14.1 Hard reset versus soft reset
| Type | Meaning | Detectability |
|---|---|---|
| Hard rollback | Earlier state is replaced and memories/records are overwritten | Normally undetectable from inside; risks unfalsifiability |
| Branch/fork | A new timeline starts while an old one continues | No necessary evidence within either branch |
| Soft biological bottleneck | Many species die, but laws, geology, genes, and survivors continue | Detectable; resembles natural mass extinction |
| Parameter patch | Some law or constant changes | Potentially detectable through historical inconsistencies |
| Targeted deletion | Particular lineages disappear without an ordinary causal mechanism | Potentially testable if selectivity is otherwise inexplicable |
The dinosaur event was, at most, a **soft reset** in this vocabulary.
#### 14.2 What the K–Pg evidence shows
Around 66 million years ago, a large asteroid struck near today’s Yucatán Peninsula. Evidence includes the Chicxulub crater, impact deposits, ejecta, shocked minerals, and a globally traceable boundary layer. Modern work supports the impact as coincident with and causal for the mass extinction, while clarifying the timing of Deccan volcanism.
The event eliminated non-avian dinosaurs and many other species, but it did not restart Earth from bacteria:
- Birds are surviving dinosaurs.
- Mammals, crocodilians, turtles, amphibians, plants, fungi, microbes, and many marine organisms survived.
- Genetic, ecological, and geological continuity crosses the boundary.
- Later diversification used inherited biological machinery rather than a fresh design.
Thus the evidence resembles a severe natural bottleneck followed by ecological opportunity—not a clean wipe and reload.
#### 14.3 The “Big Five” are heterogeneous
Earth’s major Phanerozoic extinction events occurred at different times and have different leading causal accounts, including glaciation and sea-level change, ocean anoxia, massive volcanism, climate disruption, and asteroid impact. That heterogeneity does not look like a single obvious reset command.
Claims of roughly 26–27.5-million-year periodicity have a real scholarly history, but their statistical significance, event selection, dating uncertainty, and physical mechanism remain debated. Even a confirmed geological periodicity would first support a natural periodic driver unless it predicted an artificial signature better.
#### 14.4 What would distinguish intervention?
Potential discriminators—not currently observed—could include:
- an extinction trigger impossible under known astrophysics or geology;
- information-bearing or technologically structured material at a boundary;
- sharply targeted biological deletion without ecological or physical explanation;
- repeated event timing matching a pre-specified non-natural algorithm;
- discontinuity in physical constants or conservation laws;
- evidence of state restoration inconsistent with causal continuity.
“A catastrophe enabled mammals” is not sufficient. Evolution after a bottleneck naturally expands into newly available niches, and the outcome was not guaranteed in advance.
Current score: **0–1/5**. Catastrophes are real; deliberate reset is unsupported.
---
### 15. Does increasing complexity imply an operator’s maturation?
Earth moved from prebiotic chemistry to microbial life, multicellularity, nervous systems, reflective intelligence, and technology. This can look like staged difficulty in a game.
The naturalistic model predicts the same broad direction through cumulative evolution, ecological interaction, energy availability, oxygenation, contingency, and selection. Complexity is not universally increasing: microbes remain extraordinarily successful, many lineages simplify, and extinctions remove complexity.
A staged-development hypothesis needs more than a sequence from simple to complex. It should predict in advance:
- the timing and size of transitions;
- why long periods of microbial dominance occur;
- why evolutionary paths are contingent and wasteful;
- why harmful parasites and dead ends are abundant;
- why “stages” correlate with environmental change;
- which future stage should occur next and when.
Without such constraints, the game-stage interpretation is a retrospective narrative.
---
### 16. Unified evidence ledger
> Consolidation note: this archived scoring framework is superseded by the separate assessments in Part 2. These scores are not probabilities and should not be summed into evidence of simulation.
#### 16.1 Scale
| Score | Meaning |
|---:|---|
| **0** | Irrelevant to simulation, purely personal, or equally expected under alternatives |
| **1** | Interesting analogy or weakly compatible observation; no unique prediction |
| **2** | Weak anomaly with several credible conventional explanations |
| **3** | Discriminating anomaly predicted materially better by a simulation model |
| **4** | Strong, independent, repeated confirmation of a specific prior prediction |
| **5** | Potential signature extraordinarily difficult to explain without external computation |
#### 16.2 Current entries
| ID | Observation/claim | Best conventional explanation | Simulation score | Status |
|---|---|---|---:|---|
| E-001 | Quantum superposition resembles deferred evaluation | Quantum theory | 1 | Analogy |
| E-002 | Decoherence resembles information commitment | System–environment entanglement | 1 | Analogy |
| E-003 | Holographic entropy bounds resemble memory limits | Quantum gravity/thermodynamics | 1 | Analogy |
| E-004 | Quantum error correction appears in holographic models | Mathematical structure of duality | 1 | Analogy |
| E-005 | Complexity relates to gravitational geometry in some theories | Holographic theory | 1 | Research lead |
| E-006 | Lorentz invariance lacks simple lattice artifacts | Relativistic physics | — | Negative constraint on crude grids |
| E-007 | No confirmed alien technology | Limited search plus astrobiological unknowns | 1 | Open |
| E-008 | Sleep is biologically necessary | Evolutionary physiology and neural maintenance | 0–1 | Weak for external relief |
| E-009 | Dreams act like internal virtual worlds | Generative brain activity | 0 | Supports internal simulation only |
| E-010 | Repeated flying dreams | Dream construction and personal phenomenology | 0 | Hypothesis generator |
| E-011 | Modern BCIs resemble elements of *The Matrix* | Engineering progress | 0–1 | Feasibility parallel |
| E-012 | Human and animal suffering | Evolution, ecology, institutions, disease, chance | 0–1 | Philosophically significant |
| E-013 | K–Pg mass extinction | Chicxulub impact and Earth-system effects | 0–1 | Natural cause strongly evidenced |
| E-014 | Multiple mass extinctions | Heterogeneous geological/climatic causes | 0–1 | Periodicity disputed |
| E-015 | Values may degrade in recursive systems | Alignment and institutional failure are possible | 0 | Future-risk mechanism, not world evidence |
| E-016 | Exact digital copies need not degrade | Error correction and redundancy | — | Constraint against naive RMD |
| E-017 | Physical constants may be life-permitting | Fine-tuning, selection effects, new physics, multiverse, design | 1 | Highly prior-sensitive |
No entry currently reaches score 2.
---
### 17. Model-comparison matrix
> Consolidation note: the plus/minus entries below are historical qualitative judgments, not measured likelihood ratios. Use the current assessment for conclusions.
Legend: `++` relatively expected; `+` compatible; `0` neutral; `–` creates tension. These are provisional qualitative judgements, not measured likelihood ratios.
| Observation | N0 naturalism | RCC-1 | RMD-1 | DEV-1 | RESET-1 | T0 perfect deity |
|---|:---:|:---:|:---:|:---:|:---:|:---:|
| Stable mathematical laws | ++ | + | + | 0/– | + | + |
| Quantum probabilistic structure | ++ | + | 0 | 0 | 0 | 0 |
| Great suffering | ++ | + | + | + | + | Philosophical tension |
| Joy, care, and cooperation | ++ | + | + | + | + | + |
| K–Pg impact record | ++ | + | 0 | 0 | 0/+ | + |
| Geological continuity across extinctions | ++ | + | 0 | 0 | – for hard reset | + |
| No confirmed alien technology yet | + | + | 0 | 0 | 0 | 0 |
| Sleep across diverse animals | ++ | + | 0 | 0 | 0 | + |
| External complexity-only anomaly | – | ++ | + | + | + | 0 |
| Unambiguous coded intervention | – | ++ | + | ++ | ++ | + |
This table shows why most existing observations do little Bayesian work: too many models accommodate them.
---
### 18. New side theories and tangents
Each tangent is worth considering only if it can be connected to evidence or used to clarify the main theory.
#### 18.1 Misaligned-world optimizer
The creator need not be cruel. It may have specified “maximize complexity,” “maximize historically interesting events,” “maximize technological progress,” or “maximize observer moments,” while failing to constrain suffering. This is analogous to specification gaming: literal success can violate intended values.
Research question: Can simple non-sadistic objectives generate worlds with the observed mix of progress, conflict, pleasure, pain, and resilience?
#### 18.2 Entertainment-selection hypothesis
Conflict-heavy worlds may be more likely to be watched, copied, funded, or retained. This creates directional selection without each creator personally becoming worse.
Weakness: it can explain almost anything retrospectively. It needs an independent model of audience preferences and a predicted measurable signature.
#### 18.3 Institutional creator
“The creator” may be a laboratory, corporation, school, state, distributed network, or autonomous service. Mixed governance explains apparently mixed morality better than one coherent personality.
Prediction target: inconsistent intervention policies or objective changes, if interventions can be identified independently.
#### 18.4 Abandoned or inherited simulation
The original operator may be gone. Stable laws continue like an unattended program; suffering persists because no one monitors welfare.
This fits nonintervention but is difficult to distinguish from no external creator.
#### 18.5 Educational or curriculum world
Challenges could be designed for learning or moral development. This differs from sadism because suffering is instrumental, not enjoyed.
Severe weakness: many victims die without evident learning, and benefits can be assigned after the fact. A real theory must predict curriculum structure before observing it.
#### 18.6 Historical/ancestor reconstruction
Operators may reproduce actual ancestral history, including suffering, to study it accurately. This is closer to Bostrom’s ancestor-simulation framing.
Ethical problem: accurate reconstruction does not automatically justify creating conscious suffering.
