OBSERVATORY / THE RESEARCH NOTE
THE RESEARCH NOTE · VERSION 0.2

Universal model:
two-domain schematic and tests.

This note is the reference for the Continuum adaptation. It keeps established physics and biology separate from the proposed organizing architecture, and it states plainly what the calculations do and do not show.

PRIMARY-SOURCE ANNOTATED

For Christopher Mayne · 20 November 2026 · Working version 0.2

What this investigation establishes

The proposed five dimensions can organize a useful schematic of two very different domains: biological evolution and large-scale cosmic expansion. Explicit domain models can then calculate trajectories and probability distributions. The strongest reusable structure is a graph of typed relationships, supported by nested descriptions and domain-specific equations.

The original wording needs refinement. A directional gradient is not required for every change; potential is not a universal synonym for cause. Transformation is usually a calculated result of a model, rather than another independent input. Energy content, structure, and constraints also overlap in their mathematical roles.

The revised five therefore survive as organizing dimensions, not five experimentally established, independent universal driving variables. We have tested whether they help construct models; we have not proved that reality has exactly five foundations.

This is a bounded computational investigation. Inputs below are illustrative and synthetic. No biological observations, cosmological survey data, or new experiments were fitted. Numerical consistency of an established model is different from empirical validation of either the model or our five-part framework.

1. Revised five dimensions

Original wordingRefined dimensionWhat belongs hereImportant boundary
Structure1. Structure and stateConstituents, arrangement, geometry, composition, distributions, memory, current variable valuesActual state and our evidence about it remain distinct
Potential2. Energetics and availabilityEnergy content, exchanges, thermodynamic availability, gradients where relevantNot every change requires a spatial gradient or a separate energy variable in its effective model
Interaction3. Couplings and dynamicsInteraction mechanisms, constitutive/rate rules, transition probabilities, feedbackThis supplies the domain-specific transition law; a correlation does not automatically qualify
Constraint4. Constraints and conditionsConservation, normalization, allowed ranges, geometry restrictions, initial/boundary conditions, scope assumptionsAn admissibility constraint is different from a mechanism that produces change
Transformation5. Change and persistencePredicted/observed trajectories, stationary behaviour, formation, growth, loss, decay, downstream consequencesOften an output of the other specifications; not assumed to be an independent cause

This arrangement keeps the emphasis on an operating universe while removing the suggestion that all change is growth, expansion, or improvement. Probability belongs wherever a model represents uncertainty or stochastic transitions. Value evaluates consequences against a subject or objective; this report does not turn it into a physical constant.

The distinction follows an earlier Decision OS architecture: constructing a domain's coordinates, locating its current evidential state, modelling trajectories, and assessing consequences are separate tasks. This five-dimension proposal is new and provisional.

2. The schematic: nesting plus cross-links

A tree tells us what is contained within something, or how a description is decomposed. A graph tells us which components affect, constrain, describe, or summarize other components. Both are needed. Nesting alone would miss feedback and cross-scale relationships.

The atlas connections are typed. Each type carries a different claim, and the app keeps the unresolved ones visibly dashed.

TypeMeaningExample
CompositionOne entity is part of anotherA gene is a segment within a genome
MechanismA specified process changes a stateReproduction and sampling change allele counts
ConstraintA relationship limits admissible statesAllele frequencies sum to one
DerivationEquations follow from other equations plus assumptionsFLRW symmetry and matter assumptions specialize Einstein's equations
AggregationMany details are represented by a coarser variableIndividual allele copies become a population frequency
ObservationEvidence measures or estimates a quantitySequencing counts estimate allele frequency
AssociationVariables covary, with causation not establishedA trait and an environmental measurement correlate
Hypothesis / unresolved bridgeA proposed connection needs evidence or another modelMolecular energetics quantitatively determines an organism's reproductive advantage

Each variable should have one canonical identity and units. Multiple roles reference that same variable; they do not create duplicate quantities to add together. This is the software-schema lesson applied directly to the scientific map.

