Information-Driven Phase Transition on Weighted Graphs with Spontaneous Dimensional Sensitivity

This paper introduces a weighted graph model (FIU) where information-driven topology evolution governed by spectral curvature exhibits a sharp phase transition at a critical coupling strength, revealing a stable discrete Poisson relation between curvature and information flux that spontaneously demonstrates dimensional sensitivity through distinct system-size collapse thresholds in 2D versus 3D lattices.

The Gist

Imagine you have a giant, invisible web connecting millions of people. At first, everyone is only connected to their immediate neighbors (like a grid of city blocks). But this web is alive: it changes, grows, and reshapes itself based on how much "energy" or "information" flows through it.

This paper describes a computer experiment where the author, V. Dolci, built a digital universe to see what happens when you let this web evolve on its own. The goal? To see if geometry (the shape of the web) and information (the flow of data) can create something that looks like the laws of physics, specifically gravity.

Here is the story of what they found, explained simply:

1. The Setup: A Web That "Feels" Its Shape

The researchers created a network where every node (person) has a "curvature score." Think of this score like a temperature gauge.

• If a node is in a crowded, chaotic area, it has high "curvature" (high heat).
• If it's in a calm, flat area, it has low curvature.

The rules of the game are simple:

• Energy flows between neighbors. If the "temperature" difference is huge, the connection gets weaker (it dissipates).
• New connections are made preferentially between the "hottest" (highest curvature) nodes. It's like people in a crowded room instinctively reaching out to other busy people to form a new club.

2. The Big Surprise: A "Light Switch" Moment

The researchers tweaked a single knob called the coupling parameter ($g$). This knob controls how aggressively the network tries to connect the "hot" nodes.

• Below the critical point ($g < 0.023$): The system is messy. The flow of information and the formation of new links are fighting each other. It's like a chaotic dance where everyone is stepping on each other's toes.
• Above the critical point ($g > 0.023$): Suddenly, the system snaps into order. The information flow and the new links start working in perfect harmony. It's as if the chaotic dance suddenly turned into a synchronized ballet.

This is a Phase Transition. It's like water suddenly turning into ice. The paper found that this happens at a very precise moment, and the system behaves exactly like a physical object undergoing a phase change.

3. The "Gravity" Connection: A Poisson Equation

Once the system is in this ordered state, something magical happens. The researchers discovered a mathematical rule that looks suspiciously like Newton's Law of Gravity.

In physics, gravity says: Mass tells space how to curve.
In this computer model, they found: Information flow tells the network how to curve.

They found a formula: Curvature Change = Constant × Information Flow.

• Curvature Change: How much the shape of the web is bending.
• Information Flow: How much data is moving.
• The Constant: A number that appeared out of nowhere, acting like a "gravitational constant" for this digital universe.

The Catch: The researchers realized this isn't a direct cause-and-effect (like "flow causes curvature"). Instead, both the flow and the curvature are being driven by a third, hidden force: the Curvature Field itself. It's like a conductor leading an orchestra; the violins (flow) and the brass (curvature) aren't playing each other, they are both following the conductor.

4. The Dimensional Mystery: Why 2D and 3D are Different

This is the most fascinating part. The researchers ran the simulation in two dimensions (a flat sheet) and three dimensions (a cube).

• In 2D (Flat): The system works beautifully for small networks. But as the network gets bigger (more than ~900 nodes), the magic disappears. The "curvature" smooths out, the flow becomes uniform, and the special connection between flow and shape vanishes. The system becomes "boring" and flat.
• In 3D (Cube): The system holds on to the magic much longer! It stays ordered up to ~3,300 nodes before it finally collapses.

Why?
Imagine trying to smooth out a crumpled piece of paper (2D) versus a crumpled ball of yarn (3D).

• The paper is easy to flatten out completely. Once it's flat, there's no more "bumpiness" (curvature) to drive the system.
• The ball of yarn is much harder to flatten. It has more "topological frustration"—it keeps getting stuck in knots and bumps even when it's huge. This keeps the system active and ordered for much longer.