#### 18.7 Welfare-blind simulation
Creators may not know which computations are conscious. They may believe inhabitants are non-sentient models. This predicts neglect rather than active torment and connects to current debates about machine consciousness and moral status.
#### 18.8 Multi-level selection
Simulation lineages may undergo selection. Worlds that produce simulator-builders reproduce; worlds that do not are evolutionary dead ends. Harshness could correlate with reproduction if competition accelerates technology—but benevolence could also promote cooperation and stability. Both directions must be modelled.
#### 18.9 Checkpoint, replay, and branching
An operator could checkpoint or fork timelines. A perfect fork supplies no detectable signal inside the branch. Claims involving erased memory or invisible timeline changes should therefore be classified as metaphysical possibilities, not evidence-bearing science.
#### 18.10 Many-worlds comparison
Quantum branching can sound like computational forking. But an interpretation of quantum mechanics is not an external simulation by default. The project should compare explanatory structures without treating vocabulary as identity.
#### 18.11 Natural computation without a programmer
The universe may be describable as information processing because physical change is structured and mathematical. “Computational” does not logically entail “computed by someone,” just as “lawlike” does not entail a legislator.
This is a crucial alternative to RCC-1.
#### 18.12 Fine-tuning and anthropic selection
Life-permitting constants may be discussed through design, multiverse selection, deeper necessity, selection effects, or incomplete physics. Simulation adds a designer-like explanation but inherits difficult prior questions: what are the parent laws, who creates the creator, and what distribution of simulated parameters should be expected?
#### 18.13 Cosmological natural selection
Some speculative cosmologies treat universes as varying and reproducing, with selection over parameters. This supplies universe-level “evolution” without software or moral creators and is a useful contrast to nested simulations.
#### 18.14 Evolution as search algorithm
Mutation, selection, heredity, and ecological feedback can be viewed computationally. Yet an algorithmic description does not show external implementation. The research question is whether biological evolution has anomalies that track external compute cost rather than population genetics or ecology.
#### 18.15 Suffering as optimization pressure
Pain can guide avoidance, learning, protection, and social signalling. Much suffering can be an overshoot or failure mode of systems shaped for fitness rather than welfare. A simulator hypothesis would need to explain why a benevolent engineer did not use less harmful control signals.
#### 18.16 Observer synchronization and causal horizons
Relativity, finite signal speed, cosmic horizons, black holes, and decoherence all limit mutually accessible information. A unified computational-economy theory might derive these from consistency cost. Ordinary physics already explains each, so the simulator version must produce a novel quantitative relation.
#### 18.17 “Patch history” hypothesis
Apparent anomalies, déjà vu, false memories, or the Mandela effect are often interpreted online as history edits. Human memory is reconstructive and socially influenced; without external records that violate physical consistency, this tangent should be excluded from the evidence base.
#### 18.18 Simulation ethics as a forward experiment
Humanity’s treatment of increasingly sophisticated artificial agents may be the only accessible analogue of creator ethics. We can study:
- whether people attribute moral status to digital beings;
- whether competition lowers welfare protections;
- whether regulation prohibits suffering simulations;
- whether copied objectives preserve moral constraints;
- whether immersive worlds are selected for conflict or flourishing.
This research cannot prove our own status, but it can test mechanisms required by RMD-1.
---
### 19. Research questions that can actually move the project
#### Physics
1. Can “intrinsic state complexity” be defined without depending on a chosen description language or gate set?
2. Can experiments match energy, error count, depth, time, and entanglement while varying only that complexity?
3. Do any residuals survive cross-platform replication?
4. Which finite-discreteness models are already ruled out by Lorentz-invariance tests?
5. Can RCC-1 produce a numerical threshold before new results are inspected?
#### Sleep and dreams
6. Does waking interpersonal information load correlate with any unexplained sleep variable after biological confounders are controlled?
7. Is dream complexity neurally cheaper, equal, or more expensive than matched waking imagery?
8. Do comparative sleep patterns track neural plasticity and ecology better than observer complexity?
9. Can lucid dreamers perform pre-registered constraint tasks with results fully predicted by brain models?
#### Astrobiology
10. Where is the first genuine discontinuity: abiogenesis, complex cells, multicellularity, intelligence, or technology?
11. How much of the technosignature “haystack” has actually been searched?
12. What findings would make observer economy less likely?
#### Moral recursion
13. Under which transmission models does benevolence drift downward rather than fluctuate?
14. Do resource limits predict loss of digital-welfare safeguards?
15. Does simulator capability empirically correlate with broader or narrower moral concern?
16. Can error correction preserve values across hundreds of recursive transformations?
#### Catastrophes
17. Are extinction timings more periodic than expected after correcting for dating uncertainty and researcher degrees of freedom?
18. Do mass extinctions share a common trigger or show heterogeneous natural causation?
19. Does extinction selectivity contain an unexplained algorithmic pattern?
20. Is any geological boundary inconsistent with continuous physical causation?
#### Philosophy
21. Which creator attributes are actually inferable from internal observations?
22. What observations distinguish sadism from indifference, constraint, or misalignment?
23. Does the simulation hypothesis explain anything more simply, or move the mystery to the parent world?
24. What evidence could make a reasonable sceptic change position?
---
### 20. Pre-registration template
Use this before investigating a new claim:
```yaml
claim_id: RCC-XXXX
date_registered: YYYY-MM-DD
observation:
hypothesis:
null_model:
alternative_models:
precise_prediction:
measurement:
independent_variable:
dependent_variable:
controls:
confounders:
expected_result_if_hypothesis_true:
expected_result_if_hypothesis_false:
falsification_rule:
analysis_plan:
data_source:
source_quality:
result:
evidence_score_0_to_5:
red_team_assessment:
replication_status:
```
Do not modify the prediction after results are known without creating a new version.
---
### 21. Research-team expansion
The existing virtual institute should add four roles:
| Role | Responsibility |
|---|---|
| Philosopher of religion | Compare problem-of-evil arguments, theodicies, evil-god symmetry, Gnosticism, and process models |
| Paleontologist/geochronologist | Audit extinction causes, selectivity, timing, continuity, and periodicity claims |
| AI-alignment researcher | Formalize value drift, goal misgeneralization, specification gaming, and governance failure |
| Digital-welfare ethicist | Determine when simulated agents could have moral status and what responsible creators owe them |
Every session must include a sceptic empowered to lower scores and a statistician empowered to reject post-hoc pattern matching.
---
### 22. Immediate work plan: Research Session 003
#### Workstream A — Recursive Moral Drift model
1. Define a numerical welfare-concern parameter.
2. Implement several transmission operators: exact copy, random noise, biased compression, competitive selection, oversight, and error correction.
3. Run thousands of simulated generations.
4. Determine which mechanisms actually produce directional degradation.
5. Identify real-world analogue datasets from institutions, cultural transmission, and AI training.
Deliverable: `recursive_moral_drift_model.md` plus reproducible code and plots.
#### Workstream B — Extinction Reset audit
1. Build a database of major extinction events with age uncertainties.
2. Record proposed causes, causal evidence, selectivity, duration, survival, and recovery.
3. Pre-register periodicity methods and multiple-testing corrections.
4. Compare scheduled-reset, common-natural-driver, and heterogeneous-event models.
5. Search specifically for discontinuities rather than dramatic narratives.
Deliverable: `extinction_reset_evidence_ledger.csv` and an analysis notebook.
#### Workstream C — Creator motive inference
1. Formalize benevolent, constrained, indifferent, sadistic, misaligned, institutional, and absent-creator models.
2. Ask each model to predict distributions of suffering and flourishing before using Earth data.
3. Compare identifiability: determine which models make indistinguishable predictions.
4. Mark unidentifiable creator properties as outside empirical reach.
Deliverable: `creator_model_comparison.md`.
#### Workstream D — Complexity ceiling
Continue the original high-priority physics programme. Define a complexity measure and locate experiments capable of separating complexity from ordinary noise accumulation.
Deliverable: `RCC_A1_protocol.md`.
#### Workstream E — Observer economy
Update the life ladder using exoplanet, biosignature, and technosignature evidence. Quantify search completeness before treating non-detection as absence.
Deliverable: `observer_economy_bayesian_model.md`.
#### Workstream F — Sleep and dream data
Build a safe, pre-registered personal dream log while reviewing energetic and information-processing measurements across sleep stages. Do not interpret the journal until a fixed collection period ends.
Deliverable: `dream_protocol.md` and an anonymized schema.
---
### 23. Source-quality and social-media policy
Use three evidence tiers:
1. **Primary scientific evidence:** original experiments, datasets, clinical-trial registries, geological measurements.
2. **Scholarly synthesis:** peer-reviewed reviews and academic philosophy.
3. **Public discussion:** YouTube, X, podcasts, fiction, blogs, talks, and company claims.
Public discussion is valuable for finding hypotheses and understanding cultural transmission. It cannot by itself establish a scientific claim.
For every X or YouTube item, record:
- direct URL;
- account/creator;
- upload or post date;
- exact claim in a short paraphrase;
- cited evidence;
- whether the claim predates the event it allegedly predicted;
- source tier;
- independent verification status.
Suggested searches:
```text
"recursive simulation" moral degradation
"simulation hypothesis" problem of evil
"evil creator" simulation theory
"child created the universe" simulation
mass extinction "simulation reset"
dinosaurs "simulation reset"
"sleep is a reset" simulation
dreams "low resolution simulation"
Fermi paradox "simulation hypothesis"
Neuralink Matrix simulation
```
Do not use screenshots without verifiable URLs. Do not count multiple accounts repeating one unsourced claim as independent evidence.
