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 9
Chapter 9 · 3 h · 8 quiz items · pass at 80%
This module completes IQCB Domain IV (EEG), 18% of the exam, by putting the normal and abnormal patterns into the populations and conditions a practitioner meets: sleep, development, epilepsy, TBI, the major psychiatric conditions, dementia, and disorders of consciousness. The quiz confirms the learner can identify a condition-specific EEG and QEEG signature and apply the correct age-referenced baseline.
The previous chapter taught patterns. This one teaches contexts. The same waveform means different things in a newborn, a sleeping adult, a head-injured athlete, and a patient with dementia, and a QEEG practitioner who treats every recording as an interchangeable "resting adult" will misread all four.
The thread running through the chapter is the one that runs through the whole book: a normative database is built on a defined population in a defined state, and the further your recording sits from that definition, the less the comparison means. A two-year-old, a stage N3 sleeper, and a comatose patient are not waking adults, and no z-score map pretends otherwise. So this chapter does two things at once. It gives you the condition-specific patterns the IQCB exam expects you to recognize, and it keeps flagging the line between describing a finding and interpreting a clinical EEG, which stays the province of a board-certified electroencephalographer.
Pattern descriptions follow the standard references, principally Niedermeyer and da Silva's Electroencephalography and the American Academy of Sleep Medicine scoring criteria for sleep, and the sibling research base for the condition-specific QEEG signatures (Niedermeyer & Lopes da Silva, 2005; American Academy of Sleep Medicine, 2023). Specific timings, prevalences, and prognoses carry a flag until the citation gate clears them.
You need to know sleep EEG for a reason that has nothing to do with running sleep studies: your eyes-closed resting recordings drift toward drowsiness and light sleep, and the transients of early sleep are the most over-read patterns in clinical EEG. If you cannot recognize a vertex wave, you will eventually call one a spike. And if you cannot tell that your "resting" epoch is actually stage N1, you will feed a drowsy spectrum into a database that assumed an alert subject.
The wake-to-sleep transition is a continuum, not a switch, and the classic description of its electrographic landmarks comes from Santamaria and Chiappa's account of drowsiness in normal adults (Santamaria & Chiappa, 1987). The modern staging framework, the AASM criteria that replaced the older Rechtschaffen and Kales system, scores sleep in epochs by these landmarks (American Academy of Sleep Medicine, 2023).
Wake. A well-organized posterior dominant rhythm in the alpha range, reactive to eye opening, with eye blinks and muscle tone present.
Drowsiness and NREM stage N1. The posterior rhythm attenuates and slows, the background shifts to low-voltage mixed frequencies with theta intruding frontally and centrally, and slow rolling eye movements appear, before the EEG changes are obvious (American Academy of Sleep Medicine, 2023). The transients of light sleep arrive here:
NREM stage N2. Defined by two landmark waveforms on a background of theta:
NREM stage N3 (slow-wave sleep). High-amplitude delta dominates the record, scored when delta occupies enough of the epoch (American Academy of Sleep Medicine, 2023). This is the deepest, most synchronized state, and its high-amplitude slow activity is normal here and only here in the healthy adult.
REM sleep. A low-amplitude, mixed-frequency background that resembles wake or N1, with theta and sometimes sawtooth waves, accompanied by rapid eye movements and a drop in muscle tone (American Academy of Sleep Medicine, 2023). The paradox of REM is that the cortical EEG looks activated while the body is atonic.
Arousals. Brief intrusions of faster, wake-like activity into sleep, sometimes preceded by a K-complex, marking transient lightening of sleep. In the sleep-medicine setting, a respiratory effort-related arousal (RERA) is one of these brief EEG arousals driven by increasing respiratory effort that falls short of a scored apnea or hypopnea, a pattern that matters for sleep-disordered breathing and is named here only so the term is familiar. Scoring it is polysomnography, not QEEG.
Across a night, these stages cycle in a recognizable architecture, deeper slow-wave sleep weighted toward the first third and longer REM periods toward the morning, summarized in the hypnogram. You will not record a full night, but the same staging vocabulary describes the fragment of N1 or N2 that creeps into a long eyes-closed run, and naming it is what keeps the fragment out of your analysis.
