What Is Action Potential

How the Nerve Signal Actually Works

Action potential is the way neurons communicate. It's an electrical spike that travels down an axon. If you're studying neuroscience or dealing with neurophysiology lab work, you need to understand this at a practical level, not just the textbook definition. I spent years recording single-unit activity from cortical neurons and the difference between knowing the equations and actually making sense of what your traces are telling you is enormous. The action potential is a rapid, temporary change in membrane potential. It goes from about negative 70 millivolts up to positive thirty millivolts and back down again in roughly one millisecond. This happens because voltage-gated sodium channels open first, letting sodium rush into the cell. Then those channels inactivate and voltage-gated potassium channels open, pushing potassium out. The membrane repolarizes, sometimes briefly overshooting into hyperpolarization before settling back. Here's something most introductory courses gloss over: the action potential doesn't degrade as it travels. That's because it's actively regenerated at every segment of the axon. My first year running patch clamp recordings, I assumed the signal would attenuate over distance like a cable signal. It doesn't. The all-or-nothing principle means every spike along the axon looks virtually identical. The information is in the frequency and timing, not the amplitude.

I once spent two weeks troubleshooting what I thought was a recording artifact. The spikes kept changing shape mid-experiment, and I was convinced my electrode was drifting. Turned out I was recording from a myelinated axon and the signal was jumping between nodes of Ranvier. The "drift" was actually the propagation pattern shifting as the preparation deteriorated slightly. Once I recognized the nodal firing pattern, the problem solved itself. You just need to know what you're looking at.

The Mechanism at the Channel Level

At rest, a neuron sits around negative seventy millivolts. That's maintained by the sodium-potassium pump and leak channels. When a stimulus pushes the membrane toward threshold, usually around negative fifty-five millivolts, voltage-gated sodium channels sense the change and flip open. Sodium flows in down its electrochemical gradient. The membrane depolarizes rapidly. This is the rising phase. Within about a millisecond, the sodium channels hit their inactivation gate. They close and stay closed for a few milliseconds regardless of the membrane potential. That's the refractory period, and it's critical because it limits how fast a neuron can fire. You cannot get a second action potential immediately after the first one. Absolute refractory period lasts roughly one millisecond, followed by a relative refractory period where a stronger-than-usual stimulus can trigger another spike. The falling phase comes from potassium efflux through voltage-gated potassium channels. These channels open more slowly than the sodium channels but stay open longer. Potassium leaving the cell brings the membrane potential back down. In many neurons, it undershoots the resting potential, creating the after-hyperpolarization. This tail is functionally important because it helps reset the sodium channels from their inactivated state and contributes to spike-frequency adaptation.

Common Misunderstandings and Where Beginners Trip Up

One persistent confusion is thinking the action potential is a flow of current through the cell like water through a pipe. It's not. It's a wave of changing permeability. Each segment of membrane briefly becomes permeable to sodium, then potassium, in sequence. The local currents that flow between active and inactive regions are what drive neighboring segments to threshold, but ions aren't marching from the dendrite to the axon terminal. Another pitfall is assuming conduction velocity is the same everywhere. Myelination changes everything. In myelinated fibers, the action potential effectively jumps between nodes, a process called saltatory conduction. This can increase speed by up to one hundred times compared to unmyelinated fibers of the same diameter. The largest motor neurons in your body conduct at around one hundred twenty meters per second. Your pain fibers, which are thin and unmyelinated, crawl along at less than two meters per second. I've also seen people confuse the action potential with the postsynaptic potential. They're completely different phenomena. An excitatory postsynaptic potential is a graded, local change that decays with distance. It can summate. The action potential is a regenerative, all-or-nothing event that propagates without loss. Mixing these up leads to some seriously confused interpretations of electrophysiology data. Check your labels on the traces.

Measuring Action Potentials in Practice

If you're working with extracellular recordings, the most useful thing you'll measure is the spike waveform. The shape tells you about the cell type, the recording quality, and sometimes the health of the preparation. A clean, narrow spike with a clear negative peak and a smaller positive deflection usually means you're close to the soma or a thick proximal dendrite. Broader, lower-amplitude waveforms suggest you're further away from the source. Intracellular recording is the gold standard but it's finicky. Glass microelectrodes need to be pulled to very fine tips, filled with three molar potassium chloride, and have resistances between five and fifteen megohms. When you penetrate the cell, you want to see a stable resting potential around negative sixty-five to negative seventy millivolts. If it drifts or collapses within minutes, you've likely damaged the membrane during impalement. The cell will still fire, but the resting potential won't be trustworthy. A practical tip from years of doing this: always monitor the seal resistance before and after your recording session. A good whole-cell seal starts above one gigaohm and anything below five hundred megohms is suspect. Degradation during the experiment usually means the cell is breaking down or the pipette is clogging. Documenting seal quality makes your data defensible when reviewers ask about it.

When the Model Breaks Down

The Hodgkin-Huxley model describes action potentials beautifully, but it assumes a uniform cylindrical axon with homogeneous channel distributions. Real neurons don't work like that. Dendrites have active conductances that can generate their own spikes. Axon terminals show channel clustering that varies across terminal branches. Some neurons express calcium-based action potentials that are much slower and longer than the sodium-driven kind. The basic model won't capture any of that. Pharmacology also disrupts the standard picture. Tetrodotoxin blocks voltage-gated sodium channels and abolishes the fast action potential entirely. Tetraethylammonium blocks certain potassium channels and broadens the spike. These are useful tools for dissection but they make the membrane behave in ways that don't reflect normal physiology. If you're using blockers, report the concentrations and exposure times precisely. A ten-minute application does something different than a thirty-minute one. The refractory period is another area where textbook descriptions oversimplify. In cortical neurons, the relative refractory period isn't just about sodium channel inactivation. Potassium channel accumulation, calcium-activated potassium currents, and sodium pump activation all contribute to the reduced excitability that follows a spike. The exact recovery profile varies considerably between cell types, which is why you'll see some neurons fire in high-frequency bursts while others adapt rapidly.

Key Takeaways

The action potential is a self-regenerating wave of ion channel activity that propagates along the neuron without decrement. Its duration, amplitude, and shape depend on the specific complement of voltage-gated channels expressed by that cell. Conduction velocity is determined by axon diameter and myelination status. The all-or-nothing nature means amplitude carries no information, but frequency and timing do. Recording artifacts are far more common than genuine biological phenomena, so verify your setup rigorously before interpreting unusual waveforms.