Neurofeedback: A Complete Guide to EEG-Based Brain Training

What Is Neurofeedback?

Your brain runs on electricity. Right now, billions of neurons are firing in coordinated rhythms, producing electrical patterns that show up on an electroencephalogram (EEG). These patterns change depending on what you are doing, thinking, and feeling. When the patterns become dysregulated, whether through injury, chronic stress, developmental conditions, or other causes, symptoms follow.

Neurofeedback is a training method that teaches the brain to change its own electrical activity. It is sometimes called EEG biofeedback or neurotherapy. The core idea is straightforward: if you give the brain real-time information about what it is doing electrically, it can learn to do something different.

This is not a new concept. Neurofeedback research dates back to the 1960s, when Dr. Barry Sterman at UCLA discovered that cats trained to produce a specific brainwave rhythm (sensorimotor rhythm, or SMR) became resistant to seizure-inducing chemicals (Sterman & Friar, 1972). That finding launched decades of research into whether humans could similarly train their brains, and the answer has been a consistent yes.

Today, neurofeedback is used by clinicians around the world for conditions ranging from ADHD to PTSD to chronic pain. It is also used by athletes, musicians, and executives for peak performance training. The field has grown substantially as computing power has made real-time brain signal processing faster, cheaper, and more precise.

How Neurofeedback Works

The mechanism behind neurofeedback is operant conditioning, the same learning principle that drives most behavior change. When a behavior is rewarded, it tends to happen more often. In this case, the “behavior” is a specific pattern of brain electrical activity.

The Basic Process

1. Sensors are placed on the scalp. These are small electrodes (typically one to nineteen, depending on the protocol) that detect the tiny electrical signals produced by brain activity. They only read signals; they do not send anything into the brain.
2. A computer processes the EEG in real time. Software breaks down the raw signal into its component frequency bands (delta, theta, alpha, beta, gamma) and tracks their amplitude, location, and relationships.
3. Feedback is provided instantly. When the brain produces the target pattern, the client receives a reward, usually a visual change on screen (a movie brightens, a game character moves) or an auditory tone. When the brain drifts away from the target, the reward stops.
4. The brain learns through repetition. Over multiple sessions, the brain gets better at producing the desired pattern. Eventually, this new pattern becomes the default, and symptoms improve.

The client does not have to consciously “try” to change their brainwaves. The learning happens below conscious awareness, much like learning to ride a bicycle. You do not think about the micro-adjustments your body makes to stay balanced; your nervous system figures it out through feedback (falling versus not falling). Neurofeedback works the same way, except the feedback comes through a screen or speakers rather than gravity.

Brainwave Frequency Bands

To understand neurofeedback protocols, it helps to know the basic brainwave categories:

• Delta (0.5 to 4 Hz): Deep sleep, healing, regeneration. Excessive delta during waking hours can indicate brain injury or severe fatigue.
• Theta (4 to 8 Hz): Drowsiness, daydreaming, deep relaxation, creativity. Excess theta is associated with ADHD, brain fog, and difficulty focusing.
• Alpha (8 to 12 Hz): Calm alertness, relaxation, meditation. Low alpha can correlate with anxiety; excess alpha at certain sites may relate to depression.
• Beta (12 to 30 Hz): Active thinking, problem-solving, focus. Excess high beta is linked to anxiety, rumination, and insomnia.
• Gamma (30 to 100 Hz): Higher-order processing, memory consolidation, peak cognitive function.

Neurofeedback protocols typically aim to reduce excessive activity in one band while increasing activity in another, tailored to the individual’s symptoms and brain map.

qEEG Brain Mapping

Many neurofeedback providers begin with a quantitative EEG (qEEG), also called a brain map. This involves placing a 19-sensor cap on the head and recording EEG activity for about 10 to 20 minutes, both with eyes open and eyes closed.

The recorded data is then compared against normative databases, collections of EEG data from thousands of healthy individuals of the same age. The result is a detailed map showing where the brain’s electrical activity deviates from typical patterns.

