A major UK Biobank study has found that heightened activity in the brain’s emotion centre, the amygdala, during negative emotional processing is linked to a person’s lifetime history of recurrent depression rather than their current mood, suggesting each depressive episode may imprint a lasting biological vulnerability. The research, published in Psychological Medicine, indicates the amygdala may act as a long-term trait marker for depression rather than a state marker tied to present symptoms.
Major depressive disorder is a common condition that affects millions worldwide, yet psychiatrists still rely largely on clinical interviews and patient history for diagnosis and treatment decisions. There are currently no biological tests or brain scans used routinely to guide therapy.
The study utilised functional magnetic resonance imaging, or fMRI, which tracks brain activity by measuring blood-oxygen levels. When a brain region becomes active, it requires more oxygen, leading to detectable changes in blood flow. This makes it possible to map how the brain responds in real time.
Researchers have long studied the amygdala, a limbic structure deep within the brain that processes fear and negative emotions. Earlier imaging work suggested amygdala hyperreactivity in depression when people viewed negative images, but a large UK Biobank analysis later found no clear link between amygdala activity and current depressive symptoms.
To address this contradiction, Jerke J. van den Berg of the University of Amsterdam and colleagues examined whether lifetime depression history—rather than current mood—might better explain amygdala responses. They applied the kindling theory, which posits that initial depressive episodes sensitize the brain to stress, making relapse easier to trigger with subsequent trauma.
The team analysed data from the UK Biobank, a long-running population health project. They included functional scans from more than 11,000 participants who completed a visual task during scanning, known as the Hariri task, which prompts responses to angry or fearful faces.
As brain activity can vary with age, gender and even head movement, the researchers used normative modelling to establish a baseline of typical amygdala activity. They compared each participant’s results against this baseline using data from over 6,400 healthy individuals.
Participants were then grouped according to self-reported mental health histories: healthy controls, individuals who had experienced a single depressive episode, those with two to five episodes (moderate recurrence), and a high recurrence group with six or more episodes.
In the cross-sectional analysis focusing on people in remission, initial scans did not show a clear difference. After adjusting for age and gender with normative modelling, however, a pattern emerged: those with a history of recurrent depression showed greater amygdala reactivity to negative faces than healthy controls.
When examining individuals currently experiencing a depressive episode, the researchers found no distinct rise in amygdala activity compared with controls. This strengthened the interpretation that amygdala responsiveness in this context reflects a long-term cognitive-biological trait rather than an immediate mood state.
The study also looked at how medications might influence these signals. A larger share of the severe recurrence group were taking antidepressants. Re-running the analysis after removing medicated participants strengthened the observed differences between healthy controls and those with recurrent depression, suggesting antidepressants may dampen amygdala reactivity and partially mask the underlying brain changes.
To explore changes over time, the researchers conducted a longitudinal analysis with a smaller subset who returned for a second scan about two and a half years later. They focused on individuals who were in remission at both visits. Those who initially had one depressive episode but later developed new episodes showed an increase in amygdala reactivity over time, aligning with the kindling concept that new episodes progressively alter how the brain processes negative information.
Nevertheless, the authors emphasised that the effect sizes were small. The findings do not support using functional MRI as a diagnostic tool for depression or as a precise predictor of relapse from a single scan. Limitations include reliance on self-reported depressive histories, which can be imperfect, and the use of a combined anxiety–depression measure that limits dissection of anxiety-specific effects.
Looking ahead, the researchers argue that advances in imaging and analysis could one day enable clinicians to tailor treatments more precisely to a patient’s history, moving beyond the current trial-and-error approach to antidepressant prescribing.
The study’s authors are Jerke J. van den Berg, Henricus G. Ruhé, Henk A. Marquering, Liesbeth Reneman, and Matthan W. A. Caan.
