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 7
Chapter 7 · 2.5 h · 8 quiz items · pass at 80%
This module opens IQCB Domain IV (EEG), 18% of the exam and the second-largest domain. Knowing the normal EEG band by band is the prerequisite for recognizing the abnormal: every later abnormality and phenotype is defined against the baseline established here. The quiz confirms the learner can characterize each band and its normal variants and connect them to their QEEG correlates.
Before you can call anything abnormal, you have to know cold what normal looks like, and normal in EEG is not one thing. It is a set of rhythms, each with a generator, a location where it belongs, a state in which it appears, and an age at which it is expected. A burst of slow activity is normal in a sleeping infant and ominous in a waking adult. A run of frontal theta is the signature of cognitive effort in one context and the signature of drowsiness in another. The same numbers mean different things depending on band, region, state, and age, and a practitioner who reads the numbers without those four frames will manufacture pathology out of ordinary physiology.
This chapter is the catalog of normal rhythms and the variants at their edges. It is Domain IV of the IQCB blueprint, the largest interpretive domain, and it is the foundation the abnormal-pattern chapter builds on. Work through it the way you would learn an instrument: not as a table to memorize but as a set of sounds to recognize, each with a source you can name. The physiology of how these rhythms are generated is treated in full in Neurophysiology for Neurofeedback. Here the emphasis is on reading them at the cap and knowing the boundary between a normal variant and a finding worth chasing.
A note on band boundaries before we start. The frequency ranges below are conventions, not natural law, and the boundaries between bands are soft. The most important consequence is at the theta-alpha border. An individual whose alpha peaks at eight hertz, on the slow side of normal, has alpha activity straddling the conventional theta-alpha line, so a fixed-band computation will score part of their alpha as excess theta. Before you interpret any theta-alpha boundary finding, find the individual alpha peak frequency and read the bands against it, not against a textbook cutoff. We return to this under alpha, because it is the single most common way a normal spectrum gets misread as abnormal.
Delta is the slowest band and the one whose meaning swings most violently with context. Two generators produce it in the sleeping brain. The thalamocortical loop, when its relay cells sit in deep burst mode, fires synchronized bursts at delta frequencies, and the cortex itself produces a slow oscillation near one cycle per second in which whole populations of neurons depolarize and fire together for a fraction of a second, then fall silent, and cycle at that slow rate (Steriade et al., 1993). These are the engines of the high-amplitude delta that dominates slow-wave sleep.
In its normal contexts, delta is the sign of a brain that is either deeply asleep or developmentally young. It is the dominant rhythm of slow-wave sleep, stages three and four, where it appears as high-amplitude waves across the whole head. It is normal and abundant in infants and young children, whose immature thalamocortical systems and incomplete cortical inhibition keep the resting record slow and rich in low frequencies. There is no single normal EEG, only normal for an age, and in the first years of life a great deal of delta is exactly what the age predicts.
The picture inverts in the waking adult. Normal waking cortex does not cycle at delta rates, so delta in a waking adult is the exceptional finding and demands an explanation. The first explanation to rule out is artifact, because the commonest source of apparent waking delta is not the brain at all: slow eye movements, the rolling deflections of a drowsy subject, and other movement contaminate the low band and masquerade as cortical slow activity. Clean the record first. Once artifact is excluded, genuine waking delta points to two broad possibilities that the abnormal-EEG chapter develops in full. Focal delta over a region suggests a local structural disturbance, because the cortex there is not maintaining its normal fast activity. Diffuse delta across the whole head suggests that something is pushing the entire cortex toward its sleep state: deep drowsiness, or a metabolic or toxic process compromising global function.
The morphology of pathological waking delta carries information. Polymorphic delta, irregular and varying in shape and duration, tends to reflect structural damage to the underlying cortex or its white matter. Rhythmic delta, by contrast, regular and sinusoidal, tends to reflect a disturbance at a distance from the recording site, often a deeper or more diffuse process. One rhythmic-delta pattern has a name you must know: frontal intermittent rhythmic delta activity, FIRDA, a run of rhythmic delta over the frontal regions that appears intermittently against an otherwise faster background. FIRDA is not specific to any one disease. It appears in diffuse encephalopathies, in raised intracranial pressure, and in a range of toxic and metabolic states, and its message is that something is disturbing deep midline or diffuse function rather than that a particular diagnosis is present (Niedermeyer & Lopes da Silva, 2005). Its occipital counterpart in children, occipital intermittent rhythmic delta activity, OIRDA, carries an association with generalized epilepsy in the pediatric population (Niedermeyer & Lopes da Silva, 2005). For your purposes at the cap, the rule is simple: rhythmic frontal or occipital delta in a waking record is a flag, not a diagnosis, and it routes the case toward clinical evaluation rather than toward a quantitative interpretation.
