Skip to content

PDSL Model Card (Public)

1. Model Name

Public Diplomacy Simulation Lab (PDSL)
Public documentation release

2. Version / Release Anchor

Track Version Status
Model semantics v1.1 Frozen — all formulas, regime taxonomy, fatigue rule, and compute() logic are stable and citable
Engine architecture v0.2.0 Modularized internal layout — no change to outputs or analytical behavior

For citation purposes, reference v1.1 semantics. Architecture versioning is internal and non-semantic.

3. Summary

The Public Diplomacy Simulation Lab (PDSL) is a deterministic, theory-constrained simulation framework designed for analytical demonstration and controlled scenario reasoning in public diplomacy and digitally mediated influence contexts.

PDSL is governance-first by design. It emphasizes interpretive discipline, transparency of assumptions, and explicit non-claims, rather than prediction, optimization, or empirical inference.

4. What PDSL Is

  • A deterministic simulation framework that maps structured analytical inputs to consistent, interpretable outputs
  • A controlled environment for testing regime-defined narrative dynamics and sensitivity behavior
  • A reviewable demonstration artifact intended for academic, policy, and educational audiences

5. What PDSL Is Not

PDSL is explicitly not: - Empirical causal inference - A predictive model of real-world publics - A substitute for validated surveys, experiments, or field studies - A machine-learning or black-box system

Outputs should not be interpreted as measurements of population attitudes or behavioral effects.

6. Architecture Overview (v0.2.0)

PDSL v0.2.0 adopts a modular internal architecture organized into discrete library components. This is an internal structural change only — it does not alter model semantics, output behavior, or the governance posture of the framework.

The public API surface remains unchanged: - POST /api/simulate — single-scenario simulation - POST /api/scenario — multi-phase path simulation - POST /api/sensitivity — ASPE slope analysis - POST /api/validate — hard gate: verifies semantic integrity against canonical test vector

The /api/validate hard gate enforces that v1.1 semantics are intact on every deployment. Canonical verification inputs and expected outputs are publicly committed at: docs/test-vectors/

7. Methods (Conceptual Overview)

PDSL implements a deterministic, theory-constrained simulation design for controlled scenario reasoning. Rather than estimating causal effects or forecasting outcomes, it operationalizes a regime-based interpretive framework that translates structured analytical inputs into three analytically distinct outcome channels:

  • Algorithmic Persuasion Risk (APR)
  • Identity Alignment Shift (IAS)
  • Policy Support via Threat Framing (PST)

The framework is designed so that interpretation is governed by regimes, not by raw numerical values or analyst discretion.

8. Conceptual Inputs (High-Level)

Public documentation describes inputs abstractly. These include representations of: - Exposure or distribution pressure - Framing and narrative intensity (including saturation effects) - Identity salience - AI / media literacy context - Time or iteration horizon

This release does not disclose parameter thresholds, equations, or engine-specific mechanics.

9. Outputs (Governed Outcome Channels)

APR — Algorithmic Persuasion Risk

APR captures short-run susceptibility to attitudinal change under modeled exposure and framing pressure. APR functions as a diagnostic risk signal, not as a proxy for policy support or approval.

IAS — Identity Alignment Shift

IAS captures slower, structural movement in identity alignment driven by exposure, identity salience, and time. IAS acts as a long-horizon constraint, not as a trigger for short-run interpretation.

PST — Policy Support via Threat Framing

PST represents policy support generated through threat framing, subject to explicit diminishing returns under saturation. PST is interpreted only through the regime-based framework, never as a standalone score.

10. Regime-Based Interpretation

PDSL does not interpret outputs independently. Instead, outputs are read through named analytical regimes that:

  • Constrain appropriate policy interpretation
  • Enable modular and comparable scenario reasoning
  • Prevent over-reading of marginal score changes

The same output value may imply escalation, plateau, or diminishing returns depending on regime context.

11. Interpretability Guardrails

  • APR, IAS, and PST are analytical constructs, not empirical measurements
  • Outputs are comparative signals, not predictions
  • Interpretation requires