# PBRF Layer 2 NBM v0.4 Initial Simulation Plan
## 1. Objective
This document outlines the plan for initial computational experiments using the refined PBRF Layer 2 formalism, the Dynamic Causal Influence Network (DCIN) v0.4, as defined in [[archive/projects/PBRF/0235_PBRF_L2_NBM_Definition_v0.4]]. The primary goals of these simulations are:
1. **Verification:** Re-verify the conservation mechanism (REQ-L2-06) with the modified flow calculation.
2. **Persistence Mechanism Exploration:** Observe the qualitative effects of the *new* persistence feedback mechanism (resistance to flux, controlled by `β`) on dynamics. Compare directly with v0.3 results. Does it promote stability or aggregation differently?
3. **Context Mechanism Exploration:** Observe the qualitative effects of the newly defined context mechanism (average neighbor state influencing potential, controlled by `ε`).
4. **Basic Behavior:** Explore diffusion/equilibration and potential pattern formation tendencies on simple network structures under the new rules.
## 2. Formalism Recap (DCIN v0.4)
* **Network:** Directed graph `G=(V, E)` with static edge weights `w_ji ≥ 0`.
* **States:** Node state `S_i ∈ ℝ`, Node persistence `P_i ∈ ℝ≥0`.
* **Context:** `Context_i(t) = [ Σ_{k∈N_{in}(i)} w_{ki} S_k(t) ] / [ Σ_{k∈N_{in}(i)} w_{ki} ]`.
* **Potential:** `Φ_i(t) = S_i(t) * (1 + ε * Context_i(t))`.
* **Flow:** `Flow_{ji}(t) = [ w_{ji} / (1 + β * P_i(t)) ] * (Φ_j(t) - Φ_i(t))`. (Persistence `P_i` at *receiving* node resists flow).
* **State Update:** `S_i(t+1) = S_i(t) + Δt * [Net Flow]` (inherently conservative).
* **Persistence Update:** `P_i(t+1) = P_i(t) * exp(-γ * |ΔS_i|) + δ * (1 - exp(-γ * |ΔS_i|))`.
## 3. Simulation Setup
* **Environment:** Standard scientific computing environment (e.g., Python with NumPy/SciPy).
* **Time:** Discrete sequence steps `t = 0, 1, 2, ...`. Set `Δt = 1`.
* **Default Parameters:** `γ = 0.1`, `δ = 0.1` (unless varied).
* **Initial Conditions:**
* `S_i(0)`: Specific patterns (e.g., single peak, step function) or random noise around baseline.
* `P_i(0)`: Typically initialized low (e.g., `P_i(0) = δ`).
* `w_ji`: Defined by topology (e.g., uniform weights).
## 4. Proposed Experiments
**Experiment 1: Conservation Verification (v0.4)**
* **Objective:** Numerically verify that `Σ_i S_i(t)` remains constant with the v0.4 flow rule.
* **Network:** Small, closed networks (e.g., 3-node cycle, 5-node random graph).
* **Parameters:** Arbitrary initial `S_i(0)`, `P_i(0)`, `w_ji`. Set `β > 0` and `ε ≠ 0` to ensure all mechanisms are active.
* **Procedure:** Run simulation for T steps (e.g., T=100). Calculate `TotalS(t) = Σ_i S_i(t)` at each step.
* **Expected Outcome:** `TotalS(t)` should remain constant within machine precision.
* **Analysis:** Plot `TotalS(t)` vs `t`. Check for drift.
**Experiment 2: Diffusion Baseline (v0.4: `β=0, ε=0`)**
* **Objective:** Establish baseline diffusion behavior without persistence or context effects.
* **Network:** 1D chain (N=51 nodes), reflecting boundaries. `w=1` for adjacent nodes.
* **Parameters:** Set `β = 0`, `ε = 0`. Initial condition: Single peak `S_25(0) = 1.0`, others 0. `P_i(0) = δ`.
