Neural Impulse Simulation (2D)

This 2D companion replaces the original's decorative pulsing dots with a genuine leaky integrate-and-fire spiking network: each of the five layers holds real neurons whose membrane potential leaks exponentially toward rest and jumps on every arriving synaptic impulse, a literal Poisson process drives the input layer at the rate you set, and firing, refractory locking and conduction delay all follow the same equations neuroscience textbooks use for LIF models — so raising the threshold or dropping the synaptic weight visibly changes whether a spike train actually propagates end to end.