After a concussion, studies report improvements in attention, memory and executive function with qEEG normalisation, and reductions in headache, sleep disturbance and anxiety with respiratory and HRV biofeedback. The literature is small and often uncontrolled, but treatment alternatives are scarce and adverse effects minimal. In children and adolescents the only available data concern biofeedback for post-traumatic headache.
Updated :
What the research shows
EEG neurofeedback — standard amplitude training
, including qEEG-guided amplitude protocols — persistent post-concussion symptoms
Ages
Adult (18+)
Techniques
qEEG-guided (incl. z-score), Beta training
AAPB3Probably efficaciousNeuroLogic3Probably efficacioussame vs AAPB
Level 3 (AAPB level assigned jointly to biofeedback and neurofeedback.) Surmeli 2017: 40 patients with post-concussion syndrome treated with qEEG neurometric-analysis-guided neurofeedback, with overall improvement on primary and secondary measures (SA-45, CGI, Hamilton Depression Scale, MMPI, TOVA) and a fall in the NeuroGuide Traumatic Brain Index; 39 were followed up an average of 3.1 years later and all but two were stable and off medication (AAPB ch. 14) — the strongest study in the AAPB base, and uncontrolled. Lamprecht 2019: 16 athletes randomised to neurofeedback (n = 7) or to a placebo neurofeedback condition the author did not define (n = 9), four sessions only, with significant findings on tandem gait time without cognitive loading and on SCAT-5 symptom number and severity (p < 0.05) (AAPB ch. 14) — randomised, but far below a clinical dose. The adult base remains largely observational, so Level 3 is held.
, including qEEG-guided amplitude protocols — persistent post-concussion symptoms
Ages
Adolescent (12-17)
Techniques
qEEG-guided (incl. z-score)
AAPB3Probably efficaciousNeuroLogic2Possibly efficaciouslower vs AAPB
Level 2 (AAPB level assigned jointly to biofeedback and neurofeedback, and pooled across adolescents and adults.) Lowered for a specific reason: the only neurofeedback data in an under-18 patient, in the AAPB base or in the 2022-2026 window, is Linden 2015, a single case study — a 17-year-old athlete whose qEEG and IVA normalised after 22 sessions of qEEG-guided neurofeedback and again after 40 further sessions following a second concussion (AAPB ch. 14). Dubienski 2016 recruited hockey players aged 15 to 28 but is respiratory biofeedback (see the biofeedback box). One case study cannot carry more; the level reflects the absence of an adolescent literature of its own rather than a negative result, since the protocols and training logic are those validated in adults.
Other neurofeedback methods
LORETA and variants (live z-score)
Ages
Adult (18+)
Techniques
LORETA, qEEG-guided (incl. z-score)
AAPB3Probably efficaciousNeuroLogic2Possibly efficaciouslower vs AAPB
Level 2 (AAPB level assigned jointly to biofeedback and neurofeedback, and pooled with the rest of the AAPB base: this row is a disaggregation by technique, not a disagreement.) Hershaw 2020: 38 participants enrolled and 23 completing all parts, live z-score training, 15 sessions within 6 weeks with session length progressing from 10 to 30 minutes; attention, processing speed, executive function and memory all improved, with assessments 4 weeks before, 4 weeks after and 3 months after the intervention (AAPB ch. 14). Adequately sized with well-identified outcomes but no randomised control internal to the study: Level 2 by definition. The one randomised test, Bonn 2021 (LORETA neurofeedback combined with HRV biofeedback, 24 participants), found no difference on the GAD-7 or the RPQ against either control group and none in driving simulation, while EEG z-scores normalised most in the intervention group with a large effect (d = 1.35) (AAPB ch. 14) — a neural rather than clinical result. Hershaw 2022 adds an exploratory analysis in which post-intervention theta/alpha change predicted reduction in pain intensity and cognitive symptoms and beta power change predicted affective improvement, with no EEG change predicting most physical symptoms (abstract).
