The scattering side: the theory's expected response (a calculation from the equations; nothing simulated)

The model (N = 4000 per group): 3872 pulled oscillators, slower half 1936, faster half 1936.
  b, the largest natural rate in the slower half: 0.95021; the smallest in the faster half: 0.95170.
  The faster half's median natural rate: 2.178.

  The faster half moved by cf = 0.6: median 2.178 -> 1.687; largest 19.723 -> 12.214.
  The faster half moved by cf = 0.4: median 2.178 -> 1.441; largest 19.723 -> 8.459.

Uniform narrowing, the matching estimate:
  raised pull: K + eps = 2.0095, 1.00475 times the threshold 2 Delta.
  narrowing every rate by c puts K + eps = 2.0 at 1/c times the threshold: c = 0.995272 (narrowing by 0.4728%).
  For scale: the raised pull's central rhythm grows at 0.00475 per time unit in the large-group theory (e-folding 211 units).

The faster half moved toward the edge by cf (large-group stand-in, 400,000 quantiles; b = 0.95125).
  centre: how crowded the centre is relative to the untouched spread (sets the threshold of the central rhythm);
  J: the central rhythm's response-time factor (untouched about -0.994; a smaller size means a quicker response);
  edge: how crowded the rates are just outside b, relative to the centre;
  needed: the lowest K + eps at which a turning pattern near the edge could grow (the model has K + eps = 2.0).

  cf 1.00: centre 1.0000  J -0.9941  edge 0.525  needed at least 3.81 (the pattern sits within 0.0003 of b, not resolved; bound from the edge crowding)
  cf 0.80: centre 1.0000  J -0.9737  edge 0.656  needed at least 3.05 (the pattern sits within 0.0003 of b, not resolved; bound from the edge crowding)
  cf 0.70: centre 1.0000  J -0.9610  edge 0.750  needed at least 2.67 (the pattern sits within 0.0003 of b, not resolved; bound from the edge crowding)
  cf 0.65: centre 1.0000  J -0.9538  edge 0.808  needed 2.784 (at turning rate 0.9514)
  cf 0.60: centre 1.0000  J -0.9459  edge 0.875  needed 2.377 (at turning rate 0.9523)
  cf 0.55: centre 1.0000  J -0.9371  edge 0.954  needed 2.136 (at turning rate 0.9548)
  cf 0.50: centre 1.0000  J -0.9274  edge 1.050  needed 1.945 (at turning rate 0.9573)
  cf 0.45: centre 1.0000  J -0.9166  edge 1.167  needed 1.764 (at turning rate 0.9615)
  cf 0.40: centre 1.0000  J -0.9043  edge 1.312  needed 1.587 (at turning rate 0.9656)

Growing turning patterns near the edge (large-group stand-in): growth rate per time unit, turning rate, e-folding time
  A and B together (K + eps = 2.0), cf 0.60: none growing
  A and B together (K + eps = 2.0), cf 0.55: none growing
  A and B together (K + eps = 2.0), cf 0.52: none growing
  A and B together (K + eps = 2.0), cf 0.51: growth 0.00061, turning 0.9573, e-folding 1632 units
  A and B together (K + eps = 2.0), cf 0.50: growth 0.00137, turning 0.9579, e-folding 732 units
  A and B together (K + eps = 2.0), cf 0.45: growth 0.00678, turning 0.9604, e-folding 147 units
  A and B together (K + eps = 2.0), cf 0.40: growth 0.01488, turning 0.9612, e-folding 67 units
  A and B opposite (K - eps = 1.8), cf 0.60: none growing
  A and B opposite (K - eps = 1.8), cf 0.55: none growing
  A and B opposite (K - eps = 1.8), cf 0.52: none growing
  A and B opposite (K - eps = 1.8), cf 0.51: none growing
  A and B opposite (K - eps = 1.8), cf 0.50: none growing
  A and B opposite (K - eps = 1.8), cf 0.45: growth 0.00135, turning 0.9618, e-folding 741 units
  A and B opposite (K - eps = 1.8), cf 0.40: growth 0.00856, turning 0.9651, e-folding 117 units

Growing turning patterns near the edge (the model's 4,000): growth rate per time unit, turning rate, e-folding time
  A and B together (K + eps = 2.0), cf 0.60: none growing
  A and B together (K + eps = 2.0), cf 0.55: none growing
  A and B together (K + eps = 2.0), cf 0.52: none growing
  A and B together (K + eps = 2.0), cf 0.51: growth 0.00078, turning 0.9563, e-folding 1284 units
  A and B together (K + eps = 2.0), cf 0.50: growth 0.00141, turning 0.9569, e-folding 708 units
  A and B together (K + eps = 2.0), cf 0.45: growth 0.00685, turning 0.9594, e-folding 146 units
  A and B together (K + eps = 2.0), cf 0.40: growth 0.01496, turning 0.9601, e-folding 67 units
  A and B opposite (K - eps = 1.8), cf 0.60: none growing
  A and B opposite (K - eps = 1.8), cf 0.55: none growing
  A and B opposite (K - eps = 1.8), cf 0.52: none growing
  A and B opposite (K - eps = 1.8), cf 0.51: none growing
  A and B opposite (K - eps = 1.8), cf 0.50: none growing
  A and B opposite (K - eps = 1.8), cf 0.45: growth 0.00141, turning 0.9607, e-folding 711 units
  A and B opposite (K - eps = 1.8), cf 0.40: growth 0.00864, turning 0.9641, e-folding 116 units

For the model's 4,000, growth rates below about 0.001 are near the spacing of its rates at the edge (about 0.0008 after moving) and are not reliable.