The practitioner's takeaway is sharp: vertex waves, POSTS, spindles, and K-complexes are normal sleep transients, the most common things mistaken for epileptiform discharges by inexperienced readers, and the most common reason a "resting" QEEG epoch is quietly contaminated by drowsiness (American Academy of Sleep Medicine, 2023). Score your own vigilance state before you trust your spectrum. A recording that has slid into N1 shows rising frontal-central theta and falling posterior alpha, the same spectral move a drowsiness-detection algorithm keys on, and a normative database will read that as a phenotype rather than as the sleep onset it actually is (American Academy of Sleep Medicine, 2023). The defense is procedural: select epochs from clearly alert segments with a reactive posterior rhythm, document the vigilance state, and discard anything that shows the drowsiness signature.
The single most important fact about the developing EEG is that "normal" is a function of age, and the function moves fast in the first years of life. A pattern that is healthy at thirty weeks of gestation is grossly abnormal at term, and a pattern normal at term is abnormal at two years. Neonatal EEG is its own subspecialty for exactly this reason, and reading it is not a QEEG competency. It is named here so you understand why standard databases simply do not apply to the youngest patients.
Neonatal EEG. The premature and term newborn EEG is read against gestational age (more precisely, conceptional age), and one of its defining features is discontinuity: a normal premature pattern alternates bursts of activity with intervals of relative quiet, a pattern called tracé discontinu in the more premature infant and tracé alternant nearer term (Niedermeyer & Lopes da Silva, 2005). The expected degree of discontinuity decreases as the brain matures toward continuous activity. This is the one context where a burst-then-quiet pattern is developmentally normal rather than the grave burst-suppression of the previous chapter, which is precisely why neonatal interpretation requires specialized training and conceptional-age norms, not adult pattern-matching (Niedermeyer & Lopes da Silva, 2005).
Pediatric EEG and maturational landmarks. Through infancy and childhood, the background organizes and speeds up. The posterior dominant rhythm emerges in infancy at low frequencies and climbs with age, reaching the adult alpha range and stabilizing by adolescence. Normative work places the mature PDR frequency in the 9 to 11 Hz range with maturation complete by the mid-teens (Niedermeyer & Lopes da Silva, 2005). Two features that are abnormal in an adult are normal in a child:
The QEEG consequence is non-negotiable: pediatric recordings must be compared against age-appropriate normative data, because the developmental trajectory of band power is steep, and comparing a seven-year-old against an adult-anchored norm manufactures abnormality out of normal maturation (Niedermeyer & Lopes da Silva, 2005). The lifespan normative-database chapter takes up which databases carry adequate pediatric samples and where they thin out. Here the point is the principle: in children, the norm is a moving target, and the map is only as valid as the age-match behind it.
Epilepsy is where clinical EEG earns its keep and where QEEG scope is most restricted. The recognition that a patient has epileptiform activity, the syndromic classification, the localization for surgery, and the ictal-versus-interictal call are all the electroencephalographer's, and the previous chapter set that boundary. What you should understand is the three-phase temporal structure of the epileptic EEG, because it organizes everything a referral report will say.
Interictal. Between seizures. The interictal EEG carries the discharges of seizure tendency without a clinical seizure occurring: focal spikes and sharp waves pointing at a region in the focal epilepsies, generalized spike-and-wave in the generalized epilepsies, and the named patterns of the previous chapter such as 3 Hz spike-and-wave and TIRDA (Niedermeyer & Lopes da Silva, 2005). The interictal recording is what a routine clinical EEG most often captures, and its yield rises with sleep deprivation, drowsiness, and activation procedures, which is part of why it is a specialist study.
Ictal. During a seizure. The ictal EEG evolves: it characteristically begins with a change in frequency and amplitude that builds, spreads, and progresses through the event, a rhythmic discharge that increases in amplitude and decreases in frequency as the seizure organizes and then propagates (Niedermeyer & Lopes da Silva, 2005). This evolution, a pattern that is moving rather than static, is the electrographic definition of a seizure, and capturing or interpreting it is entirely outside QEEG practice.
Postictal. After the seizure. The postictal EEG typically shows focal or diffuse slowing and attenuation that gradually recovers over minutes to hours, the electrographic correlate of the postictal state (Niedermeyer & Lopes da Silva, 2005). Postictal slowing can itself be localizing, pointing back toward the region of seizure onset.
Two procedural facts shape what a clinical epilepsy EEG looks like and why it differs from your resting acquisition. Activation procedures, hyperventilation and photic stimulation, are standard parts of a clinical epilepsy study because they provoke discharges that a brief resting record would miss. Hyperventilation classically brings out 3 Hz spike-and-wave in absence epilepsy, and photic stimulation can elicit a photoparoxysmal response in photosensitive epilepsies (Niedermeyer & Lopes da Silva, 2005). Sleep and sleep deprivation also raise the yield of interictal discharges, which is why prolonged and sleep-deprived studies exist. Your standard QEEG protocol deliberately avoids these provocations, so a resting brain map is not, and is not meant to be, a seizure-detection study.