* **Procedure:** Run simulation until near equilibrium (e.g., T=500).
* **Expected Outcome:** Standard diffusion towards uniform `S_i`. `P_i` evolves based on local `ΔS_i`. (Should replicate Exp 2 from v0.3).
* **Analysis:** Plot `S_i(t)` and `P_i(t)` profiles.
**Experiment 3: Effect of Persistence Resistance (`β > 0, ε=0`)**
* **Objective:** Observe how persistence-as-resistance affects dynamics. Compare to v0.3 (potential enhancement).
* **Network:** Same as Exp 2 (1D chain).
* **Parameters:** Set `ε = 0`. Vary `β > 0` (e.g., 0.1, 1.0, 10.0). Initial condition: Single peak `S_25(0) = 1.0`, others 0. `P_i(0) = δ`.
* **Procedure:** Run simulation (e.g., T=500 or longer if dynamics slow).
* **Expected Outcome (Hypothesis):** High `P_i` at a node `i` will reduce flow *into* and *out of* it. This should *slow down* the diffusion/equilibration process compared to `β=0`. Will this resistance allow the initial peak to persist longer? Could it lead to sharper interfaces or prevent low-`S` regions from being invaded? Will it promote aggregation by making existing high-`S` regions resistant to dissipation?
* **Analysis:** Compare `S_i(t)` and `P_i(t)` profiles for different `β` values and against the `β=0` baseline (Exp 2) and the v0.3 results (Exp 3 from [[0234]]). Measure equilibration time or peak decay rate. Look for signs of pattern stabilization or aggregation.
**Experiment 4: Effect of Context Modulation (`ε ≠ 0`)**
* **Objective:** Observe the effect of context (average neighbor state) influencing potential.
* **Network:** 1D chain (N=51) or small graph with varying connectivity.
* **Parameters:** Set `β` to 0 or a fixed positive value. Vary `ε` (e.g., -1.0, -0.1, 0.1, 1.0). Initial condition: Step function (e.g., `S_i=1` for `i<25`, `S_i=0` for `i>=25`) or random noise.
* **Procedure:** Run simulation.
* **Expected Outcome (Hypothesis):**
* Positive `ε`: Nodes with high-`S` neighbors will have higher potential `Φ`, potentially increasing outflow from those neighbors towards the node (if `S_i` is lower) or increasing outflow from the node itself (if `S_i` is higher). Might enhance contrast or accelerate dynamics near interfaces.
* Negative `ε`: Nodes with high-`S` neighbors will have lower potential `Φ`, potentially inhibiting flow from those neighbors or reducing outflow from the node. Might dampen dynamics or smooth interfaces.
* **Analysis:** Compare `S_i(t)`, `P_i(t)`, `Φ_i(t)` profiles for different `ε` values. Analyze flow patterns near interfaces or regions with varying context.
## 5. Outputs and Analysis
* Time series data for `S_i(t)`, `P_i(t)`, `Φ_i(t)`.
* Calculation of total `Σ S_i(t)` (for Exp 1).
* Plots of spatial profiles `S_i`, `P_i`, `Φ_i` at different times `t`.
* Comparison plots for different parameter values (`β`, `ε`) vs. baselines (v0.4 `β=0`, v0.3 results).
* Quantitative metrics (e.g., equilibration time, peak height decay rate, interface width) if applicable.
## 6. Scope and Limitations
* Focus on verifying core mechanics and exploring qualitative effects of new persistence/context rules on simple topologies.
* Static topology (`w_ji` fixed).
* Limited parameter exploration.
* Does not aim to reproduce specific physical phenomena yet.
Results will guide further refinement of the DCIN formalism, particularly the effectiveness of the persistence-resistance mechanism for pattern stabilization/aggregation and the role of the context term.
**Next Step:** Implement the DCIN v0.4 model and execute these experiments. Create node `0237_PBRF_L2_NBM_v0.4_Initial_Sim_Results`.