Infra-low frequency (ILF) / infra-slow (ISF)
Ages
Adult (18+)
Techniques
Infra-low frequency (ILF)
AAPB0Not ratedNeuroLogic3Probably efficacious
Level 3. Carlson 2025: randomised trial in combat veterans with post-concussive symptoms, 87 enrolled, 36 completing twenty half-hour sessions of ILF neurofeedback and 38 completing the control procedures (eight weekly 15-minute health-related discussions), both arms continuing treatment as usual, analysed on intention to treat; from baseline to end of the intervention there was clinically and statistically significant improvement in headache (p < 0.0001), sleep (p < 0.0001) and attention (p = 0.0022), as well as quality of life (p < 0.0001), depressive symptoms (p < 0.0001) and post-traumatic stress symptoms (p = 0.0001) (abstract). It is the largest neurofeedback trial in this indication and the only one at a clinical dose. It is not rated higher because contact time was markedly unequal between arms, blinding is not reported, and the p values given in the abstract are for baseline-to-end-of-intervention comparisons rather than for a between-group contrast.
Biofeedback
HRV and resonance-frequency breathing biofeedback — post-concussion symptoms (adult)
AAPB3Probably efficaciousNeuroLogic3Probably efficacioussame vs AAPB
Level 3 (AAPB level assigned jointly to biofeedback and neurofeedback.) Lu 2023: randomised trial in mild TBI, 49 recruited and 41 completers (21 psychoeducation, 20 HRV biofeedback), with a performance-based neuropsychological battery plus self-report; executive function, information processing, verbal memory, emotional functioning and HRV improved significantly in the HRV group at post-test while the psychoeducation group showed no change (abstract) — a within-group contrast rather than a between-group test. Lagos 2013: single case, a 10-week HRV protocol with two daily home breathing sessions, POMS-SF falling from 72 to 24 between sessions 1 and 10, with reported improvement in headaches and post-concussion symptoms (AAPB ch. 14). Dubienski 2016: six hockey players, 10 sessions of six-breaths-per-minute training with audiovisual feedback, a significant fall in heart rate (p < 0.05), no significant change on the other physiological measures, and qualitative improvement in anxiety, fear, attention and concentration (AAPB ch. 14). Talbert 2023 and Talbert 2026 support the modality without lifting it to Level 4.
(modality unspecified) — persistent post-traumatic headache in children and adolescents
Ages
Child (under 12) · Adolescent (12-17)
AAPB3Probably efficaciousNeuroLogic2Possibly efficaciouslower vs AAPB
Level 2 (AAPB level assigned jointly to biofeedback and neurofeedback, and pooled: AAPB contains no paediatric study of its own, so this row is a disaggregation by age.) Schwarz 2025: retrospective cohort of 74 children aged 10 to 18 receiving clinic biofeedback for persistent post-traumatic headache; 42 % met the responder criterion (at least a 50 % reduction in headache days per week or at least a 3-point drop in pain rating between the first and last visit), 36 % had at least a 50 % fall in frequency, 13 % at least a 3-point fall in severity and 8 % both; among those with daily headache, 36 % responded; school absence and use of an SSRI or a prescription preventive medication were associated with non-response (abstract). The study is uncontrolled and retrospective and does not specify the biofeedback modality: Level 2 at most. It is the only paediatric evidence in either indication, which is why one combined child and adolescent row is shown rather than two.
By age
Child (under 12)
In children, the only available data concern biofeedback for persistent post-traumatic headache: in a retrospective study of 74 young people aged 10 to 18, more than four in ten had a clear reduction in headache. No neurofeedback study exists in children with concussion.
Adolescent (12-17)
In adolescents, biofeedback for post-traumatic headache rests on the same retrospective study as in children, and neurofeedback on a single case study — a 17-year-old athlete whose qEEG and attention testing normalised after training. The levels shown rest largely on the adult evidence.
In short
Clinical reading
AAPB gives a single Level 3 rating assigned jointly to biofeedback and neurofeedback, over eight studies and roughly 150 participants (three RCTs, three quasi-experimental studies, two case studies); two of those eight studies are two reports of the same Bonn 2021 trial. NeuroLogic separates the modalities: amplitude neurofeedback 3 in adults (Surmeli 2017, Lamprecht 2019) and 2 in adolescents (a single case study); LORETA z-score 2; ILF 3, new, on Carlson 2025, the largest trial in the indication; biofeedback 3 in adults (Lu 2023, Lagos 2013, Dubienski 2016) and 2 in children and adolescents (Schwarz 2025).
