Sensorimotor-rhythm training (Sterman)

1. Introduction

Version followed: the SMR epilepsy programme as M. Barry Sterman describes it in his own reviews (Sterman & Egner 2006; Sterman 2010), with the practice standards of the International Society for Neurofeedback & Research (ISNR; Hammond et al. 2011) and the adverse-effects safeguards of Hammond & Kirk (2008). Applications to ADHD and insomnia follow Arns and colleagues (2014). Where the developer's position and current medical practice differ (see below), this guide follows current medical practice.

What it is. Sensorimotor-rhythm (SMR) training is a form of EEG operant conditioning (neurofeedback). The client sees and hears real-time feedback and is rewarded for producing more of a 12–15 Hz rhythm recorded over the sensorimotor cortex, while unwanted slower activity is held down. [source: Sterman 2010]

Brief history. Sterman, working at UCLA, first saw this rhythm in cats. It appeared during still, motionless wakefulness, when the animals were holding back a learned response. Because it looked like the sleep spindle, he proposed that both reflect a shared mechanism of motor inhibition. When cats were trained to produce SMR, their sleep spindles became denser and their sleep less fragmented. In an Air Force-funded study of monomethylhydrazine, a rocket-fuel component that causes seizures, cats that had already had SMR training seized later than untrained cats, and some did not seize at all. A later counterbalanced study confirmed this. In 1972 Sterman and Friar reported a young woman with nocturnal tonic-clonic seizures that medication had not controlled. After long-term training she was described as essentially seizure free, and her medicines were held constant during training. [source: Sterman 2010; Arns, Fetz & Birbaumer 2024] A 1988 randomised study by Lantz and Sterman compared three conditions: seizure logging only, yoked (non-contingent) feedback, and contingent SMR training. Only the contingent training reduced seizures, and that effect grew over time. [source: as reported in Sterman 2010; original not verified, [uncertain]]

How change is thought to happen. On the developers' model, rewarding SMR strengthens the thalamocortical circuits that produce it (an exchange between ventrobasal and reticular thalamic nuclei). This lowers motor-cortex excitability and so raises the threshold for seizures. [source: Sterman 2010; Sterman & Egner 2006] Arns and colleagues suggest SMR training may also strengthen the spindle-generating network that governs sleep onset and vigilance. In their account, it trains the regulation of a network rather than simply raising power in one band. [source: Arns, Feddema & Kenemans 2014] These are the authors' hypotheses, not established mechanisms. [uncertain]

Where it sits among other methods. Sibling methods in this family are 045 Alpha training (Kamiya), 046 Alpha-theta training and the Peniston protocol, 048 Theta-beta training (Lubar) and 049 Slow cortical potential training (Birbaumer). SMR rewards are also part of many ADHD protocols (see 048). Slow cortical potential training is the other self-regulation approach studied for epilepsy (see 049).

Epilepsy: the non-negotiable frame. Sterman and Egner (2006) call neurofeedback for seizures "a well-founded and viable alternative to anticonvulsant pharmacotherapy". This guide does not adopt that position. In clinical practice, SMR training for epilepsy is offered only alongside the treating neurologist, as an addition to antiseizure medicines and never as a replacement. Every medication decision stays with the neurologist. Current epileptologists describe the method as an adjunct, used mainly in drug-resistant epilepsy. [source: Marchi et al. 2024 abstract; [synthesis] for the stance]

Who may deliver it. Psychotherapists and other licensed clinicians may deliver it within the scope of their licence. Those treating diagnosable conditions should be licensed for independent practice. BCIA board certification is recommended. Unlicensed technicians may train clients only under documented, on-site supervision by a licensed professional, who first evaluates the client and sets the treatment plan. Interpreting a quantitative EEG (QEEG) calls for further qualification, or the recording should be sent to a qualified interpreter. [source: Hammond et al. 2011]

2. Framework

The phases below are a [synthesis] that orders the steps Sterman (2010) and ISNR (Hammond et al. 2011) describe. The sources do not set out formal stages.

Phase 1: Referral, intake and medical clearance (about 1–3 meetings).

