Every clean recording during an eeg test begins with the same battle: separating real cortical signals from the dozens of non-cerebral voltages that contaminate the tracing. These contaminants are called EEG artifacts, and they are arguably the single most important concept a technologist, neurologist, or polysomnographer must master. An artifact is any electrical potential recorded by the scalp electrodes that does not originate from the brain itself, whether it comes from the patient's body, the electrodes, the amplifier, or the surrounding environment.

Understanding what is an eeg test really measures helps explain why artifacts are unavoidable. Scalp electrodes pick up microvolt-level signals, typically 10 to 100 microvolts, while nearby muscle tissue can generate millivolt-level activity that is a thousand times stronger. The ambient electromagnetic field of a hospital, complete with fluorescent lights, infusion pumps, and elevators, easily dwarfs the alpha rhythm you are trying to capture. Recognizing these intruders is half the diagnostic skill.

Artifacts fall into two broad families. Physiologic artifacts come from the patient — eye movements, muscle contractions, sweat, heartbeats, tongue movement, and respiration. Extraphysiologic artifacts come from outside the patient, including 60 Hz line frequency interference, electrode pops, IV drips, cell phones, and even loose lead wires swinging during transport. Each family has a distinct fingerprint on the tracing, and seasoned readers can usually identify the source within a single page.

The clinical stakes are real. A misread muscle artifact can mimic a fast spike-and-wave discharge, leading to an incorrect epilepsy diagnosis and unnecessary anticonvulsant therapy. A poorly recognized electrode pop has fooled experienced readers into reporting a focal sharp wave. Conversely, dismissing a true epileptiform discharge as artifact delays treatment and exposes patients to ongoing seizure risk. This is why the ACNS guidelines devote entire chapters to artifact recognition.

This guide walks through the major artifact categories you will encounter in a clinical lab, the physical mechanisms behind each one, and the practical steps you can take at the bedside to minimize them. We will cover impedance checks, electrode preparation, filter settings, montage selection, and patient coaching. Whether you are studying for the ABRET R. EEG T. credential or running a busy ICU continuous monitoring service, artifact mastery is non-negotiable.

We will also touch on cost and logistics, because patients frequently ask about the eeg test experience itself — how long it takes, what it costs, and whether artifacts mean their study will need to be repeated. Spoiler: a well-prepared technologist eliminates most artifacts before the recording even begins, and that preparation is what separates a diagnostic-grade study from a wasted hour.

By the end of this article, you should be able to look at a 10-second page of EEG, point to the dominant artifact, name its likely source, and describe two interventions to reduce it. That single skill will improve your read accuracy more than any other piece of training.

Physiologic artifacts originate from the patient and are, by definition, biological signals — just not cerebral ones. Because they come from a living, breathing, blinking human, they cannot be completely eliminated. Instead, the goal is to recognize them quickly so they are not mistaken for pathology. Anyone learning what is an eeg test involves should spend significant time studying these patterns, because they appear in nearly every recording you will ever perform or interpret.

Eye movement artifact is the most common and most misread. Each eye behaves like a small battery with the cornea positive and the retina negative. When the eyes look up or blink, the positive cornea moves toward the frontal electrodes, producing a large downward (positive) deflection at Fp1 and Fp2. Lateral eye movements produce opposite deflections at F7 and F8, the classic out-of-phase lateral rectus pattern. Placing dedicated eye-monitor leads makes identification trivial.

Muscle artifact is generated by motor unit firing in scalp and facial muscles. The temporalis and frontalis are the worst offenders because they sit directly under the standard 10-20 array. Muscle spikes are brief, high-frequency, and irregular, often resembling polyspikes but lacking the rhythmic build-up of true epileptiform activity. Asking the patient to relax the jaw, drop the shoulders, and open the mouth slightly resolves most cases within seconds.

Cardiac artifact takes two forms. The first is direct ECG contamination, where the QRS complex bleeds into scalp electrodes, particularly in patients with short, thick necks or those lying on one side. The second is pulse artifact, a slow rhythmic wave produced when an electrode rests directly over a scalp artery. Both are confirmed by lining up the EEG channel with the simultaneous ECG trace, which every modern lab records as a matter of routine.