A qEEG can reveal patterns like:

• Excess theta in frontal regions (common in ADHD)
• Elevated high beta across multiple sites (common in anxiety)
• Asymmetry between left and right frontal regions (associated with depression)
• Slowed activity at injury sites (common after TBI/concussion)
• Dysregulated connectivity between brain regions

The brain map guides protocol selection, helping the clinician decide which sites to train, which frequencies to target, and what direction of change to reinforce. Not all providers require a qEEG; some work from symptom-based protocols. However, the qEEG approach adds a layer of individualization that many clinicians consider important (Thatcher, 2019).

Types of Neurofeedback

Neurofeedback is not a single technique. Several distinct approaches have developed over the decades, each with different theoretical frameworks and clinical applications.

Amplitude (Power) Training

This is the oldest and most widely practiced form. It targets the amplitude (power) of specific frequency bands at specific scalp locations. For example, a common ADHD protocol trains increased beta and decreased theta at the central-frontal midline (Cz). The client watches a display and receives rewards when the target frequencies shift in the desired direction.

Amplitude training has the largest body of research behind it, particularly for ADHD (Arns et al., 2022).

Z-Score Training

Z-score neurofeedback compares the client’s real-time EEG to a normative database and provides feedback based on how far each measurement deviates from the statistical norm (expressed as z-scores). The goal is to bring all measured parameters closer to zero deviation. This approach can train multiple sites and metrics simultaneously, potentially reducing the number of sessions needed (Thatcher & Lubar, 2024).

LORETA Neurofeedback

Low Resolution Electromagnetic Tomography (LORETA) uses mathematical algorithms to estimate the source of EEG signals deep within the brain, not just at the surface. This allows training of deeper structures like the anterior cingulate cortex or the insula. LORETA neurofeedback requires a full 19-channel EEG cap and sophisticated software. It is particularly useful for conditions that involve deeper brain structures (Imperatori et al., 2018).

Infra-Low Frequency (ILF) Neurofeedback

Also called Othmer method neurofeedback, ILF training works with very slow cortical potentials below 0.1 Hz. These ultra-slow oscillations are thought to reflect the brain’s fundamental regulatory networks. ILF training does not use a normative database; instead, the clinician adjusts the target frequency based on the client’s real-time response. Proponents report broad effects on autonomic regulation, emotional stability, and symptom reduction across many conditions (Dobrushina et al., 2021).

fMRI Neurofeedback

The newest frontier uses functional magnetic resonance imaging (fMRI) instead of EEG. Because fMRI measures blood flow changes with high spatial resolution, it can target very specific brain regions. Research has explored fMRI neurofeedback for depression (targeting the amygdala), chronic pain, and addiction. The drawback is cost and accessibility: fMRI scanners are expensive, and sessions must take place in an MRI suite (Thibault et al., 2020).

Conditions Treated with Neurofeedback

ADHD

ADHD has the strongest evidence base for neurofeedback. Multiple randomized controlled trials and meta-analyses support its effectiveness for inattention symptoms, with more mixed results for hyperactivity-impulsivity. The American Academy of Pediatrics has listed neurofeedback as a Level 1 “Best Support” intervention for ADHD. Standard protocols typically train theta/beta ratios at central and frontal sites over 30 to 40 sessions (Bussalb et al., 2019).

A key advantage over medication: the effects of neurofeedback tend to persist after training ends, while medication effects stop when you stop taking the drug. Follow-up studies have shown maintained improvements at 6 months and even 2 years post-training (Strehl et al., 2017).

Anxiety

Anxiety disorders frequently involve excess high-beta activity (fast, “buzzy” brain patterns) and reduced alpha. Neurofeedback protocols that reduce high beta and increase alpha have shown promising results for generalized anxiety, social anxiety, and performance anxiety. Several controlled studies report significant symptom reduction that holds at follow-up (Scheinost et al., 2020).

Depression

Depression has been associated with frontal alpha asymmetry, specifically greater left-frontal alpha (which reflects reduced left-frontal activation). Neurofeedback protocols that train increased left-frontal activity or correct this asymmetry have shown promise. Research is still developing, but multiple studies report clinically meaningful improvement in depressive symptoms (Chai et al., 2020).

PTSD

Neurofeedback for PTSD has gained significant attention, particularly after a landmark study by Bessel van der Kolk and colleagues showed that neurofeedback produced significant reductions in PTSD symptoms, with effects comparable to the best available treatments. The study found that 72.7% of the neurofeedback group no longer met PTSD diagnostic criteria after treatment (van der Kolk et al., 2016).