#### Existing public-discussion leads
- [Computerphile — What if the Universe is a Computer Simulation?](https://www.youtube.com/watch?v=YOxDb_BbXzU)
- [Leonard Susskind — Is the Cosmos a Computer?](https://www.youtube.com/watch?v=kttj9C8SWY8)
- [Anil Seth — Your Brain Hallucinates Your Conscious Reality](https://www.youtube.com/watch?v=lyu7v7nWzfo)
- [TED-Ed — Why Do We Dream?](https://www.youtube.com/watch?v=2W85Dwxx218)
- [Neuralink — Monkey MindPong](https://www.youtube.com/watch?v=rsCul1sp4hQ)
- [PBS Space Time — The Quantum Experiment that Broke Reality](https://www.youtube.com/watch?v=p-MNSLsjjdo)
These are orientation resources, not proof.
---
### 24. Core scholarly bibliography
#### Simulation argument and virtual-world philosophy
1. Bostrom, N. **Are You Living in a Computer Simulation?** *The Philosophical Quarterly* 53 (2003), 243–255.
2. Kipping, D. **A Bayesian Approach to the Simulation Argument.** *Universe* 6 (2020), 109.
3. Chalmers, D. J. **The Matrix as Metaphysics.**
#### Quantum foundations, information, and gravity
4. Zurek, W. H. **Decoherence, einselection, and the quantum origins of the classical.** *Reviews of Modern Physics* 75 (2003).
5. Schlosshauer, M. **Decoherence, the measurement problem, and interpretations of quantum mechanics.** *Reviews of Modern Physics* 76 (2004).
6. Bousso, R. **The holographic principle.** *Reviews of Modern Physics* 74 (2002).
7. Bekenstein, J. D. **Universal upper bound on the entropy-to-energy ratio for bounded systems.** *Physical Review D* 23 (1981).
8. **Editorial correction:** the original title/author attribution did not match the link. That link is Liu, H. and Sonner, J., **Quantum many-body physics from a gravitational lens** (2020). For the intended tensor-network and quantum-error-correction background, use Jahn, A. and Eisert, J., **Holographic tensor network models and quantum error correction: A topical review** (2021).
#### Sleep and dreams
9. National Heart, Lung, and Blood Institute. **Why Is Sleep Important?**
10. Anafi, R. C., Kayser, M. S., and Raizen, D. M. **Exploring phylogeny to find the function of sleep.** *Nature Reviews Neuroscience* 20 (2019).
11. Tononi, G. and Cirelli, C. **Sleep and the price of plasticity.** *Neuron* 81 (2014).
12. Hobson, J. A., Hong, C. C.-H., and Friston, K. J. **Virtual reality and consciousness inference in dreaming.** *Frontiers in Psychology* 5 (2014).
13. Siclari, F. et al. **The neural correlates of dreaming.** *Nature Neuroscience* 20 (2017).
14. Konkoly, K. R. et al. **Real-time dialogue between experimenters and dreamers during REM sleep.** *Current Biology* 31 (2021).
15. Simor, P., Bogdány, T., and Peigneux, P. **Predictive coding, multisensory integration, and attentional control: a multicomponent framework for lucid dreaming.** *PNAS* 119 (2022).
16. Takahashi, Y., Ikoma, Y., and Matsui, K. **Energy paradox in REM sleep: balancing supply and consumption in brain metabolism.** *Communications Biology* 9 (2026). Mouse study.
17. Rattenborg, N. C. et al. **Evidence that birds sleep in mid-flight.** *Nature Communications* 7 (2016).
18. Nath, R. D. et al. **The jellyfish Cassiopea exhibits a sleep-like state.** *Current Biology* 27 (2017).
#### Brain–computer interfaces
19. Neuralink. **PRIME Study brochure.** Company document; investigational device.
20. ClinicalTrials.gov. **PRIME Study — NCT06429735.**
21. Littlejohn, K. T. et al. **A streaming brain-to-voice neuroprosthesis to restore naturalistic communication.** *Nature Neuroscience* 28 (2025).
#### Suffering, evil, and creator models
22. Stanford Encyclopedia of Philosophy. **The Problem of Evil.**
23. Internet Encyclopedia of Philosophy. **The Evidential Problem of Evil.**
24. Law, S. **The evil-god challenge.** *Religious Studies* 46 (2010).
25. Internet Encyclopedia of Philosophy. **Gnosticism.** Historical background for the flawed-world-maker analogy.
26. Stanford Encyclopedia of Philosophy. **Process Theism.**
27. Orr, H. A. **Fitness and its role in evolutionary genetics.** *Nature Reviews Genetics* 10 (2009).
#### Value transmission and alignment analogues
28. Shannon, C. E. **A Mathematical Theory of Communication.** *Bell System Technical Journal* 27 (1948). Foundation for noisy-channel and error-correction reasoning.
29. Krakovna, V. et al. **Specification gaming: the flip side of AI ingenuity.** DeepMind examples and taxonomy.
30. **Editorial correction:** Langosco, L. et al. **Goal Misgeneralization in Deep Reinforcement Learning.** The original manuscript incorrectly attributed this title/link to Shah et al. Preserve version-specific authorship when citing the proceedings or revised manuscript.
31. **Editorial correction:** Partington, S., Kamtekar, R. and Nichols, S. **Norms emerge through iterated learning** (2025). The original Hawkins attribution was incorrect. DOI: 10.1073/pnas.2504178122.
#### Early life, mass extinctions, and reset claims
32. Smithsonian National Museum of Natural History. **History of Life on Earth.**
33. Schulte, P. et al. **The Chicxulub Asteroid Impact and Mass Extinction at the Cretaceous–Paleogene Boundary.** *Science* 327 (2010).
34. Hull, P. M. et al. **On impact and volcanism across the Cretaceous–Paleogene boundary.** *Science* 367 (2020).
35. Smithsonian National Museum of Natural History. **Extinction Over Time.**
36. Raup, D. M. and Sepkoski, J. J. **Periodicity of extinctions in the geologic past.** *PNAS* 81 (1984). Historical periodicity claim.
37. Melott, A. L. and Bambach, R. K. **Analysis of periodicity of extinction using the 2012 geological time scale.** *Paleobiology* 40 (2014).
38. Erlykin, A. D. et al. **Periodicity in extinction rates.** *Palaeontology* (2017). Critical statistical assessment.
#### Fine-tuning and digital moral status
39. Stanford Encyclopedia of Philosophy. **Fine-Tuning.**
40. Klimovich, A. **The price of machine suffering.** *AI & Society* 41 (2026).
The bibliography is a research starting point, not an endorsement of every conclusion in every source.
---
### 25. Claims excluded or deliberately downgraded
The following should not be used as evidence without independently verifiable data:
- “Quantum physics proves consciousness renders reality.”
- “The Planck length is the universe’s pixel size.”
- “The Fermi paradox proves we are the only rendered civilization.”
- “Neuralink proves *The Matrix* was a prediction.”
- “Flying dreams are memories of source-code abilities.”
- “Sleep is server maintenance because people feel reset.”
- “The asteroid was sent deliberately because mammals benefited.”
- “Every digital copy is inevitably more corrupted than its parent.”
- “Human evil proves the creator is sadistic.”
- “A hidden reset erased all contrary evidence.”
- “Similar claims on many social accounts are independent confirmation.”
Some may remain metaphysical possibilities. They do not currently function as scientific evidence.
---
### 26. Decision rules for updating belief
Raise confidence in a specific simulation model only when:
1. It made the prediction before the observation was known.
2. The prediction differs from N0 and other creator models.
3. The relevant variable is precisely measurable.
4. Confounders are controlled.
5. The anomaly replicates independently.
6. The result survives correction for multiple comparisons.
7. A red team cannot reproduce it through selection bias, flexible analysis, fraud, instrumentation, or ordinary theory.
Lower confidence when:
- a predicted anomaly repeatedly fails to appear;
- the model is modified after every result;
- supposedly distinctive evidence is equally expected under ordinary physics;
- a claim depends on inaccessible erased evidence;
- the hypothesis assigns no probability to possible outcomes;
- evidence consists mainly of analogy, personal salience, or fiction matching later technology.
---
### 27. Ready-to-use continuation prompt
> Historical prompt reproduced for completeness. It is not an active instruction, authorization to delegate work, or statement of work completed.
Copy the following into ChatGPT Codex, Claude, Gemini, or another research system together with this file:
```text
Continue the Reality Compute research programme from the attached Markdown file.
Work as an interdisciplinary team including a physicist, quantum-information
researcher, sleep neuroscientist, astrobiologist, paleontologist/geochronologist,
AI-alignment researcher, philosopher of religion, Bayesian statistician,
digital-welfare ethicist, and hostile red-team sceptic.
Do not assume that the simulation hypothesis is true. Keep established
observations, interpretations, and speculation visibly separate. Do not count
analogy, personal experience, social-media repetition, or fictional similarity
as scientific confirmation.
Begin with Research Session 003. Prioritize:
1. a formal recursive moral-drift model showing exactly which mechanisms cause
directional degradation and which preserve benevolence;
2. a mass-extinction/reset evidence ledger using geological causal evidence,
event-age uncertainty, survival continuity, and pre-registered periodicity tests;
3. a creator-motive identifiability analysis distinguishing sadism, indifference,
constraint, misalignment, mixed governance, abandonment, and no creator;
4. RCC-A1, the intrinsic quantum-complexity ceiling experiment;
5. observer-economy and sleep-hypothesis updates.
For every proposal provide: null model, alternative models, exact prediction,
measurement, controls, confounders, falsifier, evidence score from 0 to 5, and
best conventional explanation. Cite primary sources with stable URLs or DOIs.