Goals: a detailed history (medical, psychological, earlier treatments, current medicines); risk factors such as severity, chronicity, history of suicidality, dissociation, abuse and comorbidity; and a review of medical records. [source: Hammond et al. 2011; Sterman 2010]
For seizures, an accurate seizure history and classification are required. [source: Sterman 2010]
Practitioners should be cautious about taking on clients with medical conditions until appropriate medical evaluation has been done. [source: Hammond et al. 2011]
Move on when: there is a neurologist-confirmed diagnosis, written agreement from the neurologist that training may run alongside unchanged medical treatment, a baseline seizure log, and documented consent. [synthesis]

Phase 2: EEG assessment and treatment planning.

An objective EEG assessment should come before training. The most thorough form is a multichannel QEEG compared against an age-appropriate normative database. Sterman and Egner recommend mapping before and after treatment, both to choose the strategy and to track movement towards normal EEG values. [source: Sterman & Egner 2006; Hammond et al. 2011]
A licensed professional writes an individualised protocol. In Sterman's programme it then stays the same throughout treatment. [source: Sterman 2010; Hammond et al. 2011]
Device settings, recording sites and reward criteria belong in specialist training and are not given here.
Move on when: the protocol is documented and the client understands it.

Phase 3: Training.

Sterman's programme describes one or two sessions a week, each 60–90 minutes, for at least 20 weeks. Each session is a series of short feedback trials. [source: Sterman 2010]
In Arns' open-label ADHD study, courses averaged about 30 sessions. [source: Arns, Feddema & Kenemans 2014]
ISNR warns against promising a fixed number of sessions. Course length is decided jointly by clinician and client, based on diagnosis, response, comorbidity and stressors. [source: Hammond et al. 2011]
Review progress at set points: reward rate and band amplitudes across sessions, the seizure log or symptom ratings, and side effects. [source: Sterman 2010; Hammond & Kirk 2008]

Phase 4: Generalisation. Carrying the learned state into daily life is not described in detail in the cited SMR sources. See section 4, and 049 for the transfer procedures used in slow cortical potential training. [synthesis; [uncertain]]

Phase 5: Ending and follow-up.

Do not encourage a client to continue when there is no benefit. Explain the reasons for stopping, referring on or switching treatment, and involve the client in that decision. [source: Hammond et al. 2011]
Repeat the QEEG after treatment. [source: Sterman & Egner 2006]
In the 1972 case, the physician reduced medication only after training had ended. Any such step belongs entirely to the neurologist. [source: Sterman 2010]
Arrange follow-up seizure logging and booster reviews. [synthesis]

Format. Individual, in-office sessions with the practitioner or a trained coach present throughout. ISNR criticises "treatment mill" settings where clients are connected to equipment and left alone. Home or remote training is allowed only after in-office evaluation and enough in-office sessions, and only under licensed supervision with a written agreement. [source: Hammond et al. 2011]

3. Guiding principles

Adjunct, never a substitute, in epilepsy. Antiseizure medicines and their dosing belong to the neurologist. You never suggest that a client start, stop or change a medicine, and you route every question about medicines to the prescriber. [synthesis; Domain IX standard]
Assess before training. Use a history and an objective EEG assessment, not a "canned" protocol. Hammond & Kirk link generic protocols applied without assessment to adverse effects. [source: Hammond & Kirk 2008; Sterman & Egner 2006]
Contingency matters. Effects depend on accurate, contingent reward. In controlled studies, reversing the contingency made seizure control and epileptiform activity worse, and restoring it reversed the change. Check the protocol before each session. [source: Hammond & Kirk 2008; Lantz & Sterman as reported in Sterman 2010]
Shape by successive approximation. Reward small steps towards the target pattern, adjusting criteria gradually. [source: Sterman 2010]
Keep the protocol consistent. Sterman holds one individualised protocol throughout. Hammond & Kirk report lost gains after protocol switches and harm from trial-and-error changes. [source: Sterman 2010; Hammond & Kirk 2008]
Be present and coach. Watch for drowsiness, mind-wandering and loss of focus, and give encouragement and reinforcement. [source: Hammond et al. 2011]
Consent is ongoing. Discuss the evidence level, possible side effects, the fact that it does not help everyone (ISNR cites at least 20% non-response), alternatives, and the right to stop. [source: Hammond et al. 2011]
Ask about side effects every session and change training promptly when they occur. Telling clients side effects never happen is unethical. [source: Hammond & Kirk 2008; Hammond et al. 2011]
Measure, don't assume. Track seizure logs and symptom ratings alongside the EEG learning curve. Within-session learning did not reliably predict clinical gain in Arns' study. [source: Arns, Feddema & Kenemans 2014; Hammond & Kirk 2008]
Stay within competence and licence. Hold BCIA certification or work under a mentor. An introductory course does not qualify you for independent practice. [source: Hammond et al. 2011]
Don't oversell. Promise no cure and no fixed session count. [source: Hammond et al. 2011]
Treat the person, not only the EEG. Epilepsy brings anxiety, depression, stigma and restrictions on driving and work. Use your psychotherapy skills alongside training. [synthesis]

