What if people with depression could watch their attention become trapped by negative information, then practice redirecting it? A new sham-controlled neurofeedback study suggests this learning may be possible inside an MRI scanner, although it did not demonstrate greater relief from depression.
The research targets negative attentional bias: difficulty disengaging from emotionally negative material. Published as a medRxiv preprint in October 2026, the findings have not undergone peer review.
When Negative Information Holds Attention
Depression treatment commonly involves medication, psychotherapy, or neuromodulation. Yet improving mood and changing the moment-to-moment processing of negative information are related but distinct goals.
People with major depressive disorder can struggle to redirect attention once negative content captures it. This pattern may help sustain rumination. Behavioral attention-training approaches attempt to change that bias, but their effects have been inconsistent, motivating efforts to target the underlying brain processes more directly.
How Neurofeedback Makes Attention Visible
Researchers analyzed 48 adults with major depressive disorder, randomly assigned to genuine neurofeedback or sham feedback, with 24 participants in each group. Everyone completed three training sessions using real-time functional magnetic resonance imaging, or fMRI.
Participants viewed overlapping images of negative faces and neutral scenes. Their task was to identify features of the scenes while ignoring the faces.
Software decoded brain activity associated with attending to faces versus scenes. In the genuine-feedback group, attention to negative faces made those faces more visible. Successfully attending to scenes made the scenes clearer instead.
The display therefore exposed attentional lapses and rewarded successful redirection by making the task easier.
Why The Sham Comparison Matters
Control participants saw feedback previously generated by someone in the genuine-feedback group. They completed comparable tasks, but the changing images did not reflect their own current brain activity.
Participants and experimenters were blinded to allocation. This design helped distinguish learning tied to personal brain signals from effects of repeated practice, expectations, and task engagement.
The primary outcome measured how likely participants were to remain in the strongest negative attentional state approximately five seconds later, a property the researchers called “stickiness.”
A Shift In Attention, Without Added Symptom Relief
Across training, genuine neurofeedback produced a greater reduction in negative-state persistence than sham feedback. A Bayesian analysis supported this group difference.
However, fatigue complicated interpretation. Negative-state persistence increased within sessions, so researchers focused their main training comparison on the first feedback run of each session.
The outcome was also undefined whenever participants never entered the strongest negative state. The main statistical model retained 45 participants, but missing measurements and a small sample limited precision.
Most consequentially, the groups did not differ significantly in depression severity after training or at one-week and one-month follow-ups. A separate behavioral gaze task also showed no significant group difference. The study therefore supports a change in the trained neural measure, without establishing broader clinical benefit.
Brain Changes Offer A Possible Explanation
Offline imaging analyses found reduced activity in the precuneus/posterior cingulate region and medial prefrontal cortex during successful attention to scenes, relative to sham training.
These regions participate in internally directed and self-referential processing. Their reduced activity may reflect less engagement with distracting negative material, although that interpretation remains provisional.
Unlike protocols that ask participants to deliberately increase activity in a selected region, this approach decoded competing attentional representations and embedded feedback directly within the task.
Earlier peer-reviewed amygdala neurofeedback research examined a different strategy, reinforcing positive autobiographical memories, and reported symptom benefits. Those findings cannot establish clinical effectiveness for this attention-training protocol.
From Scanner Learning To Everyday Recovery
The next challenge is demonstrating transfer: whether learning inside the scanner changes attention, rumination, functioning, or mood outside it. Larger trials, alternative control conditions, and shorter sessions could clarify effectiveness while reducing fatigue.
For now, neurofeedback appears capable of modifying one depression-related attentional process under controlled conditions. Whether that retraining becomes a useful depression treatment remains an open clinical question.
Citations
- Lubianiker N, et al. Escaping The Negative Attentional Bias In Depression With Real-Time Neurofeedback. medRxiv. 2026. Preprint, not peer reviewed.
- Young KD, et al. Randomized Clinical Trial Of Real-Time fMRI Amygdala Neurofeedback For Major Depressive Disorder: Effects On Symptoms And Autobiographical Memory Recall. American Journal of Psychiatry. 2017;174:748–755.
Explore more at https://www.interventionalpsychiatry.org/