The age dimension matters here too. The epilepsy syndromes are strongly age-dependent: infantile spasms with hypsarrhythmia in the first year or two, childhood absence with 3 Hz spike-and-wave, the benign focal epilepsies of childhood with their characteristic centrotemporal discharges, and the focal epilepsies of adulthood with temporal or extratemporal foci (Niedermeyer & Lopes da Silva, 2005). The same scope boundary applies across all of them, and the syndromic classification is the electroencephalographer's.
For QEEG, epilepsy is mostly a scope lesson and an adjunct role. Quantitative methods and source localization have a supporting place in epilepsy, for instance in pre-surgical lateralization of a known focus, but always as an adjunct to clinical EEG read by qualified clinicians, never as a stand-alone diagnostic tool (Niedermeyer & Lopes da Silva, 2005; Pascual-Marqui et al., 1994). If your resting recording shows evolving rhythmic activity rather than resting rhythms, you have potentially captured an ictal pattern, and the only correct action is to stop and refer.
TBI is one of the conditions where QEEG has a genuine, evidence-supported contribution, and it is also where over-claiming is most tempting, especially in forensic contexts the later chapters cover. Hold both facts.
Acute. In the hours and early days after a significant injury, the EEG shows diffuse slowing, an increase in delta and theta with reduced alpha, reflecting global metabolic disruption, and focal delta over a contusion if there is focal structural damage (Niedermeyer & Lopes da Silva, 2005). Severity tracks with the degree of slowing and loss of reactivity. Acute TBI EEG is a clinical study in a clinical setting, not a QEEG mapping exercise.
Subacute and chronic. As metabolic recovery proceeds over days to weeks, the background reorganizes and alpha re-emerges, and in milder injuries this normalizes over weeks to a few months (Thatcher et al., 1989). The QEEG contribution lives here, in the quantitative residue that visual inspection can miss. Thatcher's longitudinal work established that TBI produces reproducible coherence abnormalities, characteristically reduced interhemispheric and long-range coherence, in the alpha and theta bands, reflecting the disruption of long white-matter connections by diffuse axonal injury (Thatcher et al., 1989). These connectivity disturbances can persist after the visual EEG has normalized, which is the basis for QEEG's role in documenting chronic post-concussive dysfunction.
Mild TBI and sports concussion. The hardest and most common case is the mild injury with a normal-looking routine EEG and a normal structural scan, where the patient still reports the cognitive fog, slowed processing, and attentional trouble of post-concussive syndrome. Connectivity research, including work in sports-related concussion, has shown measurable coherence and network disturbances in exactly these patients, the functional residue of diffuse axonal stretch that conventional imaging misses (Thatcher et al., 1989; Lewine et al., 2019). This is the strongest argument for QEEG in TBI and, simultaneously, the place where the evidence is most often overstated, because group-level connectivity differences do not translate cleanly to an individual-level diagnosis, and the discordance between maps and symptoms cuts both ways. State the finding as a quantitative abnormality consistent with the clinical picture, not as proof of injury.
As a prognostic and monitoring marker. Serial QEEG carries more information than a single recording in TBI, because the trajectory of recovery is itself the signal: progressive normalization of coherence over serial recordings suggests a favorable course, while persistent or worsening connectivity disruption beyond several months suggests entrenched reorganization (Thatcher et al., 2003). Thatcher's age-corrected coherence norms, and his demonstration that interpretation must be referenced to chronological age, are the methodological backbone of this application (Thatcher & Lubar, 2008; Thatcher et al., 2003).
The caveats are essential and load-bearing. Coherence is exquisitely sensitive to technical factors, reference montage, electrode impedance, drowsiness, and artifact, so a clean acquisition and consistent parameters across serial recordings are prerequisites, not niceties (Stam et al., 2007). And the relationship between connectivity findings and symptoms is imperfect: patients recover with abnormal maps and have abnormal maps without symptoms, so the QEEG is one input to a clinical formulation, never the formulation itself. The forensic chapter returns to this with the admissibility standards that govern presenting such findings in court.
Across the dementias, the EEG carries a recognizable and well-replicated signature, and QEEG sharpens it. The framing matters: these are group-level patterns with substantial individual overlap, useful as one strand of a workup that also includes clinical assessment, neuropsychology, imaging, and molecular biomarkers, and not a stand-alone diagnostic test (Cassani et al., 2018; Niedermeyer & Lopes da Silva, 2005).
