Real-time feedback on movement, gait and posture improves motor function — one of the oldest and most consistent rehabilitation applications of biofeedback. Five randomised trials show that adding feedback to physiotherapy does better than physiotherapy alone.
Level 2. New row: AAPB does not rate neurofeedback for cerebral palsy. The only randomised evidence is Chen 2024, a pilot trial in which 19 children with cerebral palsy and co-occurring attention deficits were randomised to 20 sessions of EEG neurofeedback or a control group: the theta/beta ratio fell relative to the control group (p = 0.04) with within-group improvement during training (p = 0.02), visual sequential memory and visual closure on the TVPS-3 improved (p = 0.02 and p = 0.01), and continuous-performance omissions showed only a trend (p = 0.08). Behboodi 2024 developed a movement-related cortical potential BCI-neurofeedback system coupled to electrical stimulation and reported a 10-session ankle dorsiflexion protocol in one child with cerebral palsy (80.8 % average detection accuracy, increased dorsiflexion velocity, walking speed and step length). Cioffi 2024 mapped 133 sensorimotor neurofeedback studies, only three of which enrolled children. One small randomised trial plus feasibility work supports Level 2 and nothing higher.
Biofeedback
Surface EMG, movement, gait and postural feedback — motor function
AAPB4EfficaciousNeuroLogic4Efficacioussame vs AAPB
Level 4 maintained. AAPB base: five randomised trials from independent settings, all adding feedback to physiotherapy and all favouring the feedback arm. Dursun 2004 (n = 36 children with spastic CP, 10 days of intervention): better muscle tone, ankle dorsiflexion, movement velocity, stride length and clinical gait assessment at post-intervention, 1 month and 3 months, cadence excepted. Gharib 2011 (n = 30, ages 10 to 13, treadmill gait-trainer feedback three times weekly for 3 months): better ambulation index and time of support on each foot. Hussein 2019 (n = 30, ages 4 to 6, proprioceptive and visual gait feedback): greater gains on spatial and temporal parameters, kinetic parameters unchanged. Alwhaibi 2020 (n = 45, gamified upper-limb feedback): physiotherapy plus biofeedback superior to either alone on visual-motor integration, visual perception and motor coordination. Huang 2020 (n = 66): greater gains on Ashworth, Berg, gross motor function, assisting-hand and Gesell scores. MacIntosh 2019 (57 studies, 817 participants) found improvement in 79 % of studies but graded the evidence 'positive very low'. 2022-2026 window: Xu 2025's network meta-analysis (81 RCTs) ranks biofeedback among the interventions that significantly reduced spasticity (SMD -3.29 to -0.73), with no ranking for gross motor function, walking or daily living. Not Level 5: no trial used a credible sham and the feedback arms generally received more intervention time.
In short
Clinical reading
AAPB Level 4; NeuroLogic Level 4. Base: five RCTs adding feedback to physiotherapy (Dursun 2004, Gharib 2011, Hussein 2019, Alwhaibi 2020, Huang 2020), all favouring the feedback arm, plus a crossover study and a systematic review of 57 studies. Surface EMG is used to reduce spasticity, then in phased feedback to activate a muscle in one gait phase and relax it in another; inertial sensors and goniometers broaden the trainable parameters. Not Level 5: no comparison with a sham, and the feedback arms received more intervention time. Neurofeedback is rated separately at Level 2 on a single paediatric pilot trial.
Protocols
Single- or dual-site SEMG with phased gait-cycle feedback, stride length, velocity and cadence feedback, postural stability work; gamified upper-limb feedback; virtual-reality treadmill training.
Limits
Wide heterogeneity of devices and trained parameters; no trial used a sham condition; limited data on long-term maintenance and transfer to daily activities — in the one long follow-up, most gains were lost at 18 to 24 months. The neurofeedback evidence is limited to one pilot trial with attention, not motor, outcomes.
Study base
Extensive international literature since the 1970s, including five RCTs and a dedicated systematic review (MacIntosh 2019: 57 studies, 817 participants). 2022-2026 base: 3 publications indexed in the archive (2 biofeedback, 1 neurofeedback).
Brendan's perspective
Level 4, unchanged, and one of the few places where the biofeedback rating sits comfortably: five randomised trials from independent groups (Dursun 2004, Gharib 2011, Hussein 2019, Alwhaibi 2020, Huang 2020) all favour adding feedback to physiotherapy. What stops it at 4 is not subtle — no sham anywhere, and the feedback arms generally received more therapy time, so part of the effect is dose. The neurofeedback row is new and small. Chen 2024 is a 19-child pilot with a respectable 20-session dose and a qEEG-derived theta/beta target: the ratio moved, two visual-perceptual scores moved, and the attention primary outcome only trended. That supports level 2 and nothing above it. I do work on co-occurring attention and regulation difficulties in children with cerebral palsy, with individualised amplitude training, and I am explicit with families that this is about attention rather than motor function — and that Seeger & Caudrey 1983 found most motor gains gone by 18 to 24 months. That said, I want to record what I have actually seen, with the label it deserves: in a handful of cases — anecdote, not evidence — I have seen motor gains substantial enough that nobody in the room doubted them. Always with neurofeedback alongside physiotherapy and never instead of it, so I cannot hand the gain to one of the two. What strikes me is the shape of it. It behaves as an all-or-nothing: the gains are either striking or absent, roughly an even split, with very little in between. Nothing I have found in the qEEG or the clinical profile lets me tell in advance which child is which. I have looked, and I have not found the marker; and until someone does, I say to families that this is worth trying and that I cannot predict the outcome.
Gharib et al. (2011) Efficacy of gait trainer as an adjunct to traditional physical therapy on walking performance in hemiparetic cerebral palsied children: A randomized controlled trial doi:10.1177/0269215511400768
Huang et al. (2020) Effect of biofeedback combined with task-oriented training on hand function, Gesell Scale Score, and balance ability in children with spastic CP doi:10.12182/20200360105
Xu et al. (2025) Nonsurgical Therapies for Spastic Cerebral Palsy: A Network Meta-Analysis doi:10.1542/peds.2024-070402
Chen et al. (2024) The effects of neurofeedback training for children with cerebral palsy and co-occurring attention deficits: A pilot study doi:10.1111/cch.13231