4. Intervention inventory

The sources describe the training itself (4.4–4.8) only in general terms. Equipment settings, recording sites, reward thresholds and trial timings are taught in BCIA-approved training and are not reproduced here. Items marked [synthesis] are this guide's own practice guidance.

4.1 Screening, referral and intake interview

Purpose: decide whether SMR training is appropriate, safe and within your competence, and what else the client needs. Use at first contact.
Steps:
1. Take a medical and psychological history: earlier treatments, current medicines, sleep, substance use. [source: Hammond et al. 2011]
2. Screen for the risk factors ISNR lists: severity, chronicity, history of suicidality, dissociation, abuse, litigiousness and comorbidity. Their presence calls for fuller assessment and more expertise. [source: Hammond et al. 2011]
3. For seizures, record the type and frequency, the diagnosis and who made it, the investigations done, the current antiseizure medicines (names only), adherence, triggers, and any nocturnal seizures. [source: Sterman 2010 on accurate seizure history and classification; [synthesis] for the list]
4. Ask for and review medical records. [source: Sterman 2010]
5. Decide whether to refer on: an undiagnosed or poorly characterised seizure disorder, new neurological symptoms, active suicidality or acute psychosis go to medical or psychiatric assessment first. [synthesis]
Example language (illustrative): "Before we think about training, I need a clear picture of your seizures and your treatment. With your permission I'll also read your neurologist's letters."
Watch for: a client hoping to come off medicines; events whose epileptic nature has never been confirmed; untreated mood or anxiety disorders. Questions about whether events are psychogenic non-epileptic seizures are for the neurologist to answer, since these call for a different treatment. [synthesis]

4.2 Medical clearance and a collaborative agreement with the neurologist

Purpose: make sure training runs alongside, not instead of, medical care. This is essential for epilepsy.
Steps (all [synthesis] unless marked):
1. With the client's written consent, write to the treating neurologist. Explain the planned course of SMR neurofeedback, its adjunctive aim and the evidence status, and ask whether there are any contraindications.
2. Agree in writing that antiseizure medicines continue unchanged during training unless the neurologist decides otherwise. In Sterman's 1972 case, medicines were held constant during training. [source: Sterman 2010]
3. Agree how you will share the seizure log and how you will report any rise in seizures, new seizure types or adverse effects.
4. Agree who the client should contact in an emergency and what the seizure first-aid plan is for sessions.
Example language (illustrative): "This training is an addition to your treatment, not a replacement. Your medicines stay exactly as your neurologist prescribes. If you ever wonder about changing them, that's a conversation for your neurologist, and I'm glad to help you prepare for it."
Watch for: a neurologist who has not been told; pressure from the client to promise medication reduction.

4.3 Informed consent (ongoing)

Purpose: an educated, documented choice, revisited as treatment goes on. [source: Hammond et al. 2011]
Steps: cover the assessment, fees, the limits of confidentiality, the recommended treatment, your qualifications, the expected frequency and duration, alternatives (including medical treatment, psychotherapy and other biofeedback), the evidence level for the condition, possible side effects, the fact that it does not work for everyone, and the right to stop without penalty. Document it in the record and revisit it at reviews. [source: Hammond et al. 2011; Hammond & Kirk 2008]
Example language (illustrative): "Most people notice nothing unpleasant. Some get tiredness, headaches, feeling wired or sleeping badly. Tell me straight away if anything like that happens, so we can adjust."
Watch for: expectations of cure. Make sure the client understands that the evidence is contested (see 4.12).