Alzheimer's disease. The canonical QEEG pattern is spectral slowing: increased power in delta and theta, decreased power in alpha and beta, a progressive decline of posterior alpha source activity, and loss of functional cortical connectivity, particularly in posterior and long-range networks (Cassani et al., 2018). The posterior dominant rhythm slows, with the individual alpha frequency dropping below its normal range, and the magnitude of slowing tracks cognitive severity (Babiloni et al., 2014). Babiloni's consortium work, replicated across laboratories, established the posterior-predominant alpha decline and the delta-alpha source ratio as among the more reliable electrophysiological markers across the AD continuum, with the pattern intensifying from mild cognitive impairment through mild to moderate dementia (Babiloni et al., 2014). The key clinical distinction is from normal aging: healthy aging brings modest slowing, but in AD theta increases where in healthy aging it tends to decrease, and reduced alpha power that predicts decline is a disease marker rather than an age effect (Babiloni et al., 2014; Scally et al., 2018).
Several differential-diagnostic refinements are worth recognizing, with the standing caveat that none of them diagnoses a dementia subtype on its own. Posterior alpha reactivity, the normal blocking of alpha when the eyes open, is blunted in AD and blunted more severely in dementia with Lewy bodies, so reduced reactivity can point toward Lewy body pathology when other features fit (Niedermeyer & Lopes da Silva, 2005). Source-space connectivity carries a related signal: interhemispheric coupling in the delta sources runs abnormally high in AD but tends to look more normal in dementia with Lewy bodies and in Parkinson's disease with dementia, a pattern that has been used to discriminate AD from the Lewy body dementias with reasonable accuracy at the group level (Triggiani et al., 2016). Frontotemporal dementia, by contrast, is comparatively less likely to show the prominent posterior slowing of AD early on, which is part of why a relatively well-preserved posterior rhythm in a patient with a clinically frontal or language-led syndrome does not rule the process out (Niedermeyer & Lopes da Silva, 2005). These are tendencies, not tests, and the differential among AD, FTD, and the Lewy body dementias remains a clinical and neuroimaging diagnosis that QEEG can inform but not settle.
Creutzfeldt-Jakob disease. The EEG signature is distinctive: generalized periodic sharp wave complexes recurring at roughly one per second, emerging as the disease progresses, in the setting of rapidly progressive dementia and myoclonus (Niedermeyer & Lopes da Silva, 2005). CJD is one of the few dementias where the EEG pattern itself is diagnostically weighty, which is exactly why recognizing the periodic complexes obligates referral rather than QEEG interpretation. A normative brain map on a patient with CJD periodic complexes is meaningless. The complexes are the finding, and they belong with a neurologist.
For QEEG practice, the AD slowing pattern is also a methodological warning. A diffusely slowed dementia EEG will deviate from a healthy-adult norm at essentially every site and band, which is a true description of the disease and a poor phenotype, and any report has to frame it as consistent-with rather than diagnostic-of.
Psychiatric QEEG is the area of largest commercial interest and weakest specificity, and it is where the practitioner-learner most needs calibrated skepticism. The findings below are real at the group level and modest at the individual level, and stating them as diagnostic markers oversells the evidence.
ADHD. The most-discussed finding is an elevated theta-to-beta ratio, frontocentral theta excess, and reduced alpha reactivity, the pattern that motivated theta-down, beta-up neurofeedback protocols (Monastra et al., 1999). The honest framing is that the theta/beta ratio's diagnostic specificity has been walked back: it separates groups on average but overlaps too much to diagnose an individual, a confound that includes drowsiness and developmental stage, and the field has revised its earlier claims accordingly (Arns et al., 2013). Present elevated theta/beta as a phenotype that informs a workup, never as a test for ADHD.
Schizophrenia. Reported findings include reduced gamma-band activity and gamma synchrony, coherence abnormalities, and increased slow-wave activity associated with negative symptoms (Hughes & John, 1999). These are research-grade group differences with no validated individual-diagnostic application.
Major depression. The most-cited marker is frontal alpha asymmetry, relatively greater left-frontal alpha (interpreted as relatively reduced left-frontal activation) in some depressed samples, the basis of the Davidson lateralization model and of alpha-asymmetry neurofeedback (Luo et al., 2025). The literature is mixed, the effect is small and inconsistent across studies, and it does not function as a diagnostic marker for an individual (Thibodeau et al., 2006).
