Every clinical neurophysiology test rests on the same underlying biophysics: voltage-gated sodium and potassium channels generating action potentials in axons, saltatory conduction along myelinated fibers, summed postsynaptic potentials producing the EEG, and the relationship between recorded waveforms and the populations of neurons or muscle fibers that produce them. Understanding this layer is what separates the technician reading a tracing from the clinician interpreting it. A demyelinating neuropathy on NCS is not just “slow” — it reflects loss of saltatory conduction; a low-amplitude motor response is not just “small” — it reflects axonal loss with reduced synchronized depolarization. This page covers the membrane biophysics, action potential generation, conduction physiology, and the recording principles that determine what we see on every clinical study.

The Resting Membrane Potential

Resting potential of a typical neuron: approximately −70 mV intracellular relative to extracellular.
Set by selective permeability to K⁺ at rest combined with electrochemical gradient established by Na⁺/K⁺-ATPase (3 Na⁺ out, 2 K⁺ in per ATP).
Nernst equation gives the equilibrium potential for each ion (E_K ≈ −90 mV, E_Na ≈ +60 mV, E_Cl ≈ −70 mV, E_Ca ≈ +120 mV).
Goldman-Hodgkin-Katz equation gives the actual membrane potential, weighted by relative permeabilities.
At rest, the membrane is most permeable to K⁺, so the resting potential sits near E_K.

Action Potential Phases

Threshold

A subthreshold depolarization opens a few voltage-gated Na⁺ channels.
At threshold (~ −55 mV), Na⁺ influx exceeds K⁺ efflux and triggers regenerative depolarization.
All-or-none: once threshold is reached, the action potential reliably propagates.

Rising Phase (Depolarization)

Voltage-gated Na⁺ channels open rapidly.
Membrane potential rises toward E_Na (+60 mV) but doesn’t reach it because of K⁺ channel opening.
Peak typically +30 to +40 mV.
Lasts ~0.5 ms in a peripheral axon.

Falling Phase (Repolarization)

Na⁺ channels enter the inactivated state (fast, voltage-dependent).
Voltage-gated K⁺ channels open (delayed rectifier).
K⁺ efflux returns membrane toward E_K.
Hyperpolarization (after-hyperpolarization) often follows because K⁺ channels stay open briefly.

Refractory Periods

Absolute refractory: while Na⁺ channels are inactivated; no stimulus can trigger another action potential. Lasts ~1 ms in peripheral axons.
Relative refractory: while membrane is hyperpolarized but Na⁺ channels have recovered; stronger-than-normal stimulus can trigger an action potential.
Refractoriness sets an upper limit on firing frequency (typically <500–1000 Hz in axons).

Voltage-Gated Channel Biology

Sodium Channels

α subunit forms the pore; β subunits modulate kinetics.
Three states: closed (rest), open (active), inactivated.
Inactivation is via the “ball and chain” intracellular IFM motif blocking the pore.
Mutations cause channelopathies (paramyotonia congenita, hyperkalemic periodic paralysis, Dravet syndrome via SCN1A).
Targeted by Na⁺-channel-blocking ASMs (phenytoin, carbamazepine, lamotrigine, lacosamide), local anesthetics, mexiletine.

Potassium Channels

Voltage-gated (Kv) for repolarization; calcium-activated (BK, SK) for after-hyperpolarization.
Multiple subtypes with different kinetics and tissue distribution.
Mutations cause myotonia congenita (KCNQ), episodic ataxia type 1 (KCNA1), some forms of epilepsy.
Blocked therapeutically by 4-aminopyridine and 3,4-diaminopyridine (LEMS, MS walking impairment).

Calcium Channels

Voltage-gated (L, N, P/Q, R, T types).
Presynaptic terminal Ca²⁺ entry triggers neurotransmitter release.
Mutations cause familial hemiplegic migraine (CACNA1A), episodic ataxia type 2, congenital myasthenic syndromes.
Targeted by α2δ ligands (gabapentin, pregabalin), ethosuximide (T-type, in thalamus).

Propagation in Unmyelinated Axons

Continuous propagation: each segment of membrane must depolarize the next.
Conduction velocity proportional to √(axon diameter), roughly 0.5–2 m/s for typical C fibers.
Limits firing frequency, conduction speed, and metabolic efficiency (entire membrane participates).
C fibers (pain, temperature, postganglionic autonomic) are unmyelinated; this is why pain is slow.