Sweat artifact appears as very slow undulating baseline drift, usually below 0.5 Hz, that affects multiple adjacent electrodes. It happens when perspiration creates a salt bridge between gel pools or when sodium chloride in sweat alters electrode-skin potential. Cooling the room, drying the scalp, and switching to a sintered Ag/AgCl electrode reduces it dramatically. A high-pass filter of 1 Hz can clean the display but may distort genuine delta activity.

Glossokinetic (tongue) artifact is the surprise package. The tongue is also a dipole, and movements like swallowing or speaking generate slow frontal-dominant deflections that can mimic frontal intermittent rhythmic delta activity, or FIRDA. Asking the patient to say the word lilt while you watch the tracing reproduces the artifact instantly and confirms its source. Note it in the technologist log.

Respiration produces subtle, slow oscillations synchronized with breathing. In ventilated patients, the rhythmic chest expansion can mechanically tug on electrode wires, creating a regular slow wave across temporal or occipital chains. The fix is to secure the wires with tape, route them away from the chest, and document the ventilator rate so the reader can correlate frequencies.

Filter settings and montage selection are the two software-side tools every reader uses to manage artifacts. Used correctly, they reveal the underlying cerebral activity. Used carelessly, they can hide pathology or create artifact patterns that did not exist in the raw data. Trainees often ask what is eeg test filtering supposed to accomplish, and the honest answer is: it shapes what you see without changing what was actually recorded.

The low-frequency filter, also called the high-pass filter, removes slow drift below a set cutoff. The standard routine EEG setting is 1 Hz, which eliminates sweat and respiration artifact while preserving delta activity down to about 1.5 Hz. For neonatal recordings, the cutoff is dropped to 0.3 Hz or lower to preserve genuine slow activity. Setting the filter too high cuts real delta and theta, while setting it too low lets baseline drift dominate the page.

The high-frequency filter, or low-pass filter, removes fast activity above a set cutoff. The default is usually 70 Hz, which preserves all clinically relevant brain frequencies while removing the worst muscle artifact above that band. Aggressive 35 Hz or 15 Hz filters dramatically clean up muscle but also distort spike morphology, turning sharp transients into rounded blobs. Use them sparingly and document any changes from the standard.

The notch filter targets the 60 Hz line frequency specifically. Modern digital systems implement it as a narrow band-stop centered at 60 Hz, with a width of one or two Hz. It should be the last resort, applied only after physical sources have been addressed. Engaging the notch by default trains technologists to ignore correctable problems, and it can distort fast gamma activity in research recordings.

Montage selection changes how channels are derived without changing the raw data. The bipolar longitudinal (double banana) montage compares adjacent electrodes and is excellent for localizing focal abnormalities through phase reversal. The referential montage compares each electrode to a common reference (usually averaged ears or Cz) and is better for seeing absolute amplitudes and diffuse abnormalities. Switching montages during review is a powerful artifact-detection technique.

The Laplacian or source derivation montage subtracts the average of surrounding electrodes from each channel, emphasizing local activity and suppressing distant signals. It is excellent for separating a true focal spike from a distant generator like ECG bleed-through. It is also useful for highlighting electrode pops, which become exaggerated because they have no surrounding field to subtract.

The average reference montage uses the average of all electrodes as the reference. It works well for diffuse abnormalities but can spread a single high-amplitude artifact across every channel, creating the illusion of widespread pathology. Always confirm findings in at least two different montages before reporting them, especially when the original montage shows something dramatic at only one electrode.

Documenting artifacts is not optional paperwork — it is a clinical skill that directly affects patient outcomes. The reading neurologist was not in the room when the recording happened, so the technologist's notes are the only record of what was actually going on with the patient and equipment. A few extra minutes of documentation can prevent a misdiagnosis, save a repeat study, and protect the lab from liability. Many patients ask about eeg test price when a repeat is ordered, and avoidable repeats are a real cost driver.

Standard documentation includes the patient's state of arousal (awake, drowsy, asleep, agitated), any medications that may affect the recording, the room environment (lights on or off, monitors running), and a chronological log of activation procedures. Every artifact-generating event — a cough, a swallow, a position change, a nurse entering the room — should be marked in real time with a brief annotation on the recording itself, not written later from memory.

