An EEG (electroencephalogram) records the electrical activity of your brain through small discs stuck to your scalp. It is painless. Its main use is in epilepsy. It can support a diagnosis and help show which type of epilepsy someone has. It is also used for some sleep conditions and for encephalitis, an inflammation of the brain.
The most important thing to know is that a normal EEG does not rule out epilepsy. Many people with epilepsy have normal brain activity between seizures. NICE, which sets guidance in England, is direct about this: "Do not use EEG to exclude a diagnosis of epilepsy." A routine EEG takes about an hour, including set-up. Longer types can run for hours or days.
Brain cells communicate using tiny electrical signals. An EEG picks up the combined pattern of those signals through the skull. A trained specialist, usually a clinical neurophysiologist, reads the trace.
Some patterns point strongly towards epilepsy. These are often called epileptiform discharges, such as spikes or sharp waves. Epilepsy Action explains that the location of unusual activity helps doctors tell focal seizures from generalised ones.
The NHS lists the main uses:
Epilepsy Action adds that EEG can show non-convulsive status epilepticus. This is a prolonged seizure that changes awareness or causes confusion, and it may not look like a seizure at all.
Durations come from Epilepsy Action and the NHS.
Routine EEG. You sit or lie down while electrodes are attached with sticky paste. You may be asked to open and close your eyes, breathe deeply for a few minutes, and look at a flashing light. These are called provoking manoeuvres. They can bring out abnormal activity.
Sleep EEG. Some seizures happen mainly in sleep, and brainwave patterns change in sleep. Epilepsy Action says you may be given medicine to help you sleep.
Sleep-deprived EEG. You may be asked not to sleep at all the night before, or to wake much earlier than usual. Tiredness makes unusual activity more likely to show. The NHS says you will not be able to drive afterwards, so arrange a lift.
Ambulatory EEG. Electrodes connect to a small recorder you carry in a bag. You go about your normal day and night. You keep a diary of activities, sleep, meals and any events, so the team can match them to the trace.
Video EEG (video telemetry). You are filmed while the EEG records, usually over a few days. Epilepsy Action says epilepsy medicine is sometimes reduced to make a seizure more likely to happen during the recording. That is done under close supervision. In England, NICE lists video EEG telemetry among the tests offered by tertiary (specialist) epilepsy services (NICE NG217).
Most EEGs are recorded between seizures. Epilepsy Action puts it plainly: many people with epilepsy only show unusual activity during a seizure. The rest of the time, their trace can look normal.
The numbers bear this out. A 2016 meta-analysis looked at routine EEG after a first unprovoked seizure (Bouma et al., Eur J Neurol 2016). It pooled 15 studies with 1,799 participants. The comparison was whether people went on to have a second seizure within at least a year.
In plain terms, a routine EEG missed most adults who went on to have more seizures. But when an adult's EEG did show epileptiform discharges, it was a fairly strong signal. NICE reached the same view from its own review. It found low sensitivity for routine EEG but specificity high enough for a positive result to support a diagnosis (NICE NG217).
Somewhat. NICE says provoking manoeuvres, sleep recordings and ambulatory EEG could "slightly increase sensitivity". It adds that this is still not enough to exclude epilepsy when the result is negative.
The evidence on sleep deprivation is mixed. A 2024 meta-analysis of 12 studies found no significant overall diagnostic effect for sleep-deprived EEG (Zhang et al., Seizure 2024). Results varied widely between studies. After excluding outliers, the authors saw a trend towards a modest benefit.
Longer recordings may pick up more. In one Canadian clinic study, 100 adults with a first unprovoked seizure had three EEGs each (Hernandez-Ronquillo et al., Neurol Clin Pract 2023). Against the clinical epilepsy diagnosis, sensitivity was 11% for the first routine EEG and 22% for a second. A 24-hour ambulatory EEG reached 72%. Specificity did not differ significantly between the three. This was a single study, so the figures may not apply everywhere.
