Sign in to Peak Brain Path
Sign in to access your courses, books, and progress tracker. New here? Signing in creates your account automatically.
Want to explore courses first?
Browse courses and booksSign in to access your courses, books, and progress tracker. New here? Signing in creates your account automatically.
Want to explore courses first?
Browse courses and booksModule 2
Chapter 2 · 2 h · 10 quiz items · pass at 80%
This module satisfies BCIA II.A.1 and II.A.3 and the IQCB synapses/signaling and neurophysiology-of-EEG topics by fixing the central misconception: the scalp signal is summed post-synaptic potentials, not spikes. The quiz proves the learner can defend that claim from timescale and geometry and can read amplitude as synchrony.
A trainee looks at a burst of beta on the screen and says the neurons are firing fast. The sentence is almost right and importantly wrong, and the gap between almost-right and wrong is the single most useful thing to understand about what the EEG actually records. Scalp EEG does not record neurons firing. It records something slower, and the difference governs what the method can and cannot see.
A neuron produces two distinct sorts of electrical signal, and they live on different timescales. Keeping them separate is the whole point of this chapter.
The action potential is the fast one. When a neuron's membrane voltage climbs past a threshold, voltage-gated sodium channels snap open, sodium floods in, and the inside of the cell briefly swings positive. Potassium channels then open and restore the negative resting state, often with a brief overshoot. The entire spike is over in one to two thousandths of a second, and it travels down the axon as a self-regenerating wave to tell the next cell that something happened. The action potential is all-or-none: it either fires at full size or it does not fire at all. It is the brain's way of sending a discrete message over distance.
The post-synaptic potential is the slow one. When an axon's signal arrives at a synapse, it releases neurotransmitter onto the receiving cell, and that cell responds with a graded shift in its membrane voltage. An excitatory post-synaptic potential nudges the cell toward threshold, and an inhibitory one nudges it away. These shifts are not all-or-none. They are graded, meaning their size scales with the input, and they are slow, lasting tens to hundreds of thousandths of a second. Most importantly, they summate. Many small post-synaptic potentials arriving close together in time, or close together on the cell's surface, add up.
That word, summate, decides everything that follows. Two properties determine which electrical events can be read at the scalp: how long they last and how well they line up. Post-synaptic potentials win on both counts.
The post-synaptic potential begins with a handoff worth naming once, because the neurotransmitter systems of Chapter 15 build on it. When an action potential reaches the axon terminal it opens calcium channels, and the calcium that enters makes small packets called vesicles fuse with the membrane and spill neurotransmitter into the narrow gap, the synaptic cleft. The transmitter crosses and binds receptors on the receiving cell, and those receptors come in two broad kinds. Ionotropic receptors are themselves ion channels: binding opens them at once, and the fast excitatory and inhibitory post-synaptic potentials of this chapter are their work. Metabotropic receptors act indirectly, through slower internal messengers that modulate the cell over a longer span. The fast channels carry the moment-to-moment signal the EEG sums. The slower, modulatory chemistry, and the specific transmitters that run each system, is the subject of Chapter 15.
The spike deserves a closer look, because it is the unit of neural communication even if it is not the unit of the EEG. At rest the cell sits near minus seventy millivolts (Chapter 1). When summed inputs push the membrane to a threshold, around minus fifty-five, voltage-gated sodium channels open, and here the key feature is positive feedback: the sodium that enters depolarizes the membrane further, which opens more sodium channels, which lets in more sodium. That runaway is why the spike is all-or-none. Once threshold is crossed the event goes to completion at full size. There is no half-spike. The membrane shoots toward positive values, then the sodium channels inactivate and slower voltage-gated potassium channels open, letting potassium out and driving the membrane back down, usually past rest into a brief undershoot before settling.
Two consequences matter. First, the refractory period: for a short interval after a spike the sodium channels are inactivated and cannot reopen, so the cell cannot immediately fire again. This caps the maximum firing rate and, importantly, forces the action potential to travel in one direction down the axon rather than back on itself. Second, propagation: the spike regenerates itself at each point along the axon, so it travels without weakening, unlike the passively spreading, decaying post-synaptic potentials. Myelin, the fatty insulation around many axons, lets the spike jump between gaps and travel far faster. Its loss, as in demyelinating disease, slows conduction.
None of this, to repeat the chapter's thesis, is what the scalp records. But it is the neuron physiology a certification candidate must know, and it sets up the contrast: the action potential is the fast, all-or-none, self-regenerating message, and the slow, graded, summating potentials it triggers downstream are what the EEG actually hears.
A post-synaptic potential does not leap instantly from the synapse to the trigger zone at the axon hillock. It spreads along the dendrite passively, the way current spreads down a leaky cable, and it decays as it goes. This passive spread is called electrotonic conduction, and it has a characteristic scale: the length constant, usually written with the Greek letter lambda. The length constant is the distance over which a voltage change falls to about 37 percent of its starting value. Beyond roughly two length constants, a potential has decayed to almost nothing.