In the quantitative record, all of this shows up as elevated absolute power in the delta band, and the same caution applies with more force. A z-score map that lights up in delta is meaningless if the underlying epochs contained eye movement or drowsiness, because the database comparison faithfully quantifies the artifact. Garbage in, garbage out is never more true than in the delta band.
Theta has three separable origins, and conflating them produces systematic misreadings (Buzsáki & Wang, 2012). Thalamic relay cells produce theta-frequency bursting in the early stages of drowsiness and light sleep. Frontal midline theta, generated over the anterior cingulate, reflects cognitive control and working memory load in the alert waking brain. And hippocampal theta, generated by the circuitry of the medial temporal structures, reflects memory encoding and spatial navigation. The scalp cannot read hippocampal theta directly, because the hippocampus is deep and its field is largely closed at the surface, but the cortical projections of the memory system contribute to the theta you do record. Topography and state together tell you which source is speaking: theta in a drowsy subject is the thalamus sliding toward sleep, while theta across frontal and temporal sites in an alert subject under cognitive load is the cortical and hippocampal-projection system at work.
Frontal midline theta is the theta finding a practitioner most wants to read correctly. It sits at the midline frontal sites, it rises with working memory load and sustained concentration, and in trait form it is a stable individual feature rather than a sign of pathology. The critical clinical distinction is three-way: trait frontal theta, a stable feature of the person's resting record, is different from state theta, the drowsiness that pulls theta up acutely, which is different again from artifact, the eye-blink and movement deflections that contaminate the frontal channels. A run of frontal theta means nothing until you have placed it in one of those three boxes, and the way you place it is by attending to the recording state and to the cleanliness of the frontal electrodes.
Temporal theta requires its own judgment, because intermittent theta over the temporal regions sits close to the boundary between normal and abnormal. A modest amount of temporal theta is a normal finding in adults, particularly in older adults and particularly over the left temporal region, where benign temporal slowing of the elderly is a recognized normal variant (Niedermeyer & Lopes da Silva, 2005). The same finding, if it is persistent, focal, and asymmetric, can instead reflect underlying temporal dysfunction. The discriminators are the ones you will use throughout the abnormal-EEG chapter: amount, persistence, focality, and asymmetry. A little, intermittent, and symmetric leans normal. A lot, persistent, focal, and asymmetric leans toward a finding worth pursuing.
In children, theta is partly a developmental fact rather than a clinical one. The young brain runs slow, and posterior theta is abundant and expected in childhood, declining across adolescence as the alpha rhythm establishes and speeds up. Reading a child's theta against an adult template is the classic developmental error, and it manufactures abnormality wholesale. Age-matched norms encode the expected decline; your job is to remember that the decline is steep in the early years and to compare the child against children.
In the quantitative record, theta drives two of the most cited and most fragile metrics in the field. The theta-beta ratio, elevated on average in groups with attention difficulties, is a real group-level finding that does not license an individual diagnosis, and its history as a cautionary tale about reverse inference belongs to the methodology chapters. Frontal theta asymmetry is the second. Both are useful for characterization and treatment stratification and neither is diagnostic, and the reason the theta band sits at the center of these debates is that its three generators are so easy to confuse with each other and with artifact.
Alpha is the most reliable feature in human EEG and the rhythm against which a practitioner calibrates the whole record. Close the eyes and the back of the head fills with a rhythm near ten cycles per second. Open them and it drops away. That rhythm is the posterior dominant rhythm, and a clean, reactive posterior alpha that blocks crisply on eye opening is, in effect, a working thalamocortical loop reporting in (Steriade, 2006).
The posterior dominant rhythm, the PDR, sits over the occipital and parietal cortex, is strongest with the eyes closed, and is the resting idle of the visual system. Its generation is the thalamocortical engine in its resting cadence: eyes closed, the posterior cortex idles, its cells fall into the common thalamocortical rhythm, the aligned dipoles summate into a tall alpha wave. Three properties of the PDR are the working facts of normal reading. Its presence: a well-formed PDR is the hallmark of a normal waking adult record. Its frequency: in the healthy adult it sits in the alpha band, and a PDR that has slowed below eight hertz in an awake adult is itself an abnormal finding. Its reactivity, treated next, is the property you test directly.