4.4 EEG / QEEG assessment

Purpose: an individualised basis for the protocol, and a baseline for measuring change. [source: Sterman & Egner 2006]
Steps: arrange a multichannel recording compared with an age-appropriate normative database, collected by someone with documented training. If you are not qualified to interpret it, send it to a qualified interpreter. Repeat after treatment. [source: Hammond et al. 2011; Sterman & Egner 2006] A QEEG is not a diagnostic clinical EEG: diagnosis remains with neurology. [synthesis]
Watch for: recordings spoiled by artefact; findings that suggest a need for neurological review, which you pass to the neurologist. [synthesis]
Example language (illustrative): "The recording is a starting point. If anything in it needs a medical opinion, I'll pass it to your neurologist." [synthesis]

4.5 Psychoeducation and orientation to training

Purpose: make the rationale clear and build motivation for a long, effortful learning process.
Steps:
1. Explain in plain words that SMR is a rhythm linked to bodily stillness and to the brain systems that also produce sleep spindles, and that training it is thought to make the brain less prone to seizures. [source: Sterman 2010]
2. Explain operant learning: the screen rewards the brain whenever it moves in the right direction, and learning builds over many sessions. [source: Sterman 2010; Sterman & Egner 2006]
3. Show the display and the reward signals. Sterman describes simple visual and tone rewards and a running score. [source: Sterman 2010]
Example language (illustrative): "You won't control this by trying hard. It works more like learning to balance on a bike. Your brain picks it up from the feedback, a little at a time."
Watch for: performance anxiety and over-effort (see 4.7).

4.6 The training session

Purpose: contingent reinforcement of SMR, with slower unwanted activity held down, through repeated short trials. [source: Sterman 2010]
Steps:
1. Check-in. Ask about seizures and side effects since the last session, sleep, any medication changes the neurologist has made, illness and alertness. Record 0–10 symptom ratings. [source: Hammond & Kirk 2008; [synthesis] for the order]
2. Set-up. Apply the sensors as trained and with good hygiene. Check signal quality and artefact before starting. Confirm that the documented protocol is the one loaded. [source: Hammond & Kirk 2008 on reversed contingencies; [synthesis] for the hygiene and checks]
3. Feedback trials. The client watches a simple display. Success shows as a colour change, a moving object and a chime, and a score builds. Trials are short and repeated, with brief rests between them. Criteria are shaped by successive approximation. Sterman tracks the reward rate per trial. [source: Sterman 2010]
4. Coaching. Stay present throughout. Watch for drowsiness, mind-wandering, movement and tension, and give encouragement and reinforcement. [source: Hammond et al. 2011]
5. Children. Sterman describes a parent giving small tangible rewards to a child during training. [source: Sterman 2010]
6. Close. Ask how the client feels and whether any side effects have appeared. Check that they are clear-headed enough to drive or travel. Record session data. [source: Hammond & Kirk 2008]
Example language (illustrative): "Let your body go still and your eyes rest on the screen. Notice what was going on when the chime came, without forcing it."
Watch for: a rising reward rate across sessions and changes in band amplitude, as Sterman tracks them [source: Sterman 2010]; signs of fatigue or agitation; artefact being rewarded in place of EEG. [synthesis]

4.7 Coaching the SMR state (cognitive and behavioural strategies)

Purpose: help the client find the still, relaxed, outwardly attentive state in which SMR appears, without strain.
Steps ([synthesis], drawing on Sterman's description of SMR as a rhythm of motionless wakefulness and behavioural quiescence):
1. Encourage physical stillness and calm, open attention to the display.
2. After a run of rewards, invite the client to notice what they were doing inside. Don't prescribe a single strategy.
3. Discourage effortful straining, breath-holding and tensing of face or jaw, which also produce artefact.
4. Note the strategies the client reports, to use in generalisation (4.9).
Example language (illustrative): "What were you doing in your mind during that last good stretch?"
Watch for: sleepiness. SMR training should not slide into dozing. [synthesis; Hammond et al. 2011 on drowsiness]