For a typical cortical dendrite, the length constant is on the order of a few hundred micrometers, and the dendrites of pyramidal cells can stretch one to two millimeters from tip to soma. That means a synapse on the outer end of a distal branch produces a potential that arrives at the trigger zone much smaller than it started. The rules of cable theory, applied to the neuron's cylindrical branches, describe exactly how steeply the voltage drops along any given segment depending on membrane resistance and the diameter of the branch.
Two consequences follow for understanding the scalp signal. First, amplitude is not proportional to distance from the action-potential trigger zone but to how well the dendritic cable conducts: thick, well-sealed branches deliver a synapse's influence more effectively than thin, leaky ones. Second, and more important for the EEG, slow potentials survive the cable far better than fast ones. A brief perturbation at a distal synapse decays before it travels far, but a sustained, slowly-changing potential has time to spread along more of the dendritic tree, set up a current across a larger stretch of membrane, and contribute to the dipole the scalp can read.
This is the dendritic basis of the chapter's central claim: the EEG records slow, sustained synaptic currents rather than brief action potentials. It is not simply that spikes are brief and happen to cancel across the population. It is that a brief event on a distal dendrite dies electrically before it can contribute to the aligned, widespread current that summates to the scalp. Sustained post-synaptic potentials fill the dendrite's length and establish the oriented current flow that Chapter 3 describes as the current dipole.
[[FIG: FIG-33 – Cable properties and electrotonic spread – HALF PAGE – a pyramidal-cell dendrite with length constant marked, showing a fast transient decaying steeply versus a sustained PSP spreading far along the tree, with voltage plotted against distance HERE]]
The summation that makes post-synaptic potentials matter comes in two forms, and both are worth holding clearly. Temporal summation is summation in time: a single synapse firing rapidly, before each small potential has decayed, so the potentials stack on top of one another and build. Spatial summation is summation in space: many synapses across the cell firing close together in time, their potentials adding because they overlap. A neuron is constantly performing both, integrating a barrage of excitatory and inhibitory inputs into a moment-to-moment membrane voltage that rises and falls.
This integration is slow by design. Where an action potential is a brief, decisive event, the post-synaptic potential is a lingering nudge, and lingering is exactly what lets potentials from different cells overlap in time long enough to add into a field. The slowness that makes a single post-synaptic potential seem unimpressive is the very property that makes the population signal readable. Speed would defeat summation. The brain's scalp signal exists because these potentials are slow enough to coincide.
There is a further consequence. Because the EEG sums excitatory and inhibitory post-synaptic currents together, the trace reflects the net balance of excitation and inhibition in the tissue, not excitation alone. A quiet-looking stretch of EEG is not a quiet brain. It can be a brain in which excitation and inhibition are closely matched and largely canceling. This is why inhibition, taken up in Chapter 15, is so central to rhythm: well-timed inhibition shapes when the net current swings, and the swing is what the electrode reads.
[[FIG: FIG-02 – Action potential versus post-synaptic potential timescale – HALF PAGE – a brief all-or-none spike beside slow, graded, summating EPSPs and IPSPs on a shared millisecond axis HERE]]
It surprises most newcomers that the dramatic event, the spike, is nearly invisible to the EEG, while the quiet event, the synaptic shift, is what we record. There are two reasons, one about time and one about geometry.
The first is brevity and asynchrony. An action potential lasts a millisecond or two, and across a population of cells the spikes occur at slightly different instants. For many tiny fields to add into one measurable field, they must overlap in time. Spikes that are this brief and this scattered do not overlap enough. Their fields point in different directions at any given instant and largely cancel. Add the smearing and the steep voltage loss of conduction through cerebrospinal fluid, skull, and scalp, and the action potential's contribution to the surface recording falls away to almost nothing.
The second reason is the shape of the field a spike makes. As the action potential travels along the axon, it creates a small moving region with current flowing in just ahead of it and out just behind it. Those two opposing currents sit close together, and at any distance they tend to cancel, the way the two poles of a very short magnet cancel when you stand back from it. A field built from two nearly overlapping opposite sources is a closed field; it does not project far. The post-synaptic potential, by contrast, sets up a more sustained and more spatially separated current along the cell, which does project. Chapter 3 builds that geometry in full.
[[FIG: FIG-03 – Why action potentials cancel and post-synaptic potentials summate – HALF PAGE – scattered, asynchronous spike fields canceling beside aligned, sustained PSP fields adding HERE]]
Post-synaptic potentials reach the scalp because they last long enough for many cells to be in the same electrical state at the same time, and because the cortex's principal cells are physically aligned so their currents point the same way. When a population of aligned cells receives synchronized synaptic input, their individual currents add into a field large enough to conduct to the scalp and be recorded. To a first approximation, the EEG is a running measure of the summed post-synaptic activity of cortical tissue (Olejniczak, 2006).
This is why the trainee's sentence needs correcting. A burst of beta does not mean the neurons underneath are firing faster. It means a population of synaptic currents is oscillating in the beta range, and that those currents are synchronized and aligned enough to summate into a rhythm the electrode can pick up. EEG amplitude is mostly a story about synchrony, not about firing rate.