The PDR has a developmental trajectory you must hold. In young children the dominant posterior rhythm is slow, often in what would be the theta range for an adult. It climbs through childhood and adolescence into the adult alpha band, plateaus through adulthood, and slows again in later life (Scally et al., 2018). A posterior rhythm of eight and a half hertz is broadly age-appropriate in a person of sixty-eight and a flag worth chasing in a person of twenty-four. The same number reads as near-normal in one brain and as a finding in the other, and the difference is entirely the age frame.
Reactivity is the property you can test in seconds and the one that tells you the loop works. With the eyes closed the PDR is present. On eye opening the visual cortex engages, the posterior population desynchronizes, and the alpha amplitude collapses. This is alpha blocking, and intact blocking indicates intact thalamocortical function. When you record the standard eyes-closed and eyes-open conditions, you are testing reactivity directly, and the comparison between the two conditions is one of the most informative things in the whole recording.
The failure of reactivity is a finding. An alpha rhythm that does not attenuate on eye opening, that persists largely unchanged from eyes-closed to eyes-open, indicates that the expected desynchronization is not happening, and it appears in conditions of altered arousal and in certain clinical states. When eyes-open and eyes-closed records collapse into each other, losing the difference that should separate them, the loss of state-dependent modulation is itself the message. The methodology chapters treat what it means when these conditions stop differing. The reading skill here is to always compare them and to treat a non-reactive alpha as abnormal until proven otherwise.
The frequency at which a person's alpha peaks is a meaningful number in its own right, and it is distinct from how much alpha there is. This individual alpha peak frequency varies from person to person, generally between eight and thirteen hertz, and it behaves like a stable trait that reflects the speed of the thalamocortical clock (Klimesch, 1999). A faster peak tends to accompany quicker cognitive processing, and the peak slows with age. It is stable enough across sessions to serve as an individual marker.
The clinical payoff is twofold. First, the peak frequency and the alpha power are different readings with different meanings, and collapsing them into a single alpha finding loses the more useful of the two. Excess alpha power asks how much the region is idling. A slow alpha peak asks how fast the underlying clock is running. A practitioner who reports an alpha problem without separating the two has thrown away half the information. Second, the peak frequency is the fix for the theta-alpha boundary problem raised at the start of this chapter. An individual with an eight-hertz peak will have alpha power spilling across the conventional theta line, and a fixed-band metric will misclassify it as excess theta. A fast-alpha individual at twelve hertz may have alpha spilling into low beta. Before you interpret any boundary finding, find the peak and read the bands against it.
Several normal rhythms sit in or near the alpha band and must not be over-read. Mu rhythm is a central alpha-band rhythm over the sensorimotor strip; it has an arch-shaped morphology, it is suppressed by movement or the intention to move, and it gets its own treatment later in this chapter. Lambda waves are sharp posterior transients that appear during active visual scanning in a subject with open eyes looking at a patterned field. They are a normal response to visual exploration, not epileptiform sharp activity, and the discriminator is that they are time-locked to saccades and present only with the eyes open. Slow alpha and fast alpha are individual variants of peak frequency, normal in their own right once read against the person's age and baseline. Positive occipital sharp transients of sleep, POSTS, are sharp posterior waveforms that appear in light sleep and are entirely normal, despite a morphology that an inattentive reader might flag. The common thread across these variants is that each has a morphology, a location, and a state in which it is normal, and each gets over-read by a practitioner who has not learned to recognize it.
Alpha is not only a posterior rhythm. The relative amount of alpha over the left versus the right frontal regions has been studied as a marker of motivational style, in a lateralization model associated with Davidson, in which relatively less left-frontal alpha, taken to indicate relatively greater left-frontal activation, accompanies approach motivation, while the reverse pattern accompanies withdrawal (Niedermeyer & Lopes da Silva, 2005). Frontal alpha asymmetry appears in the depression literature and in the affective-style literature, and it is a characterization metric, not a diagnostic one. The reading caution is the usual one for asymmetry measures: it is sensitive to the reference montage and to electrode placement, and a small asymmetry is easy to produce artifactually, so the finding earns weight only when it is strong and replicated across conditions.
