This trial tested fMRI neurofeedback aimed at one frontal brain area, the right inferior frontal cortex. It was double-blind and compared real feedback with a fake, or sham, version.
The study included 88 boys aged 10 to 18 with confirmed ADHD. Half got real feedback and half got sham. Most of the boys were taking ADHD medication (about 65%). Each boy did 15 runs over four 1-hour sessions. Before training, 6 randomized boys refused scanning and dropped out (6.4%). The trial was stopped early by the COVID-19 lockdown.
Real feedback did not lower parent-rated ADHD symptoms more than sham. This held right after training and six months later. Symptoms dropped from the starting point in both groups, but rose again between the end of training and follow-up. Right after training, the sham group had a bigger drop in parent-rated irritability. The sham group was also better at holding back a movement on a go/no-go task. Neither gap remained at follow-up.
The real group did activate the target area more than the sham group. But this did not build up over the sessions, and it was not linked to symptom change. No serious side effects were found, and side effects did not differ between groups.
The authors note limits. Only boys and parent reports were used. Medication may have masked effects. The study may have been too small to find smaller effects. The groups also differed in ADHD type at the start. The authors conclude the method does not appear to improve symptoms or thinking skills in boys with ADHD.
The numbers Stats and what the terms mean
The statistics behind the plain-language claims above.
- Lam, 2022: ADHD symptom scores rising in both groups from the end of training to the 6-month follow-up, F [1,82.7]=8.44; p=.005.
- Lam, 2022: the two groups changing differently over time in parent-rated irritability, F [1,83.1]=7.73; p=.028.
- Lam, 2022: lower irritability in the sham group than the real group right after training, F [1,136.5]=4.04; p=.046.
- Lam, 2022: the two groups changing differently over time in holding back a movement on the go/no-go task, F [1,75.8]=8.78; p=.048.
- Lam, 2022: poorer holding back in the real group than the sham group right after training, F [1,142.7]=4.45; p=.037.
- Lam, 2022: link between change in target-area activation and change in ADHD symptoms, Pearson's r=.001, p=.99, n=41.
- Lam, 2022: difference in side effects between groups, F [1,81]=.22; p=.64.
- Lam, 2022: anxiety or worry (14%) and distractibility (20.9%) in the sham group compared with the real group (2.3% and 4.7%), F [1,82]=3.75; p=.06.
What the terms mean:
- F is a test score for whether groups or time points differ by more than chance would suggest; the numbers in brackets describe the data used to compute it.
- p is the chance of seeing a difference this large if there were truly no difference; below .05 is usually treated as unlikely to be down to chance.
- r shows how closely two measures move together, from 0 (not at all) to 1 (perfectly).
- n is the number of participants in that analysis.
Source: Sheut-Ling Lam, Marion Criaud, Steve Lukito, et al.. Double-Blind, Sham-Controlled Randomized Trial Testing the Efficacy of fMRI Neurofeedback on Clinical and Cognitive Measures in Children With ADHD. American Journal of Psychiatry (2022). doi:10.1176/appi.ajp.21100999