4.8 Session-by-session monitoring and progress reviews

Purpose: know whether the client is learning, improving and safe.
Steps:
1. Each session: ask about side effects and record ratings. [source: Hammond & Kirk 2008]
2. Keep a daily seizure log throughout, starting before training as a baseline. Sterman calls for accurate seizure history, and seizure logging was itself a control condition in Lantz and Sterman's study. [source: Sterman 2010; [synthesis] for the daily log]
3. For sleep complaints, monitor sleep-onset latency and sleep quality. [source: Arns, Feddema & Kenemans 2014]
4. Hold a structured review roughly every 10 sessions: EEG learning curves, the seizure or symptom trend and side effects. Share the summary with the neurologist. [synthesis]
Watch for: EEG learning without clinical change, or clinical change without EEG learning. Arns found that learning did not clearly predict outcome, so judge by both. [source: Arns, Feddema & Kenemans 2014]
Example language (illustrative): "Your seizure log and the session graphs together tell us whether this is helping. Let's look at both." [synthesis]

4.9 Generalisation to daily life

Purpose: carry the learned self-regulation beyond the clinic.
The cited SMR sources do not describe a specific transfer procedure. Slow cortical potential programmes use explicit transfer trials and cue cards (see 049). The steps below are [synthesis]. [uncertain]
Steps:
1. Late in each session, add brief trials with reduced or no feedback, if your equipment and training support this. The client recalls the state and the results are then shown. [synthesis]
2. Turn the client's own strategies (4.7) into a short daily practice and a "stillness cue". Use these at known seizure triggers or warning signs (auras), and also for sleep onset. [synthesis]
3. Address lifestyle factors the neurologist identifies as seizure triggers, such as sleep loss and missed medicines, without giving medical advice. [synthesis]
Example language (illustrative): "When you notice your usual warning sign, try settling into the same still, steady state we practise here, and then follow your seizure plan."
Watch for: the client using self-regulation instead of their medical seizure plan or medicines. Correct this at once. [synthesis]

4.10 Home and remote training (if used)

Purpose / when: only for selected clients, after in-office evaluation, with close remote supervision. [synthesis]
Only after in-office evaluation and enough in-office sessions. Train the on-site supervisor (for example a parent) in the equipment, sensor placement and coaching. Get data and progress reports at least weekly, with live online supervision as the ideal. Use equipment that does not let the home user change protocols, and a written agreement covering how often data are sent, office visits, use by the client only, costs and damage, and immediate return of the equipment if misused. Check licensure rules for clients in other jurisdictions. Home training is never done without a licensed professional supervising it. [source: Hammond et al. 2011]
Do not endorse consumer "brain-training" headsets or self-directed protocols for epilepsy. Unsupervised home training is a recognised cause of adverse effects. [source: Hammond & Kirk 2008]
Example language (illustrative): "Home training only starts once we've done enough sessions here, and I'll be reviewing your data every week." [synthesis]
Watch for: missed data uploads, changes in seizures, side effects, or the device being used outside the agreed plan; pause and review. [synthesis]

4.11 Ending, medication questions after training, and follow-up

Purpose: a planned, jointly decided end with gains consolidated and medical care unchanged unless the neurologist decides otherwise.
Steps:
1. Decide jointly when to end, based on response, comorbidity and stressors. Stop or refer on when there is no benefit, and explain why. [source: Hammond et al. 2011]
2. Repeat the QEEG and outcome measures. [source: Sterman & Egner 2006; Hammond et al. 2011]
3. Send the neurologist a closing summary. If the client asks whether they can now reduce their medicines, that decision is wholly the neurologist's. In Sterman's 1972 case, reductions came only after training ended, under medical care. [source: Sterman 2010; [synthesis] for the routing]
4. Plan follow-up: continued seizure logging, a review at intervals, and a route back for booster sessions if gains slip. [synthesis] Lantz and Sterman reported that the treatment effect grew over time. [source: as reported in Sterman 2010]
Example language (illustrative): "Let's write down what you've learned and what you'll keep practising. Any change to your medicines is for you and your neurologist to decide. I'll send them a summary to support that conversation."
Watch for: pressure to reduce medicines after good results; route all such questions to the neurologist. [synthesis]