A 2015 systematic review found ambulatory EEG was more likely to capture events than sleep-deprived EEG (Lawley et al., Epilepsy Behav 2015). It found too little data to compare modern ambulatory EEG with inpatient video telemetry.
Some healthy people have EEG patterns that look epileptic. A UK study screened 13,658 young men applying for Royal Air Force aircrew training. Of these, 0.5% had clear epileptiform discharges (Gregory et al., 1993). None had a history of significant illness. Of 43 followed for 5 to 29 years, only one developed epilepsy.
Epilepsy Action also says misreading an EEG is one of the most common reasons people are wrongly diagnosed with epilepsy. That is why the trace should be read by a specialist, alongside your history.
Epilepsy is a clinical diagnosis. NICE says it starts with a detailed history and examination, ideally with an eyewitness account or video of the event (NICE NG217). NICE also recommends:
The HSE says much the same for Ireland. A specialist may arrange an EEG, an ECG, brain scans, blood tests and genetic tests. It notes that having one seizure does not always mean epilepsy.
Functional (dissociative) seizures. These look like epileptic seizures but are not caused by epileptic brain activity. An international task force called video EEG the "gold standard" for diagnosing them (LaFrance et al., Epilepsia 2013). Because it is not available everywhere, the task force set out levels of diagnostic certainty for when it is not.
Sleep disorders. EEG is one part of an overnight sleep study. Our guide to why narcolepsy diagnosis takes years covers the sleep tests used there. Our page on parasomnias explains why REM sleep behaviour disorder needs a sleep study.
The NHS suggests:
Keep taking your usual medicines unless the team tells you otherwise. Bring a list of them. If someone has seen your episodes, bring them too, or bring a phone video. Epilepsy Action notes the paste can be washed out with baby shampoo, baby oil or conditioner.
You can ask for a chaperone. The NHS says this can be someone you know, a nurse or a trained staff member.
The recording itself cannot. Epilepsy Action says there is a very small risk of a seizure during the flashing lights or deep breathing. Staff watch you closely. NICE says the benefits and risks of these manoeuvres should be discussed with you first. If you are worried, ask whether they can be left out. Sleep deprivation can also raise seizure risk around the time of the test.
If a seizure during testing could affect your driving licence, ask your doctor before the test.
The NHS says results may take a few days or weeks, because a specialist needs to analyse them. You may need a follow-up appointment. If you have heard nothing after a few weeks, contact the doctor who referred you.
A report usually describes background rhythms and any abnormal activity, and where in the brain it appeared. Ask your doctor what it means alongside your history. You may need further EEGs.
In Ireland. EEGs are usually arranged by a specialist after a referral from your GP or the emergency department. The HSE describes EEG as one of the tests a specialist doctor may do after a seizure.
In the UK. NICE recommends urgent referral after a first suspected seizure, for an appointment within two weeks. The specialist decides whether an EEG is needed.
In the US. EEG is ordered by a doctor, usually a neurologist. We have not checked US access or costs.
If you live with epilepsy, our articles on the ketogenic diet and on St John's wort and epilepsy medication cover treatment questions EEG cannot answer.
No. The NHS says it is painless. Electrodes are stuck on with paste, and your hair may feel sticky afterwards.
No. NICE says EEG should not be used to rule out epilepsy. In adults after a first seizure, routine EEG missed most people who went on to have more seizures. Your doctor will weigh your history, any witness accounts and other tests.
Not necessarily. A small number of people without epilepsy have unusual EEGs. The result has to be read alongside what actually happened during your episodes.
Video EEG, recording an actual event, is considered the gold standard for this. A routine EEG between events usually cannot tell them apart.
No. EEG is not used to diagnose ADHD, autism, depression or anxiety. It is mainly used for epilepsy, some sleep conditions and encephalitis.
This article is general information, not medical advice. It does not replace advice from your GP, neurologist or EEG department. Do not change any medicine before a test unless your doctor tells you to.
Reviewed by: Proco Team
Last updated: November 2026