Bottleneck Fragility Profile¶
Definition & Conceptual Goal¶
The Bottleneck Fragility Profile is a Micro-Topological Composite Index that evaluates whether a theory graph is overly reliant on a few centralized dogmas or hubs, and quantifies the structural damage if those hubs are falsified.
It unifies three related properties to form a complete fragility diagnostic:
- Structural Homogeneity (The Presence of Hubs): Do extremely central bottleneck nodes exist?
- \(k\)-connectivity (The Collapse Threshold): What is the absolute minimum number of nodes that must be falsified to shatter the theory?
- Structural Refutability (The Damage Impact): If the top-\(k\) most central bottlenecks are removed, how much of the theory's explanatory power (shortest paths) is destroyed?
Theoretical Grounding & Model Formulation¶
A robust scientific theory should distribute its explanatory burden. According to Schurz (2024), a theory-holon should possess structural homogeneity, avoiding pathological skewness.
When Homogeneity is very low, the theory structure resembles a "star network" heavily reliant on a single central hub. In such cases, the theory becomes highly structurally refutable ([Novacek, 2015]), as the falsification of that single hub causes a massive drop in logical claim paths, splitting the theory into heterogeneous, disconnected parts.
By unifying these metrics, the Bottleneck Fragility Profile transitions from merely identifying a hub (Homogeneity) to proving its vulnerability (\(k\)-connectivity) and quantifying the actual consequence of its failure (Refutability).
Mathematical Specification¶
Let \(G = (V, E)\) be the theory graph. The Fragility Profile is a composite report combining:
- Homogeneity (\(H_{\text{node}}\)): The normalized Shannon entropy of the node degree centralities. Low entropy indicates severe bottlenecks.
- Absolute Threshold (\(\kappa\)): The vertex connectivity. Defines exactly how many core concepts must fall to disconnect the graph.
- Impact Score (\(R_k\)): The Top-\(k\) Refutability score.
$$ R_k (G) = \frac{|\Pi (G)|}{|\Pi (G)| + \sum_{i=1}^k |\Pi (G \setminus {v_i})|} $$ (Where \(\Pi (G)\) is the number of shortest paths and \(v_i\) are the nodes with the highest Betweenness Centrality).
Diagnostic & Metascientific Value¶
| Profile Diagnosis | \(H_{\text{deg}}\) (Homogeneity) | \(\kappa\) (Threshold) | \(R_k\) (Impact) | Interpretation |
|---|---|---|---|---|
| Highly Resilient (Quinean Web) | High | High | Low | The theory is structurally redundant. No single node acts as a critical point of failure. |
| Fragile Bottleneck (Dogmatic) | Low | Low (often \(\kappa=1\)) | High | The theory relies entirely on a central dogma. Falsifying it shatters the framework. |
| Modular but Resilient | Moderate | Moderate | Moderate | The theory has clear structural centers, but with enough crossover pathways to survive isolated falsifications. |
Sub-Metric Drill Down¶
To understand the mathematical components driving the Fragility Profile, inspect the standalone metrics:
- Structural Homogeneity: For node degree entropy and Gini coefficient calculations.
- Structural Refutability: For shortest-path drop calculations based on Betweenness Centrality.
- Connectedness & Cohesion: For the foundational \(k\)-connectivity bounds.
Grounding References¶
- [Schurz, 2024] Schurz, G. (2024). Philosophy of Science: A Unified Approach. Routledge, sec. 5.1.
- [Novacek, 2015] Novacek, V. (2015). Formalising Hypothesis Virtues in Knowledge Graphs.