Topology-governed attractor dynamics.

The same dynamical object admits four readings: as a Sanskrit cognitive ontology, as a mathematical structure, as a Legitimation Code Theory (LCT) construct, and as an engineering artefact. The mapping is not metaphorical decoration — it is operational. A “samskara” is not like a learned weight; under this framing, it is one, with the same equations of motion.

Why this table comes first
Mathematicians dismiss the Sanskrit. Theologians dismiss the math. LCT scholars wonder why their framework is being used for individual cognition rather than institutional knowledge practices. The Rosetta Stone is the survival condition of the bridge.

Drag the sliders. Watch the trajectory move. The dynamics below are the same equation as the full system — dq/dt = -q + tanh(β·W·q) + α·F — projected to two dimensions with two compositional patterns (Rajas, Tamas) and the origin (Sattva) emerging as the quiescent attractor.

Lower β with both forces near zero and the trajectory settles toward Sattva. Push α_R high and the system flows into Rajas. Hold both forces simultaneously and the landscape’s topology — not energy alone — decides where you land. Save state vectors as you experiment. Comparing them turns a toy into a diagnostic.

This work was not a single clean “paper experiment”. It was a tight loop of building a landscape, watching it misbehave, killing hypotheses, and turning every surviving claim into a pre-registered test. The milestone codes (M1, M10, M24…) are the actual internal sequence of that search.

The story arc is simple: compositions create structure → structure creates plateaus → plateaus force the question of escape → every local explanation fails → topology remains.

Plateau states are metastable slow regions: trajectories can dwell for 300+ steps before converging. The question is whether a given plateau can escape to a different basin under unforced dynamics (α = 0).

Proposition (tested)
deg⁺(v) > 0 ⇔ escape(v)

Across 5 seeds × 80 plateaus (400 total), correspondence is 399/400 (99.75%). Seed 42 achieves 80/80 (100%); one false negative appears across the multi-seed sweep, interpreted as a protocol sensitivity limit rather than a “governance failure”.

Alternative predictors are systematically tested: residual velocity has moderate AUC (~0.67), energy difference yields ROC ≈ 0.75 (partial), eigenvalue analysis finds no unstable modes, and injected noise does not change escape rates across tested σ values. The graph out-degree criterion dominates feasibility prediction in the tested regime.

Five plausible local explanations were tested against the topology criterion and rejected. This is the “negative result” archive — the predictors that shouldhave worked under standard dynamical-systems intuition, and didn’t. M20–M23, run before the M24 star result was accepted.

The paper distinguishes two non-overlapping roles. Topology answers whether a transition can exist (feasibility). Energy and alignment influence how hard it is (the forcing threshold).

Given that feasibility is graph-structural, transition cost is modeled as an α threshold. A simple linear form with three features predicts thresholds with LOO-CV R² = 0.676 (n = 262).

Nonlinear models and extra feature families do not exceed the linear baseline, suggesting the ceiling is feature-level (about 32% variance unexplained) rather than model-level.

The correspondence holds across multiple compositional domains (including non-philosophical structured constructions) and across dimensions. A key boundary: PCA-derived patterns from real data behave differently and show reduced correspondence (10–40%).

The framework is packaged into an end-to-end pipeline that takes a CSV and produces a five-dimensional root cause analysis (WHAT/WHY/HOW/WHEN/WHERE). It auto-detects types, selects an encoding strategy, calibrates ODE parameters, maps an atlas, and reports validations and failure modes.

This page is a narrative version; the project landing page for the full system is io-gita.

The pipeline was applied to eight independent public datasets across medicine, population health, manufacturing, finance, education, HR, and chronic disease — with zero domain-specific configuration. Pass rate: 7 of 8 at 5/5; one failure (chronic disease, AUC 0.33) reported, not hidden. The failure mode is data (128:1 class imbalance), not pipeline.

Honest framing
Topology adds +0.0002 AUC over a strong XGBoost baseline on SUPPORT2. The value is interpretability and routing, not accuracy lift. WHO replication delivered 24/40 (60%) clustering, not the unverified 23/23 (100%) originally claimed — published as the lower number.

Burnout is not stress depletion. It is topological capture. The state vector has converged to a fixed-point attractor in a high-semantic-gravity basin where the self fuses with the work role. At accessible forcing magnitudes α, out-degree(work-self) is zero. No directed edge leads to another basin. The trajectory cannot exit because the graph offers no path, whatever the available energy.

Feasibility and cost are separable. Sleep, rest, and reduced workload lower the energetic barrier, but they do not add an escape edge to the transition graph. They change the Lyapunov value, not the adjacency structure. If out-degree remains zero at the maximum affordable α, the system is clinically stable but dynamically imprisoned. Recovery requires topological intervention: a new edge must be constructed, not merely a cheaper toll paid.

Deterministic seeds (primary: 42) and a reproducibility envelope (Python 3.10+, NumPy/SciPy versions) with stable hashing for determinism. Euler is the reported baseline; RK2 is available as an option. The paper PDF is open at the link below; the source repo and validation harness are archived under the Zenodo DOI in Cite & Reproduce. For early code access pre-acceptance, email info@adaptive-mind.com.

All artefacts are public. License: CC-BY-4.0. Replication needs only the seed, dimension, and pattern set declared in the paper.