3. Domain A: biology, with evolution as the computed slice

The bounded calculation concerns a fixed-size haploid population with two inherited variants, A and B. There are discrete, nonoverlapping generations, constant relative fitness, and no mutation or migration. The model traces one genetic frequency, not the origin of life or an organism's complete development.

  • Structure and state: population size N, A count X_g, frequency p_g = X_g/N; individuals, cells, genome, sequence and regulatory regions nest beneath it.
  • Energetics and availability: cellular energy exchanges and usable chemical free energy. In this population model these are not separate simulated state variables; their influence is compressed into reproduction parameters.
  • Couplings and dynamics: inheritance and reproductive weighting (w_A = 1 + s, w_B = 1), then Wright–Fisher sampling of the next generation.
  • Constraints and conditions: X_g is an integer from 0 to N; 0 ≤ p_g ≤ 1; the two frequencies sum to one; initial count, horizon, fixed N and constant fitness are assumptions.
  • Change and persistence: the distribution of future allele frequencies, the probability of loss, persistence or fixation by a horizon, and expected diversity.

The molecular and cellular branches identify where deeper explanations belong. They are not numerically solved here. Regulatory mechanisms govern expression and responses to conditions; a sequence alone does not specify a complete operating cell.

Computed relationships

Selection-weighted frequency follows from normalizing reproductive contributions:

q_g = p_g w_A / (p_g w_A + (1 − p_g) w_B) = p_g (1 + s) / (1 + s p_g)

The next generation has a distribution rather than a single guaranteed value:

X₍g+1₎ | p_g ∼ Binomial(N, q_g)

The implementation propagates the full finite-state distribution with a transition matrix, rather than relying on a few random example paths. The paths drawn in the lab are seeded illustrations and never determine the probability cards.

Numerical test

All three cases begin at p₀ = 0.10, with a horizon of 100 generations. The 5% advantage means a relative reproductive weight of 1.05 in this specific model; it is not a claim about a real species.

Synthetic caseMean A frequency at generation 100Probability A is lost by thenProbability A is fixed by then
Neutral, N = 20010.00%64.31%0.10%
Neutral, N = 80010.00%19.13%approximately 1.10 × 10⁻⁹%
5% A advantage, N = 20074.99%13.75%28.27%

The remaining probability is neither lost nor fixed by the horizon. These are finite-horizon model probabilities, not measured biological rates or confidence intervals. For the selected case, omitting drift gives a deterministic forecast of 93.59%, versus the stochastic model's 74.99% mean. In a nonlinear model, evolving the average does not generally equal averaging evolved populations.

For neutral haploid sampling, independent closed-form checks are available:

E[p_g] = p₀, Var(p_g) = p₀(1 − p₀)[1 − (1 − 1/N)ᵍ]
E[2p_g(1 − p_g)] = 2p₀(1 − p₀)(1 − 1/N)ᵍ

The finite-state calculations agreed with these checks within 10⁻¹⁰. Expected diversity started at 0.18 and ended at 0.10904 for N = 200, versus 0.15884 for N = 800. The quantity is gene diversity; in a haploid population it is not a count of heterozygous individuals.

What biology challenges

  1. A directional driving gradient is not needed for every effective state change. Under neutrality, random sampling changes individual populations while their average frequency remains constant. This does not mean reproduction happens without physical energy.
  2. Potential is not one universal quantity. Chemical availability, standing genetic variation, and possible future outcomes have different definitions and units.
  3. A growth or advantage narrative is insufficient. Even a positively weighted variant can be lost in this finite model.
  4. The deeper bridge is incomplete. We supplied s; we did not derive it from molecular physics, regulation, metabolism, or ecology. The map exposes that interface instead of silently claiming it is solved.

Cellular chemistry can be represented with a stoichiometric balance ċ = S·v, where concentrations, reaction rates and the stoichiometric matrix have explicit identities. Thermodynamic conditions constrain possible processes; rate mechanisms determine how quickly they proceed. They do not automatically yield the reproductive parameter s.

4. Domain B: macro physics, with cosmic expansion as the computed slice

The calculation uses a spatially flat, homogeneous and isotropic FLRW model containing noninteracting pressureless matter, radiation, and a constant vacuum-energy component. It does not simulate galaxies, individual stars, quantum gravity, or an absolute beginning.