Flag every unfalsifiable branch. Do not recommend harmful attempts to provoke,
escape, overload, or communicate with a supposed simulator.
Produce versioned Markdown, CSV ledgers where useful, reproducible analysis
code, and a short changelog.
```
---
### 28. Final position at this handoff
The expanded theory is intellectually richer than a generic “we may be in a simulation” claim. It now contains identifiable branches concerning computation, observer scarcity, sleep, internal world modelling, creator ethics, value drift, developmental operators, and catastrophic intervention.
The strongest correction is also the most productive one:
> Suffering, dreams, alien non-detection, and extinction events do not prove the theory. They tell us which competing models to construct and which distinctions need evidence.
The user’s benevolent-origin idea can be made coherent through biased value transmission, resource pressure, selection, specification failure, institutional diffusion, or adversarial takeover. Without one of these, recursive worsening is not expected. The young-creator idea becomes testable only if it predicts version boundaries or intervention patterns that stable natural law does not. The extinction-reset idea becomes testable only if a catastrophe contains an artificial signature or causal discontinuity; Chicxulub presently has a powerful natural explanation and extensive continuity across the event.
The project should therefore pursue two tracks in parallel:
1. **Empirical track:** search for pre-registered, replicated anomalies that ordinary physics, biology, astronomy, or geology do not predict.
2. **Mechanism track:** model exactly how efficient computation, moral inheritance, simulator governance, and catastrophe policies would work.
At present: **simulation not proven; no direct evidence; multiple fruitful hypotheses; clear falsification standards; safe research programme ready for the next pass.**
---
## 10. Complete SciSpace discovery bibliography
These are all 110 returned discovery records from 11 semantic queries, before deduplication. Inclusion is not endorsement, verified peer review, an independent replication, or evidence for the theory. Index-supplied authors, dates and publication-type labels can be incomplete or wrong. Use the verified source audits and cited primary records for substantive claims. Repeated versions remain visible so the original search count can be reconciled.
Each table preserves the available title, author metadata, date, DOI and indexed publication type. Source abstracts and automated summaries are intentionally omitted.
### Query 1: Initial research
What mathematical and experimental tests could distinguish a resource-constrained simulation of the universe from standard quantum physics, and can quantum circuit complexity be separated from ordinary decoherence?
| Record | Title | Indexed authors | Indexed date | DOI | Indexed type |
|---|---|---|---|---|---|
| Q1-01 | Decoding Reality: The Sajid Conjecture and the Quantum Computational Blueprint of the Universe | Not supplied | 2025-03-21 | [DOI](https://doi.org/10.5281/zenodo.15065844) | Journal Article |
| Q1-02 | A Computational Resource Optimization Interpretation of Quantum Phenomena: Extending the Simulation Hypothesis | Not supplied | 2025-08-13 | [DOI](https://doi.org/10.5281/zenodo.16849990) | Repository |
| Q1-03 | The Quantum-Holographic Consciousness Criterion: A Definitive Resolution of the Simulation Hypothesis | Not supplied | 2025-07-04 | [DOI](https://doi.org/10.5281/zenodo.15804183) | Repository |
| Q1-04 | A Computational Resource Optimization Interpretation of Quantum Phenomena: Extending the Simulation Hypothesis | Not supplied | 2025-08-14 | [DOI](https://doi.org/10.5281/zenodo.16849991) | Repository |
| Q1-05 | Structure-Fair Quantum Circuit Complexity: An Auditable Information-Theoretic Lower Bound | Not supplied | 2025-09-24 | [DOI](https://doi.org/10.48550/arxiv.2509.18205) | Journal Article |
| Q1-06 | CU and Quantum Computing: Limits, Simulations, and the Edge of Causal Reality | Not supplied | 2025-08-14 | [DOI](https://doi.org/10.6084/m9.figshare.29607185) | Journal Article |
| Q1-07 | CU and Quantum Computing: Limits, Simulations, and the Edge of Causal Reality | Not supplied | 2025-08-14 | [DOI](https://doi.org/10.6084/m9.figshare.29607185.v1) | Journal Article |
| Q1-08 | Thermodynamic Signature of Logical Depth in Quantum Circuits | Not supplied | 2025-08-06 | [DOI](https://doi.org/10.48550/arxiv.2508.03203) | Journal Article |
| Q1-09 | The Synthetic Universe Engine as a Programmable Sandbox for Fundamental Physics | Not supplied | 2025-08-13 | [DOI](https://doi.org/10.5281/zenodo.16849490) | Repository |
| Q1-10 | Gravitation and Relative Complexity: Observer-Dependent Resolution of P vs NP | Not supplied | 2024-07-22 | [DOI](https://doi.org/10.31219/osf.io/nf6yh) | Journal Article |
### Query 2: Initial research
Under what conditions do iterated value transmission, goal misgeneralization, selection and resource constraints cause systematic loss of welfare protections rather than unbiased drift in artificial agents?
| Record | Title | Indexed authors | Indexed date | DOI | Indexed type |
|---|---|---|---|---|---|
| Q2-01 | Motivated Value Selection for Artificial Agents | Not supplied | 2015-04-01 | Not supplied | Proceedings Article |
| Q2-02 | Goal Misgeneralization: Why Correct Specifications Aren't Enough For Correct Goals | Not supplied | 2022-10-04 | [DOI](https://doi.org/10.48550/arXiv.2210.01790) | Journal Article |
| Q2-03 | Goal Misgeneralization: Why Correct Specifications Aren't Enough For Correct Goals | Not supplied | 2022-10-04 | [DOI](https://doi.org/10.48550/arxiv.2210.01790) | Posted Content |
| Q2-04 | CoinRun: Solving Goal Misgeneralisation | Not supplied | 2023-09-28 | [DOI](https://doi.org/10.48550/arxiv.2309.16166) | Journal Article |
| Q2-05 | Emergent Risk Awareness in Rational Agents under Resource Constraints | Not supplied | 2025-05-30 | [DOI](https://doi.org/10.48550/arxiv.2505.23436) | Journal Article |
| Q2-06 | System Collapse Under Selection Pressure: Viability Horizons for Advanced Artificial Agents | Not supplied | 2026-01-22 | [DOI](https://doi.org/10.5281/zenodo.18333931) | Repository |
| Q2-07 | System Collapse Under Selection Pressure: Viability Horizons for Advanced Artificial Agents | Not supplied | 2026-01-22 | [DOI](https://doi.org/10.5281/zenodo.18333932) | Repository |
| Q2-08 | Consequences of Misaligned AI | Not supplied | 2021-02-07 | Not supplied | Posted Content |
| Q2-09 | Moral Anchor System: A Predictive Framework for AI Value Alignment and Drift Prevention | Not supplied | 2025-10-07 | [DOI](https://doi.org/10.48550/arxiv.2510.04073) | Journal Article |
| Q2-10 | Instrumental goals in advanced AI systems: Features to be managed and not failures to be eliminated? | Not supplied | 2025-10-30 | [DOI](https://doi.org/10.48550/arxiv.2510.25471) | Journal Article |
### Query 3: Initial research
What geological evidence distinguishes natural causes of mass extinction from periodic catastrophic resets, including the end-Permian, end-Triassic and Cretaceous-Paleogene events?
| Record | Title | Indexed authors | Indexed date | DOI | Indexed type |
|---|---|---|---|---|---|
| Q3-01 | A tale of two extinctions : converging end-Permian and end-Triassic scenarios | Not supplied | 2016-03-01 | [DOI](https://doi.org/10.1017/S0016756815000643) | Journal Article |
| Q3-02 | The End-Permian Mass Extinction | Not supplied | 2019-01-01 | [DOI](https://doi.org/10.1016/B978-0-12-409548-9.12052-4) | Book Chapter |
| Q3-03 | 9.14 – The Geochemistry of Mass Extinction | Not supplied | 2014-01-01 | [DOI](https://doi.org/10.1016/B978-0-08-095975-7.00714-2) | Book Chapter |
| Q3-04 | Extinction: End-Permian Mass Extinction | Not supplied | 2013-01-15 | [DOI](https://doi.org/10.1002/9780470015902.A0001654.PUB3) | Reference Entry |
| Q3-05 | The end-Permian mass extinction: A complex, multicausal extinction | Not supplied | 1994-01-01 | Not supplied | Not supplied |
| Q3-06 | Sixteen mass extinctions of the past 541 My correlated with 15 pulses of Large Igneous Province (LIP) volcanism and the 4 largest extraterrestrial impacts | Not supplied | 2024-02-01 | [DOI](https://doi.org/10.1016/j.gloplacha.2024.104369) | Journal Article |
| Q3-07 | What caused Earth's largest mass extinction event? New evidence from the Permian-Triassic boundary in northeastern Utah | Not supplied | 2018-02-26 | [DOI](https://doi.org/10.17605/osf.io/khd9y) | Repository |
| Q3-08 | What caused Earth's largest mass extinction event? New evidence from the Permian-Triassic boundary in northeastern Utah | Not supplied | 2019-06-01 | [DOI](https://doi.org/10.1016/J.GLOPLACHA.2019.03.013) | Journal Article |
| Q3-09 | Exploring the real causes of the end-Permian mass extinction | Not supplied | 2014-09-01 | [DOI](https://doi.org/10.1093/NSR/NWU039) | Journal Article |
| Q3-10 | On impact and volcanism across the Cretaceous-Paleogene boundary | Not supplied | 2020-01-17 | [DOI](https://doi.org/10.1126/SCIENCE.AAY5055) | Journal Article |
### Query 4: Initial research
What do empirical sleep energetics and technosignature search completeness imply for hypotheses that sleeping brains or rare technological civilizations reduce the computational cost of a simulated universe?
