❶ The Question

How does raw distinguishability — the bare capacity of one state to differ from another — organize itself into structured, persistent patterns?

❷ The Observation

A cyclone is not an object in the way a rock is an object. No single air molecule "belongs" to it; the molecules flow in and out continuously. Yet the cyclone persists for days. It has a location, a direction, an internal structure. It can be tracked, predicted, and named. What makes it a thing is not the stuff it is made of but the organized pattern of differences that maintains itself through energy flow.

Similar persistent-pattern phenomena appear across many scales: ocean eddies that maintain coherent structure as water flows through them; standing waves in a vibrating medium, where the wave pattern persists even though the medium's particles oscillate back and forth; chemical reaction-diffusion patterns (Turing patterns) that self-organize and maintain spatial structure in an active medium.

In none of these cases is the pattern "made of" permanent components. The pattern is a relational structure — a specific organization of differences — that persists because it is continuously maintained. ECI asks: could something similar apply to information itself?

❸ What We Already Know

Several established scientific frameworks study how persistent organized structures emerge from local interactions:

Dynamical systems and attractors. A dynamical system can settle into stable patterns called attractors — fixed points, limit cycles, or strange attractors — that persist over time and resist small perturbations. The attractor is not a physical object; it is a region of state space toward which trajectories converge. This provides a mathematical framework for "persistent pattern" without requiring any single component to be permanent.

Dissipative structures (Prigogine). Far-from-equilibrium thermodynamic systems can spontaneously develop organized spatial or temporal structures (convection cells, chemical oscillations) that are maintained by continuous energy throughput. These structures are orderly, persistent, and arise from disordered substrates — but they require sustained energy flow to exist.

Autocatalytic networks and the hypercycle. Eigen and Schuster (1977) showed that self-replicating molecular units can form hypercycles — closed loops of catalytic coupling where each molecule catalyzes the replication of the next. These hypercycles are more stable and evolvable than individual replicators. The hypercycle is a specific biochemical model, not a proof that all information structures form persistent loops; ECI uses it as a bridge / analogy — an example of how simple units with local interactions can form higher-order persistent organizations.

Self-organization in complex systems. Pattern formation in neural networks, ant colonies, flocking behavior, and market dynamics all demonstrate that local rules can produce global structures without central coordination. These phenomena are well-studied in complexity science and do not require any metaphysical assumptions about the nature of information.

What these results collectively show: Persistent organized patterns can and do emerge from local interactions in physical, chemical, and biological systems. The "persistence through flow" motif — where the pattern endures even as its components are replaced — is a real and well-documented phenomenon. However, none of these examples require or imply that information itself has an independent ontological existence.

❹ The Framework Interpretation

ECI defines an Information Vector (𝐈) as a structured relational configuration of Information Units with possible asymmetry, recurrence, closure, or coupling potential. Persistence, recurrence and closure are possible properties of an Information Vector, not prerequisites for being one. An Information Vector that fails to persist is still an Information Vector — it is simply one that did not survive Persistence Filtering (see C2).

⚠️ Terminological note. "Vector" here is a provisional working term. It does not mean a vector in the Euclidean or linear-algebraic sense (an element of a vector space with defined addition and scalar multiplication). It was chosen to convey "structured information with organization and possible asymmetry" (directionality being one form of asymmetry), but a more precise term may be adopted as the formalism matures. Readers with physics or mathematics backgrounds should treat this as a label, not a claim about algebraic structure.

How an Information Vector forms. Starting from minimal distinguishable differences (Information Units; see A1), ECI proposes that when multiple units enter into relational configurations — forming feedback loops, maintaining closure, exhibiting recurrence — the result is a higher-order structure that can persist beyond the lifetime of any individual component. The formal expression is:

I₁, I₂, …, Iₙ + relations → 𝐈 (Information Vector)

This is a definition within the ECI framework, not a derived result.

What makes it persist? Not permanence of components, but continuous maintenance of the relational pattern. This requires: (a) energy flow to sustain the pattern against dissipation; (b) some form of error correction, redundancy, or self-repair; (c) compatibility between the pattern and the Channel in which it exists (see B1). If any of these fail, the Vector degrades.

