How Neurons Fire: The Action Potential Explained

In the summer of 1949, on a floating laboratory pontoon at the Plymouth Marine Biological Laboratory in Devon, two physiologists hunched over a chilled chamber holding a freshly dissected piece of squid. Alan Hodgkin and Andrew Huxley had a home-built voltage clamp warming up beside them and a single nerve fiber from a squid's stellate ganglion suspended in cold seawater. It was an unglamorous setup, salt spray and improvised electronics, and yet over the next three summers it would yield the first quantitative theory of how a nerve impulse actually works. The five papers they published in 1952 in the Journal of Physiology earned them the Nobel Prize in 1963, shared with John Eccles.

What they were chasing was something every animal with a nervous system depends on, second by second, and almost no one had ever measured directly. When you flinch from a hot stove, when your heart beats, when you read this sentence, the underlying event is the same: a brief electrical pulse racing down the length of a nerve cell. The question Hodgkin and Huxley answered is deceptively simple to ask and was fiendishly hard to solve. How does a living cell, made of fatty membrane and salty water, generate and propagate an electrical signal without losing it along the way?

A spike that refuses to fade

The pulse they were studying has a name: the action potential. It is a brief, stereotyped, all-or-nothing voltage spike that travels down a neuron's axon without weakening, no matter how far it has to go. That last property is the remarkable one. If you send an ordinary electrical signal down a wet, leaky cable like a nerve fiber, it fades within millimeters. The action potential does not fade. It regenerates itself at every step, arriving at the far end of a meter-long axon just as strong as it began.

A resting neuron sits at roughly minus 70 millivolts, meaning the inside of the cell is negatively charged relative to the outside. When the membrane is nudged up to a threshold near minus 55 millivolts, something dramatic happens: the voltage shoots up to a peak around plus 30 millivolts and then crashes back down, the whole event lasting only one to two milliseconds. That is the spike. It is the basic currency of the brain, and every thought, sensation, and movement you have ever had is encoded in patterns of these millisecond pulses.

The word stereotyped matters. Every action potential in a given axon looks essentially identical. The neuron does not signal a stronger stimulus with a bigger spike; it signals it with more spikes, fired more often. Information in the nervous system is carried not in the size of each pulse but in their timing and frequency.

Why a squid solved the problem

To measure what happens during a spike, you need to get an electrode inside the cell, and in the 1930s the nerve fibers physiologists knew about were impossibly thin. The breakthrough was zoological rather than technical. The squid, it turns out, has a small number of giant axons that control the rapid jet-propulsion escape it uses to flee predators. In Loligo forbesii, the species Hodgkin and Huxley worked with, these axons reach about 500 micrometers in diameter, roughly half a millimeter, hundreds of times wider than a mammalian nerve fiber.

That width changed everything. An axon that size is large enough to thread a fine wire electrode straight down its core, allowing the experimenters to measure and even control the voltage across the membrane from the inside. Hodgkin and Huxley recorded from the squid giant axon at Plymouth from the late 1930s through the early 1950s, with the work interrupted in the middle by the Second World War, during which both men served on radar research. When they returned to the bench, they brought with them the voltage clamp, an electronic feedback circuit that holds the membrane at a chosen voltage and reports exactly how much current flows in response. By stepping the voltage to different values and watching the currents, they could pull apart the separate ionic contributions hiding inside the single spike.

The four phases of a single spike

What they found was that the action potential is not one event but a tightly choreographed sequence of four phases, each driven by ions moving through gated pores in the membrane called voltage-gated ion channels. These channels open and close in response to voltage itself, which is the key to the whole self-regenerating trick.

The first phase is depolarization. When the membrane reaches threshold, voltage-gated sodium channels snap open, and sodium ions, which are far more concentrated outside the cell than inside, rush inward. Each sodium ion that enters makes the inside more positive, which opens still more sodium channels, which lets in still more sodium. That self-reinforcing loop drives the voltage upward at breathtaking speed.

The second phase is the peak, reached around plus 30 millivolts. Here the sodium channels do something clever: they inactivate. A built-in molecular gate swings shut and plugs the channel, halting the sodium flood even though the channel has not yet fully closed. The spike stops climbing.

The third phase is repolarization. Slower voltage-gated potassium channels, which began opening as the voltage rose, now carry potassium ions outward. Since potassium is more concentrated inside the cell, it flows out, and every departing potassium ion makes the interior more negative again, dragging the membrane back down toward rest.

The fourth phase is afterhyperpolarization. The potassium channels are sluggish to close, so for a brief moment they let out a little too much potassium, dipping the voltage slightly below the resting level before everything settles back to minus 70 millivolts. The sodium-potassium pump and ordinary ion gradients then quietly restore the original balance, ready for the next spike.

The threshold and the all-or-nothing rule

The single most important number in this story is the threshold, near minus 55 millivolts. It is the tipping point at which the inward sodium current first exceeds the outward leak of potassium, the moment when positive feedback takes over and the spike becomes inevitable. Below threshold, a stimulus produces only a small local wobble in voltage that decays back to rest; nothing propagates. At or above threshold, the sodium loop runs away with itself and a full-amplitude action potential fires.

