A resting neuron sits at about −70 millivolts, a little negative inside. Nudge it with a small current and the membrane just sags up and leaks back down — a subthreshold bump, no signal. Push past a tipping point — the threshold — and it runs away into a full spike to about +35 mV and back, like a toilet flushing: once it starts, it goes all the way. That is all-or-nothing.
The spike is always the same height. So the brain can't signal "louder" by making a bigger spike — it signals intensity by firing faster, more pulses per second. But there is a ceiling: right after every spike the neuron is briefly refractory — deaf, resetting — so there is a maximum rate no amount of current can beat. Push way too hard and it stops firing altogether (the depolarization block).
Under the hood is the FitzHugh-Nagumo model: two numbers, a fast voltage v and a slow recovery variable w, the textbook simplification of Hodgkin and Huxley's squid-axon equations. The threshold and the refractory pause aren't rules we wrote in — they fall out of the two equations. It is fully deterministic: the same push gives the same spike, every time. The millivolt and picoamp numbers are illustrative, calibrated so the readouts match the lesson.
Something in the simulation stopped unexpectedly — the lesson continues without it. You can move on; nothing you did was wrong.