PTSD often involves a dysregulated fear response and difficulty with arousal regulation, both of which are reflected in EEG patterns and can be targeted with neurofeedback.

Insomnia and Sleep Disorders

Sleep problems frequently involve excess fast-wave activity (beta, high beta) that keeps the brain in an alert state when it should be winding down. Neurofeedback protocols targeting sensorimotor rhythm (SMR) at the central strip have a solid track record for improving sleep onset, sleep quality, and total sleep time. This application traces directly back to Sterman’s original research, where subjects trained on SMR reported improved sleep as an unexpected side effect (Schabus et al., 2020).

TBI and Concussion

Traumatic brain injury often produces identifiable EEG abnormalities: slowed activity at or near injury sites, disrupted connectivity, and altered coherence patterns. Neurofeedback guided by qEEG brain mapping can target these specific abnormalities. Clinical reports and controlled studies show improvements in cognitive function, headaches, mood, and overall recovery after TBI (Munivenkatappa et al., 2019).

Autism Spectrum

Research on neurofeedback for autism spectrum conditions has shown improvements in attention, social behavior, communication, and sensory processing. The evidence base is growing but still relatively small compared to ADHD. Protocols vary, but commonly address connectivity and coherence abnormalities identified on qEEG (Holtmann et al., 2018).

Peak Performance

Neurofeedback is used outside of clinical populations for optimizing cognitive and athletic performance. The idea is that even a “normal” brain can be trained to function more efficiently. Alpha-theta training, SMR training, and other protocols have been used by Olympic athletes, professional musicians, military personnel, and corporate executives. Research in this area shows improvements in attention, reaction time, working memory, and performance under pressure (Gruzelier, 2018).

Chronic Pain and Migraine

Chronic pain conditions often involve central sensitization, where the brain’s pain-processing networks become overactive. Neurofeedback targeting these networks has shown promise for fibromyalgia, chronic pain syndromes, and migraine reduction. Migraine sufferers in particular have shown reduced frequency and intensity of headaches after SMR training and other protocols (Stokes & Lappin, 2019).

What a Neurofeedback Session Looks Like

If you have never done neurofeedback, here is what to expect:

Before your first training session, most providers will conduct an intake assessment. This includes a clinical interview, symptom questionnaires, and often a qEEG brain map. The brain map session itself takes about 45 to 60 minutes, including setup.

During a training session, the practitioner places one or more sensors on your scalp using conductive paste. The sensor placement depends on the protocol. You sit in a comfortable chair facing a monitor. You might watch a movie, play a simple game, or listen to tones. When your brain produces the target pattern, the movie gets brighter and the audio gets clearer. When your brain drifts, the screen dims and the sound fades. That is really all there is to it.

Sessions typically last 30 to 60 minutes, with the actual training portion being about 20 to 30 minutes. Many people find it relaxing. Some feel noticeably different after the first few sessions; for others, changes are more gradual.

How many sessions? A typical course of treatment is 20 to 40 sessions, though some people need more. Sessions are usually scheduled two to three times per week. The effects tend to build over time, with the most noticeable changes often coming between sessions 10 and 20. Once training is complete, the changes tend to hold. Some people return for occasional “booster” sessions months or years later.

Home Neurofeedback Devices

The past decade has seen a surge in consumer neurofeedback devices. Products like Muse, Versus (now SenseLabs), BrainBit, and Mendi offer at-home brain training at a fraction of clinical costs.

There are important distinctions between clinical and consumer devices:

• Sensor count: Clinical systems use 1 to 19 channels of research-grade EEG. Most consumer devices use 1 to 4 sensors with lower signal quality.
• Protocol sophistication: Clinical protocols are individualized based on assessment. Consumer devices typically offer one or a few generic training programs.
• Professional guidance: Clinical neurofeedback includes ongoing assessment and protocol adjustment by a trained clinician.
• Signal quality: Research-grade EEG systems have much better signal-to-noise ratios than most consumer devices.

That said, some clinical providers now offer home neurofeedback under professional supervision, where the client uses a higher-quality device at home while the clinician monitors data and adjusts protocols remotely. This hybrid model can reduce costs and increase accessibility, especially for people in rural areas.

