Real-time fMRI neurofeedback of left dorsolateral prefrontal cortex to enhance emotional cognitive control: A randomized controlled trial

First published online October 5, 2026. Participants receiving left-prefrontal feedback showed stronger activity in the intended region during the first two feedback runs, but the difference was not sustained in the third run or when feedback was removed. The study found no improvement in self-reported or task-based emotional cognitive control. It demonstrates an early target-engagement signal, not a clinical benefit.

McDermott TJ, Tsuchiyagaito A, Ramirez S, Cannon MJ, Akeman E, Touthang J, Kirlic N, Misaki M, Paulus MP, Aupperle RL.
NeuroImage · Publication date: 2026-10-05
Added to site: 2026-10-30
DOI: 10.1016/j.neuroimage.2026.122269

First published online October 5, 2026; assigned to the November 1, 2026 issue. Added to Neurofeedback.com October 30, 2026 after current indexing surfaced the record. It is not presented as newly published on October 30.

Modalities

Real-time fMRI neurofeedback, Functional MRI, Emotional cognitive control

Clinical focus

Neural target engagement, Transfer without feedback, Cognitive and behavioral outcomes

Population

Seventy healthy adults randomized to left-dlPFC feedback or left-postcentral-gyrus comparison feedback

Sample

70 enrolled and randomized: 37 active-target and 33 comparison-target participants; the abstract did not state the analyzed denominator after any exclusions

Intervention / method

Participants practiced emotional cognitive control during real-time fMRI feedback across practice, three feedback runs and a transfer run without feedback

Comparator

Real-time feedback from the left postcentral gyrus rather than the left dorsolateral prefrontal cortex

Clinician summary

The time-by-feedback interaction for left-dlPFC activity was significant (p=.031). Active-target feedback exceeded comparison feedback in run 1 (p=.030, d=.66) and run 2 (p=.042, d=.62), but not in run 3 or the no-feedback transfer run. No feedback effect appeared on emotional-cognitive-control measures.

How to interpret the findings

This was a single-session study in healthy adults rather than patients with depression, anxiety or PTSD. Initial regional activation did not persist through later or transfer runs and did not translate to behavioral or self-report improvement. The abstract did not report analyzed denominators after exclusions, adverse events, detailed masking assessment or longer follow-up. Results should not be generalized to EEG neurofeedback, clinical populations, other fMRI targets or repeated training courses.

Equipment and disclosures

The abstract did not identify a commercial fMRI or feedback system. No equipment is attributed to BrainMaster or another vendor.

The PubMed record lists NIMH grant F31 MH122090. Martin Paulus reported unrelated UpToDate royalties; the other authors declared no conflicts of interest. No correction, retraction or expression-of-concern relation appeared in bounded checks of PubMed, publisher-index and DOI results on October 30, 2026. AI-assisted source review, not independent human clinical review.

Related evidence

Compare findings from the same clinical area, including mixed results.