Protocols
qEEG-guided amplitude protocols matched to the symptom profile; source-space z-score where the qEEG warrants it; ILF over courses of about twenty sessions; resonance-frequency HRV training (around six breaths per minute) over ten weeks with daily home practice, for post-concussion autonomic dysfunction and headache.
Limits
Small samples, few RCTs and insufficient use of standardised neuropsychological measures as outcomes. No trial shows neurofeedback superiority over a credible sham on a clinical outcome; the randomised evidence that neurofeedback changes symptoms is weaker than the evidence that it changes the EEG (Bonn 2021). Nothing supports neurofeedback in children with concussion, the only paediatric evidence being retrospective, uncontrolled biofeedback for post-traumatic headache. No study uses return to play or return to learn as an outcome.
Study base
Eight studies and roughly 150 participants in the AAPB base, mostly small and uncontrolled, plus one adequately dosed 2025 neurofeedback RCT, one biofeedback RCT and the first paediatric cohort. 2022-2026 base: 5 publications indexed in the archive (4 neurofeedback or reviews, 1 paediatric biofeedback), together with the Lu 2023 HRV trial.
Brendan's perspective
Amplitude training at 3 in adults, same as the AAPB, but look at what carries the number: Surmeli 2017 is the strongest study in the AAPB base and it is uncontrolled, and Lamprecht 2019 delivered four sessions. Then the row I did not expect to write — ILF at 3, on Carlson 2025, the largest trial in the indication and the only one at a clinical dose. I have put my scepticism about ILF in print, and this level is not there to please me; it is not higher because the control arm received less than a quarter of the contact time and the p values reported are within-arm. I should declare an interest here: I have been concussed three times myself, so this is not a neutral indication for me. It is also one where I think neurofeedback shows real promise. I have worked with a lot of people carrying post-concussive symptoms and seen results that are hard to dismiss. In clinic I train qEEG-guided amplitude matched to the symptom profile, plus resonance-frequency HRV breathing for the headache and autonomic side. The obstacle to proving any of this is structural: no two concussions are alike. Mechanism, site, force, what was already there before — the heterogeneity that makes each case its own problem also makes every research sample and every clinical series difficult to extrapolate from. What I can offer as a pattern rather than a finding: when the qEEG markers line up with the functional picture the person describes, the training tends to go well; when they do not, I am much less confident, and I say so before we start. The honest limit: no trial shows superiority over a credible sham, the EEG changes more reliably than the symptoms do (Bonn 2021), nothing supports neurofeedback in children after concussion, and no study has used return to play as an outcome.
Bonn et al. (2021) Biofeedback as an intervention for persistent post-concussive symptoms: A randomized feasibility trial doi:10.1177/20597002211046459
Conder & Conder (2014) Heart rate variability interventions for concussion and rehabilitation doi:10.3389/fpsyg.2014.00890
Surmeli et al. (2017) Quantitative EEG neurometric analysis-guided neurofeedback treatment in postconcussion syndrome (PCS): forty cases doi:10.1177/1550059416654849
Hershaw et al. (2020) Changes in attentional processing following neurofeedback in patients with persistent post-concussive symptoms: a pilot study doi:10.1080/02699052.2020.1812720
Lagos, Thompson & Vaschillo (2013) A preliminary study: heart rate variability biofeedback for treatment of postconcussion syndrome doi:10.5298/1081-5937-41.3.02
Carlson et al. (2025) Infra-low frequency neurofeedback impact on post-concussive symptoms of headache, insomnia and attention disorder: Results of a randomized control trial doi:10.1016/j.explore.2025.103137
Lu et al. (2023) Heart Rate Variability Biofeedback for Mild Traumatic Brain Injury: A Randomized-Controlled Study doi:10.1007/s10484-023-09592-4
Schwarz et al. (2025) Retrospective Study of the Efficacy of Biofeedback Therapy for Pediatric Persistent Posttraumatic Headache doi:10.1177/08830738241312942