When unusual patterns appear, the technologist should attempt to reproduce or eliminate them at the bedside. If a sharp transient appears in T4, gently tap the electrode and watch the screen. If it changes amplitude or polarity, it is an electrode artifact. If a rhythmic pattern appears, ask the patient about the activity at that exact moment. Was she chewing gum? Was the IV pump beeping? Was someone using a cell phone in the next room?

For continuous ICU recordings, artifact documentation is even more critical because the bedside team is constantly changing. A shift-change note that reads ventilator-induced movement artifact, channels T3-T5, periodic at 12 breaths per minute saves the reading neurologist from incorrectly reporting periodic lateralized epileptiform discharges. The clinical impact of that distinction can be hours of unnecessary anticonvulsant escalation.

Artifact reduction techniques should also be documented. If you re-prepped an electrode, switched montages, or engaged a notch filter, note the exact time and the reason. Modern digital systems timestamp every filter change automatically, but the reason still has to come from a human. A note that simply reads engaged 60 Hz notch at 14:32 due to confirmed line noise from broken IV pump in adjacent bay is gold standard.

Photographs are increasingly accepted as part of the record. A quick phone photo of the room setup, with permission, can help the reader understand spatial relationships between the patient, equipment, and environmental sources of interference. Save the images to the patient record per your institution's HIPAA-compliant workflow. Some labs have begun including bedside video for ambulatory studies, which is the gold standard for behavioral and movement artifact correlation.

Finally, communicate proactively with the reading neurologist about anything unusual. A quick message that reads I want to flag possible breach rhythm versus mu rhythm in left central region — please review with the bedside video at 11:42 saves the reader time and increases diagnostic accuracy. Strong technologist-reader communication is one of the defining features of a high-performing EEG lab.

Practical artifact troubleshooting at the bedside follows a predictable workflow that every technologist eventually internalizes. When something looks wrong on the screen, do not freeze the recording — work the problem live so you can verify the fix in real time. Start by identifying whether the artifact is affecting one channel, a region, or the entire montage. Single-channel artifact almost always points to an electrode issue, while diffuse artifact points to the patient or environment.

For single-channel problems, the first move is to re-check impedance on that specific lead. If it has climbed above 10 kΩ, re-prep the site with abrasive paste, refresh the conductive gel, and re-secure the electrode. If impedance reads zero or shows as a salt bridge with an adjacent electrode, you have gel runoff between sites — clean both electrodes and re-prep with less paste. Persistent single-channel pops after re-prepping suggest a damaged electrode disc or a frayed lead wire that needs replacement.

For regional artifacts affecting two to four adjacent electrodes, look first for muscle activity. Ask the patient about jaw tension, neck position, or shoulder tightness, and try gentle repositioning. Frontal-dominant artifacts in eyes-closed segments are almost always residual eye movement or roving eye drift; ask the patient to keep eyes relaxed without forcing them closed. Temporal artifacts in patients chewing gum or swallowing frequently resolve with a sip of water and a verbal reminder.

For diffuse artifacts that affect the entire montage, suspect the environment or the ground electrode. Step through a mental checklist: Is the ground securely attached? Is there new equipment in the room? Has anyone opened a door to a hallway with bright fluorescent lights? Is a cell phone charging on the bedside table? Resolving environmental artifacts often requires walking around the room with the recording running, watching the screen as you unplug or relocate one device at a time.

During activation procedures, expect predictable artifacts and prepare for them. Hyperventilation produces deep breathing, which mechanically moves wires and may pull at electrodes. Photic stimulation produces a photomyogenic response in some patients — small muscle twitches synchronized with the flash — which can be mistaken for a photoparoxysmal response. Documenting the patient's facial appearance during photic stimulation, ideally with video, helps the reader make this distinction confidently.

Sleep recordings introduce their own artifact spectrum. As patients drift into stage N2 and N3 sleep, the EMG drops dramatically and artifacts often clean up substantially. However, snoring, mouth breathing, and position changes generate new artifacts that did not exist during the awake baseline. Document each position change and note any prolonged movement, because sleep-related movement artifact can sometimes mimic sleep myoclonus or other movement disorders that require neurology follow-up.

The final practical tip: build a personal artifact library. Save 10-second clips of common artifacts as you encounter them, with annotations explaining what you saw and what fixed it. After a year of clinical work, you will have a personalized atlas that beats any textbook, because every example comes from real patients you remember. Trade clips with colleagues and you will all become faster, better readers.