Code and data for "Plasticity Scars and Acoustic Diagnostics in Adaptive Kuramoto Networks"

Reproduction code and data accompanying the paper "Plasticity Scars and Acoustic Diagnostics in Adaptive Kuramoto Networks: Bifurcation, Spectral and Amplitude Signature, and Subpopulation Amplification" (J. Pearcey).

This deposit regenerates every figure and table in the paper from the model equations. A scarred subpopulation is an isolated Kuramoto cluster whose internal coupling was raised by a prior plasticity-saturating event and then decays passively. The central result is that the real part of the order parameter carries a 1/f² (brown-noise) spectrum in every regime — the locked and incoherent states are separated by order-parameter amplitude and corner frequency, not by spectral exponent — and that this supports a robust amplitude-based detection of plastic history over a short observation window.

Contents:

• akm_core.py — stochastic Ott–Antonsen integrator for Eq. (8) (Euler–Maruyama), Welch spectral-exponent estimation, short-window integrator, and detection statistics
• akm_shaping.py — spectral-shaping construction for the idealised reference forms (white/pink/brown)
• akm_full_model.py — full microscopic network, Eqs. (1)–(3), with adaptive coupling and saturated plasticity
• make_fig1.py … make_figS5.py, make_tables.py — one generator per figure and table
• run_all.py — regenerates all outputs in a single command
• figures/, data/ — pre-generated figures, the barrier and exponent tables, and the underlying per-coupling spectral data (CSV)

Requirements: Python 3.9+ with numpy, scipy, and matplotlib (pip install -r requirements.txt). Run python run_all.py to regenerate everything (a few minutes). All randomness uses fixed seeds, so results are reproducible; the spectral results reproduce the paper's reported values, and the README documents the reconstructed detection protocol and its effect sizes in full.