Drift-Diffusion Decision Model (2D)

This 2D companion to the drift-diffusion decision simulator swaps the racing sample paths of the 3D version for the model's other, equally standard mathematical form: a probability-density field evolved directly by numerically solving the Fokker-Planck (forward-Kolmogorov) equation, with absorbing walls at the two decision boundaries. Instead of counting how many independent racers finish where, this version reads the choice probability and mean decision time straight off the flux of probability mass leaking into each wall over time — the same continuous-population view used to derive the model's analytic predictions in the literature. Tune drift rate, boundary separation, noise and starting bias to watch the density field spread, drift and drain into one boundary faster than the other, and see the flux-based statistics converge on the closed-form gambler's-ruin prediction live.