Saltatory Conduction in Myelinated Axons

Myelin is high-resistance, low-capacitance — Schwann cells (PNS) or oligodendrocytes (CNS) wrap concentric layers of plasma membrane around the axon.
Action potentials regenerate only at the nodes of Ranvier, where Na⁺ channels are clustered at high density.
Local circuit currents flow through the internode passively, “jumping” from node to node.
Conduction velocity in large myelinated fibers: 40–80 m/s (vs 0.5–2 m/s unmyelinated).
Conduction velocity scales linearly with axon diameter in myelinated fibers (vs square root in unmyelinated).
Metabolic efficiency: only nodes consume ATP for ion pumping.

Demyelination Disrupts Saltatory Conduction

Loss of myelin lowers internodal resistance and increases capacitance — current “leaks” through what used to be insulated.
Conduction slows; in severe cases conduction block develops because the next node can’t reach threshold.
NCS hallmarks: prolonged distal latency, slow conduction velocity, prolonged F-wave latency, conduction block, temporal dispersion.
Examples: Guillain-Barré (acute demyelination), CIDP (chronic), Charcot-Marie-Tooth type 1 (hereditary).

Axonal Loss Reduces Recorded Amplitude

Each surviving axon still conducts at near-normal velocity.
The compound muscle (CMAP) or sensory nerve action potential (SNAP) amplitude is the sum of all axonal contributions.
Loss of axons → smaller amplitude with preserved (or near-normal) conduction velocity.
Examples: diabetic axonal neuropathy, alcoholic neuropathy, chemotherapy-induced neuropathy, vasculitic neuropathy.

The Neuromuscular Junction

Presynaptic terminal action potential opens voltage-gated Ca²⁺ channels.
Ca²⁺ influx triggers vesicle fusion via SNARE proteins (synaptobrevin/VAMP, syntaxin, SNAP-25).
Acetylcholine diffuses across the cleft, binds nicotinic AChR on the muscle fiber.
End-plate potential triggers muscle action potential if depolarization exceeds threshold.
Acetylcholinesterase rapidly hydrolyzes ACh; choline is reuptaken.

Implications for NMJ Disease

Myasthenia gravis (postsynaptic): autoantibodies reduce functional AChRs; safety factor drops; some end-plates fail to reach threshold, especially with repetitive activity → decremental response on RNS.
LEMS (presynaptic): autoantibodies against P/Q Ca²⁺ channels reduce ACh release per impulse; baseline CMAPs are small; high-frequency RNS or post-exercise produces facilitation (Ca²⁺ accumulates faster than it clears).
Botulism: SNARE proteins cleaved, vesicle release fails; severe weakness with autonomic involvement.

Muscle Fiber Activation

End-plate potential triggers action potential along the sarcolemma.
Sarcolemmal depolarization spreads into T-tubules.
Dihydropyridine receptor senses voltage; mechanically couples to ryanodine receptor on sarcoplasmic reticulum.
Ca²⁺ release from SR enables actin-myosin cross-bridging.
Each muscle fiber innervated by one motor neuron; one motor neuron + all the fibers it innervates = one motor unit.
Recorded as one motor unit action potential (MUAP) on needle EMG.

Generation of the EEG Signal

EEG records summed extracellular field potentials from cortical pyramidal neurons.
Source: postsynaptic potentials (excitatory EPSPs, inhibitory IPSPs), not action potentials, because PSPs are slower and more synchronized.
Apical dendrites of pyramidal neurons aligned perpendicular to cortical surface — when activated synchronously, they create dipoles measurable at the scalp.
To produce a recordable scalp signal, several cm² of cortex must be synchronously active (a single cell or small cluster contributes essentially nothing detectable).
Skull and scalp act as low-pass filters and spatial smoothers: high-frequency activity is attenuated, focal sources spread over several centimeters at the scalp.
Polarity at the scalp depends on the orientation of the dipole — negative at the surface for sinks in superficial cortical layers; positive for sinks in deeper layers.

Volume Conduction (Briefly)

Covered in detail on a dedicated page. Key principle for now: an electrical signal recorded at the scalp or skin surface is not generated AT the electrode; it is the field potential of activity at some distance. Scalp electrodes record from broad cortical areas, not a 1 cm² spot under the electrode. NCS recording electrodes pick up activity from the underlying muscle but also adjacent muscles (cross-talk) and even distant sources via the volume conductor (the body’s tissues conduct current relatively freely).

Stimulation Principles

Electrical Stimulation in NCS

Surface electrodes apply a brief (0.05–1 ms) current pulse.
The cathode is the active stimulator; depolarization is greatest under the cathode.
The fastest-conducting fibers respond first at low intensity; raising intensity recruits slower fibers.
Supramaximal stimulation: increase intensity until the recorded response stops growing; ensures all fibers are activated.
Submaximal stimulation underestimates conduction velocity (slower fibers excluded) and amplitude.

