We're unable to sign you in at this time. Please try again in a few minutes.
We were able to sign you in, but your subscription(s) could not be found. Please try again in a few minutes.
There may be a problem with your account. Please contact the AMA Service Center to resolve this issue.
Contact the AMA Service Center:
Telephone: 1 (800) 262-2350 or 1 (312) 670-7827  *   Email: subscriptions@jamanetwork.com
Error Message ......
Original Investigation |

Self-blame–Selective Hyperconnectivity Between Anterior Temporal and Subgenual Cortices and Prediction of Recurrent Depressive Episodes

Karen E. Lythe, PhD1; Jorge Moll, MD, PhD2; Jennifer A. Gethin, MRes1,3; Clifford I. Workman, BS1,4; Sophie Green, PhD1; Matthew A. Lambon Ralph, PhD1; John F. W. Deakin, PhD4; Roland Zahn, MD1,5
[+] Author Affiliations
1Neuroscience and Aphasia Research Unit, School of Psychological Sciences, The University of Manchester and Manchester Academic Health Sciences Centre, Manchester, England
2Cognitive and Behavioral Neuroscience Unit, D’Or Institute for Research and Education, Rio de Janeiro, Brazil
3doctoral student at The University of Manchester, Manchester, England
4Neuroscience and Psychiatry Unit, Institute of Brain, Behaviour and Mental Health, The University of Manchester and Manchester Academic Health Sciences Centre, Manchester, England
5Department of Psychological Medicine, Centre for Affective Disorders, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, London, England
JAMA Psychiatry. 2015;72(11):1119-1126. doi:10.1001/jamapsychiatry.2015.1813.
Text Size: A A A
Published online

Importance  Patients with remitted major depressive disorder (MDD) were previously found to display abnormal functional magnetic resonance imaging connectivity (fMRI) between the right superior anterior temporal lobe (RSATL) and the subgenual cingulate cortex and adjacent septal region (SCSR) when experiencing self-blaming emotions relative to emotions related to blaming others (eg, “indignation or anger toward others”). This finding provided the first neural signature of biases toward overgeneralized self-blaming emotions (eg, “feeling guilty for everything”), known to have a key role in cognitive vulnerability to MDD. It is unknown whether this neural signature predicts risk of recurrence, a crucial step in establishing its potential as a prognostic biomarker, which is urgently needed for stratification into pathophysiologically more homogeneous subgroups and for novel treatments.

Objective  To use fMRI in remitted MDD at baseline to test the hypothesis that RSATL-SCSR connectivity for self-blaming relative to other-blaming emotions predicts subsequent recurrence of depressive episodes.

Design, Setting, and Participants  A prospective cohort study from June 16, 2011, to October 10, 2014, in a clinical research facility completed by 75 psychotropic medication–free patients with remitted MDD and no relevant comorbidity. In total, 31 remained in stable remission, and 25 developed a recurring episode over the 14 months of clinical follow-up and were included in the primary analysis. Thirty-nine control participants with no personal or family history of MDD were recruited for further comparison.

Main Outcomes and Measures  Between-group difference (recurring vs stable MDD) in RSATL connectivity, with an a priori SCSR region of interest for self-blaming vs other-blaming emotions.

Results  We corroborated our hypothesis that during the experience of self-blaming vs other-blaming emotions, RSATL-SCSR connectivity predicted risk of subsequent recurrence. The recurring MDD group showed higher connectivity than the stable MDD group (familywise error–corrected P < .05 over the a priori SCSR region of interest) and the control group. In addition, the recurring MDD group also exhibited RSATL hyperconnectivity with the right ventral putamen and claustrum and the temporoparietal junction. Together, these regions predicted recurrence with 75% accuracy.

Conclusions and Relevance  To our knowledge, this study is the first to provide a robust demonstration of an fMRI signature of recurrence risk in remitted MDD. Additional studies are needed for its further optimization and validation as a prognostic biomarker.

Figures in this Article


Place holder to copy figure label and caption
Figure 1.
Regions Showing Functional Connectivity Group Differences With the Right Superior Anterior Temporal Lobe (RSATL) Seed Region for Self-blaming vs Other-Blaming Emotions Between the Recurring Episode Major Depressive Disorder (MDD) Group and the Stable Remission MDD Group

