The analysis framework provides the tools for systematic exploration of the model’s parameter space. It uses a fast vectorised simulation (bypassing per-step ring attractor dynamics) to enable large-scale sweeps: 200 bugs × 100+ parameter points per sweep.

Key Analysis Products

Contrast (singlet yield anisotropy) vs. compass noise → phase boundary separating navigating from lost regimes. The critical contrast threshold (~0.02) determines which radical-pair models support navigation.

Robustness Budget

Suppression mechanisms that reduce compass contrast:

  • Spin relaxation (T₁, T₂ in the radical pair)
  • Rate asymmetry (unequal singlet/triplet recombination)
  • Orientational disorder (molecules not perfectly aligned)

Each mechanism has a safety margin — the factor by which it can increase before navigation fails.

Anomaly Sweeps

Dipole and fault anomalies of increasing strength, testing whether same-frame bias cancellation keeps the bug on course. Result: robust to ~500 nT anomalies.

Path Integration Phase Diagram

Homing error as a function of exploration duration and memory leak parameter. Reveals the optimal exploration horizon for each noise level.

Model Discrimination

At low contrast (C ~ 0.1), different radical-pair models produce statistically distinguishable navigation signatures — suggesting that behavioural experiments could discriminate between molecular mechanisms.

Figures

The analysis produces 36+ diagnostic figures covering all aspects of the model. These are archived as PNGs in the experiment/ directory.

Source: modules/mayajiva/experiment/analysis.py (2,359 lines, ~94 KB), experiment/sim.py (434 lines), experiment/paper.org (results sections)