5. The Big Picture: "Topological Frustration"

The authors call this phenomenon Topological Frustration.

• Small systems are too small to smooth themselves out, so they stay "frustrated" (bumpy and active). This is where the "gravity-like" behavior lives.
• Huge systems eventually smooth themselves out into a perfect, flat plane. The "gravity" disappears because there are no more bumps to create it.

The Takeaway

This paper suggests that gravity might not be a fundamental force written into the universe from the start. Instead, it might be an emergent phenomenon—a side effect of how information flows through a network, similar to how temperature emerges from the movement of atoms.

The model shows that if you have a network that reacts to information flow, it naturally organizes itself into a shape that looks like spacetime. However, this "gravity" is a mesoscopic (middle-sized) phenomenon. It works great in the "Goldilocks zone" of size, but if the universe gets too big, the effect might fade away, just like the magic in their 2D simulation.

In short: The universe might be a giant, self-organizing web where the shape of space is determined by the flow of information, but this relationship is fragile and depends heavily on the "dimension" of the web.

Technical Summary

1. Problem Statement and Motivation

The paper investigates the dynamics of evolving networks where both node states and topological connections change simultaneously. The primary motivation is to explore the emergent gravity hypothesis, which posits that spacetime geometry and gravitational dynamics may arise from an underlying information-processing substrate.

The authors ask: What structural and dynamical phenomena emerge when a graph is driven by curvature-sensitive information flow? Specifically, they aim to determine if a discrete graph model driven by spectral curvature can exhibit phase transitions, self-organization, and analogies to gravitational physics (such as the Einstein equations) without explicitly imposing geometric rules.

2. Methodology: The Unified Informational Framework (FIU)

The authors propose a model called the Unified Informational Framework (FIU), defined by three minimal rules on a weighted undirected graph $G=(V, E, w)$:

1. Spectral Curvature Measure ($R$):

   • A local curvature $R(i)$ is defined for each node based on the diagonal of the graph Green function (scalar propagator).
   • $R(i) = \Gamma (1 - G_{ref}/G(i,i))$, where $G(i,i)$ is the diagonal element of the Green function derived from the spectral decomposition of the graph Laplacian, and $G_{ref}$ is a flat-space reference.
   • $R$ acts as a measure of local spectral inhomogeneity, not a differential geometric Ricci curvature.

2. Energy Propagation and Dissipation:

   • Nodes carry energy $E_i$. Energy flows between neighbors, and edge weights decay based on the average energy of the endpoints and the curvature contrast $|R(i) - R(j)|$.
   • High curvature contrast leads to stronger dissipation, creating a feedback loop between information flux and graph geometry.

3. Link Formation (g-links):

   • New long-range links are created preferentially between nodes with high curvature $R$.
   • The rate of creation is controlled by a coupling parameter $g$. The new link weight is proportional to $g \cdot R_{source}$.

Key Parameters:

• $\alpha$: Energy decay rate.
• $B$: Energy budget per node.
• $g$: Coupling parameter for long-range link formation.
• $\tau$: Temporal delay in measuring information flux to avoid circularity.

3. Key Contributions and Experimental Design

The study employs a rigorous statistical approach to avoid circular reasoning (where flux defines links and links define flux). Five specific tests were designed:

• Test C (Primary): Correlates delayed information flux ($\sigma_{i}(t-\tau)$) with the number of long-range links.
• Test A: Counts topological events (link creation/deletion) to check for "freezing" in high-curvature regions.
• Test B (Null Test): Uses an external scalar field independent of curvature to ensure correlations are not generic artifacts.
• Test D (Discrete Poisson Relation): Tests for a linear relationship $\nabla^2 R(i) = \kappa \sigma_{prev}(i)$, analogous to the Poisson equation in gravity.
• Test E (Mediator Analysis): Determines if the correlation between flux and curvature Laplacian is direct or mediated entirely by the curvature field $R$ itself.