This is the origin of the all-or-nothing rule, a principle that Edgar Adrian and Keith Lucas had extended from muscle to nerve back in 1912, decades before anyone could explain it mechanistically. A neuron either fires a complete spike or it fires nothing at all, with no half-measures in between. A stimulus just barely over threshold produces exactly the same spike as one far above it. The neuron is, in this sense, a digital device built from analog parts, and Hodgkin and Huxley's channel kinetics finally showed why the threshold and the all-or-nothing behavior emerge from the physics of the membrane.

It is worth holding a few of their headline numbers in one place, because together they sketch the entire event: threshold near minus 55 millivolts, peak near plus 30 millivolts, sodium channels open from roughly zero to two milliseconds into the spike, potassium channels open from about one to four milliseconds, an absolute refractory period of one to two milliseconds, and, in the fastest fibers, conduction velocities approaching 120 meters per second.

Refractory periods and the arrow of conduction

After every action potential, the inactivated sodium channels cannot reopen for about one to two milliseconds, no matter how strong the stimulus. This interval is the absolute refractory period, and it does two essential jobs. First, it sets a ceiling on how fast a neuron can fire, capping the maximum firing rate of an axon somewhere around 500 to 1,000 hertz. Second, and more subtly, it gives the action potential a direction.

As a spike travels down an axon, the patch of membrane just behind it is left in the refractory state, its sodium channels temporarily locked shut. The spike therefore cannot turn around and re-excite the region it just came from; it can only push forward into membrane that is still rested and ready. The refractory period is what makes nerve conduction a one-way street, ensuring that signals march from the cell body toward the terminals rather than sloshing back and forth.

Leaping between the nodes

There is one more trick that vertebrates evolved to make conduction faster, and it transforms the leisurely propagation of the squid axon into something far quicker. Many of our axons are wrapped in myelin, a fatty insulating sheath laid down by supporting glial cells, with the wrapping interrupted at regular intervals by tiny bare gaps called the nodes of Ranvier.

The voltage-gated sodium channels cluster densely at these nodes, while the long stretches of membrane buried under the myelin are electrically silent. Instead of regenerating itself continuously along every micrometer of membrane, the action potential effectively jumps from one node to the next, a mode of propagation known as saltatory conduction, from the Latin saltare, to leap. Because the signal only has to be rebuilt at the widely spaced nodes rather than everywhere, it travels dramatically faster, reaching peak velocities of about 120 meters per second in the fastest A-alpha motor fibers that drive your muscles. That is roughly 270 miles per hour, which is why you can yank your hand off a hot pan before you consciously feel the burn.

From a frog nerve to a single channel

The action potential's discovery is a long arc spanning most of the twentieth century, and each chapter sharpened the resolution. Adrian and Lucas established the all-or-nothing rule for nerve in 1912, working with whole nerves and crude recordings. Hodgkin and Huxley quantified the squid spike in 1952 and won the Nobel in 1963, modeling the membrane currents with equations so accurate that their formalism is still taught and simulated today. Then, in 1976, Erwin Neher and Bert Sakmann invented the patch clamp, a technique delicate enough to record the current through a single ion channel, the very molecular gates Hodgkin and Huxley had inferred but never seen. That work earned its own Nobel in 1991, closing the loop from whole nerve to individual molecule.

The clinical stakes of all this become vivid when the machinery breaks. The genes encoding voltage-gated sodium and potassium channels can carry mutations, and when they do the result is an inherited disorder of excitability called a channelopathy. Certain sodium-channel mutations underlie particular forms of epilepsy and rare inherited pain syndromes; potassium-channel mutations are implicated in episodic ataxia and in long-QT cardiac arrhythmias, where a faulty repolarization in heart muscle can prove dangerous. The same physics that lets a neuron fire, when slightly miswired, can produce a seizure or a skipped heartbeat, which is a sobering reminder of how much rides on a few milliseconds of ion flow.

Key Takeaways

The action potential is a brief, all-or-nothing voltage spike, about one to two milliseconds long, that propagates down an axon without weakening, and it is the fundamental signal of the entire nervous system. It fires only when the membrane is pushed past a threshold near minus 55 millivolts, at which point voltage-gated sodium channels open and drive the inside of the cell from a resting minus 70 millivolts up to a peak near plus 30 millivolts; sodium channels then inactivate while slower potassium channels open and carry the membrane back down through repolarization and a brief afterhyperpolarization. An absolute refractory period of one to two milliseconds, caused by sodium-channel inactivation, both caps the firing rate around 500 to 1,000 hertz and forces the spike to travel in only one direction, while myelin and saltatory conduction between the nodes of Ranvier push the fastest signals to roughly 120 meters per second. Alan Hodgkin and Andrew Huxley worked all of this out on the half-millimeter giant axon of the squid at Plymouth, publishing their quantitative model in 1952, and the long chain of discovery, from Adrian and Lucas in 1912 through the patch-clamp recordings of Neher and Sakmann in 1976, shows both how a single nerve impulse works and how much human health depends on getting those milliseconds right.