Escaping the Negative Attentional Bias in Depression with Real-Time Neurofeedback
- 1Department of Psychology, Yale University, New Haven, CT, USA
Abstract
Individuals with major depressive disorder (MDD) show an attentional bias toward negatively valenced stimuli and thoughts. In this study, we applied a closed-loop neurofeedback procedure designed to reduce this bias. Participants were shown composite negative faces and neutral scenes with variable opacity and were instructed to attend to the neutral scene while ignoring the negative face. Internal attentional states were decoded in real time from functional magnetic resonance imaging (fMRI) data. When a participants decoded attentional state indicated a failure to ignore the negative faces, the faces became more visible (higher opacity), thus externalizing the brains attentional lapse in that moment. Forty-eight individuals with MDD were randomly assigned to a real neurofeedback training group (N = 24) or a sham control group (N = 24); the control group received feedback yoked to a participant in the real group. All participants completed three fMRI neurofeedback sessions. The main outcome quantified the extent to which participants got "stuck" in the most negative attentional state of focusing strongly on the negative faces. We hypothesized that the real neurofeedback group would learn to escape that state by the end of training, relative to the sham control group. Consistent with our hypothesis, neurofeedback training reduced the probability of getting stuck in the most negative attentional state. In offline analyses, training reduced fMRI activity in the precuneus/posterior cingulate and the medial prefrontal cortex when participants successfully attended to scenes and ignored faces. These results demonstrate the efficacy of remediating the negative attentional bias in depression with closed-loop neurofeedback from real-time fMRI.
Related
- Temporal attentional bias to food stimuli in adolescent anorexia nervosa - an ERP study.
- No association between body dissatisfaction and attentional bias towards lower-weight bodies in a non-clinical sample of women: an eye-tracking study
- Negative evaluation bias in person perception is linked to interpersonal dysfunction
- Baseline Response Criterion and Changes in Emotional Bias as Markers of Esketamine Response in Treatment-Resistant Depression
- Beyond Negative Bias: Disrupted Positive Valence Processing and Asynchronous Hot–Cool Development in Youth Internalizing Symptoms
BibTeX
@misc{lubianiker2026escaping,
title = {{Escaping the Negative Attentional Bias in Depression with Real-Time Neurofeedback}},
author = {N. Lubianiker and B. Woods and F. Nitchie and A. Batzdorf and A. Mennen and Q. Lin and K. A. Norman and N. B. Turk-Browne and Y. I. Sheline},
howpublished = {medRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.30.26363896},
url = {https://www.medrxiv.org/content/10.64898/2026.09.30.26363896v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Escaping the Negative Attentional Bias in Depression with Real-Time Neurofeedback
AU - N. Lubianiker
AU - B. Woods
AU - F. Nitchie
AU - A. Batzdorf
AU - A. Mennen
AU - Q. Lin
AU - K. A. Norman
AU - N. B. Turk-Browne
AU - Y. I. Sheline
PY - 2026
DA - 2026/09/30
PB - medRxiv
DO - 10.64898/2026.09.30.26363896
UR - https://www.medrxiv.org/content/10.64898/2026.09.30.26363896v1
AB - Individuals with major depressive disorder (MDD) show an attentional bias toward negatively valenced stimuli and thoughts. In this study, we applied a closed-loop neurofeedback procedure designed to reduce this bias. Participants were shown composite negative faces and neutral scenes with variable opacity and were instructed to attend to the neutral scene while ignoring the negative face. Internal attentional states were decoded in real time from functional magnetic resonance imaging (fMRI) data. When a participants decoded attentional state indicated a failure to ignore the negative faces, the faces became more visible (higher opacity), thus externalizing the brains attentional lapse in that moment. Forty-eight individuals with MDD were randomly assigned to a real neurofeedback training group (N = 24) or a sham control group (N = 24); the control group received feedback yoked to a participant in the real group. All participants completed three fMRI neurofeedback sessions. The main outcome quantified the extent to which participants got "stuck" in the most negative attentional state of focusing strongly on the negative faces. We hypothesized that the real neurofeedback group would learn to escape that state by the end of training, relative to the sham control group. Consistent with our hypothesis, neurofeedback training reduced the probability of getting stuck in the most negative attentional state. In offline analyses, training reduced fMRI activity in the precuneus/posterior cingulate and the medial prefrontal cortex when participants successfully attended to scenes and ignored faces. These results demonstrate the efficacy of remediating the negative attentional bias in depression with closed-loop neurofeedback from real-time fMRI.
ER -