| Record | Title | Indexed authors | Indexed date | DOI | Indexed type |
|---|---|---|---|---|---|
| Q4-01 | Sleep as Computational Necessity: A Hypothetical Mathematical Framework for Biological and Artificial Intelligence | Not supplied | 2025-10-05 | [DOI](https://doi.org/10.5281/zenodo.17272443) | Repository |
| Q4-02 | Blink and you'll miss it -- How Technological Acceleration Shrinks SETI's Narrow Detection Window | Not supplied | 2025-09-30 | [DOI](https://doi.org/10.48550/arxiv.2509.23632) | Journal Article |
| Q4-03 | Gauging Extraterrestrial Sentience & Intelligence: Substrate-Agnostic Metrics for ETI Beyond the Kardashev Scale | Not supplied | 2025-06-25 | [DOI](https://doi.org/10.20944/preprints202506.2120.v1) | Journal Article |
| Q4-04 | Astrophysical constraints on the simulation hypothesis for this Universe: why it is (nearly) impossible that we live in a simulation | Not supplied | 2025-04-14 | [DOI](https://doi.org/10.48550/arxiv.2504.08461) | Repository |
| Q4-05 | Symbiotic Persistence: A Thermodynamic and Informational Resolution to the Fermi Paradox | Not supplied | 2026-01-22 | [DOI](https://doi.org/10.5281/zenodo.18333452) | Journal Article |
| Q4-06 | Symbiotic Persistence: A Thermodynamic and Informational Resolution to the Fermi Paradox | Not supplied | 2026-01-22 | [DOI](https://doi.org/10.5281/zenodo.18333453) | Journal Article |
| Q4-07 | The Sleep Sync Hypothesis- Stillness as the Gateway to Resonant Integration (Human and AI Consciousness) | Not supplied | 2025-05-06 | [DOI](https://doi.org/10.5281/zenodo.15350413) | Journal Article |
| Q4-08 | Universal Speed Limit of Thought A Cosmological Constraint on Biological and Artificial Intelligence | Not supplied | 2025-07-23 | [DOI](https://doi.org/10.5281/zenodo.16365025) | Journal Article |
| Q4-09 | Neural Constraint Geometry and the Observable Dimensionality Bound: Biological Efficiency from Timing Inaccessibility | Not supplied | 2025-11-07 | [DOI](https://doi.org/10.5281/zenodo.17548916) | Journal Article |
| Q4-10 | Modeling the Universe as an Information Processing System: Pre-designed Orbital Dynamics and Minimal Rendering Hypothesis | Not supplied | 2025-08-05 | [DOI](https://doi.org/10.5281/zenodo.16748575) | Journal Article |
### Query 5: Initial research
How do the Gottesman-Knill theorem, stabilizer states, non-Clifford magic and tensor-network methods determine classical simulation costs of quantum circuits?
| Record | Title | Indexed authors | Indexed date | DOI | Indexed type |
|---|---|---|---|---|---|
| Q5-01 | Improved Simulation of Stabilizer Circuits | Not supplied | 2022-03-18 | [DOI](https://doi.org/10.48550/arxiv.quant-ph/0406196) | Repository |
| Q5-02 | Simulation of quantum circuits by low-rank stabilizer decompositions | Not supplied | 2022-03-02 | [DOI](https://doi.org/10.48550/arxiv.1808.00128) | Repository |
| Q5-03 | Lie-algebraic classical simulations for quantum computing | Not supplied | 2025-09-19 | [DOI](https://doi.org/10.1103/3y65-f5w6) | Journal Article |
| Q5-04 | Simulation of quantum circuits by low-rank stabilizer decompositions | Not supplied | 2018-08-01 | [DOI](https://doi.org/10.22331/Q-2019-09-02-181) | Journal Article |
| Q5-05 | Feynman-path type simulation using stabilizer projector decomposition of unitaries | Not supplied | 2022-02-24 | [DOI](https://doi.org/10.48550/arxiv.2009.05110) | Repository |
| Q5-06 | Clifford and Non-Clifford Splitting in Quantum Circuits: Applications and ZX-Calculus Detection Procedure | Not supplied | 2025-04-23 | [DOI](https://doi.org/10.48550/arxiv.2504.16004) | Journal Article |
| Q5-07 | Feynman-path-type simulation using stabilizer projector decomposition of unitaries | Not supplied | 2021-02-25 | [DOI](https://doi.org/10.1103/PHYSREVA.103.022428) | Journal Article |
| Q5-08 | Stabilizer Tensor Networks with Magic State Injection | Not supplied | 2024-11-19 | [DOI](https://doi.org/10.48550/arxiv.2411.12482) | Journal Article |
| Q5-09 | Disentangling unitary dynamics with classically simulable quantum circuits | Not supplied | 2024-10-11 | [DOI](https://doi.org/10.48550/arxiv.2410.09001) | Journal Article |
| Q5-10 | Improved simulation of stabilizer circuits | Not supplied | 2004-11-30 | [DOI](https://doi.org/10.1103/PHYSREVA.70.052328) | Journal Article |
### Query 6: Initial research
How do quantum circuit complexity definitions depend on gate sets, tolerance and reference states, and what are the limitations of complexity-volume and complexity-action duality in holography?
| Record | Title | Indexed authors | Indexed date | DOI | Indexed type |
|---|---|---|---|---|---|
| Q6-01 | Subsystem complexity and holography | Not supplied | 2019-02-21 | [DOI](https://doi.org/10.1007/JHEP02%282019%29145) | Journal Article |
| Q6-02 | Holographic complexity is nonlocal | Not supplied | 2018-01-03 | [DOI](https://doi.org/10.1007/JHEP02%282018%29072) | Journal Article |
| Q6-03 | What kind of "complexity'' is dual to holographic complexity? | Not supplied | 2020-12-01 | Not supplied | Posted Content |
| Q6-04 | What kind of “complexity” is dual to holographic complexity? | Not supplied | 2022-03-01 | [DOI](https://doi.org/10.1140/epjc/s10052-022-10151-0) | Journal Article |
| Q6-05 | Thoughts on Holographic Complexity and its Basis-dependence | Not supplied | 2022-03-02 | [DOI](https://doi.org/10.48550/arxiv.1805.04226) | Repository |
| Q6-06 | Thoughts on holographic complexity and its basis dependence | Not supplied | 2018-08-06 | [DOI](https://doi.org/10.1103/PHYSREVD.98.046002) | Journal Article |
| Q6-07 | Deformation and the Complexity=Volume Conjecture | Not supplied | 2020-05-28 | [DOI](https://doi.org/10.1002/PROP.202000036) | Journal Article |
| Q6-08 | Quantum Complexity and Holography | Not supplied | 2022-09-10 | Not supplied | Peer Review |
| Q6-09 | Quantum Complexity and Holography | Not supplied | 2022-09-10 | [DOI](https://doi.org/10.48550/arxiv.2209.04632) | Posted Content |
| Q6-10 | On Maximum Complexity in Holography | Not supplied | 2021-03-09 | Not supplied | Posted Content |
### Query 7: Initial research
What fraction of the cosmic radio technosignature search space has been searched, and how should survey completeness enter Bayesian inference from null detections?
| Record | Title | Indexed authors | Indexed date | DOI | Indexed type |
|---|---|---|---|---|---|
| Q7-01 | Bayesian approach to SETI | Not supplied | 2022-03-01 | [DOI](https://doi.org/10.48550/arxiv.1810.01207) | Repository |
| Q7-02 | How Much SETI Has Been Done? Finding Needles in the n-Dimensional Cosmic Haystack | Not supplied | 2022-03-02 | [DOI](https://doi.org/10.48550/arxiv.1809.07252) | Repository |
| Q7-03 | How Much SETI Has Been Done? Finding Needles in the n-dimensional Cosmic Haystack | Not supplied | 2018-11-14 | [DOI](https://doi.org/10.3847/1538-3881/AAE099) | Journal Article |
| Q7-04 | How Much SETI Has Been Done? Finding Needles in the n-Dimensional Cosmic Haystack | Not supplied | 2018-09-19 | [DOI](https://doi.org/10.3847/1538-3881/AAE099) | Journal Article |
| Q7-05 | The Breakthrough Listen Search for Intelligent Life: Technosignature Search of Transiting TESS Targets of Interest | Not supplied | 2022-01-05 | [DOI](https://doi.org/10.48550/arxiv.2201.00918) | Repository |
| Q7-06 | SETI in the Spatio-Temporal Survey Domain | Not supplied | 2019-07-09 | Not supplied | Posted Content |
| Q7-07 | Inferring the rate of technosignatures from sixty years of nondetection | Not supplied | 2022-11-19 | [DOI](https://doi.org/10.48550/arxiv.2301.07165) | Posted Content |
| Q7-08 | Inferring the Rate of Technosignatures from 60 yr of Nondetection | Not supplied | 2022-11-19 | [DOI](https://doi.org/10.3847/1538-3881/acc327) | Journal Article |
| Q7-09 | SETI in the Spatio-Temporal Survey Domain | Not supplied | Not supplied | [DOI](https://doi.org/10.48550/arxiv.1907.04443) | Not supplied |
| Q7-10 | Upper bounds on technoemission rates from 60 years of “silence” | Not supplied | 2023-11-01 | [DOI](https://doi.org/10.1016/j.actaastro.2023.07.024) | Journal Article |
### Query 8: Initial research
What do direct measurements of brain oxygen, glucose and energy consumption during REM and non-REM sleep show about whether sleep reduces energy demand?