The most important reason to internalize that alpha means resting wakefulness is that there is a state in which widespread alpha-frequency activity means almost the opposite. In alpha coma, a comatose patient shows diffuse, monotonous alpha-frequency activity across the head, and the activity is not reactive: unlike the normal resting alpha it does not block to stimulation, and unlike the normal PDR it is often widespread and frontally distributed rather than posterior (Niedermeyer & Lopes da Silva, 2005). The pattern can follow anoxic brain injury or brainstem lesions, and it carries a serious prognostic weight that depends on its cause. The lesson for a practitioner is conceptual and it is a warning against pattern-matching on frequency alone: an alpha-frequency rhythm is not automatically a sign of the relaxed waking state. State, reactivity, distribution, and clinical context determine whether alpha-frequency activity is the healthy resting rhythm or its grim impersonator. You will rarely record an unresponsive patient, but you must know the pattern exists, because it is the cleanest demonstration in all of EEG that frequency without context means nothing.
Beta is the signature of an active, engaged cortex, and it is the band most contaminated by muscle. Its main cortical source is the sensorimotor strip, and frontal beta accompanies sustained cognitive effort. Read by sub-band, it carries different associations. Low beta, roughly thirteen to fifteen hertz, includes the sensorimotor rhythm, the SMR, a rhythm of relaxed sensorimotor inhibition over the central strip in a still but alert subject. Mid beta, roughly fifteen to twenty hertz, accompanies active cognition and effortful processing. High beta, above about twenty hertz, accompanies intense focus and, when excessive in a resting record, associates with anxiety, hyperarousal, and rumination.
The dominant hazard in the beta band is muscle. True cortical beta is typically modest in amplitude and topographically organized over its generators. Muscle-contaminated beta is large, broadband, and distributed wherever the contaminating muscle sits, most often the frontalis and temporalis muscles over the frontal and temporal sites. The contamination rises with frequency and is severe in high beta and above. Before you read any beta excess as a brain finding, you have to exclude muscle, and the methods for doing so, the morphology of EMG, its topography, its spectral shape, belong to the artifact chapter. The clinical translation here is blunt: an excess of high beta over the temporal regions is muscle until proven otherwise, and a practitioner who reports it as cortical hyperarousal without excluding EMG is reporting an artifact.
Beta has a sleep counterpart worth distinguishing from waking beta. Sleep spindles, the brief waxing-and-waning bursts around twelve to fifteen hertz that mark stage-two sleep, are produced by the same thalamocortical engine that produces alpha, running in a different state (McCormick & Pape, 1990). A spindle is a sleep grapheme with a defined morphology and a defined stage. A waking beta burst is a different thing produced in a different state. Confusing the two is a state error, and the fix is to know what stage the subject was in.
In the quantitative record, beta asymmetries and beta excesses are reportable characterization findings, always read after artifact exclusion. Benzodiazepines and related drugs drive beta up in a recognizable way, which is one reason the medication chapter insists that you document every drug a client is taking before you interpret their beta.
Gamma reflects fast, local cortical computation. Its cortical generator is the fast-spiking interneuron circuit: timed inhibition from parvalbumin-expressing interneurons creates rhythmic windows of excitability at gamma frequencies, so a cortical population oscillates in that range when the local computation demands it (Cardin et al., 2009). Functionally, gamma is associated with perceptual binding, the integration of features into a unified percept, with working memory, and with attention, and the forty-hertz oscillation in particular has been studied as a correlate of conscious perceptual processing. Gamma is a local rhythm. It rarely organizes large regions of cortex at the same frequency at once, which is part of why it does not dominate the scalp power spectrum the way alpha does.
The reason gamma sits near the end of this chapter rather than at its center is the EMG problem, and for the scalp practitioner the EMG problem is decisive. The gamma band overlaps directly with the frequency range of muscle activity, and scalp muscle, particularly the cranial muscles, produces broadband activity that fills the gamma band with non-cerebral signal. A scalp recording's apparent gamma power is contaminated by muscle to a degree that makes naive interpretation untrustworthy, and any claim about scalp gamma deserves careful scrutiny on these grounds. Separating true cortical gamma from muscle requires specialized methods, independent component analysis to isolate the cerebral sources, or recording modalities that sidestep scalp muscle altogether, and even then the claims are made cautiously. For the IQCB-level practitioner the operational rule is the one the field has settled on: treat scalp gamma findings with deep suspicion, exclude muscle before reading anything in the band, and do not build an interpretation on gamma power from a standard scalp montage without the methods to defend it.