Pushed one step further, the same principle explains a pattern every brain trainer sees daily: when a region engages in active processing its cells fall out of lockstep and the rhythm in that band shrinks, and when it idles they fall back into a common rhythm and the amplitude grows. This is event-related desynchronization and synchronization, the synchrony story of Chapter 3 watched in real time and worked through at the end of this chapter.
If action potentials are nearly invisible at the scalp, why does muscle tension ruin a recording? The answer sharpens the whole chapter. Muscle fibers fire action potentials too, but three things make muscle the loud exception that brain action potentials are not. The muscle is millimeters from the electrode, not across the skull, so almost none of its signal is lost. Motor units fire in sustained, overlapping volleys rather than scattered single spikes, so they summate instead of canceling. And the fibers are large and numerous. The result is electromyographic activity that is high in frequency and often large in amplitude, and it lands squarely on top of the beta and gamma bands a brain trainer cares about.
This is not a footnote. It is one of the highest-stakes practical facts in the field. A jaw clench, a furrowed brow, or a tense neck produces fast, high-amplitude activity that looks, to the untrained eye, like beta. A practitioner who mistakes that muscle for brain may conclude a client is hyperaroused and train their "beta" down, chasing a signal the cortex never produced. The difference is one of origin, not of reading: genuine cortical fast activity is modest in amplitude because it is summed synaptic current seen through the skull, while muscle is large because its action potentials come from a source too close and too synchronous to cancel, arising at the big muscles of the jaw, brow, and neck. Spotting and removing it on a real recording is the work of The QEEG Field Guide. Understanding why it dominates the high frequencies is the work of this chapter.
Muscle is the loudest intruder, not the only one, and the same question, what reaches the scalp and why, sorts the rest. A brain trainer needs to recognize their origins; catching and clearing them is the work of The QEEG Field Guide. Each has a source and a signature.
The unifying lesson is the muscle lesson generalized: the scalp records whatever electrical or mechanical event is large enough and close enough to reach it, cerebral or not. This book's job is only that "why," the electrophysiology of each source. Recognizing them on a real recording, with visual examples, montage and phase analysis, and the methods for removing them (independent component analysis, artifact subspace reconstruction, and the rest), is the entire subject of The QEEG Field Guide, Chapter 4, and is not repeated here. Knowing the origin is what lets a practitioner refuse to train an artifact. Learning to catch and clear it is the Field Guide's work.
Picture a client at rest, eyes closed, with a tall, steady posterior alpha rhythm on the screen. Ask them to open their eyes and do mental arithmetic, and the alpha collapses, the trace flattening into low-voltage, faster activity. Nothing about the number of neurons changed in that second, and the cells did not stop working; quite the opposite, they got busier. What changed is synchrony. At rest, the posterior population oscillated together, its aligned post-synaptic currents summing into a large rhythm. Engaged by the task, the population desynchronized, each region working on its own schedule, and the currents that had been adding began to cancel. The amplitude fell because the cells fell out of step, not because they powered down.
This is event-related desynchronization, one of the most reliable phenomena in the field (Pfurtscheller & Lopes da Silva, 1999), and it is the cleanest demonstration of this chapter's thesis that amplitude tracks synchrony. A practitioner who internalizes it stops reading a flattening rhythm as the brain doing less and starts reading it as the brain coordinating less, which is usually the brain doing more. Run in reverse, the same logic explains why a drowsy or idling cortex throws large, slow waves: with nothing to coordinate around, its cells drift into a common rhythm and the amplitude climbs.
Consider a practitioner who reads a client's elevated high beta over the frontal cortex and concludes the frontal neurons are overactive, working too hard. The inference feels natural and it is the wrong shape. High beta amplitude reflects a population of synaptic currents oscillating, synchronously, in a fast band. It can accompany worry or hyperarousal, but it is not a measure of how hard the neurons are working or how fast they are firing. The correct reading is about the state of the synaptic population, its rhythm and its synchrony, not about cellular effort. Keeping the distinction straight is what separates a physiological interpretation from a metaphor.
What this means for the signal: when you read amplitude, you are reading how many aligned cells are doing the same synaptic thing at the same time, not how hard any neuron is working. Hold that one correction and a large part of EEG interpretation stops being mysterious and starts being mechanism.
Key points
In one sentence: the EEG hears the slow, summating synaptic chorus, not the fast individual spikes, and it also hears any loud neighbor.
Check yourself
Ch 1 (membrane potential), Ch 3 (dipole geometry), Ch 12 (synchrony across populations = coherence).
Post-synaptic potentials (not action potentials) of cortical neurons; summed activity from millions of neurons acting in synchrony; volume-conducted signals through brain, CSF, skull, scalp; 70-80% voltage loss; sensitivity primarily to tangential dipoles in cortical sulci.
The volume-conduction and attenuation points stay in the Field Guide
(recording-relevant); only the generative explanation was pulled here. See
qeeg-field-guide/meta/PRUNE-AFTER-PHYSIOLOGY-TRANSPLANT.md.