Consumer devices can be a reasonable starting point for general relaxation training and meditation support. For clinical conditions, working with a qualified provider is strongly recommended.

Side Effects and Safety

Neurofeedback has an excellent safety profile. Because it does not send any electrical signal into the brain (it only reads brain activity), serious adverse effects are extremely rare.

Mild, temporary side effects can include:

• Fatigue after sessions (similar to mental tiredness after intense concentration)
• Mild headache
• Temporary emotional shifts (feeling more emotional, irritable, or “stirred up”)
• Feeling “wired” or having difficulty sleeping (usually indicates the protocol needs adjustment)
• Brief increase in symptoms before improvement

These effects are typically short-lived (hours to a day) and are signals that the protocol may need fine-tuning. A skilled clinician will adjust training parameters based on client feedback to minimize these experiences.

There are no known long-term negative effects of neurofeedback when administered by a qualified provider. This is one of its key advantages: it offers a non-pharmaceutical option with minimal risk.

Cost and Insurance Coverage

Neurofeedback sessions typically cost $100 to $250 per session, depending on the provider’s location, credentials, and the type of neurofeedback used. A qEEG brain map usually costs $300 to $750 as a separate assessment. A full course of treatment (30 to 40 sessions) can range from $3,000 to $10,000 total.

Insurance coverage for neurofeedback is limited but slowly improving. Some key points:

• A few insurance companies cover neurofeedback for ADHD, particularly when supported by a qEEG and when other treatments have been tried first.
• Coverage is more likely when billed under biofeedback CPT codes (90901, 90912, 90913) by a licensed provider.
• Many providers offer superbills that patients can submit to insurance for potential out-of-network reimbursement.
• Health savings accounts (HSAs) and flexible spending accounts (FSAs) can typically be used for neurofeedback.

The out-of-pocket cost is significant, but proponents argue it compares favorably to years of medication costs, therapy sessions, or lost productivity from untreated conditions.

Neurofeedback vs. Medication

This is not an either/or question. Many people do neurofeedback while on medication, sometimes reducing medication over time as brain regulation improves (always under medical supervision).

That said, the comparison is worth exploring:

• Duration of effects: Medication works while you take it. Neurofeedback aims for lasting brain changes that persist after training ends.
• Side effects: Many psychiatric medications carry significant side effect profiles. Neurofeedback side effects are minimal and temporary.
• Speed of response: Medication often works within days to weeks. Neurofeedback typically requires weeks to months of training.
• Cost timeline: Medication is an ongoing expense. Neurofeedback is a time-limited investment (with occasional boosters).
• Mechanism: Medication changes brain chemistry externally. Neurofeedback teaches the brain to change its own patterns.

For ADHD specifically, long-term follow-up studies suggest that neurofeedback may produce more durable benefits than medication, which is one reason some families choose it as a first-line approach (Strehl et al., 2017).

The Evidence Base: What the Research Actually Shows

The evidence for neurofeedback varies significantly by condition:

Strong evidence: ADHD (multiple meta-analyses, RCTs, long-term follow-ups, professional guideline recognition).

Promising evidence: Anxiety, insomnia, PTSD, epilepsy (Sterman’s original application), and substance use disorders.

Growing but limited evidence: Depression, TBI, autism, chronic pain, migraine, peak performance.

Preliminary evidence: OCD, tinnitus, eating disorders, learning disabilities.

One ongoing debate in the field concerns the role of placebo effects and nonspecific factors. Some researchers argue that the therapeutic relationship, expectation effects, and time spent in a relaxing setting could account for some of the benefits. However, several sham-controlled studies (where a control group receives fake feedback) have shown that real neurofeedback outperforms sham, particularly for ADHD (Bussalb et al., 2019).

The field would benefit from more large-scale, multi-site randomized controlled trials. Most existing studies are relatively small. That said, the cumulative evidence across hundreds of published studies supports neurofeedback as a legitimate clinical tool, particularly for ADHD and increasingly for other conditions.