Cropped images are displayed at an uncorrected voxel-level threshold of P = .005, with no cluster-size threshold. A predictive linear discriminant analysis31 using cluster average regression coefficients from the subgenual cingulate cortex and adjacent septal region (SCSR), right temporal junction (RTPJ), and right ventral putamen and claustrum (RVP), including cases with lower magnetic resonance imaging quality (n = 64), resulted in 79% correctly classified patients into recurring vs stable groups, which could be cross-validated using the well-established leave-one-out method in 75% of the cases (positive predictive value of 74%, negative predictive value of 76%, Wilks λ = .681, P < .001, estimating prior probabilities from subgroup sizes, 1000 bootstrap samples). In contrast, repeating this analysis only based on clinical variables (number of previous episodes, time in remission, time since stopping antidepressant medication, Beck Depression Inventory, Montgomery-Åsberg Depression Rating Scale, and Global Assessment of Functioning scores), prediction accuracy was at chance levels (61% cross-validated, Wilks λ = .855, P = .16). The cluster size of the SCSR effect is 56 (Table), which is large compared with the relatively small size of the anatomical region. Our statistical inference for this finding is based on the peak voxel effect size that survives voxel-based familywise error correction at P = .05 over our a priori region of interest. The other clusters survive familywise error correction over the whole brain. Note that some functional magnetic resonance imaging findings appear to fall on white matter; this is at least partly because of individual anatomical variability. We used a gray matter mask for inclusive masking to only retain voxels with gray matter density values on averaged segmentations of greater than 0.10. Therefore, gray matter is present in at least some participants in all the voxels of each cluster, even if the projection onto the template gives the impression of falling on white matter. The peak voxel (Cohen d = 0.98, corresponding to t54 = 3.59) and cluster average (d = 1.07) effect sizes for the subgenual region are large. They are even larger for the RTPJ (d = 1.22) and RVP (d = 1.63) cluster averages. On a cautionary note, these effect size estimates are post hoc and may therefore overestimate the true effect size, which needs to be determined in future independent studies. In contrast, the large multivariate effect size (Wilks λ = 0.681) of the linear discriminant analysis, which was independently cross-validated, provides a robust estimate that can inform power calculations for future studies. Blue line indicates relatively higher connectivity; yellow dashed line, relatively lower connectivity.

Graphic Jump Location
Place holder to copy figure label and caption
Figure 2.
Connectivity Coefficients for Posterior Right Superior Anterior Temporal Lobe–Subgenual Cingulate Cortex and Adjacent Septal Region (RSATL-SCSR) for Self-blaming and Other-Blaming Emotions vs Baseline

Patients with stable major depressive disorder (MDD) and recurring MDD are shown. The RSATL-SCSR connectivity differences were driven by an interaction between emotion (self-blaming vs other-blaming) and group (recurring episode MDD [n = 25] vs stable MDD [n = 31]) (F1,54 = 16.23, P < .001). As can be seen, the interaction arose by higher connectivity during self-blame and by lower connectivity during other-blame in the recurring episode MDD group and the opposite pattern in the stable remission MDD group. There were no significant main effects of emotion (F1,54 = 2.303, P = .14) or group (F1,54 = 0.016, P = .90).

Graphic Jump Location
Place holder to copy figure label and caption
Figure 3.
Connectivity Coefficients for Posterior Right Superior Anterior Temporal Lobe–Subgenual Cingulate Cortex and Adjacent Septal Region (RSATL-SCSR) for Self-blaming vs Other-Blaming Emotions

Patients with stable major depressive disorder (MDD), subthreshold MDD, and recurring MDD are shown. A secondary data analysis on the extracted SCSR regression coefficients (cluster averages) in the larger data set, including the subthreshold MDD group and patients with nonoptimal quality of functional magnetic resonance imaging data, confirmed a connectivity difference for self-blaming vs other-blaming emotions between the MDD groups (F2,74 = 6.39, P = .003). Post hoc pairwise comparisons showed increased RSATL-SCSR connectivity for self-blaming vs other-blaming emotions in the recurring episode MDD group compared with both the stable remission MDD group (mean [SE] difference, 1.65 [0.48]; 95% CI, 0.69-2.61; P = .001) and the subthreshold MDD group (mean [SE] difference, 1.59 [0.68]; 95% CI, 0.24-2.95; P = .001), with no difference between the stable remission and subthreshold MDD groups (mean [SE] difference, −0.05 [0.65]; 95% CI, −1.25 to 1.36; P = .93).

Graphic Jump Location




Also Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
Please click the checkbox indicating that you have read the full article in order to submit your answers.
Your answers have been saved for later.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.


Some tools below are only available to our subscribers or users with an online account.

3 Citations

Sign in

Purchase Options

• Buy this article
• Subscribe to the journal
• Rent this article ?

Related Content

Customize your page view by dragging & repositioning the boxes below.

Articles Related By Topic
Related Collections
PubMed Articles

Care at the Close of Life: Evidence and Experience
Guilt and Self-Blame