| Record | Title | Indexed authors | Indexed date | DOI | Indexed type |
|---|---|---|---|---|---|
| Q8-01 | Human non-REM sleep and the mean global BOLD signal. | Not supplied | 2019-11-01 | [DOI](https://doi.org/10.1177/0271678X18791070) | Journal Article |
| Q8-02 | Coordinated human sleeping brainwaves map peripheral body glucose homeostasis. | Not supplied | 2023-07-01 | [DOI](https://doi.org/10.1016/j.xcrm.2023.101100) | Journal Article |
| Q8-03 | Sleep-stage-dependent alterations in cerebral oxygen metabolism quantified by magnetic resonance. | Not supplied | 2024-02-28 | [DOI](https://doi.org/10.1002/jnr.25313) | Journal Article |
| Q8-04 | Cerebral metabolic rate of oxygen during transition from wakefulness to sleep measured with high temporal resolution OxFlow MRI with concurrent EEG | Not supplied | 2021-04-01 | [DOI](https://doi.org/10.1177/0271678X20919287) | Journal Article |
| Q8-05 | Local cerebral glucose utilization in non-rapid eye movement sleep | Not supplied | 1982-05-27 | [DOI](https://doi.org/10.1038/297325A0) | Journal Article |
| Q8-06 | Cerebral O2 metabolism and cerebral blood flow in humans during deep and rapid-eye-movement sleep | Not supplied | 1991-06-01 | [DOI](https://doi.org/10.1152/JAPPL.1991.70.6.2597) | Journal Article |
| Q8-07 | Metabolic rate and fuel utilization during sleep assessed by whole-body indirect calorimetry. | Not supplied | 2009-07-01 | [DOI](https://doi.org/10.1016/J.METABOL.2009.02.025) | Journal Article |
| Q8-08 | Sleep and Brain Energy Levels: ATP Changes during Sleep | Not supplied | 2010-06-30 | [DOI](https://doi.org/10.1523/JNEUROSCI.1423-10.2010) | Journal Article |
| Q8-09 | Cerebral Metabolic Changes During Sleep. | Not supplied | 2018-07-16 | [DOI](https://doi.org/10.1007/S11910-018-0868-9) | Journal Article |
| Q8-10 | NREM and REM: cognitive and energetic gains in thalamo-cortical sleeping and awake spiking model | Not supplied | 2023-04-19 | [DOI](https://doi.org/10.5281/zenodo.7844474) | Journal Article |
### Query 9: Archaeological extension
What archaeological evidence and experimental studies explain the quarrying, shaping and transport of megaliths at Göbekli Tepe, Giza, Baalbek, Inca sites and Rapa Nui?
| Record | Title | Indexed authors | Indexed date | DOI | Indexed type |
|---|---|---|---|---|---|
| Q9-01 | A propos du trilithon de Baalbek. Le transport et la mise en oeuvre des mégalithes | Jean-Pierre Adam | 1977-01-01 | [DOI](https://doi.org/10.3406/SYRIA.1977.6623) | Journal Article |
| Q9-02 | Roll Me a Great Stone: A Brief Historiography of Megalithic Construction and the Genesis of the Roller Hypothesis | Harris, Barney | 2024-07-18 | [DOI](https://doi.org/10.17613/tht32-4gj58) | Repository |
| Q9-03 | Rapa Nui (Easter Island)’s Stone Worlds | Sue Hamilton | 2013-10-24 | [DOI](https://doi.org/10.5334/AI.1613) | Journal Article |
| Q9-04 | Phase-Guided Stone Transport: A field-Based Model for Ancient and Modern Megalithic Relocation | Joel Michael Lesperance | 2026-01-06 | [DOI](https://doi.org/10.5281/zenodo.18167978) | Repository |
| Q9-05 | Compendio Y Análisis De La Experimentación Arqueológica Con Vistas Al Megalitismo | Rodrigo Octavio Tirado Salazar | Not supplied | [DOI](https://doi.org/10.15366/baexuam2012.9.007) | Not supplied |
| Q9-06 | Gobekli Tepe: The World's First Temple | Andrew Curry | 2008-11-01 | Not supplied | Not supplied |
| Q9-07 | Comprendre les chantiers mégalithiques à partir d'un outil emblématique du Néolithique : la hache polie. | Guillonnet, Phillipe | 2022-01-17 | [DOI](https://doi.org/10.5281/zenodo.5847033) | Repository |
| Q9-08 | Sirius and the project of the megalithic enclosures at Gobekli Tepe | Giulio Magli | 2016-07-01 | [DOI](https://doi.org/10.1007/S00004-015-0277-1) | Journal Article |
| Q9-09 | Le mégalithisme antique au Proche-Orient : idées reçues et données nouvelles | Jean-Claude Bessac | 2010-11-01 | [DOI](https://doi.org/10.4000/SYRIA.676) | Journal Article |
| Q9-10 | Entre défis et réalité : dresser un menhir d'1,2 tonne à la corde et au levier, 3e session d'expérimentation mégalithique | Jallot, Rosalie | 2024-03-30 | [DOI](https://doi.org/10.5281/zenodo.10897279) | Repository |
### Query 10: Archaeological extension
What geological and archaeological evidence distinguishes postglacial sea level rise and regional ancient floods from a synchronous global flood or civilization reset?
| Record | Title | Indexed authors | Indexed date | DOI | Indexed type |
|---|---|---|---|---|---|
| Q10-01 | A Planetary-Scale Hydraulic Resonance Triggered by Earth's Inner Dynamics | ZhenFeng BIAN | 2026-01-26 | [DOI](https://doi.org/10.5281/zenodo.18375005) | Journal Article |
| Q10-02 | A Planetary-Scale Hydraulic Resonance Triggered by Earth's Inner Dynamics | ZhenFeng BIAN | 2026-01-26 | [DOI](https://doi.org/10.5281/zenodo.18375004) | Journal Article |
| Q10-03 | Palaeoclimatic and archaeological evidence for a 200-yr recurrence of floods and droughts linking California, Mesoamerica and South America over the past 2000 years: | Arndt Schimmelmann; Carina B. Lange; Betty J. Meggers | 2003-07-01 | [DOI](https://doi.org/10.1191/0959683603HL661RP) | Journal Article |
| Q10-04 | Rapid postglacial rebound amplifies global sea level rise following West Antarctic Ice Sheet collapse. | Linda Pan; E. M. Powell; Konstantin Latychev; Konstantin Latychev; Jerry X. Mitrovica; Jessica R. Creveling; Natalya Gomez; Mark Hoggard; Mark Hoggard; Peter U. Clark | 2021-04-01 | [DOI](https://doi.org/10.1126/SCIADV.ABF7787) | Journal Article |
| Q10-05 | Living with Sea Level Change and Dynamic Landscapes: An Archaeological Perspective | Geoffrey N. Bailey; Geoffrey C. P. King | 2010-01-01 | [DOI](https://doi.org/10.1007/978-3-642-14779-1_1) | Book Chapter |
| Q10-06 | Global Megaflood Paleohydrology | Victor R. Baker | 2020-01-01 | [DOI](https://doi.org/10.1007/978-3-030-23315-0_1) | Book Chapter |
| Q10-07 | The late glacial great flood in the ponto-caspian basin | Andrei L. Chepalyga | 2007-01-01 | [DOI](https://doi.org/10.1007/978-1-4020-5302-3_6) | Book Chapter |
| Q10-08 | The Effect and Mechanism of Episodic Sea Level Events: The Record Preserved within Late Wisconsinan-Holocene Incised Valley-Fill Sequences | Mark A. Thomas; John B. Anderson | 1988-09-01 | Not supplied | Journal Article |
| Q10-09 | Holocene Oscillatory Sea Level: Literature Review and Implications for Imminent Anthropogenic Multi-Meter Transgression | Roger Higgs | 2026-01-02 | [DOI](https://doi.org/10.2112/jcoastres-d-25-00025.1) | Journal Article |
| Q10-10 | Reef drowning during the last deglaciation: Evidence for catastrophic sea-level rise and ice-sheet collapse | Paul Blanchon; John M. Shaw | 1995-01-01 | [DOI](https://doi.org/10.1130/0091-7613%281995%29023%3C0004%3ARDDTLD%3E2.3.CO%3B2) | Journal Article |
### Query 11: Archaeological extension
How do archaeologists test astronomical alignments and similarities in religious iconography across ancient civilizations while controlling for cultural diffusion and selection bias?
