Print Close

Elusive Robustness in Brain-Depression Connectivity Findings

Poster No:

1429

Submission Type:

Abstract Submission

Authors:

Madleen Stenger¹, Sarah Alizadeh¹, Nooshin Javaheripour¹, Lea Teutenberg¹, Frederike Stein¹, Florian Thomas-Odenthal¹, Paula Usemann¹, Nils Winter², Janik Goltermann², Carlotta Barkhau², Daniel Emden², Jan Ernsting², Maximilian Konowski², Ramona Leenings², Tiana Borgers², Kira Flinkenflügel², Dominik Grotegerd², Anna Kraus², Elisabeth J. Leehr², Susanne Meinert², Andreas Jansen¹, Igor Nenadić¹, Benjamin Straube¹, Udo Dannlowski², Tim Hahn², Tilo Kircher¹, Hamidreza Jamalabadi¹

Institutions:

¹Department of Psychiatry and Psychotherapy, University of Marburg, Marburg, Germany, ²Institute for Translational Psychiatry, University of Münster, Münster, Germany

First Author:

Madleen Stenger
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Germany

Co-Author(s):

Sarah Alizadeh
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Germany

Nooshin Javaheripour
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Germany

Lea Teutenberg
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Germany

Frederike Stein
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Germany

Florian Thomas-Odenthal
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Germany

Paula Usemann
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Germany

Nils Winter
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Janik Goltermann
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Carlotta Barkhau
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Daniel Emden
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Jan Ernsting
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Maximilian Konowski
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Ramona Leenings
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Tiana Borgers
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Kira Flinkenflügel
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Dominik Grotegerd
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Anna Kraus
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Elisabeth J. Leehr
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Susanne Meinert
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Andreas Jansen
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Germany

Igor Nenadić
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Germany

Benjamin Straube
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Germany

Udo Dannlowski
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Tim Hahn
Institute for Translational Psychiatry, University of Münster
Münster, Germany

Tilo Kircher
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Germany

Hamidreza Jamalabadi
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Germany

Introduction:

Identifying robust neural correlates of depressive symptomatology remains a significant challenge, as no single network abnormality consistently accounts for the complex etiology of Major Depressive Disorder (MDD)[Winter et al., 2022]. Instead, subtle perturbations in functional connectivity among large-scale brain networks appear to underlie depressive states [Braun et al., 2018], with evidence pointing towards abnormalities in the Default Mode Network (DMN) and several other large-scale networks, including Frontoparietal (FPN), Dorsal Attention (DAN), and Salience (SAL) [Javaheripour et al., 2021]. While multivariate approaches increasingly reveal brain-behavior associations at the level of connectivity and psychiatric symptom severity (e.g., Beck Depression Inventory [BDI; Beck et al., 1996], the reliability of these findings under changing sample sizes and subsamples remains unclear. It is unknown if larger samples truly yield more stable canonical correlations between connectivity and symptom severity, and whether covariate loadings remain consistent across runs and subsets of subjects.

Methods:

In the present study, we employed Canonical Correlation Analysis (CCA) to examine the association between resting-state functional connectivity of DMN, FPN, DAN, and SAL networks and depressive symptom severity. We analyzed data from samples of n = 100, n = 400, and n = 841 subjects, treating the largest sample as a "gold standard" reference. We assessed whether correlations between BDI scores and connectivity patterns increased with larger sample sizes, and whether matching similar covariate loadings across runs would influence these correlations. Further, we computed cosine similarity to quantify how closely the loadings derived from smaller subsamples and various runs approximated the patterns obtained from the N=841 reference. We conducted separate cosine similarity analyses for BDI- and connectivity-related loadings, focusing on how these measures varied across sample sizes and networks.