  • Structure and state: FLRW geometry, scale factor a, expansion rate H, with separate matter, radiation and vacuum components, each with a density and pressure relation.
  • Energetics and availability: component energy densities and pressures. Matter dilutes as a⁻³, radiation as a⁻⁴, and the vacuum component stays constant. There is no external reservoir or spatial pressure gradient in this model.
  • Couplings and dynamics: Einstein's field equations, their FLRW specialization, and continuity equations for the chosen components.
  • Constraints and conditions: spatial flatness, homogeneity and isotropy, Friedmann consistency, an initial scale factor and a branch choice.
  • Change and persistence: the scale-factor trajectory, changing component importance, expansion rate, acceleration or deceleration, and redshift.

Let τ = H_ref Δt and h = H/H_ref. Dimensionless coefficients C_m, C_r, C_v specify densities relative to a fixed reference density. They are illustrative inputs, not a new fit to observed cosmological parameters.

h² = C_m a⁻³ + C_r a⁻⁴ + C_v
da/dτ = a h, dh/dτ = −½(3 C_m a⁻³ + 4 C_r a⁻⁴)

The implementation evaluates the Friedmann relation and integrates the elapsed reference time by numerical quadrature in ln a, checked against the same constraint. The coefficients use one fixed reference scale across scenarios; changing the vacuum coefficient changes the expansion rate at a = 1. They must not be described as scenarios with an identical measured present-day Hubble rate.

Numerical result

Both calculations run from a = 0.1 to a = 2. The initial positive expansion rate is calculated consistently for each case.

Synthetic modelC_mC_rC_vElapsed H_ref Δt
With constant vacuum component0.30.00010.69991.69011012
Without vacuum component0.30.000103.40349401

Within these assumptions, the vacuum-containing model reaches the target scale factor sooner. This is a comparison of two hypothetical models, not an intervention performed on the universe. The largest relative Friedmann-constraint residual across both integrations was approximately 6.1 × 10⁻⁹. Differential integration and independent quadrature agreed to approximately 3.5 × 10⁻⁹ in reference-time units. These quantify numerical consistency, not accuracy against observations.

At a = 1, the first model's constraint permits h = +1 or h = −1. The same instantaneous density relation admits an expanding or contracting branch. Initial conditions and evolution equations are necessary to specify a trajectory. An equation containing a relationship between quantities cannot always be read as an arrow of causation.

As a doubles, the energy of a fixed comoving region changes differently for each component: matter's is unchanged, radiation's halves, and the vacuum component's increases eightfold. This is compatible with local relativistic energy accounting; it warns against assuming one ordinary globally conserved cosmic energy sum.

What cosmology challenges

  1. Change need not originate in a spatial gradient. Homogeneous expansion has none; an expanding initial state and its dynamics still evolve.
  2. Energy availability and energy content differ. A cosmological energy density is not automatically an extractable work supply.
  3. Conservation requires its proper domain. Local relativistic conservation cannot casually be replaced by a global bookkeeping rule copied from an isolated laboratory system.
  4. Expansion is not biological growth. Their mathematical descriptions and mechanisms differ; a shared trajectory category does not establish one mechanism.

6. Dynamic, static, and principle-guided modes

ModeWhat it can legitimately doWhat it needsExample
Dynamic computationCalculate a trajectory or a distribution of trajectoriesAdequate state, transition law, conditions, parameters, time horizonBoth implemented models
Static / relational modelCheck compatibility, composition, constraints, or equilibrium relationshipsDefined quantities and valid relationshipsFrequency normalization; Friedmann consistency on a time slice
Principle-guided reasoningIdentify likely dependencies, missing evidence, plausible explanations, or needed testsExplicit assumptions and qualitative/domain knowledgeMapping cellular chemistry to reproductive success without a calibrated multi-scale model

Static means the representation does not calculate time evolution. It does not necessarily mean the underlying physical system is motionless. Equilibrium, a maintained steady state, and a single snapshot are also different cases. When evidence is insufficient, retain a typed hypothesis or a qualitative relationship rather than manufacture a transition coefficient.