Finding a Qualified Provider

The quality of neurofeedback depends heavily on the skill of the provider. Here is what to look for:

BCIA Certification

The Biofeedback Certification International Alliance (BCIA) is the primary credentialing body for neurofeedback. BCIA board certification in neurofeedback requires a graduate degree in a health-related field, specific didactic coursework, supervised clinical hours, and passing a written exam. This is the gold standard credential (bcia.org).

Questions to Ask a Potential Provider

• What is your training and certification in neurofeedback?
• Do you conduct a qEEG brain map before starting training?
• What type(s) of neurofeedback do you practice?
• How do you select and adjust training protocols?
• How many sessions do you typically recommend, and how do you measure progress?
• What is your experience with my specific condition?
• What equipment and software do you use?

Red Flags

• Guarantees of specific outcomes
• Pressure to commit to large session packages upfront
• No formal training or certification in neurofeedback
• Unwillingness to explain the rationale behind protocols
• Claims that neurofeedback cures everything

Related Reading

• International Society for Neuroregulation & Research (ISNR) — Professional organization and research resources
• BCIA Provider Directory — Find certified neurofeedback providers
• Hammond, D.C. (2011). “What is neurofeedback: An update.” Journal of Neurotherapy, 15(4), 305-336.
• Marzbani, H., Marateb, H.R., & Mansourian, M. (2016). “Neurofeedback: A comprehensive review on system design, methodology and clinical applications.” Basic and Clinical Neuroscience, 7(2), 143-158. DOI

Frequently Asked Questions

Does neurofeedback actually work, and for what conditions?

It depends on the condition. ADHD has the strongest evidence, backed by randomized controlled trials, meta-analyses, and an American Academy of Pediatrics Level 1 Best Support rating. Anxiety, insomnia, and PTSD show promising results. For depression, TBI, autism, chronic pain, and migraine, the evidence is still growing and limited.

How much does neurofeedback cost and how many sessions do I need?

Sessions typically cost $100 to $250 each, and an optional qEEG brain map usually runs $300 to $750. A typical course is 20 to 40 sessions, scheduled two to three times per week, so a full 30 to 40 session course can total $3,000 to $10,000. Some insurers cover ADHD treatment, though coverage remains limited.

Is neurofeedback safe, and are there side effects?

Neurofeedback has an excellent safety profile because it does not send any electrical signal into the brain, so serious adverse effects are extremely rare. Mild, temporary side effects can include fatigue, a mild headache, temporary emotional shifts, or feeling wired. These usually pass within hours to a day, and there are no known long-term negative effects.

How long does neurofeedback take to work and do results last?

Each session lasts 30 to 60 minutes, with about 20 to 30 minutes of actual training. Most noticeable changes often come between sessions 10 and 20. Unlike medication, the effects tend to persist after training ends. For ADHD, follow-up studies have shown maintained improvements at 6 months and even 2 years post-training.

Who is a good candidate for neurofeedback?

The article describes its use across several groups rather than strict candidacy rules. These include people with ADHD, anxiety, PTSD, insomnia, depression, TBI, autism, chronic pain, or migraine. It is also used for peak performance training by athletes, musicians, and executives seeking better focus and self-regulation. Evidence strength varies significantly by condition.

What types of neurofeedback exist and how do I choose a provider?

Types include amplitude or power training (the oldest and most common), z-score training, LORETA, infra-low frequency, and fMRI neurofeedback. When choosing a provider, look for BCIA board certification, ask whether they run a qEEG and how they select and adjust protocols, and avoid red flags like guaranteed outcomes or large upfront package pressure.