Mu is a central alpha-and-beta-band rhythm with a distinct identity. It sits over the sensorimotor cortex at the central sites, it shares the alpha frequency range and often carries a beta-frequency harmonic, and it has a characteristic arch or comb-shaped morphology. Its defining feature is its reactivity, and the reactivity is what separates it from posterior alpha: mu suppresses with movement, with the intention to move, and with observation of movement, while it persists when the eyes open, exactly the opposite of the posterior PDR. That double dissociation is the test. If a central alpha-band rhythm blocks to movement but not to eye opening, it is mu; if a posterior rhythm blocks to eye opening, it is the PDR. Mu reflects the idle state of the sensorimotor system the way posterior alpha reflects the idle state of the visual system, and reading it as displaced posterior alpha, or over-reading its presence as abnormal, is the error its distinct reactivity exists to prevent.
Posterior slow waves of youth are a normal variant that exists precisely to be misread. They are slow waves, in the delta-theta range, that appear intermittently over the posterior regions and fuse with the alpha rhythm of children and adolescents, and they are entirely normal in that age group (Niedermeyer & Lopes da Silva, 2005). Against an adult template they look like posterior slowing, the kind of finding that would prompt concern in an adult record. In a child or adolescent they are an expected developmental feature that resolves with maturation. The variant is the clearest case of the chapter's organizing principle: the same waveform is normal at one age and abnormal at another, and the age frame, not the waveform, decides which. A practitioner who does not know this variant will flag healthy young brains, and a practitioner who knows it will read the same record as ordinary for the age.
Every rhythm in this chapter moves across the lifespan, and the movement is the reason age-matched normative comparison is mandatory rather than optional. The shape of the change is consistent. In infancy and early childhood the record is slow and high in amplitude, dominated by delta and theta, because the thalamocortical system is immature and, at the cellular level, the cortex's inhibitory circuitry is not yet operating at adult strength, so the population cannot organize the fast, precisely timed oscillations of the mature brain (Ben-Ari, 2002). Across childhood and adolescence the brain myelinates its tracts and prunes its synapses, the thalamocortical clock speeds up, the posterior rhythm climbs from the theta range into the adult alpha band, slow activity recedes, and beta develops. Through adulthood the bands are stable: a well-formed alpha at rest, appropriate beta on engagement. In later life the individual alpha peak slows, slow activity may re-emerge modestly, and the aperiodic background of the spectrum flattens (Scally et al., 2018; Donoghue et al., 2020).
Two consequences for reading follow. The first is that the aperiodic shape of the spectrum, the way power falls off with increasing frequency, changes with age on its own, and a shift in that background can masquerade as a change in a specific band if the spectrum is not parameterized into its periodic and aperiodic parts (Donoghue et al., 2020). An apparent band change in an older adult may be a background change, not a band change. The second is the rule the entire chapter has been building toward: a child is not a small adult, and an older adult is not a young one whose brain has gone wrong. Each is normal for its stage. The normative databases of Chapter 11 are binned by age for exactly this reason, and the discipline of reading a rhythm is to hold the age frame first and the clinical question second.
The reading skill this chapter teaches is a habit of four questions asked of every finding. Which band is it, and what is that band's generator. Where is it, and does it belong there. In what state was it recorded, eyes open or closed, alert or drowsy. And at what age, against which the finding has to be judged. A run of delta is sleep in an infant, structure or metabolism in a waking adult, and artifact more often than either. A run of frontal theta is cognitive control, drowsiness, or eye movement, depending on state. An alpha-frequency rhythm is the healthy resting idle, or mu, or the paradox of alpha coma, depending on location and reactivity. Gamma at the scalp is muscle until the methods prove otherwise. The numbers do not interpret themselves, and the four frames are what turn a spectrum into a reading.
For the IQCB exam, Domain IV expects you to characterize each band, to distinguish a normal variant from an abnormal pattern, and to know how frequency content changes with development. The variants that exist to be misread, mu against posterior alpha, lambda and POSTS against epileptiform sharps, posterior slow waves of youth against pathological slowing, alpha coma against resting alpha, are favorite material precisely because they test whether you read context or only frequency. The next chapter takes the same four frames into the abnormal record, where the cost of dropping a frame is no longer a mislabeled normal variant but a missed or invented pathology.