7. Stress tests of the five-part hypothesis

QuestionResultRefinement
Can the five categories organize these two bounded domains?Yes, with explicit scope and overlapsRetain as a schema hypothesis
Is potential/gradient the necessary source of every change?No at the effective-model levelNarrow the energetic meaning; keep inertia, stochastic sampling, and state dynamics distinct
Are they five independent universal causal coordinates?Not established; some contain or describe othersDo not advertise them as a discovered five-variable law
Does a current snapshot determine a trajectory by itself?No in the tested representationsAdd laws, parameters, boundary/branch conditions, and distributions where needed
Does nesting alone recover causality?NoMaintain explicit cross-links and relation types
Does a common curve or category show a common mechanism?NoRequire a derivation or a specified mechanism
Can formation and decay occur together?Yes in suitable systems and descriptionsKeep multiple observables; avoid one universal progress score
Did these tests reconstruct fitness from molecular physics?NoThat interface is an explicit next modelling problem
Did the five labels add predictive information beyond domain equations?Not demonstratedForecasts came from specified domain kernels; measure the framework's practical contribution separately

A simpler competing model

A generic model can be recorded using state, transition law, admissibility constraints, and an observation model. Energetics may be part of state or dynamics; trajectory is then calculated. That simpler organization can represent both implemented examples too.

Consequently, this investigation provides no evidence that five is the uniquely correct or minimal count. The five-part presentation may still improve understanding and expose omitted energetics or lost cross-links. To establish that advantage, compare models built with and without it on previously withheld problems: missing assumptions detected, correct predictions, clarity, complexity, and revision cost.

8. Reusable minimum schema

RecordRequired fields
DomainIdentity, system boundary, scale, horizon, purpose of the model
EntityCanonical ID, type, parent/composition links
VariableDefinition, units, admissible range, scale; state/parameter/derived/observed role
EvidenceSource, observation time, uncertainty, provenance, which variable it estimates
RelationshipSource and target IDs, type, mechanism, equation, assumptions, lag/range, evidential status
Transition kernelDeterministic or stochastic law, parameters, conditions, valid regime
PredictionHorizon, scenario, trajectory/distribution, observable consequences
EvaluationSubject/objective, value criterion, trade-offs; no assumed universal purpose
Check and revisionConsistency tests, prediction error, missing variables, conditions that falsify the model

A compact deterministic representation is ẋ = f(x; θ, b, t), subject to constraints, with observables y = h(x). A stochastic representation is X₍g+1₎ ∼ K_θ(· | X_g, b_g). These are containers for a domain model, not new laws of nature. A supplied equation, parameter, or boundary condition must not be confused with an empirically established current truth.

9. Where to deepen the nesting next

The most informative next biology step would unpack a single supplied rate: connect a measurable environmental condition, a cellular response, a reproductive rate, and then population change. That would directly test whether deeper nesting improves prediction over treating fitness as a fitted number.

For physics, the corresponding extension would add one controlled level of inhomogeneity to the background expansion and examine how perturbations evolve. That would test whether the map carries relationships between a background model and a nested structure model without confusing them.

Both should eventually face observations with uncertainty and a held-out prediction. This version stops at explicit schematics and internally checked calculations. No claim is made about new fundamental physics, a complete theory of biology, or the metaphysical status of love, wisdom, or mathematical truth.

Sources and scope of attribution

The equation families and biological/cosmological mechanisms are established material. The five-part classification, schema, cross-link typing, chosen synthetic scenarios, numerical results, and framework critique are this investigation's construction. No quoted passages are reproduced.

Appendix: reproducible calculations

The original appendix is a Python script requiring NumPy and SciPy. In this adaptation the same finite-state Wright–Fisher distribution, closed-form neutral moments, Friedmann quadrature, matter-only analytic duration and input validation are implemented once in JavaScript and used by both labs and the check suite.

Every stated reference value is recomputed live on the checks page, with an absolute tolerance of 1e-8 for the scalar regressions and tighter tolerances where an analytic identity justifies one. Random example paths are seeded illustrations and are deliberately not required to match another language's random-number generator.