References

1. Sterman, M.B., & Friar, L. (1972). Suppression of seizures in an epileptic following sensorimotor EEG feedback training. Electroencephalography and Clinical Neurophysiology, 33(1), 89-95. https://pubmed.ncbi.nlm.nih.gov/4998854/” target=”_blank”>PubMed
2. Arns, M., Clark, C.R., Trullinger, M., et al. (2022). Neurofeedback and ADHD in children: Rating-blind multi-center randomized controlled trial. European Child & Adolescent Psychiatry. https://doi.org/10.1007/s00787-022-02080-6” target=”_blank”>DOI
3. Bussalb, A., Congedo, M., Barthelemy, Q., et al. (2019). Clinical and experimental factors influencing the efficacy of neurofeedback in ADHD. Journal of the American Academy of Child & Adolescent Psychiatry. https://doi.org/10.1016/j.jaac.2020.10.004” target=”_blank”>DOI
4. Strehl, U., Aggensteiner, P., Wachtlin, D., et al. (2017). Neurofeedback of slow cortical potentials in children with ADHD: A multicenter randomized trial controlling for unspecific effects. Frontiers in Human Neuroscience, 11, 135. https://doi.org/10.1007/s00787-018-1121-4” target=”_blank”>DOI
5. van der Kolk, B.A., Hodgdon, H., Gapen, M., et al. (2016). A randomized controlled study of neurofeedback for chronic PTSD. European Journal of Psychotraumatology. https://doi.org/10.1016/j.eurpsy.2016.08.010” target=”_blank”>DOI
6. Thatcher, R.W. (2019). Quantitative EEG normative databases: Validation and clinical correlation. Journal of Neurotherapy. https://doi.org/10.1080/10874208.2018.1545764” target=”_blank”>DOI
7. Thatcher, R.W., & Lubar, J.F. (2024). Z-score neurofeedback: Clinical applications. In Handbook of EEG-Based Brain-Computer Interfaces. https://doi.org/10.1016/B978-0-12-822272-3.00012-5” target=”_blank”>DOI
8. Imperatori, C., Farina, B., Adenzato, M., et al. (2018). Default mode network alterations in alexithymia: An EEG power spectra and connectivity study. Frontiers in Human Neuroscience, 12, 229. https://doi.org/10.3389/fnhum.2018.00229” target=”_blank”>DOI
9. Dobrushina, O.R., Vlasova, R.M., Rumshiskaya, A.D., et al. (2021). Modulation of intrinsic brain connectivity by EEG infra-low frequency neurofeedback. Frontiers in Human Neuroscience. https://doi.org/10.3389/fnhum.2021.648574” target=”_blank”>DOI
10. Thibault, R.T., MacPherson, A., Bhatt, M., et al. (2020). Neurofeedback with fMRI: A critical systematic review. NeuroImage, 116114. https://doi.org/10.1016/j.neuroimage.2019.116114” target=”_blank”>DOI
11. Scheinost, D., Stoica, T., Saksa, J., et al. (2020). Neurofeedback and anxiety: A systematic review. Neuroscience & Biobehavioral Reviews. https://doi.org/10.1016/j.neubiorev.2020.09.015” target=”_blank”>DOI
12. Chai, M.T., Amin, H.U., Izhar, L.I., et al. (2020). Exploring EEG effective connectivity network in depression. Journal of Affective Disorders. https://doi.org/10.1016/j.jad.2020.03.056” target=”_blank”>DOI
13. Schabus, M., Griessenberger, H., Gnjezda, M.T., et al. (2020). Neurofeedback for insomnia: State of the science. Sleep Medicine Reviews. https://doi.org/10.1016/j.smrv.2019.101246” target=”_blank”>DOI
14. Munivenkatappa, A., Rajeswaran, J., Indira Devi, B., et al. (2019). EEG neurofeedback therapy: Can it attenuate brain changes in TBI? Frontiers in Neurology. https://doi.org/10.3389/fneur.2019.00998” target=”_blank”>DOI
15. Holtmann, M., Steiner, S., Hohmann, S., et al. (2018). Neurofeedback in autism spectrum disorders. Journal of Autism and Developmental Disorders. https://doi.org/10.1007/s10803-017-3153-z” target=”_blank”>DOI
16. Gruzelier, J.H. (2018). Enhancing performance with neurofeedback. Brain and Cognition. https://doi.org/10.1016/j.bandc.2018.04.003” target=”_blank”>DOI
17. Stokes, D.A., & Lappin, M.S. (2019). Neurofeedback and biofeedback for mood and chronic pain. Applied Psychophysiology and Biofeedback. https://doi.org/10.1007/s10484-019-09432-y” target=”_blank”>DOI

This article is for informational purposes only. It is not medical advice. Always consult a qualified healthcare provider before beginning any new treatment.

About the medical reviewer

Dr. Bronwyn Holmes, MD, FAARFM is a physician specialising in regenerative medicine, advanced peptide therapeutics, exosome and stem cell biology, hormonal health, and longevity. Last reviewed July 5, 2026.