Results:

Correlations between BDI and network connectivity increased as sample size grew from 100 to 841 subjects, ranging from approximately 0.2 to 0.6. However, after matching similar covariates across runs and sample sizes, we found no significant differences in their correlations (p > .2), suggesting a lack of generalizable covariate patterns. Cosine similarity analyses revealed that both BDI and connectivity loadings derived from smaller samples and different runs were only weakly similar to the N=841 reference (cosine similarity ~ -0.2 to 0.3). The notable exception was the DMN, where similarity stabilized around 0.2 once sample sizes exceeded 400 subjects. Thus, while larger samples enhanced the strength of canonical correlations, they did not ensure stable or replicable covariate loadings across different subsamples or runs.

Conclusions:

Our findings highlight that increasing sample size can amplify the observed connectivity-symptom correlations but does not guarantee the robustness or reproducibility of covariate loadings, a finding very much in line with previous large scale studies using predictive machine learning [Winter et al., 2024]. The low cosine similarities underscore potential overfitting or instability in multivariate analyses of MDD-related connectivity. These results caution against overreliance on even moderately large datasets and leave open the question of whether substantially larger samples might ultimately resolve these stability concerns.

Disorders of the Nervous System:

Psychiatric (eg. Depression, Anxiety, Schizophrenia) ²

Modeling and Analysis Methods:

Connectivity (eg. functional, effective, structural)

fMRI Connectivity and Network Modeling ¹

Multivariate Approaches

Task-Independent and Resting-State Analysis

Keywords:

Affective Disorders

Computational Neuroscience

Data analysis

DISORDERS

FUNCTIONAL MRI

MRI

Multivariate

Psychiatric Disorders

^1|2Indicates the priority used for review

Abstract Information

By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.

I accept

The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:

I do not want to participate in the reproducibility challenge.

Please indicate below if your study was a "resting state" or "task-activation” study.

Resting state

Healthy subjects only or patients (note that patient studies may also involve healthy subjects):

Patients

Was this research conducted in the United States?

Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.

Yes

Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.

Not applicable

Please indicate which methods were used in your research:

Functional MRI

Behavior

Computational modeling

For human MRI, what field strength scanner do you use?

3.0T

Which processing packages did you use for your study?

FSL

Other, Please list - ICA-FIX

Provide references using APA citation style.

Beck AT, Steer RA, Brown GK (1996): Manual for the beck depression inventory-II. San Antonio, TX: Psychological Corporation 1:10.1037.
Braun U, Schaefer A, Betzel RF, Tost H, Meyer-Lindenberg A, Bassett DS (2018): From Maps to Multi-dimensional Network Mechanisms of Mental Disorders. Neuron 97:14–31.
Javaheripour N, Li M, Chand T, Krug A, Kircher T, Dannlowski U, Nenadić I, Hamilton JP, Sacchet MD, Gotlib IH (2021): Altered resting-state functional connectome in major depressive disorder: a mega-analysis from the PsyMRI consortium. Translational psychiatry 11:1–9.
Winter NR, Blanke J, Leenings R, Ernsting J, Fisch L, Sarink K, Barkhau C, Emden D, Thiel K, Flinkenflügel K (2024): A Systematic Evaluation of Machine Learning–Based Biomarkers for Major Depressive Disorder. JAMA psychiatry.
Winter NR, Leenings R, Ernsting J, Sarink K, Fisch L, Emden D, Blanke J, Goltermann J, Opel N, Barkhau C, Meinert S, Dohm K, Repple J, Mauritz M, Gruber M, Leehr EJ, Grotegerd D, Redlich R, Jansen A, Nenadic I, Nöthen MM, Forstner A, Rietschel M, Groß J, Bauer J, Heindel W, Andlauer T, Eickhoff SB, Kircher T, Dannlowski U, Hahn T (2022): Quantifying Deviations of Brain Structure and Function in Major Depressive Disorder Across Neuroimaging Modalities. JAMA Psychiatry 79:879–888.

UNESCO Institute of Statistics and World Bank Waiver Form

I attest that I currently live, work, or study in a country on the UNESCO Institute of Statistics and World Bank List of Low and Middle Income Countries list provided.