The unifying caution: psychiatric EEG findings are group-level associations with wide individual overlap, sensitive to state, medication, and vigilance, and they do not diagnose. Their legitimate use is as one input among many, and a QEEG report that calls frontal alpha asymmetry "consistent with depression" has already gone further than the evidence and the scope allow.
Anesthetic depth has a recognizable electrographic progression, and you should understand it both because the patterns recur and because they make the limits of "normal" vivid. The intraoperative monitoring itself is an anesthesiology and neuromonitoring domain, not QEEG, but the depth continuum is instructive.
As a patient goes from awake through light and then deep anesthesia, the EEG progresses in a direction paralleling the descent into sleep and into encephalopathy: from a low-amplitude, fast, alert background, through an early stage of paradoxical fast activity and disinhibition that some agents produce on the way down, into increasing higher-amplitude slow activity, with many agents generating prominent frontal alpha and beta at surgical planes, then deepening toward burst-suppression and, at extreme depth, suppression (Niedermeyer & Lopes da Silva, 2005). Different agents leave different spectral signatures, an anteriorized alpha with one class of drug, a markedly different profile with another, which is why processed-EEG depth monitors and their spectrogram displays are built around agent-specific patterns rather than a single universal index (Niedermeyer & Lopes da Silva, 2005).
The lesson for QEEG is the one anesthesia makes unmissable. The EEG state is set by the pharmacologic and physiologic context, the same alpha frequency can mean alert rest or a surgical plane depending on its reactivity, distribution, and the drug on board, and a spectrum read without that context is read wrong. It is the strongest possible argument for the medication-documentation discipline the pharmacology chapter insists on.
At the severe end of the arousal continuum, the EEG helps characterize states of impaired consciousness, and several of the previous chapter's grave patterns recur here: diffuse slowing graded by depth, burst-suppression, periodic discharges, and the alpha-coma pattern of unreactive, often anteriorly-distributed alpha that mimics waking rhythm while the patient is unresponsive (Niedermeyer & Lopes da Silva, 2005). The single most useful bedside-EEG question across these states is reactivity: whether the background changes at all in response to stimulation, because preserved reactivity carries more favorable implications than an unreactive record (Niedermeyer & Lopes da Silva, 2005).
After cardiac arrest, the EEG is one of the tools used to estimate prognosis, and the features that carry weight are recognizable as a hierarchy from the patterns of the previous chapter: a continuous, reactive background is comparatively reassuring, while suppression, burst-suppression with certain burst morphologies, and unreactive periodic patterns sit at the malignant end (Niedermeyer & Lopes da Silva, 2005). The point of naming this is not to enable you to prognosticate, which you must not do, but to make concrete why these recordings are read by neurologists under defined criteria and why the timing, sedation state, and temperature at the time of recording all change the meaning. The bedside reactivity test recurs as the single most informative qualitative question: does the background change at all when the patient is stimulated.
The clinical taxonomy of disorders of consciousness, coma, the vegetative or unresponsive-wakefulness state, and the minimally conscious state, is defined behaviorally and neurologically, with EEG and evoked-potential studies as supporting evidence, and distinguishing among them is a specialist neurological assessment (Niedermeyer & Lopes da Silva, 2005). It is named here so you recognize the terms and the gravity, and so the boundary is unambiguous: prognostication after anoxic injury and the classification of disorders of consciousness are physician work, governed by separate expertise and protocol, and a normative QEEG has no role in it.
The recurring point lands hardest here: a normative database models a healthy waking adult, and a comatose patient is the furthest thing from one, so the only honest use of the EEG in this setting is qualitative, descriptive, and referred.
The skill this chapter builds is reading the EEG in context, and the context determines both the meaning and your scope. A high-amplitude run of delta is normal slow-wave sleep in a healthy adult, normal discontinuity in a premature newborn, diffuse encephalopathy in a confused elder, and a postictal residue in a patient who just seized, and only the population and state tell you which. For the IQCB exam, Domain IV expects you to name the stage-defining transients of sleep, the maturational landmarks of the developing EEG, the three-phase structure of the epileptic EEG, the slowing signatures of TBI and dementia, and the depth patterns of anesthesia and coma, and to know that several of these are referral findings rather than QEEG calls. In the clinic, the same contextual reading does two concrete things: it keeps drowsiness, development, medication, and disease from being quietly fed into a database that assumed none of them, so the brain map you produce describes a phenotype and not an artifact, and it tells you the moment a recording belongs not on your screen but with a board-certified electroencephalographer.