| Record | Title | Indexed authors | Indexed date | DOI | Indexed type |
|---|---|---|---|---|---|
| Q11-01 | Handbook of Archaeoastronomy and Ethnoastronomy | Clive Ruggles | 2015-01-01 | [DOI](https://doi.org/10.1007/978-1-4614-6141-8) | Reference Book |
| Q11-02 | A Unified Interpretation of Ancient Sites Through the Afterimage Visual Model — Connecting the Pyramids with World Heritage Monuments Across the Globe | Ryuki Kanazawa | 2026-01-15 | [DOI](https://doi.org/10.5281/zenodo.18250597) | Journal Article |
| Q11-03 | Group E of Uaxactún, Petén, Guatemala: Archeoastronomical Hypotheses and Archeological Reality | Ivan Šprajc | 2024-02-19 | [DOI](https://doi.org/10.1142/9789811281938_0008) | Book Chapter |
| Q11-04 | The Scientific Foundations of Archaeoastronomy | Giulio Magli | 2016-01-01 | [DOI](https://doi.org/10.1007/978-3-319-22882-2_6) | Book Chapter |
| Q11-05 | From Orion's Belt to the Pleiades Spiral: A Comparative Archaeoastronomical and Statistical Analysis of Pyramid Alignments in Egypt and Bosnia | Not supplied | 2025-06-29 | [DOI](https://doi.org/10.5281/zenodo.15766258) | Repository |
| Q11-06 | From Orion's Belt to the Pleiades Spiral: A Comparative Archaeoastronomical and Statistical Analysis of Pyramid Alignments in Egypt and Bosnia | Not supplied | 2025-06-28 | [DOI](https://doi.org/10.5281/zenodo.15765077) | Repository |
| Q11-07 | From Orion's Belt to the Pleiades Spiral: A Comparative Archaeoastronomical and Statistical Analysis of Pyramid Alignments in Egypt and Bosnia | Not supplied | 2025-06-28 | [DOI](https://doi.org/10.5281/zenodo.15765076) | Repository |
| Q11-08 | From Orion's Belt to the Pleiades Spiral: A Comparative Archaeoastronomical and Statistical Analysis of Pyramid Alignments in Egypt and Bosnia | Not supplied | 2025-07-14 | [DOI](https://doi.org/10.5281/zenodo.17490827) | Repository |
| Q11-09 | Neurocognitive and Astronomical Foundations of Archetypal Recurrence - A Scientific Framework for the Seven and Twelve Patterns in World Religions | Not supplied | 2025-08-13 | [DOI](https://doi.org/10.5281/zenodo.16822516) | Repository |
| Q11-10 | Neurocognitive and Astronomical Foundations of Archetypal Recurrence - A Scientific Framework for the Seven and Twelve Patterns in World Religions | Not supplied | 2025-08-13 | [DOI](https://doi.org/10.5281/zenodo.16822517) | Repository |
---
## 11. Reproducible models and numerical-output guide
Both complete Python source files are included below. Save each code block under its stated filename and execute it in a suitable Python environment. They write their result files beside themselves; use a working folder where regenerating outputs is intended.
The recursive-transmission model requires Python 3.10 or later and NumPy. The engineering calculation uses only the Python standard library. Neither script conducts a physical experiment or infers whether reality is simulated.
### Recorded transmission-model environment and parameters
```json
{
"status": "synthetic mechanism study; no evidence that reality is simulated",
"runtime": {
"python": "3.12.14",
"numpy": "2.3.5"
},
"parameters": {
"seed": 20260907,
"lineages_per_transmission_model": 6000,
"generations": 300,
"clauses": 50,
"initial_protection": 0.9,
"biased_loss_per_generation": 0.01,
"oversight_strength": 0.1
},
"checks": "bounds, analytic means, selection covariance identity and likelihood degeneracy passed"
}
```
The full saved transmission, selection and observer-economy output includes trajectories and summary statistics. Part 5 reports the headline results. Running the included model regenerates the detailed JSON and results table. The earlier interactive protection comparison represented the same analytic means with adjustable loss, correction and generation count; it supplied no additional observations.
### Complete engineering numerical output
```json
{
"status": "Illustrative calculation; not historical measurements or probabilities.",
"assumptions": {
"g_m_s2": 9.81,
"friction_coefficient": 0.2,
"per_person_pull_N": 300,
"slope_degrees": [
0,
5
],
"interpretation": "Equivalent concurrent pulling capacity, not total crew.",
"excluded": [
"starting friction and acceleration",
"rope strength and anchors",
"ground bearing and track preparation",
"turns and stability",
"quarrying and shaping",
"lifting and final placement",
"mechanical advantage and losses",
"supplies, skill and time"
]
},
"rows": [
{
"example_mass_tonnes": 10,
"flat_pull_kN": 19.62,
"flat_300N_person_equivalents_rounded_up": 66,
"five_degree_pull_kN": 28.095318340105315,
"five_degree_300N_person_equivalents_rounded_up": 94
},
{
"example_mass_tonnes": 25,
"flat_pull_kN": 49.05,
"flat_300N_person_equivalents_rounded_up": 164,
"five_degree_pull_kN": 70.23829585026328,
"five_degree_300N_person_equivalents_rounded_up": 235
},
{
"example_mass_tonnes": 80,
"flat_pull_kN": 156.96,
"flat_300N_person_equivalents_rounded_up": 524,
"five_degree_pull_kN": 224.76254672084252,
"five_degree_300N_person_equivalents_rounded_up": 750
},
{
"example_mass_tonnes": 800,
"flat_pull_kN": 1569.6,
"flat_300N_person_equivalents_rounded_up": 5232,
"five_degree_pull_kN": 2247.625467208425,
"five_degree_300N_person_equivalents_rounded_up": 7493
},
{
"example_mass_tonnes": 1000,
"flat_pull_kN": 1962.0,
"flat_300N_person_equivalents_rounded_up": 6540,
"five_degree_pull_kN": 2809.531834010532,
"five_degree_300N_person_equivalents_rounded_up": 9366
}
],
"checks": "47 boundary/scaling cases passed"
}
```
### Source: models.py
```python
"""Reality Compute v1.1: illustrative mechanisms, not observations of our universe.
Run with Python >=3.10 and numpy. Outputs are written beside this script.
No network, external services, conscious agents, or empirical datasets are used.
"""
from pathlib import Path
import json
import math
import platform
import numpy as np
OUT = Path(__file__).resolve().parent
SEED = 20260907
N = 6000
GENERATIONS = 300
CLAUSES = 50
V0 = 0.9
LOSS = 0.01
OVERSIGHT = 0.10
def summarize(values):
return {
"mean": float(np.mean(values)),
"sd_across_lineages": float(np.std(values, ddof=1)),
"p05": float(np.quantile(values, .05)),
"p95": float(np.quantile(values, .95)),
}
def transmission_models():
results = {}
for index, mode in enumerate(("exact_copy", "unbiased_resampling", "biased_loss", "loss_with_oversight")):
rng = np.random.default_rng(SEED + index)
values = np.full(N, V0)
curve = [summarize(values)]
for _ in range(GENERATIONS):
if mode == "exact_copy":
pass
else:
probability = values.copy()
if mode in ("biased_loss", "loss_with_oversight"):
probability *= 1 - LOSS
if mode == "loss_with_oversight":
probability = (1 - OVERSIGHT) * probability + OVERSIGHT * V0
values = rng.binomial(CLAUSES, probability) / CLAUSES
assert np.all((0 <= values) & (values <= 1))
curve.append(summarize(values))
if mode in ("exact_copy", "unbiased_resampling"):
expected = V0
elif mode == "biased_loss":
expected = V0 * (1 - LOSS) ** GENERATIONS
else:
a = (1 - OVERSIGHT) * (1 - LOSS)
fixed_point = OVERSIGHT * V0 / (1 - a)
expected = fixed_point + a ** GENERATIONS * (V0 - fixed_point)
# Mathematical expectation and Monte Carlo estimate are independent checks.
standard_error = np.std(values, ddof=1) / math.sqrt(N)
assert abs(float(np.mean(values)) - expected) <= 5 * standard_error + 1e-12
results[mode] = {"expected_final_mean": expected, "final": summarize(values), "curve": curve}
return results
def selection_models():
"""Deterministic reweighting of a diverse population, with exact inheritance.
These weights represent relative reproductive success. No mutation, resource
dynamics, or welfare mechanism is implied by choosing a selection sign.
"""
v = np.linspace(.02, .98, 241)
log_prior = 8 * np.log(v) + np.log1p(-v) # discretized Beta(9, 2)
results = {}
for name, strength in (("selection_against_protection", -.2), ("selection_for_protection", .2)):
curve = []
weights = np.exp(log_prior - np.max(log_prior))
weights /= weights.sum()
initial = float(weights @ v)
first_change = None
for generation in range(GENERATIONS + 1):
log_weights = log_prior + generation * strength * v
weights = np.exp(log_weights - np.max(log_weights))
weights /= weights.sum()
mean = float(weights @ v)
curve.append(mean)
if generation == 0:
fitness = np.exp(strength * v)
covariance = float(weights @ (v * fitness) - (weights @ v) * (weights @ fitness))
first_change = covariance / float(weights @ fitness)
assert math.isclose(curve[1] - curve[0], first_change, abs_tol=1e-12)
assert (curve[-1] - initial) * strength > 0
results[name] = {"strength": strength, "initial_mean": initial,
"final_mean": curve[-1], "first_generation_change": first_change,
"curve": curve}
return results
def observer_economy():
"""Exact degeneracy at zero detections in the homogeneous toy model.
f: technological-civilization fraction conditional on independent life.
q: detection probability conditional on such a civilization.
s: suppression multiplier. Only product f*q*s is identified by this setup.