4.12 Evidence-status conversation

Purpose: honest, attributed information for consent.
Epilepsy:
- Sterman (2010) reports pooled data from 24 studies, with most patients showing more than 50% seizure reduction.
- Tan et al. (2009) found significant seizure reduction in most of the studies they included. [source: as reported in Sterman 2010; the meta-analysis was not verified]
- Hassanvandi et al. (2023), a meta-analysis of randomised trials, call the evidence "preliminary". [source: Hassanvandi et al. 2023 abstract]
- Marchi et al. (2024) describe use as an adjunct in drug-resistant epilepsy that "remains under-explored". [source: Marchi et al. 2024 abstract]
- ISNR (2011) lists uncontrolled epilepsy among conditions with considerable research support. [source: Hammond et al. 2011]
- The US insurer Anthem's 2026 guideline CG-MED-97 classes neurofeedback, including for epilepsy, as not medically necessary. [source: Anthem CG-MED-97]
ADHD and insomnia:
- Anthem cites the American Academy of Pediatrics (2019), which places EEG biofeedback among ADHD interventions with too little evidence to recommend. [source: Anthem CG-MED-97]
- Arns' open-label study linked SMR training to shorter sleep-onset latency in ADHD, but the authors call this preliminary. [source: Arns, Feddema & Kenemans 2014]
NICE: no NICE statement on SMR neurofeedback for epilepsy was found in the cited sources. [uncertain; check the current NICE epilepsy guideline, NG217]
Example language (illustrative): "The research is encouraging but disputed. Some reviews and the developers report fewer seizures, while some insurers and guideline groups don't consider it proven. That's why we use it only as an addition to your medical treatment."
Steps: (1) Give the evidence for the client's condition as attributed above, including the disagreements. (2) Name the alternatives and that training is an add-on. (3) Check understanding and record the discussion. [synthesis]
Watch for: hope that outruns the evidence, or a client hearing the training as a replacement for medicines. [synthesis]

4.13 Integrating psychotherapy

Purpose: treat the psychological load of epilepsy or ADHD alongside training. [synthesis]
Steps ([synthesis]): screen for depression, anxiety and suicidality and treat them with your usual evidence-based methods. Work on stigma, fear of seizures and avoidance. Help with adherence to medical treatment by exploring ambivalence, without ever advising a change; route medication questions to the prescriber. Include a seizure safety plan, and make sure the client knows the local driving rules, which the neurologist or licensing authority explains.
Watch for: hopelessness after a poor response. Have a suicide-risk review and a crisis plan in place.
Example language (illustrative): "Living with seizures is hard emotionally. Alongside the training, we can work on the worry and the things you've stopped doing." [synthesis]

4.14 Training route, certification and equipment safety

Purpose / when: the training, certification and safety requirements to meet before offering SMR training. [synthesis]
BCIA "Board Certified in Neurofeedback" (BCN), clinical route [source: Wikipedia summary of BCIA requirements; check current figures on bcia.org [uncertain: secondary]]:
- A bachelor's degree or higher in a clinical health-care field, plus BCIA's ethics agreement.
- A 36-hour didactic course covering the Neurofeedback Blueprint: orientation to EEG biofeedback; basic neurophysiology and neuroanatomy; instrumentation and electronics; research; psychopharmacological considerations; treatment planning; professional conduct.
- A course in human anatomy and physiology, or BCIA's own examination in it.
- Mentoring by a BCIA-approved mentor, with a set number of contact hours, supervised client sessions and case conference presentations.
- A written examination.
- Recertification every four years, through continuing education or re-examination.
- Academic and technician routes also exist; technicians work under a supervisor's licence.
Further qualifications: QEEG interpretation calls for further qualification, such as QEEG diplomate status. [source: Hammond et al. 2011]
Licence: certification does not replace a licence to treat. [source: Hammond et al. 2011]
Equipment and electrical safety ([synthesis], general terms; the cited sources do not cover this):
- Use equipment intended and marketed for clinical neurofeedback, with patient isolation, and keep it maintained and safety-tested as local electrical-safety rules require.
- Never connect a client to mains-powered equipment that is not isolated.
- Clean or replace sensors between clients and follow infection-control routines.
- Check skin under sensors for irritation.
- Keep the room safe for a seizure: a reclining chair or mat, no hard edges, and a trained first-aider present.
- The regulatory status of specific devices was not verified in the cited sources. [uncertain]
Example language (illustrative): "I'm trained and supervised in neurofeedback and work within my professional licence. I'm happy to tell you about my training." [synthesis]
Watch for: working beyond your training, and equipment that is not isolated or maintained. [synthesis]

5. Handling problems

Guidance is marked by source; most of it is [synthesis] built on the ISNR standards and the Hammond & Kirk safeguards.