4. Key Results

A. Sharp Phase Transition

• Critical Point: The system undergoes a sharp continuous (second-order) phase transition at a critical coupling $g_c \approx 0.023$.
• Ordered vs. Disordered:
  • Below $g_c$: Information flux and topology are anti-correlated ($\text{gravC} \approx -0.7$). High flux leads to link erosion.
  • Above $g_c$: Strong positive correlation ($\text{gravC} \approx 0.75$, $p < 10^{-38}$). High flux nodes attract new links, forming an ordered structure.
• Critical Exponent: The onset follows a power law with a mean-field exponent $\nu \approx 0.54$.
• Fluctuations: Variance peaks near $g_c$, indicating critical fluctuations.

B. Emergent Discrete Poisson Relation

• In the ordered phase, a node-level linear regression reveals a stable relationship:
  $$\nabla^2 R(i) = \kappa \cdot \sigma_{prev}(i)$$
• Emergent Coupling ($\kappa$): The slope $\kappa$ is remarkably stable across different parameter configurations ($\kappa \approx 67.85 \pm 1.74$ in 2D) and independent of $g$ within the ordered plateau.
• Analogy: This mirrors the Newtonian limit of the Einstein equation ($\nabla^2 \phi = 4\pi G \rho$), where $R$ is the potential, $\sigma$ is the source density, and $\kappa$ acts as an emergent gravitational constant.

C. Spontaneous Dimensional Sensitivity

• 2D vs. 3D: The ordered phase and the Poisson relation exist only in a mesoscopic window.
  • 2D Lattices: The signal collapses (decays to zero) at $N \approx 576$–$900$.
  • 3D Lattices: The signal persists to larger sizes ($N \approx 1728$) and collapses at $N \approx 3375$.
• Ratio: The critical size ratio $N_c^{3D} / N_c^{2D} \approx 5.9$ cannot be explained by coordination numbers alone, suggesting a deeper geometric/topological cause.
• Mechanism: The collapse is driven by curvature homogenization. At large $N$, the graph "regularizes" into a flat geometry where $\nabla^2 R \to 0$, while information flux fluctuations remain finite, breaking the correlation.

D. Mediator Analysis (Test E)

• The correlation between $\nabla^2 R$ and $\sigma$ is almost entirely mediated by the curvature field $R$.
• When the component of $\sigma$ orthogonal to $R$ is isolated, the correlation vanishes. This implies $R$ is the central self-organizing variable driving both energy dynamics and topology, rather than a direct causal link from flux to curvature.

5. Significance and Interpretation

• Topological Frustration: The ordering signal is interpreted as a topological frustration phenomenon. In the mesoscopic regime, the graph lacks the resources to smooth out curvature everywhere, forcing fluctuations that correlate with information flow. In the thermodynamic limit, the graph can fully regularize, eliminating the signal.
• Gravity Analogies:
  • The 2D collapse is structurally analogous to 2+1 gravity, which is topological and lacks local propagating degrees of freedom (no gravitons).
  • The persistence in 3D is analogous to 3+1 gravity, which supports local dynamics and propagating modes.
• Emergent Constants: The stability of $\kappa$ suggests that "gravitational" constants can emerge from purely informational and topological rules in a finite-size regime, though they vanish in the infinite limit for this specific model.

6. Limitations

• Mediation: The Poisson relation is not a direct causal link from matter (flux) to geometry (curvature) as in Jacobson's thermodynamic derivation of gravity; it is a self-consistent organization driven by $R$.
• Finite-Size Nature: The emergent coupling $\kappa$ vanishes in the thermodynamic limit ($N \to \infty$) for the current rules, meaning the model describes mesoscopic ordering rather than macroscopic gravity.
• Lack of Analytical Proof: The existence of the Poisson relation and the specific value of $\kappa$ are empirical observations without a current analytical derivation.

Conclusion

The paper demonstrates that a simple set of local rules governing information flow and spectral curvature on a graph can spontaneously generate a phase transition, a discrete Poisson equation, and a dimensional dependence that mimics the qualitative differences between 2+1 and 3+1 gravity. The findings suggest that geometric order and "gravitational" behavior can emerge from information-theoretic constraints in the mesoscopic regime, driven by topological frustration.