What it is NOT. An Information Vector is not a soul, not a Platonic form, not a disembodied mind. It is a structured information pattern that may or may not have persistence properties depending on its dynamics and environment. The question of whether an Information Vector can persist independently of a specific Carrier is a separate, much more speculative hypothesis (see Soul Hypothesis, E3).

❺ If This Were True...

If organized information naturally forms persistent vector-like structures, several implications follow:

Identity becomes a pattern question, not a substance question. The Ship of Theseus problem — whether an entity remains "the same" after all its components are replaced — would have a natural information-theoretic answer: identity resides in the relational structure, not in the specific material substrate. This connects to long-standing philosophical discussions of personal identity and has practical relevance for questions about neural prosthetics, AI consciousness, and biological cell turnover.

"Life" might be definable in informational terms. Rather than defining life by chemistry (carbon, water, DNA), one could define it by the type of informational organization: self-maintaining, self-replicating, adaptable Information Vectors instantiated in Carriers. This would allow for non-carbon-based life in principle (see Life, D1).

Evolutionary dynamics apply to information patterns, not just organisms. If Information Vectors can persist, compete for Carrier resources, and vary, then selection-like dynamics might operate at the information level even before biological evolution begins (see Persistence Filtering, C2).

These are extrapolations, not predictions. They indicate directions for future work, not established consequences.

❻ How Could We Test It?

The Information Vector concept can be partially tested through simulation and controlled experiments, though full ontological validation is beyond current reach.

Simulation approach. Construct artificial systems (recurrent neural networks, coupled oscillator arrays, agent-based models) with varying levels of internal relational structure. Measure whether systems with more closure, recurrence, and feedback exhibit more persistent organized patterns under perturbation. This tests the dynamical prediction — that relational structure promotes persistence — not the ontological status of Information Vectors.

Biological comparison. Compare the persistence of organized information patterns (gene regulatory networks, neural activity patterns, immune memory) across systems with different degrees of internal feedback and closure. If systems with more Vector-like properties show greater pattern persistence after controlling for size and energy budget, this supports the framework.

Critical methodological requirement: The "Vector-ness" metrics (recurrence, feedback, closure) must be measured independently of and prior to observing persistence. If we first identify which patterns persist and then retroactively label them as "high in relational structure," the reasoning is circular. This is an instance of the persistence tautology described in Falsifiability (F3), Trap 4 — compatibility and survival must be independently operationalized.

What would weaken this claim: If relational structure (closure, recurrence, feedback), measured independently, does not predict pattern persistence better than simpler variables (system size, energy input alone).

What would kill this claim: If persistent organized patterns can be fully explained by component-level properties without any reference to relational structure — i.e., if reductionism works perfectly at every scale with no emergent organization.

❼ Connected Nodes

→ Information Substrate (A1): The foundational layer from which Information Units arise — the raw material for Vectors. → Information Ontology (A3): The philosophical question of whether these patterns have independent ontological status. → ECI Unit: Where Information Vectors meet Carriers and Energy to form operational systems.

❽ Mathematical Detail

Definition: Information Vector

I₁, I₂, …, Iₙ + 𝓡 → 𝐈

where Iₖ are Information Units and 𝓡 denotes the set of relations among them (feedback, closure, recurrence, possible asymmetry — of which directionality is one form).

• Status: Definition (framework notation).
• Assumptions: That Information Units can form stable relational configurations; that these configurations can be described independently of specific physical substrate.
• Variables: n = number of units; 𝓡 = relational structure (currently not formally specified — this is a gap that future work must address).
• Falsifiable consequence: If relational structure 𝓡 has no measurable effect on pattern persistence, the concept of "Vector" as distinct from "collection of units" adds nothing.

Proposed measure (sketch):

The "Vector-ness" of an information structure could potentially be quantified by metrics such as: recurrence rate, feedback loop density, closure (fraction of information flows that return to origin), and attractor dimensionality. These have not been formally unified into a single measure within ECI. Doing so is a priority for the mathematical development of the framework (see Key Equations, F2).