Magnetic Stimulation in TMS

A rapidly changing current in a coil generates a magnetic field that penetrates the skull and induces an electrical current in the underlying cortex.
Activates intracortical interneurons rather than directly depolarizing pyramidal cells; pyramidal cells are activated trans-synaptically.
Useful clinically because the magnetic field is not attenuated by the skull (unlike transcranial electrical stimulation).

Recording Filters

High-pass (low-frequency) filter: removes slow drift (sweat, electrode polarization). Typical: 1 Hz for EEG, 10 Hz for NCS, 20 Hz for EMG, 30 Hz for muscle.
Low-pass (high-frequency) filter: removes muscle artifact, line noise, high-frequency noise. Typical: 70 Hz for EEG, 2 kHz for NCS, 10 kHz for EMG.
Notch filter: targets specific line noise (60 Hz in US, 50 Hz in Europe). Use sparingly — it can attenuate true biological activity in that frequency band.
Inappropriate filtering distorts measurements: high low-pass cutoff makes everything look slower; aggressive notch can create ringing artifacts.

🔍 Did You Know?

The Hodgkin-Huxley equations, published in 1952 and based on voltage-clamp experiments in the squid giant axon, remain the foundation of every clinical neurophysiology test today. Hodgkin and Huxley measured how Na⁺ and K⁺ conductances change with membrane voltage and time, then built a quantitative model that predicted the action potential waveform. Their model — four differential equations describing the kinetics of channel gating particles — accurately reproduces the rising phase, peak, and repolarization of the action potential, and predicts conduction velocity along an axon. This work won the 1963 Nobel Prize. Why does it matter clinically? Because the same biophysics governs human pathology: a sodium channel mutation that slows inactivation by 10% produces myotonia — and the predicted repetitive firing pattern matches what’s recorded on EMG; a potassium channel that fails to repolarize quickly enough produces episodic ataxia, and the predicted attack pattern matches clinical observation. Modern channelopathies are diagnosed by genetic testing but understood through the Hodgkin-Huxley framework: which channel, what gating change, what physiologic consequence. For the clinical neurophysiologist, the lesson is that every recorded waveform is a window into ion channel function — and many clinical syndromes can be predicted (and treated) from channel physiology principles alone.

Pitfalls and Pearls

All-or-none: action potentials don’t gradate in amplitude at a single fiber — they either fire or don’t.
Refractoriness limits firing rate: peripheral axons can fire up to ~1000 Hz briefly; sustained firing is limited by metabolic and channel recovery.
Saltatory conduction: nodes of Ranvier regenerate the action potential; internodes carry it passively.
Demyelination → slow conduction or conduction block: classic NCS findings of GBS, CIDP, hereditary CMT1.
Axonal loss → reduced amplitude with preserved velocity: classic NCS findings of diabetic, alcoholic, chemo neuropathies.
NMJ disease patterns: MG (decrement on low-frequency RNS), LEMS (facilitation on high-frequency or post-exercise).
EEG signal is summed PSPs, not action potentials; needs several cm² of synchronous cortex to be detectable at scalp.
Volume conduction: surface electrodes record from broad regions, not a 1 cm² spot.
Supramaximal stimulation required for reliable NCS measurements.
Cathode is the active stimulator; orientation matters for accurate placement.
TMS works via trans-synaptic activation of pyramidal cells through interneurons; skull does not attenuate magnetic field.
Filter settings shape the recorded waveform: never compare across labs without matching filter settings.
Temperature matters: cooler nerves conduct slower; warm the limb to 32°C distally for NCS.
Sodium channel mutations produce repetitive firing patterns (myotonia, paramyotonia) — the EMG waveform matches the Hodgkin-Huxley prediction for slowed inactivation.
Calcium channel mutations (P/Q, presynaptic) reduce neurotransmitter release per impulse — produces LEMS or familial hemiplegic migraine.

References

Hodgkin AL, Huxley AF. A quantitative description of membrane current and its application to conduction and excitation in nerve. J Physiol. 1952;117(4):500-544.
Kandel ER, Schwartz JH, Jessell TM, Siegelbaum SA, Hudspeth AJ, eds. Principles of Neural Science. 6th ed. McGraw-Hill; 2021.
Hille B. Ion Channels of Excitable Membranes. 3rd ed. Sinauer Associates; 2001.
Preston DC, Shapiro BE. Electromyography and Neuromuscular Disorders. 4th ed. Elsevier; 2021.
Kimura J. Electrodiagnosis in Diseases of Nerve and Muscle: Principles and Practice. 4th ed. Oxford University Press; 2013.
Catterall WA, Kalume F, Oakley JC. NaV1.1 channels and epilepsy. J Physiol. 2010;588(Pt 11):1849-1859.