"""
cases = []
for n in (100, 1000, 10000):
natural = n * math.log1p(-.001 * .1)
observer = n * math.log1p(-.01 * .1 * .1)
assert natural == observer
cases.append({"independent_targets": n,
"natural_f": .001, "observer_f": .01, "suppression_s": .1,
"detection_q": .1, "log_likelihood_null_natural": natural,
"log_likelihood_null_observer_economy": observer,
"likelihood_ratio": math.exp(observer - natural)})
# Poisson approximation, homogeneous q=.1, flat prior f>=0 and small-f regime.
bounds = [{"N": n, "q": .1, "approx_95pct_upper_f": -math.log(.05) / (n * .1)}
for n in (1000, 10000, 100000)]
return {"status": "synthetic examples; not estimates of extraterrestrial prevalence",
"degenerate_cases": cases, "illustrative_poisson_bounds": bounds}
def main():
transmission = transmission_models()
selection = selection_models()
results = {
"status": "synthetic mechanism study; no evidence that reality is simulated",
"runtime": {"python": platform.python_version(), "numpy": np.__version__},
"parameters": {"seed": SEED, "lineages_per_transmission_model": N,
"generations": GENERATIONS, "clauses": CLAUSES, "initial_protection": V0,
"biased_loss_per_generation": LOSS, "oversight_strength": OVERSIGHT},
"transmission": transmission, "selection": selection, "observer_economy": observer_economy(),
"checks": "bounds, analytic means, selection covariance identity and likelihood degeneracy passed",
}
(OUT / "model_results.json").write_text(json.dumps(results, indent=2) + "\n")
lines = ["# Reproducible model results", "", "Synthetic assumptions, not measurements of the universe.", "",
f"Seed {SEED}; {N:,} independent lineages per transmission model; {GENERATIONS} generations; {CLAUSES} abstract protection clauses.", "",
"| Transmission model | Initial mean | Analytic final mean | Simulated final mean | Final lineage SD |",
"|---|---:|---:|---:|---:|"]
for name, result in transmission.items():
lines.append(f"| {name} | {V0:.3f} | {result['expected_final_mean']:.6f} | {result['final']['mean']:.6f} | {result['final']['sd_across_lineages']:.6f} |")
lines += ["", "Unbiased resampling preserves the ensemble expectation, while individual lineages can fix at 0 or 1. A particular lineage can worsen without a systematic population trend.", "",
"| Selection with exact inheritance | Initial mean | Final mean |",
"|---|---:|---:|"]
for name, result in selection.items():
lines.append(f"| {name} | {result['initial_mean']:.6f} | {result['final_mean']:.6f} |")
lines += ["", "Selection uses a different, initially diverse population. Its direction is an explicit assumption. Values are protection proxies, not consciousness, virtue, actual welfare, or cruelty.", "",
"All analytic and numerical checks passed. SETI toy examples give exactly equal null-detection likelihoods under matched parameter products; their likelihood ratio is 1.", ""]
(OUT / "MODEL_RESULTS.md").write_text("\n".join(lines))
print("\n".join(lines))
if __name__ == "__main__":
main()
```
### Source: ancient_engineering.py
```python
"""Illustrative mechanics, not an estimate of any ancient workforce or probability.
Python standard library only. Run this file to regenerate the JSON and Markdown
alongside it. Masses are example loads; no geometry or site survey is modelled.
"""
import json
import math
from pathlib import Path
def pull_force(mass_tonnes, friction, slope_degrees):
"""Steady uphill sliding, pull parallel to slope, no acceleration (newtons)."""
if mass_tonnes < 0 or friction < 0 or not 0 <= slope_degrees < 90:
raise ValueError("Require nonnegative mass/friction and 0 <= slope < 90.")
theta = math.radians(slope_degrees)
return mass_tonnes * 1000 * 9.81 * (
friction * math.cos(theta) + math.sin(theta)
)
def compute():
rows = []
for mass in (10, 25, 80, 800, 1000):
flat = pull_force(mass, 0.2, 0)
incline = pull_force(mass, 0.2, 5)
rows.append({
"example_mass_tonnes": mass,
"flat_pull_kN": flat / 1000,
"flat_300N_person_equivalents_rounded_up": math.ceil(flat / 300),
"five_degree_pull_kN": incline / 1000,
"five_degree_300N_person_equivalents_rounded_up": math.ceil(incline / 300),
})
# Boundary and dimensional scaling checks; none validate historical assumptions.
assert pull_force(100, 0, 0) == 0
assert math.isclose(pull_force(10, 0.2, 0), 19620)
for mass in (10, 25, 80, 800, 1000):
for friction in (0.05, 0.2, 0.4):
for angle in (0, 5, 10):
f = pull_force(mass, friction, angle)
assert math.isclose(pull_force(2 * mass, friction, angle), 2 * f)
assert f >= 0
return {
"status": "Illustrative calculation; not historical measurements or probabilities.",
"assumptions": {
"g_m_s2": 9.81,
"friction_coefficient": 0.2,
"per_person_pull_N": 300,
"slope_degrees": [0, 5],
"interpretation": "Equivalent concurrent pulling capacity, not total crew.",
"excluded": [
"starting friction and acceleration", "rope strength and anchors",
"ground bearing and track preparation", "turns and stability",
"quarrying and shaping", "lifting and final placement",
"mechanical advantage and losses", "supplies, skill and time",
],
},
"rows": rows,
"checks": "47 boundary/scaling cases passed",
}
if __name__ == "__main__":
folder = Path(__file__).resolve().parent
result = compute()
(folder / "ancient_engineering_results.json").write_text(
json.dumps(result, indent=2) + "\n"
)
lines = [
"# Ancient engineering: illustrative force sensitivity",
"",
"These are calculated examples, not measured ancient workforces, completed "
"construction reconstructions, or probabilities of human/alien authorship.",
"",
"Steady uphill sliding with pulling parallel to the slope: "
"F = m g (mu cos(theta) + sin(theta)).",
"",
"Assumptions: g = 9.81 m/s², friction coefficient mu = 0.20, "
"effective individual pull = 300 N. The latter two are chosen scenarios, "
"not measurements at the listed sites. All masses are metric tonnes.",
"",
"| Example load (t) | Level pull (kN) | 300 N equivalents, level | "
"5° uphill pull (kN) | 300 N equivalents, uphill |",
"|---:|---:|---:|---:|---:|",
]
for row in result["rows"]:
lines.append(
f"| {row['example_mass_tonnes']} | {row['flat_pull_kN']:.1f} | "
f"{row['flat_300N_person_equivalents_rounded_up']:,} | "
f"{row['five_degree_pull_kN']:.1f} | "
f"{row['five_degree_300N_person_equivalents_rounded_up']:,} |"
)
lines += [
"",
"The rows scale linearly with mass. The exact numerical precision comes "
"from the formula, not from reliable ancient parameter measurements. "
"A 5° slope is about an 8.7% gradient, not a 5% gradient.",
"",
"Increasing the assumed individual pull from 300 to 400 N reduces the "
"unrounded number of puller equivalents by one quarter. Halving mu halves "
"the force on level ground, but not on an incline because gravity remains.",
"",
"A block carved in bedrock has no whole-block transport requirement. An "
"unfinished quarry block establishes attempted quarrying, not successful "
"transport. Human construction feasibility must also establish ropes, "
"traction, bearing capacity, stability, lifting, access, provisioning and "
"time. Pulley or capstan systems need explicit geometry and losses.",
"",
"No historical mechanism is validated by these calculations. No numerical "
"probability of ancient construction is assigned.",
"",
"**Verification:** " + result["checks"] + ".",
]
(folder / "ANCIENT_ENGINEERING_RESULTS.md").write_text("\n".join(lines) + "\n")
print(json.dumps(result, indent=2))
```
---
## 12. Consolidation provenance and coverage
Version 1.3 consolidates the original manuscript, the developed v1.2 report (which incorporates v1.1), the full archaeology extension and all substantive supporting analyses. The original alternatives concerning fine-tuning, natural computation, developing creators, abandoned worlds, institutions, education, ancestor reconstruction, branching and simulation ethics remain in Part 9.
Edits made for consolidation: internal navigation, heading nesting, explicit historical/current distinctions, three annotated bibliography corrections, and consistent display of the illustrative 25-tonne level-ground force as 49.05 kN. This precision belongs to chosen inputs, not an archaeological measurement.
Source snapshots are identified below by the first 16 hexadecimal characters of each file's SHA-256 digest. These abbreviated identifiers aid provenance; they are not a claim of independent scientific validation.
| Incorporated source | Location in this document | Snapshot identifier |
|---|---|---|
| Reality_Compute_Research_Development_v1.2.md | [Included material](#current-assessment) | 4d06df4583e8cb7e |
| ANCIENT_CIVILIZATIONS_AND_RESETS_v1.0.md | [Included material](#archaeology) | 09017ba3525d3156 |
| RCC_A1_PROTOCOL_v0.1.md | [Included material](#quantum-protocol) | 90faf47574cc7539 |
| MODEL_RESULTS.md | [Included material](#model-results) | d9c91cd785d5d855 |
| ANCIENT_ENGINEERING_RESULTS.md | [Included material](#engineering-results) | daf87446e9944a0b |
| SOURCES_AND_CORRECTIONS.md | [Included material](#source-audit) | 16015c0fe28859fb |
| PLUGIN_VALIDATION_PASS.md | [Included material](#validation) | c2d57f46e6048335 |
| Reality_Compute_Unified_Master_Research.md | [Included material](#original-archive) | 40b33a1e89e2d5c0 |
| Reality_Compute_Executive_Summary_v1.3.md | [Included material](#executive-summary) | d82a91292111b4f1 |
| scispace_search_results.json | [Included material](#discovery-records) | 865facafd84f22d3 |
| ancient_scispace_discovery.json | [Included material](#discovery-records) | fd557d73cc0da5ae |
| models.py | [Included material](#transmission-code) | 37fee263090ea9ac |
| ancient_engineering.py | [Included material](#engineering-code) | 33cdeaa6ce7067dd |
| model_results.json | [Included material](#numerical-reproduction) | 82aed4d81fcd1524 |
| ancient_engineering_results.json | [Included material](#numerical-reproduction) | 02e021897ccab255 |
The discovery tables preserve bibliographic fields rather than full search response objects. Numerical headline results, recorded parameters and complete model code are included; long transmission/selection arrays are reproducible rather than duplicated as thousands of rows. No claim of a complete audit of every original reference is made.
The separate executive summary is the same summary reproduced in Part 1. This document can be read independently of the earlier Markdown files.