Seizures increase or a new seizure type appears.
- Stop the training plan and inform the neurologist the same day. Review the protocol for errors, including a reversed contingency. [source: Hammond & Kirk 2008 on reversal effects; [synthesis] for the response]
- Resume only with the neurologist's agreement and a revised plan.
- A seizure during a session: follow standard seizure first aid and your agreed emergency plan, and record it. [synthesis]
Side effects (fatigue, headache, anxiety, feeling wired, insomnia, emotional lability, tics, fogginess, dissociation).
- Document them, discuss them with the client and act promptly. Options are modifying the protocol, reducing how much or how often you train, adding other treatment, or seeking consultation. [source: Hammond et al. 2011; Hammond & Kirk 2008]
- Do not let a groggy client drive. [source: Hammond & Kirk 2008]
The client wants to cut or stop antiseizure medicines.
- Explore the reasons: side effects, cost, stigma, hope of cure. Do not agree or disagree with a change. Say clearly that decisions about medicines belong to the neurologist, and help the client prepare questions for that appointment. [synthesis; Domain IX standard]
- If the client has already stopped, tell them to contact their neurologist urgently. With consent, or where risk requires it, inform the neurologist. [synthesis]
No EEG learning (a "non-learner").
- Check signal quality and artefact, sleepiness, over-effort and the reward display. Use 4.7 coaching and consult your mentor. [synthesis]
- Stop rather than prolong a course without benefit. [source: Hammond et al. 2011]
- Remember that learning and clinical outcome can come apart. [source: Arns, Feddema & Kenemans 2014]
Learning without clinical benefit. Hold a joint review with the client and neurologist, consider other treatments, and end openly. [source: Hammond et al. 2011; [synthesis]]
Gains lost after a protocol change. Return to the documented protocol and avoid trial-and-error changes. [source: Hammond & Kirk 2008]
Drowsiness or boredom during long courses.
- Shorten trials, vary the displays, coach actively and check the client's sleep. [source: Hammond et al. 2011 on drowsiness; [synthesis]]
- For children, use reward systems as Sterman describes. [source: Sterman 2010]
Non-attendance or dropout. Training needs regular sessions over months. Review practical barriers and motivation, use motivational interviewing, and agree a realistic schedule or a planned pause. [synthesis]
Unrealistic expectations or disappointment (a rupture risk).
- Revisit consent and the evidence (4.12). Acknowledge disappointment without defending the method, and repair the alliance by deciding next steps together. [synthesis]
- Never promise cure or a fixed number of sessions. [source: Hammond et al. 2011]
Events that may not be epileptic, or a diagnosis in doubt. Pause and ask the neurologist to clarify the diagnosis. Psychogenic non-epileptic seizures need a different treatment pathway. [synthesis]
Psychiatric comorbidity, dissociation, fragile clients. These call for more thorough assessment and a higher level of expertise, and Hammond & Kirk report decompensation. Treat comorbidity in psychotherapy and go slowly. [source: Hammond et al. 2011; Hammond & Kirk 2008]
Home equipment misuse or consumer devices. Apply the written agreement and recall the equipment. Explain the risks of unsupervised training. [source: Hammond et al. 2011; Hammond & Kirk 2008]
Pressure from the client or family to skip the neurologist. Decline to train for seizures without medical involvement. [synthesis]
Unlicensed practitioners. If you learn of unlicensed people offering neurofeedback, report them to the state or provincial regulator. [source: Hammond et al. 2011]

Sources

Online sources:

Cited as reported in the sources above, not verified [uncertain]: Sterman & Friar (1972) case report; Lantz & Sterman (1988) randomised study; Tan et al. (2009) meta-analysis of EEG biofeedback in epilepsy; Egner & Sterman (2006), Expert Review of Neurotherapeutics 6: 247–257; Wolraich et al. (2019), AAP ADHD guideline.

← Previous · IX-046Alpha-theta training and the Peniston protocol Next · IX-048 →Theta-beta training (Lubar)